Tumor targeting platform and its use

Abstract

Provided herein are RAS-selective effector proteins that can be used to treat cancer. These proteins can be delivered with the Anthrax toxin protective antigen (PA)-based delivery systems disclosed herein. The PA-based systems achieve high tumor specificity, by targeting at least two tumor markers and delivering the effector protein across the cell membrane of the targeted tumor cell. The effector proteins specifically cleaves RAS in tumor cells, achieving potent antitumor effects and minimized side effects. Nucleic acids, compositions, kits, and methods of producing and using the effector and / or delivery proteins are also disclosed.

Description

[0001] 8123-112570-02

[0002] TUMOR TARGETING PLATFORM AND ITS USE CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority to U.S. Provisional Application No. 63 / 791,332, filed April 18, 2025, which is herein incorporated by reference in its entirety.

[0003] STATEMENT OF GOVERNMENT SUPPORT

[0004] This invention was made with government support under grant # CA254938 awarded by the National Institutes of Health. The government has certain rights in the invention.

[0005] FIELD

[0006] This application relates to the field of tumor targeting effector proteins and delivery platforms, particularly RAS-selective effector proteins, and Anthrax toxin-based delivery platforms.

[0007] SEQUENCE LISTING

[0008] The nucleic and amino acid sequences described herein and listed in the accompanying sequence listing are shown using standard letter abbreviations for nucleotide bases, and single letter code for amino acids, as defined in 37 C.F.R. 1.822. Only one strand of each nucleic acid sequence is shown, but the complementary strand is understood as included by any reference to the displayed strand. The Sequence Listing is submitted as an XML file in the form of the file named “8123-112570-02_Sequence Listing. xml” (78, 118 bytes), which was created on April 13, 2026, which is incorporated by reference herein.

[0009] BACKGROUND

[0010] Human epidermal growth factor receptor-2 (HER2, also known as EGFR2, ERBB2) is a member of the EGFR family of receptor tyrosine kinases. HER2 overexpression or gene amplification is a driving cause of 20% of breast cancers and, to a lesser extent, various other types of cancers (e.g., gastric and uterus cancers). This has encouraged the successful development of HER2 antibodies (such as trastuzumab and pertuzumab) as targeted therapeutics for patients with metastatic HER2 -positive breast cancers. However, since HER2 blockade is not tumoricidal, the antitumor efficacy of anti-HER2 therapeutics is modest. To improve the efficacy, anti-HER2 antibody-cytotoxic drug conjugates (ADCs), trastuzumab-emtansine (tubulin binder) (T-DM1) and the more recent trastuzumab-8123-112570-02

[0011] deruxtecan (topoisomerase I inhibitor) (T-DXd), have been developed. T-DXd demonstrated utility in treating patients with HER2-low breast cancers and other HER2-low solid tumors. However, the efficacy of T-DXd is modest, limited by the small amount of the payload it can carry (< 8 deruxtecan molecules per trastuzumab). In addition, the side effect profile of T-DXd is comparable to that of standard chemotherapies: a significant portion of patients developed unacceptable interstitial lung disease due to targeting of certain HER2-low normal tissues by T-DXd.

[0012] Many bacterial pathogens have evolved potent protein toxins which have been shown to disrupt cellular pathways that are dysregulated in cancers. One example is Anthrax lethal toxin (LT), a virulent factor of Bacillus anthracis (BA). LT has two polypeptide components: a receptor-binding and delivering component protective antigen (PA), and an enzymatic effector component named lethal factor (LF). The action of this system requires PA to bind to cellular receptors and to undergo proteolytic activation by furin proteases on the cell surface, allowing PA to translocate LF into the cytosol of target cells to exert their cytotoxic effects.

[0013] SUMMARY

[0014] Provided herein are novel antitumor effector proteins that specifically cleave rat sarcoma virus (RAS) proteins (including KRAS, NRAS, and HRAS) over RAS-related protein- 1 (RAP1), thereby achieving potent antitumor effects in cancers driven by hyperactive RAS activities while minimizing cytotoxicity to normal cells that express RAP1 for important cellular functions. Coupled with a tumor-specific delivery platform, the RAS-selective effector proteins achieve highly specific tumor targeting.

[0015] In some aspects, the effector protein is a recombinant DUF5 comprising a L418W substitution. In some aspects, the recombinant DUF5 protein is fused to a heterologous protein, such as a domain of LF that binds to PA.

[0016] In some aspects, the delivery platform is a PA-based delivery platform. PA used as a delivery protein herein is engineered to increase specificity toward tumor cells, by, for example, replacing the furin cleavage site to MMP cleavage site or uPA cleavage site; fusing PA to a tumor targeting component (e.g., EGFR, HER2, or HER3 affibody or antibody); mutating PA so that it is not recognized by native receptors; and / or mutating a pair of PAs so that an effector protein can only bind to the hetero-oligomer formed by both of the PAs, but not to the homo-oligomer formed by only one of the PAs.8123-112570-02

[0017] Also provided are compositions, kits, and methods for treating cancers based on the use of the effector proteins and / or delivery proteins provided herein.

[0018] The foregoing and other objects, features, and advantages of the invention will become more apparent from the following detailed description, which proceeds with reference to the accompanying figures.

[0019] SEQUENCE

[0020] The nucleic and amino acid sequences listed herein are shown using standard letter abbreviations for nucleotide bases and amino acids. Only one strand of each nucleic acid sequence is shown, but the complementary strand is understood as included by any reference to the displayed strand.

[0021] I. Delivery Proteins

[0022] SEQ ID NO: 1 is an exemplary amino acid (aa) sequence of wild-type PA.

[0023] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIPS ENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNKIRLEKG RLYQ1K1QYQRENPTEKGLDFKLYWTDSQNKKEV1SSDNLQLPELKQKSSNSRKKRST SAGPTVPDRDNDGIPDSLEVEGYTVDVKNKRTFLSPWISNIHEKKGLTKYKSSPEKW STASDPYSDFEKVTGR1DKNVSPEARHPLVAAYP1VHVDMEN11LSKNEDQSTQNTDS QTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSLSLAGE RTWAETMGLNTADTARLNANIRYVNTGTAPIYNVLPTTSLVLGKNQTLATIKAKEN QLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTDQVYG NIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPSDPLE TTKPDMTLKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQNIKNQLAELNAT NIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTEGLLLN IDKDIRKILSGYIVEIEDTEGLKEVINDRYDMLNISSLRQDGKTFIDFKKYNDKLPLYIS NPNYKVNVYAVTKENTIINPSENGDTSTNGIKKILIFSKKGYEIG (SEQ ID NO: 1)

[0024] SEQ ID NO: 2 is an exemplary amino acid sequence of PA3M. 3M refers to 3 mutations, 683 A, 685E, and 688K (shown with underlined and bolded), that impede binding to PA receptors.

[0025] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIPS ENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNKIRLEKG RLYQIKIQYQRENPTEKGLDFKLYWTDSQNKKEVISSDNLQLPELKQKSSNSRKKRST SAGPTVPDRDNDGIPDSLEVEGYTVDVKNKRTFLSPWISNIHEKKGLTKYKSSPEKW STASDPYSDFEKVTGRIDKNVSPEARHPLVAAYPIVHVDMENIILSKNEDQSTQNTDS QTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSLSLAGE RTWAETMGLNTADTARLNANTRYVNTGTAPIYNVLPTTSLVLGKNQTLATIKAKEN QLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTDQVYG NIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPSDPLE8123-112570-02

[0026] TTKPDMTLKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQNIKNQLAELNAT NIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTEGLLLN IDKDIRKILSGYIVEIEDTEGLKEVINDRYDMLNISSLRQDGKTFIDFKKYNAKEPLKIS NPNYKVNVYAVTKENTIINPSENGDTSTNGIKKILIFSKKGYEIG (SEQ ID NO: 2)

[0027] SEQ ID NO: 3 is an exemplary amino acid sequence of a matrix metalloproteinase (MMP) cleavage site.

[0028] GPLGMLSQ (SEQ ID NO: 3)

[0029] SEQ ID NO: 4 is an exemplary amino acid sequence of an urokinase-type plasminogen activator (uPA) cleavage site.

[0030] PGSGRSA (SEQ ID NO: 4)

[0031] SEQ ID NO: 5 is an amino acid sequence of an exemplary HER2 specific affibody (zHER2:343).

[0032] VDNKFNKEMRNAYWEIALLPNLNNQQKRAFIRSLYDDPSQSANLLAEAKKL NDAQAPK (SEQ ID NO: 5)

[0033] SEQ ID NO: 6 is an amino acid sequence of an exemplary HER3 specific affibody (zHER3:8699).

[0034] VDNKFNKEKYNAYYEIWQLPNLTKYQKAAFIGKLQDDPSQSSELLSEAKKLN DSQAPK (SEQ ID NO: 6)

[0035] SEQ ID NO: 7 is an amino acid sequence of an exemplary EGFR specific affibody (zEGFR2377).

[0036] VDNKFNKEMWIAWEEIRDLPNLNGWQMTAFIASLLDDPSQSANLLAEAKKL NDAQAPK (SEQ ID NO: 7)

[0037] SEQ ID NO: 8 is an amino acid sequence of an exemplary GS linker.

[0038] GGGSGTRGGGSG (SEQ ID NO: 8)

[0039] SEQ ID NO: 9 is an amino acid sequence of an exemplary PA signal peptide that can be added to the N-terminus of a delivery protein or an effector protein disclosed herein for expression in Bacillus anthracis. The signal peptide will be removed from the mature protein upon secretion from the host strain, such as BH500.8123-112570-02

[0040] MKKRKVLIPLMALSTILVSSTGNLEVIQA (SEQ ID NO: 9)

[0041] SEQ ID NO: 10 is an exemplary amino acid sequence of PA3MzHER2 (mature: 90.20 kDa) (the original furin activated version, toxic to mice), aa 1-735 is a PA3M sequence. The furin cleavage site (RKKR) is underlined and bolded. The 3M mutations are underlined and bolded, aa 736-747 (underlined) is a GS linker, aa 748-805 is zHER2:343.

[0042] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIP SENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNKIRLEK GRLYOIKIOYORENPTEKGLDFKLYWTDSONKKEVISSDNLOLPELKOKSSNSRKKR STSAGPTVPDRDNDGIPDSLEVEGYTVDVKNKRTFLSPWISNIHEKKGLTKYKSSPEK WSTASDPYSDFEKVTGRIDKNVSPEARHPLVAAYPIVHVDMENIILSKNEDQSTQNT DSQTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSLSLA GERTWAETMGLNTADTARLNANIRYVNTGTAPIYNVLPTTSLVLGKNQTLATIKAK ENQLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTDQ VYGNIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPS DPLETTKPDMTLKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQNIKNQLAE LNATNIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTE GLLLNIDKDIRKILSGYIVEIEDTEGLKEVINDRYDMLNISSLRQDGKTFIDFKKYNAK EPLKISNPNYKVNVYAVTKENTIINPSENGDTSTNGIKKILIFSKKGYEIGGGGSGTRG GGSGVDNKFNKEMRNAYWEIALLPNLNNQQKRAFIRSLYDDPSQSANLLAEAKKLN DAQAPK (SEQ ID NO: 10)

[0043] SEQ ID NO: 11 is an exemplary amino acid sequence of PA3MzHER2-Ll (mature: 90.43 kDa) (the MMP-activated version, not toxic to mice, also referred to as PA3MzHER2- MMP herein), aa 1-739 is a PA3M sequence with the furin cleavage site replaced with an MMP cleavage site. The MMP cleavage site (GPLGMLSQ, SEQ ID NO: 3) is underlined and bolded. The 3M mutations are underlined and bolded, aa 740-751 (underlined) is a GS linker, aa 752-809 is zHER2:343.

[0044] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIPS ENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNKIRLEKG RLYOIKIOYORENPTEKGLDFKLYWTDSONKKEVISSDNLOLPELKOKSSNSGPLGM LSQSTSAGPTVPDRDNDGIPDSLEVEGYTVDVKNKRTFLSPWISNIHEKKGLTKYKSS PEKWSTASDPYSDFEKVTGRIDKNVSPEARHPLVAAYPIVHVDMENIILSKNEDQSTQ NTDSQTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSLS LAGERTWAETMGLNTADTARLNANIRYVNTGTAPIYNVLPTTSLVLGKNQTLATIKA KENQLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTDQ VYGNIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPSK PLETTKPDMTLKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQNIKNQLAELN ATNIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTEGLL LNIDKDIRKILSGYIVEIEDTEGLKEVINDRYDMLNISSLRQDGKTFIDFKKYNAKEPL KISNPNYKVNVYAVTKENTIINPSENGDTSTNGIKKILIFSKKGYEIGGGGSGTRGGGS GVDNKFNKEMRNAYWEIALLPNLNNQQKRAFIRSLYDDPSQSANLLAEAKKLNDAQ APK (SEQ ID NO: 11)8123-112570-02

[0045] SEQ ID NO: 61 is an exemplary amino acid sequence of PA3MzHER2-Ll (mature: 90.43 kDa) (the MMP-activated version, not toxic to mice, also referred to as PA3MzHER2- MMP herein), aa 1-739 is a PA3M sequence with the furin cleavage site replaced with an MMP cleavage site. The MMP cleavage site (GPLGMLSQ, SEQ ID NO: 3) is underlined and bolded. The 3M mutations are underlined and bolded, aa 740-751 (underlined) is a GS linker, aa 752-809 is zHER2:343.

[0046] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIPS ENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNKIRLEKG RLYOIKIOYORENPTEKGLDFKLYWTDSONKKEVISSDNLOLPELKOKSSNSGPLGM LSQSTSAGPTVPDRDNDGIPDSLEVEGYTVDVKNKRTFLSPWISNIHEKKGLTKYKSS PEKWSTASDPYSDFEKVTGRIDKNVSPEARHPLVAAYPIVHVDMENIILSKNEDQSTQ NTDSQTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSLS LAGERTWAETMGLNTADTARLNANIRYVNTGTAPIYNVLPTTSLVLGKNQTLATIKA KENQLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTDQ VYGNIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPSD PLETTKPDMTLKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQNIKNQLAELN ATNIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTEGLL LNIDKDIRKILSGYIVEIEDTEGLKEVINDRYDMLNISSLRQDGKTFIDFKKYNAKEPL KISNPNYKVNVYAVTKENTIINPSENGDTSTNGIKKILIFSKKGYEIGGGGSGTRGGGS GVDNKFNKEMRNAYWEIALLPNLNNQQKRAFIRSLYDDPSQSANLLAEAKKLNDAQ APK (SEQ ID NO: 61)

[0047] SEQ ID NO: 12 is an exemplary amino acid sequence of PA3MzHER2-U2 (mature: 90.24 kDa) (the uPA-activated version, not toxic to mice), aa 1-738 is a PA3M sequence with the furin cleavage site replaced with an uPA cleavage site. The uPA cleavage site (PGSGRSA, SEQ ID NO: 4) is underlined and bolded. The 3M mutations are underlined and bolded, aa 739-750 (underlined) is a GS linker, aa 751-808 is zHER2:343.

[0048] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIPS ENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNKIRLEKG RLYOIKIOYORENPTEKGLDFKLYWTDSONKKEVISSDNLOLPELKQKSSNSPGSGR SASTSAGPTVPDRDNDGIPDSLEVEGYTVDVKNKRTFLSPWISNIHEKKGLTKYKSSP EKWSTASDPYSDFEKVTGRIDKNVSPEARHPLVAAYPIVHVDMENIILSKNEDQSTQ NTDSQTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSLS LAGERTWAETMGLNTADTARLNANIRYVNTGTAPIYNVLPTTSLVLGKNQTLATIKA KENQLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTDQ VYGNIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPSD PLETTKPDMTEKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQN1KNQEAELN ATNIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTEGLL LNIDKDIRKILSGYIVEIEDTEGLKEVINDRYDMLNISSLRQDGKTFIDFKKYNAKEPL KISNPNYKVNVYAVTKENTIINPSENGDTSTNGIKKILIFSKKGYEIGGGGSGTRGGGS GVDNKFNKEMRNAYWEIALLPNLNNQQKRAFIRSLYDDPSQSANLLAEAKKLNDAQ APK (SEQ ID NO: 12)8123-112570-02

[0049] SEQ ID NO: 13 is an exemplary amino acid sequence of PA3MzHER3-Ll (MMP activated, HER3 targeted, not toxic to mice), aa 1-739 is a PA3M sequence with the furin cleavage site replaced with an MMP cleavage site. The MMP cleavage site (GPLGMLSQ, SEQ ID NO: 3) is underlined and bolded. The 3M mutations are underlined and bolded, aa 740-751 (underlined) is a GS linker, aa 752-809 is zHER3:8699.

[0050] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIPS ENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNKIRLEKG RLYOIKIOYORENPTEKGLDFKLYWTDSONKKEVISSDNLOLPELKOKSSNSGPLGM LSQSTSAGPTVPDRDNDGIPDSLEVEGYTVDVKNKRTFLSPWISNIHEKKGLTKYKSS PEKWSTASDPYSDFEKVTGRIDKNVSPEARHPLVAAYPIVHVDMENIILSKNEDQSTQ NTDSQTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSLS LAGERTWAETMGLNTADTARLNANIRYVNTGTAPIYNVLPTTSLVLGKNQTLATIKA KENQLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTDQ VYGNIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPSK PLETTKPDMTLKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQNIKNQLAELN ATNIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTEGLL LNIDKDIRKILSGYIVEIEDTEGLKEVINDRYDMLNISSLRQDGKTFIDFKKYNAKEPL KISNPNYKVNVYAVTKENTIINPSENGDTSTNGIKKILIFSKKGYEIGGGGSGTRGGGS G VDNKFNKEK YN A Y YEIWQLPNLTKYQKA AFIGKLQDDPS QS SELLS EAKKLNDS Q APK (SEQ ID NO: 13)

[0051] SEQ ID NO: 62 is an exemplary amino acid sequence of PA3MzHER3-Ll (MMP activated, HER3 targeted, not toxic to mice), aa 1-739 is a PA3M sequence with the furin cleavage site replaced with an MMP cleavage site. The MMP cleavage site (GPLGMLSQ, SEQ ID NO: 3) is underlined and bolded. The 3M mutations are underlined and bolded, aa 740-751 (underlined) is a GS linker, aa 752-809 is zHER3:8699.

[0052] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIPS ENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNKIRLEKG RLYOIKIOYORENPTEKGLDFKLYWTDSONKKEVISSDNLOLPELKQKSSNSGPLGM LSQSTSAGPTVPDRDNDGIPDSLEVEGYTVDVKNKRTFLSPWISNIHEKKGLTKYKSS PEKWSTASDPYSDFEKVTGRIDKNVSPEARHPLVAAYPIVHVDMENIILSKNEDQSTQ NTDSQTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSLS LAGERTWAETMGLNTADTARLNANIRYVNTGTAPIYNVLPTTSLVLGKNQTLATIKA KENQLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTDQ VYGNIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPSD PLETTKPDMTLKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQNIKNQLAELN ATNIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTEGLL LNIDKDIRKILSGYIVEIEDTEGLKEVINDRYDMLNISSLRQDGKTFIDFKKYNAKEPL KISNPNYKVNVYAVTKENTIINPSENGDTSTNGIKKILIFSKKGYEIGGGGSGTRGGGS GVDNKFNKEKYNAYYEIWQLPNLTKYQKAAFIGKLQDDPSQSSELLSEAKKLNDSQ APK (SEQ ID NO: 62)

[0053] SEQ ID NO: 14 is an exemplary amino acid sequence of PA3MzHER3-U2 (uPA activated, HER3 targeted, not toxic to mice), aa 1-738 is a PA3M sequence with the furin8123-112570-02

[0054] cleavage site replaced with an uPA cleavage site. The uPA cleavage site (PGSGRSA, SEQ ID NO: 4) is underlined and bolded. The 3M mutations are underlined and bolded, aa 739-750 (underlined) is a GS linker, aa 751-808 is zHER3:8699.

[0055] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIPS ENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNKIRLEKG RLYOIKIOYORENPTEKGLDFKLYWTDSONKKEVISSDNLOLPELKQKSSNSPGSGR SASTSAGPTVPDRDNDGIPDSLEVEGYTVDVKNKRTFLSPWISNIHEKKGLTKYKSSP EKWSTASDPYSDFEKVTGRIDKNVSPEARHPLVAAYPIVHVDMENIILSKNEDQSTQ NTDSQTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSLS LAGERTWAETMGLNTADTARLNANIRYVNTGTAPIYNVLPTTSLVLGKNQTLATIKA KENQLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTDQ VYGNIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPSD PLETTKPDMTLKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQNIKNQLAELN ATNIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTEGLL LNIDKDIRKILSGYIVEIEDTEGLKEVINDRYDMLNISSLRQDGKTFIDFKKYNAKEPL KISNPNYKVNVYAVTKENTIINPSENGDTSTNGIKKILIFSKKGYEIGGGGSGTRGGGS GVDNKFNKEKYNAY YEIWQLPNLTKYQKA AFIGKLQDDPS QS SELLS EAKKLNDS Q APK (SEQ ID NO: 14)

[0056] SEQ ID NO: 15 is an exemplary amino acid sequence of PA3MzEGFR-Ll (MMP activated, EGFR targeted, not toxic to mice), aa 1-739 is a PA3M sequence with the furin cleavage site replaced with an MMP cleavage site. The MMP cleavage site (GPLGMLSQ, SEQ ID NOG) is underlined and bolded. The 3M mutations are underlined and bolded, aa 740-751 (underlined) is a GS linker, aa 752-809 is zEGFR2377.

[0057] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIPS ENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNKIRLEKG RLYOIKIOYORENPTEKGLDFKLYWTDSONKKEVISSDNLOLPELKQKSSNSGPLGM LSQSTSAGPTVPDRDNDGIPDSLEVEGYTVDVKNKRTFLSPWISNIHEKKGLTKYKSS PEKWSTASDPYSDFEKVTGRIDKNVSPEARHPLVAAYPIVHVDMENIILSKNEDQSTQ NTDSQTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSLS LAGERTWAETMGLNTADTARLNANIRYVNTGTAPIYNVLPTTSLVLGKNQTLATIKA KENQLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTDQ VYGNIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPSK PLETTKPDMTLKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQNIKNQLAELN ATNIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTEGLL LNIDKDIRKILSGYIVEIEDTEGLKEVINDRYDMLNISSLRQDGKTFIDFKKYNAKEPL KISNPNYKVNVYAVTKENTIINPSENGDTSTNGIKKILIFSKKGYEIGGGGSGTRGGGS GVDNKFNKEMWIAWEEIRDLPNLNGWQMTAFIASLLDDPSQSANLLAEAKKLNDA QAPK (SEQ ID NO: 15)

[0058] SEQ ID NO: 63 is an exemplary amino acid sequence of PA3MzHER3-Ll (MMP activated, HER3 targeted, not toxic to mice), aa 1-739 is a PA3M sequence with the furin cleavage site replaced with an MMP cleavage site. The MMP cleavage site (GPLGMLSQ,8123-112570-02

[0059] SEQ ID NO: 3) is underlined and bolded. The 3M mutations are underlined and bolded, aa 740-751 (underlined) is a GS linker, aa 752-809 is zHER3:8699.

[0060] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIPS ENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNKIRLEKG RLYOIKIOYORENPTEKGLDFKLYWTDSONKKEVISSDNLOLPELKOKSSNSGPLGM LSQSTSAGPTVPDRDNDGIPDSLEVEGYTVDVKNKRTFLSPWISNIHEKKGLTKYKSS PEKWSTASDPYSDFEKVTGRIDKNVSPEARHPLVAAYPIVHVDMENIILSKNEDQSTQ NTDSQTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSLS LAGERTWAETMGLNTADTARLNANIRYVNTGTAPIYNVLPTTSLVLGKNQTLATIKA KENQLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTDQ VYGNIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPSD PLETTKPDMTLKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQNIKNQLAELN ATNIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTEGLL LNIDKDIRKILSGYIVEIEDTEGLKEVINDRYDMLNISSLRQDGKTFIDFKKYNAKEPL KISNPNYKVNVYAVTKENTIINPSENGDTSTNGIKKILIFSKKGYEIGGGGSGTRGGGS GVDNKFNKEKYN A Y YEIWQLPNLTKYQKA AFIGKLQDDPS QS SELLS EAKKLNDS Q APK (SEQ ID NO: 63)

[0061] SEQ ID NO: 16 is an exemplary amino acid sequence of PA3MzEGFR-U2 (uPA activated, EGFR targeted, not toxic to mice), aa 1-738 is a PA3M sequence with the furin cleavage site replaced with an uPA cleavage site. The uPA cleavage site (PGSGRSA, SEQ ID NO:4) is underlined and bolded. The 3M mutations are underlined and bolded, aa 739-750 (underlined) is a GS linker, aa 751-808 is zEGFR2377.

[0062] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIPS ENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNKIRLEKG RLYOIKIOYORENPTEKGLDFKLYWTDSONKKEVISSDNLOLPELKQKSSNSPGSGR SASTSAGPTVPDRDNDGIPDSLEVEGYTVDVKNKRTFLSPWISNIHEKKGLTKYKSSP EKWSTASDPYSDFEKVTGRIDKNVSPEARHPLVAAYPIVHVDMENIILSKNEDQSTQ NTDSQTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSLS LAGERTWAETMGLNTADTARLNANIRYVNTGTAPIYNVLPTTSLVLGKNQTLATIKA KENQLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTDQ VYGNIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPSD PLETTKPDMTLKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQNIKNQLAELN ATNIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTEGLL LNIDKDIRKILSGYIVEIEDTEGLKEVINDRYDMLNISSLRQDGKTFIDFKKYNAKEPL KISNPNYKVNVYAVTKENTIINPSENGDTSTNGIKKILIFSKKGYEIGGGGSGTRGGGS GVDNKFNKEMWIAWEEIRDLPNLNGWQMTAFIASLLDDPSQSANLLAEAKKLNDA QAPK (SEQ ID NO: 16)

[0063] Intermolecular complementing PA (ICPA) proteins refer to a pair of PA variants that do not have any activity when separated, but when combined with each other can form active oligomers to deliver an effector protein into cancer cells, thereby achieving higher levels of tumor targeting (e.g., simultaneously targeting at least three targets selected from MMP,8123-112570-02

[0064] uPA / uPAR, EGFR, HER2, and HER3). The below tables summarize the PA variant sequences that can be paired together and the targets of each pair.

[0065] SEQ ID NO: 17 SEQ ID NO: 18 SEQ ID NO: 19 SEQ ID NO: 20 uPA / uPAR, MMPs, uPA / uPAR, MMPs, uPA / uPAR, MMPs, and HER2 HER2, and HER3 HER2, and EGFR SEQ ID NO: 21 uPA / uPAR, MMPs, uPA / uPAR, MMPs, uPA / uPAR, MMPs, HER2, and HER3 and HER3 HER3, and EGFR SEQ ID NO: 22 uPA / uPAR, MMPs, uPA / uPAR, MMPs, uPA / uPAR, MMPs,

[0066]

[0067] HER2, and EGFR HER3, and EGFR and EGFR

[0068] SEQ ID NO: 23 SEQ ID NO: 24 SEQ ID NO: 25 SEQ ID NO: 26 uPA / uPAR, MMPs, uPA / uPAR, MMPs, uPA / uPAR, MMPs, and HER2 HER2, and HER3 HER2, and EGFR SEQ ID NO: 27 uPA / uPAR, MMPs, uPA / uPAR, MMPs, uPA / uPAR, MMPs, HER2, and HER3 and HER3 HER3, and EGFR SEQ ID NO: 28 uPA / uPAR, MMPs, uPA / uPAR, MMPs, uPA / uPAR, MMPs,

[0069]

[0070] HER2, and EGFR HER3, and EGFR and EGFR

[0071] SEQ ID NO: 17 is an exemplary amino acid sequence of PA3MzHER2-Ll I207R (mature: 90.46 kDa). aa 1-739 is a PA3M sequence with the furin cleavage site replaced with an MMP cleavage site and comprising an I207R substitution. The MMP cleavage site (GPLGMLSQ, SEQ ID NO: 3) is underlined and bolded. The 3M mutations are underlined and bolded. I207R refers to a substitution from I to R (underlined and bolded) at position 207 (numbering according to a furin-activated PA sequence, e.g., SEQ ID NO: 1 or 2). aa 740- 751 (underlined) is a GS linker, aa 752-809 is zHER2:343.

[0072] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIPS ENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNKIRLEKG RLYOIKIOYORENPTEKGLDFKLYWTDSONKKEVISSDNLOLPELKQKSSNSGPLGM LSQSTSAGPTVPDRDNDGIPDSLEVEGYTVDVKNKRTFLSPWRSNIHEKKGLTKYKS SPEKWSTASDPYSDFEKVTGRIDKNVSPEARHPLVAAYPIVHVDMENIILSKNEDQST QNTDSQTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSL SLAGERTWAETMGLNTADTARLNANIRYVNTGTAPIYNVLPTTSLVLGKNQTLATIK AKENQLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTD QVYGNIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPS DPLETTKPDMTLKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQNTKNQLAEL NATNIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTEGL LLNIDKDIRKILSGYIVEIEDTEGLKEVTNDRYDMLNISSLRQDGKTFIDFKKYNAKEP LKISNPNYKVNVYAVTKENTIINPSENGDTSTNGIKKILIFSKKGYEIGGGGSGTRGGG SGVDNKFNKEMRNAYWEIALLPNLNNQQKRAFIRSLYDDPSQSANLLAEAKKLNDA QAPK (SEQ ID NO: 17)8123-112570-02

[0073] SEQ ID NO: 18 is an exemplary amino acid sequence of PA3MzHER3-Ll I207R (mature: 90.46 kDa). aa 1-739 is a PA3M sequence with the furin cleavage site replaced with an MMP cleavage site and comprising an I207R substitution. The MMP cleavage site (GPLGMLSQ, SEQ ID NO: 3) is underlined and bolded. The 3M mutations are underlined and bolded. I207R refers to a substitution from I to R (underlined and bolded) at position 207 (numbering according to a furin-activated PA sequence, e.g., SEQ ID NO: 1 or 2). aa 740- 751 (underlined) is a GS linker, aa 752-809 is zHER3:8699.

[0074] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIPS ENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNKIRLEKG RLYOIKIOYORENPTEKGLDFKLYWTDSONKKEVISSDNLOLPELKQKSSNSGPLGM LSQSTSAGPTVPDRDNDGIPDSLEVEGYTVDVKNKRTFLSPWRSNIHEKKGLTKYKS SPEKWSTASDPYSDFEKVTGRIDKNVSPEARHPLVAAYPIVHVDMENIILSKNEDQST QNTDSQTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSL SLAGERTWAETMGLNTADTARLNANIRYVNTGTAPIYNVLPTTSLVLGKNQTLATIK AKENQLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTD QVYGNIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPS DPLETTKPDMTLKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQNIKNQLAEL NATNIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTEGL LLNIDKDIRKILSGYIVEIEDTEGLKEVINDRYDMLNISSLRQDGKTFIDFKKYNAKEP LKISNPNYKVNVYAVTKENTIINPSENGDTSTNGIKKILIFSKKGYEIGGGGSGTRGGG SGVDNKFNKEKYNAYYEIWQLPNLTKYQKAAFIGKLQDDPSQSSELLSEAKKLNDS QAPK (SEQ ID NO: 18)

[0075] SEQ ID NO: 19 is an exemplary amino acid sequence of PA3MzEGFR-Ll I207R (mature: 90.46 kDa). aa 1-739 is a PA3M sequence with the furin cleavage site replaced with an MMP cleavage site and comprising an I207R substitution. The MMP cleavage site (GPLGMLSQ, SEQ ID NO: 3) is underlined and bolded. The 3M mutations are underlined and bolded. I207R refers to a substitution from I to R (underlined and bolded) at position 207 (numbering according to a furin-activated PA sequence, e.g., SEQ ID NO: 1 or 2). aa 740- 751 (underlined) is a GS linker, aa 752-809 is zEGFR2377.

[0076] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIPS ENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNKIRLEKG RLYOIKIOYORENPTEKGLDFKLYWTDSONKKEVISSDNLOLPELKQKSSNSGPLGM LSQSTSAGPTVPDRDNDGIPDSLEVEGYTVDVKNKRTFLSPWRSNIHEKKGLTKYKS SPEKWSTASDPYSDFEKVTGRIDKNVSPEARHPLVAAYPIVHVDMENIILSKNEDQST QNTDSQTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSL SLAGERTWAETMGLNTADTARLNANIRYVNTGTAPIYNVLPTTSLVLGKNQTLATIK AKENQLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTD QVYGNIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPS DPLETTKPDMTLKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQNIKNQLAEL NATNIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTEGL LLNIDKDIRKILSGYIVEIEDTEGLKEVINDRYDMLNISSLRQDGKTFIDFKKYNAKEP LKISNPNYKVNVYAVTKENTIINPSENGDTSTNGIKKILIFSKKGYEIGGGGSGTRGGG8123-112570-02

[0077] SGVDNKFNKEMWIAWEEIRDLPNLNGWQMTAFIASLLDDPSQSANLLAEAKKLNDA QAPK (SEQ ID NO: 19)

[0078] SEQ ID NO: 20 is an exemplary amino acid sequence of PA3MzHER2-U2 R200A (mature: 90.16 kDa). aa 1-738 is a PA3M sequence with the furin cleavage site replaced with an uPA cleavage site and comprising an R200A substitution. The uPA cleavage site (PGSGRSA, SEQ ID NO: 4) is underlined and bolded. The 3M mutations are underlined and bolded. R200A refers to a substitution from R to A (underlined and bolded) at position 200 (numbering according to a furin-activated PA sequence, e.g., SEQ ID NO: 1 or 2). aa 739-750 (underlined) is a GS linker, aa 751-808 is zHER2:343.

[0079] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIPS ENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNKIRLEKG RLYOIKIOYORENPTEKGLDFKLYWTDSONKKEVISSDNLOLPELKQKSSNSPGSGR SASTSAGPTVPDRDNDGIPDSLEVEGYTVDVKNKATFLSPWISNIHEKKGLTKYKSSP EKWSTASDPYSDFEKVTGRIDKNVSPEARHPLVAAYPIVHVDMENIILSKNEDQSTQ NTDSQTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSLS LAGERTWAETMGLNTADTARLNANIRYVNTGTAPIYNVLPTTSLVLGKNQTLATIKA KENQLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTDQ VYGNIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPSD PLETTKPDMTLKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQNIKNQLAELN ATNIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTEGLL LNIDKDIRKILSGYIVEIEDTEGLKEVINDRYDMLNISSLRQDGKTFIDFKKYNAKEPL KISNPNYKVNVYAVTKENTIINPSENGDTSTNGIKKILIFSKKGYEIGGGGSGTRGGGS GVDNKFNKEMRNAYWEIALLPNLNNQQKRAFIRSLYDDPSQSANLLAEAKKLNDAQ APK (SEQ ID NO: 20)

[0080] SEQ ID NO: 21 is an exemplary amino acid sequence of PA3MzHER3-U2 R200A (mature: 90.16 kDa). aa 1-738 is a PA3M sequence with the furin cleavage site replaced with an uPA cleavage site and comprising an R200A substitution. The uPA cleavage site (PGSGRSA, SEQ ID NO: 4) is underlined and bolded. The 3M mutations are underlined and bolded. R200A refers to a substitution from R to A (underlined and bolded) at position 200 (numbering according to a furin-activated PA sequence, e.g., SEQ ID NO: 1 or 2). aa 739-750 (underlined) is a GS linker, aa 751-808 is zHER3:8699.

[0081] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIPS ENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNK1RLEKG RLYOIKIOYORENPTEKGLDFKLYWTDSONKKEVISSDNLOLPELKQKSSNSPGSGR SASTSAGPTVPDRDNDG1PDSLEVEGYTVDVKNKATFLSPWISNIHEKKGLTKYKSSP EKWSTASDPYSDFEKVTGRIDKNVSPEARHPLVAAYPIVHVDMENIILSKNEDQSTQ NTDSQTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSLS LAGERTWAETMGLNTADTARLNANIRYVNTGTAPIYNVLPTTSLVLGKNQTLATIKA KENQLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTDQ VYGNIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPSD8123-112570-02

[0082] PLETTKPDMTLKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQNIKNQLAELN ATNIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTEGLL LNIDKDIRKILSGYIVEIEDTEGLKEVINDRYDMLNISSLRQDGKTFIDFKKYNAKEPL KISNPNYKVNVYAVTKENTIINPSENGDTSTNGIKKILIFSKKGYEIGGGGSGTRGGGS GVDNKFNKEKYN AY YEIWQLPNLTKYQKA AFIGKLQDDPS QS SELLS EAKKLNDS Q APK (SEQ ID NO: 21)

[0083] SEQ ID NO: 22 is an exemplary amino acid sequence of PA3MzEGFR-U2 R200A (mature: 90.16 kDa). aa 1-738 is a PA3M sequence with the furin cleavage site replaced with an uPA cleavage site and comprising an R200A substitution. The uPA cleavage site (PGSGRSA, SEQ ID NO: 4) is underlined and bolded. The 3M mutations are underlined and bolded. R200A refers to a substitution from R to A (underlined and bolded) at position 200 (numbering according to a furin-activated PA sequence, e.g., SEQ ID NO: 1 or 2). aa 739- 750 (underlined) is a GS linker, aa 751-808 is zEGFR2377.

[0084] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIPS ENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNKIRLEKG RLYOIKIOYORENPTEKGLDFKLYWTDSONKKEVISSDNLOLPELKQKSSNSPGSGR SASTSAGPTVPDRDNDGIPDSLEVEGYTVDVKNKATFLSPWISNIHEKKGLTKYKSSP EKWSTASDPYSDFEKVTGRIDKNVSPEARHPLVAAYPIVHVDMENIILSKNEDQSTQ NTDSQTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSLS LAGERTWAETMGLNTADTARLNANIRYVNTGTAPIYNVLPTTSLVLGKNQTLATIKA KENQLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTDQ VYGNIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPSD PLETTKPDMTLKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQNIKNQLAELN ATNIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTEGLL LNIDKDIRKILSGYIVEIEDTEGLKEVINDRYDMLNISSLRQDGKTFIDFKKYNAKEPL KISNPNYKVNVYAVTKENTIINPSENGDTSTNGIKKILIFSKKGYEIGGGGSGTRGGGS GVDNKFNKEMWIAWEEIRDLPNLNGWQMTAFIASLLDDPSQSANLLAEAKKLNDA QAPK (SEQ ID NO: 22)

[0085] SEQ ID NO: 23 is an exemplary amino acid sequence of PA3MzHER2-Ll GN. aa 1-739 is a PA3M sequence with the furin cleavage site replaced with an MMP cleavage site and comprising K245G and R252N substitutions. The MMP cleavage site (GPLGMLSQ, SEQ ID NO: 3) is underlined and bolded. The 3M mutations are underlined and bolded. K245G refers to a substitution from K to G (underlined and bolded) at position 245, and R252N refers to a substitution from R to N (underlined and bolded) at position 252 (numbering according to a furin-activated PA sequence, e.g., SEQ ID NO: 1 or 2). aa 740- 751 (underlined) is a GS linker, aa 752-809 is zHER2:343.

[0086] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIPS ENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNKTRLEKG RLYOIKIOYORENPTEKGLDFKLYWTDSONKKEVISSDNLOLPELKQKSSNSGPLGM LSQSTSAGPRVPDRDNDGIPDSLEVEGYTVDVKNKRTFLSPWISNIHEKKGLTKYKSS8123-112570-02

[0087] PEKWSTASDPYSDFEKVTGRIDGNVSPEANHPLVAAYPIVHVDMENIILSKNEDQSTQ NTDSQTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSLS LAGERTWAETMGLNTADTARLNANIRYVNTGTAPIYNVLPTTSLVLGKNQTLATIKA KENQLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTDQ VYGNIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPSD PLETTKPDMTLKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQNIKNQLAELN ATNIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTEGLL LNIDKDIRKILSGYIVEIEDTEGLKEVINDRYDMLNISSLRQDGKTFIDFKKYNAKEPL KISNPNYKVNVYAVTKENTIINPSENGDTSTNGIKKILIFSKKGYEIGGGGSGTRGGGS GVDNKFNKEMRNAYWEIALLPNLNNQQKRAFIRSLYDDPSQSANLLAEAKKLNDAQ APK (SEQ ID NO: 23)

[0088] SEQ ID NO: 24 is an exemplary amino acid sequence of PA3MzHER3-Ll GN. aa 1-739 is a PA3M sequence with the furin cleavage site replaced with an MMP cleavage site and comprising K245G and R252N substitutions. The MMP cleavage site (GPLGMLSQ, SEQ ID NO: 3) is underlined and bolded. The 3M mutations are underlined and bolded. K245G refers to a substitution from K to G (underlined and bolded) at position 245, and R252N refers to a substitution from R to N (underlined and bolded) at position 252 (numbering according to a furin-activated PA sequence, e.g., SEQ ID NO: 1 or 2). aa 740-751 (underlined) is a GS linker, aa 752-809 is zHER3:8699.

[0089] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIPS ENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNKIRLEKG RLYOIKIOYORENPTEKGLDFKLYWTDSONKKEVISSDNLOLPELKQKSSNSGPLGM LSQSTSAGPRVPDRDNDGIPDSLEVEGYTVDVKNKRTFLSPWISNIHEKKGLTKYKSS PEKWSTASDPYSDFEKVTGRIDGNVSPEANHPLVAAYPIVHVDMENIILSKNEDQSTQ NTDSQTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSLS LAGERTWAETMGLNTADTARLNANIRYVNTGTAPIYNVLPTTSLVLGKNQTLATIKA KENQLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTDQ VYGNIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPSD PLETTKPDMTLKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQNIKNQLAELN ATNIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTEGLL LNIDKDIRKILSGYIVEIEDTEGLKEVINDRYDMLNISSLRQDGKTFIDFKKYNAKEPL KISNPNYKVNVYAVTKENTIINPSENGDTSTNGIKKILIFSKKGYEIGGGGSGTRGGGS GVDNKFNKEKYN A Y YEIWQLPNLTKYQKA AFIGKLQDDPS QS SELLS EAKKLNDS Q APK (SEQ ID NO: 24)

[0090] SEQ ID NO: 25 is an exemplary amino acid sequence of PA3MzEGFR-Ll GN. aa 1-739 is a PA3M sequence with the furin cleavage site replaced with an MMP cleavage site and comprising K245G and R252N substitutions. The MMP cleavage site (GPLGMLSQ, SEQ ID NO: 3) is underlined and bolded. The 3M mutations are underlined and bolded. K245G refers to a substitution from K to G (underlined and bolded) at position 245, and R252N refers to a substitution from R to N (underlined and bolded) at position 2528123-112570-02

[0091] (numbering according to a furin-activated PA sequence, e.g., SEQ ID NO: 1 or 2). aa 740- 751 (underlined) is a GS linker, aa 752-809 is zEGFR2377.

[0092] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIPS ENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNKIRLEKG RLYOIKIOYORENPTEKGLDFKLYWTDSONKKEVISSDNLOLPELKOKSSNSGPLGM LSQSTSAGPRVPDRDNDGIPDSLEVEGYTVDVKNKRTFLSPWISNIHEKKGLTKYKSS PEKWSTASDPYSDFEKVTGRIDGNVSPEANHPLVAAYPIVHVDMENIILSKNEDQSTQ NTDSQTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSLS LAGERTWAETMGLNTADTARLNANIRYVNTGTAPIYNVLPTTSLVLGKNQTLATIKA KENQLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTDQ VYGNIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPSD PLETTKPDMTLKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQNIKNQLAELN ATNIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTEGLL LNIDKDIRKILSGYIVEIEDTEGLKEVINDRYDMLNISSLRQDGKTFIDFKKYNAKEPL KISNPNYKVNVYAVTKENTIINPSENGDTSTNGIKKILIFSKKGYEIGGGGSGTRGGGS GVDNKFNKEMWIAWEEIRDLPNLNGWQMTAFIASLLDDPSQSANLLAEAKKLNDA QAPK (SEQ ID NO: 25)

[0093] SEQ ID NO: 26 is an exemplary amino acid sequence of PA3MzHER2-U2 D512K. aa 1-738 is a PA3M sequence with the furin cleavage site replaced with an uPA cleavage site and comprising a D512K substitution. The uPA cleavage site (PGSGRSA, SEQ ID NO: 4) is underlined and bolded. The 3M mutations are underlined and bolded. D512K refers to a substitution from D to K (underlined and bolded) at position 512 (numbering according to a furin-activated PA sequence, e.g., SEQ ID NO: 1 or 2). aa 739-750 (underlined) is a GS linker, aa 751-808 is zHER2:343.

[0094] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIPS ENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNKIRLEKG RLYOIKIOYORENPTEKGLDFKLYWTDSONKKEVISSDNLOLPELKOKSSNSPGSGR SASTSAGPTVPDRDNDGIPDSLEVEGYTVDVKNKRTFLSPWISNIHEKKGLTKYKSSP EKWSTASDPYSDFEKVTGRIDKNVSPEARHPLVAAYPIVHVDMENIILSKNEDQSTQ NTDSQTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSLS LAGERTWAETMGLNTADTARLNANIRYVNTGTAPIYNVLPTTSLVLGKNQTLATIKA KENQLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTDQ VYGNIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPSK PLETTKPDMTLKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQNIKNQLAELN ATNIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTEGLL LNIDKDIRKILSGYIVEIEDTEGLKEVINDRYDMLNISSLRQDGKTFIDFKKYNAKEPL KISNPNYKVNVYAVTKENTHNPSENGDTSTNG1KK1EIFSKKGYE1GGGGSGTRGGGS GVDNKFNKEMRNAYWEIALLPNLNNQQKRAFIRSLYDDPSQSANLLAEAKKLNDAQ APK (SEQ ID NO: 26)

[0095] SEQ ID NO: 27 is an exemplary amino acid sequence of PA3MzHER3-U2 D512K. aa 1-738 is a PA3M sequence with the furin cleavage site replaced with an uPA cleavage site and comprising a D512K substitution. The uPA cleavage site (PGSGRSA, SEQ ID NO: 4) is8123-112570-02

[0096] underlined and bolded. The 3M mutations are underlined and bolded. D512K refers to a substitution from D to K (underlined and bolded) at position 512 (numbering according to a furin-activated PA sequence, e.g., SEQ ID NO: 1 or 2). aa 739-750 (underlined) is a GS linker, aa 751-808 is zHER3:8699.

[0097] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIPS ENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNKIRLEKG RLYOIKIOYORENPTEKGLDFKLYWTDSONKKEVISSDNLOLPELKOKSSNSPGSGR SASTSAGPTVPDRDNDG1PDSLEVEGYTVDVKNKRTFLSPWISNIHEKKGLTKYKSSP EKWSTASDPYSDFEKVTGRIDKNVSPEARHPLVAAYPIVHVDMENIILSKNEDQSTQ NTDSQTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSLS LAGERTWAETMGLNTADTARLNANIRYVNTGTAPIYNVLPTTSLVLGKNQTLATIKA KENQLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTDQ VYGNIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPSK PLETTKPDMTLKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQNIKNQLAELN ATNIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTEGLL LNIDKDIRKILSGYIVEIEDTEGLKEVINDRYDMLNISSLRQDGKTFIDFKKYNAKEPL KISNPNYKVNVYAVTKENTIINPSENGDTSTNGIKKILIFSKKGYEIGGGGSGTRGGGS GVDNKFNKEKYNAYYEIWQLPNLTKYQKAAFIGKLQDDPSQSSELLSEAKKLNDSQ APK (SEQ ID NO: 27)

[0098] SEQ ID NO: 28 is an exemplary amino acid sequence of PA3MzEGFR-U2 D512K. aa 1-738 is a PA3M sequence with the furin cleavage site replaced with an uPA cleavage site and comprising a D512K substitution. The uPA cleavage site (PGSGRSA, SEQ ID NO: 4) is underlined and bolded. The 3M mutations are underlined and bolded. D512K refers to a substitution from D to K (underlined and bolded) at position 512 (numbering according to a furin-activated PA sequence, e.g., SEQ ID NO: 1 or 2). aa 739-750 (underlined) is a GS linker, aa 751-808 is zEGFR2377.

[0099] EVKQENRLLNESESSSQGLLGYYFSDLNFQAPMVVTSSTTGDLSIPSSELENIPS ENQYFQSAIWSGFIKVKKSDEYTFATSADNHVTMWVDDQEVINKASNSNKIRLEKG RLYOIKIOYORENPTEKGLDFKLYWTDSONKKEVISSDNLOLPELKOKSSNSPGSGR SASTSAGPTVPDRDNDGIPDSLEVEGYTVDVKNKRTFLSPWISNIHEKKGLTKYKSSP EKWSTASDPYSDFEKVTGRIDKNVSPEARHPLVAAYPIVHVDMENIILSKNEDQSTQ NTDSQTRTISKNTSTSRTHTSEVHGNAEVHASFFDIGGSVSAGFSNSNSSTVAIDHSLS LAGERTWAETMGLNTADTARLNANIRYVNTGTAPIYNVLPTTSLVLGKNQTLATIKA KENQLSQILAPNNYYPSKNLAPIALNAQDDFSSTPITMNYNQFLELEKTKQLRLDTDQ VYGNIATYNFENGRVRVDTGSNWSEVLPQIQETTARIIFNGKDLNLVERRIAAVNPSK PLETTKPDMTLKEALKIAFGFNEPNGNLQYQGKDITEFDFNFDQQTSQN1KNQLAELN ATNIYTVLDKIKLNAKMNILIRDKRFHYDRNNIAVGADESVVKEAHREVINSSTEGLL LNIDKDIRK1LSGY1VE1EDTEGLKEV1NDRYDMLN1SSLRQDGKTFIDFKKYNAKEPL KISNPNYKVNVYAVTKENTIINPSENGDTSTNGIKKILIFSKKGYEIGGGGSGTRGGGS GVDNKFNKEMWIAWEEIRDLPNLNGWQMTAFIASLLDDPSQSANLLAEAKKLNDA QAPK (SEQ ID NO: 28)8123-112570-02

[0100] IL Effector Proteins

[0101] SEQ ID NO: 29 is an exemplary amino acid sequence of DUF5 (WT) (which cleaves both RAS and RAP1 proteins).

[0102] DKTKVVVDLAQIFTVQELKERAKVFAKPIGASYQGILDQLDLVHQAKGRDQI AASFELNKKINDYIAEHPTSGRNQALTQLKEQVTSALFIGKMQVAQAGIDAIAQTRPE LAARIFMVAIEEANGKHVGLTDMMVRWANEDPYLAPKHGYKGETPSDLGFDAKYH VDLGEHYADFKQWLETSQSNGLLSKATLDESTKTVHLGYSYQELQDLTGAESVQM AFYFLKEAAKKADPISGDSAEMILLKKFADQSYLSQLDSDRMDQIEGIYRSSHETDID AWDRRYSGTGYDELTNKLASATGVDEQLAVLLDDRKGLLIGEVHGSDVNGLRFVN EQMDALKKQGVTVIGLEHLRSDLAQPLIDRYLATGVMSSELSAMLKTKHLDVTLFE NARANGMRIVALDANSSARPNVQGTEHGLMYRAGAANNIAVEVLQNLPDGEKFVA IYGKAHLQSHKGIEGFVPGITHRLDLPALKVSDSNQFTVEQDDVSLRVVYDDVANKP KITFKGSL (SEQ ID NO: 29)

[0103] SEQ ID NO: 30 is an exemplary amino acid sequence of DUF5 L418W (which cleaves RAS but not RAP1 proteins). The L418W substitution is underlined and bolded.

[0104] DKTKVVVDLAQIFTVQELKERAKVFAKPIGASYQGILDQLDLVHQAKGRDQI AASFELNKKINDYIAEHPTSGRNQALTQLKEQVTSALFIGKMQVAQAGIDAIAQTRPE LAARIFMVAIEEANGKHVGLTDMMVRWANEDPYLAPKHGYKGETPSDLGFDAKYH VDLGEHYADFKQWLETSQSNGLLSKATLDESTKTVHLGYSYQELQDLTGAESVQM AFYFLKEAAKKADPISGDSAEMILLKKFADQSYLSQLDSDRMDQIEGIYRSSHETDID AWDRRYSGTGYDELTNKLASATGVDEQLAVLLDDRKGLLIGEVHGSDVNGLRFVN EQMDALKKQGVTVIGLEHLRSDLAQPLIDRYLATGVMSSELSAMLKTKHLDVTLFE NARANGMRIVALDANSSARPNVQGTEHGWMYRAGAANNIAVEVLQNLPDGEKFV AIYGKAHLQSHKGIEGFVPGITHRLDLPALKVSDSNQFTVEQDDVSLRVVYDDVANK PKITFKGSL (SEQ ID NO: 30)

[0105] SEQ ID NO: 31 is an exemplary amino acid sequence of a domain of Anthrax lethal factor that binds to Anthrax protective antigen (LFn).

[0106] AGGHGDVGMHVKEKEKNKDENKRKDEERNKTQEEHLKEIMKHIVKIEVKGE EAVKKEAAEKLLEKVPSDVLEMYKAIGGKIYIVDGDITKHISLEALSEDKKKIKDIYG KDALLHEHYVYAKEGYEPVLVIQSSEDYVENTEKALNVYYEIGKILSRDILSKINQPY QKFLDVLNTIKNASDSDGQDLLFTNQLKEHPTDFSVEFLEQNSNEVQEVFAKAFAYY IEPQHRDVLQLYAPEAFNYMDKFNEQEINLS (SEQ ID NO: 31)

[0107] SEQ ID NO: 32 is an exemplary amino acid sequence of LFn-DUF5 (WT) (which cleaves both RAS and RAP1 proteins), aa 1-255 is an LFn domain, aa 260-768 is DUF5 (WT).

[0108] AGGHGDVGMHVKEKEKNKDENKRKDEERNKTQEEHLKEIMKHIVKIEVKGE EAVKKEAAEKLLEKVPSDVLEMYKAIGGKIYIVDGDITKHISLEALSEDKKKIKDIYG KDALLHEHYVYAKEGYEPVLVIQSSEDYVENTEKALNVYYEIGKILSRDILSKINQPY QKFLDVLNTIKNASDSDGQDLLFTNQLKEHPTDFSVEFLEQNSNEVQEVFAKAFAYY IEPQHRDVLQLYAPEAFNYMDKFNEQEINLSTRGSDKTKVVVDLAQIFTVQELKERA KVFAKPIGASYQGILDQLDLVHQAKGRDQIAASFELNKKINDYIAEHPTSGRNQALT8123-112570-02

[0109] QLKEQVTSALFIGKMQVAQAGIDAIAQTRPELAARIFMVAIEEANGKHVGLTDMMV RWANEDPYLAPKHGYKGETPSDLGFDAKYHVDLGEHYADFKQWLETSQSNGLLSK ATLDESTKTVHLGYSYQELQDLTGAESVQMAFYFLKEAAKKADPISGDSAEMILLKK FADQSYLSQLDSDRMDQIEGIYRSSHETDIDAWDRRYSGTGYDELTNKLASATGVDE QLAVLLDDRKGLLIGEVHGSDVNGLRFVNEQMDALKKQGVTVIGLEHLRSDLAQPL IDRYLATGVMSSELSAMLKTKHLDVTLFENARANGMRIVALDANSSARPNVQGTEH GLMYRAGAANNIAVEVLQNLPDGEKFVAIYGKAHLQSHKGIEGFVPGITHRLDLPAL KVSDSNQFTVEQDDVSLRVVYDDVANKPKITFKGSL (SEQ ID NO: 32)

[0110] SEQ ID NO: 33 is an exemplary amino acid sequence of LFn-DUF5 L418W (which cleaves RAS but not RAP1 proteins), aa 1-255 is an LFn domain, aa 260-768 is DUF5 L418W. The L418W substitution is underlined and bolded.

[0111] AGGHGDVGMHVKEKEKNKDENKRKDEERNKTQEEHLKEIMKHIVKIEVKGE EAVKKEAAEKLLEKVPSDVLEMYKAIGGKIYIVDGDITKHISLEALSEDKKKIKDIYG KDALLHEHYVYAKEGYEPVLVIQSSEDYVENTEKALNVYYEIGKILSRDILSKINQPY QKFLDVLNTIKNASDSDGQDLLFTNQLKEHPTDFSVEFLEQNSNEVQEVFAKAFAYY IEPQHRDVLQLYAPEAFNYMDKFNEQEINLSTRGSDKTKVVVDLAQIFTVQELKERA KVFAKPIGASYQGILDQLDLVHQAKGRDQIAASFELNKKINDYIAEHPTSGRNQALT QLKEQVTSALFIGKMQVAQAGIDAIAQTRPELAARIFMVAIEEANGKHVGLTDMMV RWANEDPYLAPKHGYKGETPSDLGFDAKYHVDLGEHYADFKQWLETSQSNGLLSK ATLDESTKTVHLGYSYQELQDLTGAESVQMAFYFLKEAAKKADPISGDSAEMILLKK FADQSYLSQLDSDRMDQIEGIYRSSHETDIDAWDRRYSGTGYDELTNKLASATGVDE QLAVLLDDRKGLLIGEVHGSDVNGLRFVNEQMDALKKQGVTVIGLEHLRSDLAQPL IDRYLATGVMSSELSAMLKTKHLDVTLFENARANGMRIVALDANSSARPNVQGTEH GWMYRAGAANNIAVEVLQNLPDGEKFVAIYGKAHLQSHKGIEGFVPGITHRLDLPA LKVSDSNQFTVEQDDVSLRVVYDDVANKPKITFKGSL (SEQ ID NO: 33)

[0112] BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1: A diagram showing the anthrax protein delivery system as a platform for targeting HER2 -expressing tumors with high specificity. Tumor specificity is achieved by reengineering the delivering vehicle protective antigen (PA) to bind to HER2 and also rely on tumor-associated proteases (MMPs) for activation. Effectors like LFn-DUF5-L418W (pan-RAS inactivator) or LF-W271 A (MEK1 / 2 inactivator) can be selectively delivered into tumor cells to cleave RAS and MEK, respectively.

[0113] FIGs.2A-2B: HER2 expression profile in cancer cells. FIG.2A shows the results of real-time PCR analysis of HER2 expression in a set of cancer cells. HER2 expression levels are presented as relative levels to that seen in HT1080 cells (mean ± SD). FIG.2B shows gel images of Western blotting analysis of HER2 expression using an anti-HER2 antibody. The cancer cells expressed various levels of HER2, whereas no HER2 expression was seen in HER2-knockout (KO) cells.8123-112570-02

[0114] FIGs. 3A-3C: Graphs showing HER2 -dependent cytotoxicity of HER2 -targeting toxins to cancer cells. FIGs 3A and 3B show viabilities of cancer cells (mean ± SD) that had been incubated with various concentrations of PA3MzHER2 (A) or PA3MzHER2-MMP (B) in the presence of FP59 (1.9 nM) for 48 hours. FIG. 3C shows viabilities of cancer cells, and their respective HER2-KO cells (mean ± SD) that had been incubated with various concentrations of PA3MzHER2 in the presence of FP59 (1.9 nM) for 48 hours. Cell viability was assessed by an MTT assay. HER2-KO cells were resistant to the HER2-targeting toxin.

[0115] FIGs. 4A-4B: Results of Schild plot analysis to estimate the apparent binding affinity of PA3MzHER2-MMP to HER2 expressing cells. FIG. 4A shows viabilities of MDA-MB-231 cells (low HER2) (mean ± SD) that had been incubated with PA3MzHER2-MMP at various concentrations plus FP59 at a constant concentration (1.9 nM) in the presence of a nontoxic competitor PA3MzHER2-MMP-GN at various fixed concentrations. Cell viability was assessed by an MTT assay. Addition of the binding competitor shifted the cytotoxicity dose-response curves of PA3MzHER2-MMP rightward. FIG. 4B shows the regression curve obtained using Prism program. Apparent binding affinity Kd = 0.21 nM.

[0116] FIGs. 5A-5B: Graphs showing reduced unwanted toxicity of PA3MzHER2-MMP to normal tissues. FIG. 5A shows viabilities of noncancerous BEAS-2B cells (mean ± SD) that had been treated with various concentrations of PA3MzHER2 or PA3MzHER2-MMP in the presence of FP59 (1.9 nM) for 48 hours. Cell viability was assessed by an MTT assay. FIG.

[0117] 5B shows a survival plot of mice that were being injected (I.P. route) with multiple doses of PA3MzHER2 or PA3MzHER2-MMP combined with FP59 or LFn-DUF5-L418W at days indicated by the arrows, and monitored for signs of malaise and survival. Log-rank test was used for survival comparisons. PA3MzHER2-MMP group vs. PA3MzHER2 group: p< 0.0001.

[0118] FIG. 6: A graph showing that PA3MzHER2-MMP requires MMPs for activation. Viabilities are shown for HT1080 cells (mean ± SD) that had been treated with PA3MzHER2-MMP (1 nM) or PA3MzHER2 (1 nM), plus FP59 (1.9 nM), in the presence of various concentrations of a specific MMP inhibitor GM6001 for 48 hours. Cell viability was assessed by an MTT assay. Notably, GM6001 could diminish the cytotoxicity of PA3MzHER2-MMP but not that of PA3MzHER2 in a dose-dependent manner.

[0119] FIGs. 7A-7B: Graphs showing sensitivity of human cancer cells with oncogenic RAS mutations to RAS inhibition. FIG. 7A shows viabilities of cancer cells with oncogenic RAS mutations or the BRAFV600Emutation (mean ± SD) that had been incubated with various concentrations of PA3MzHER2 in the presence of LFn-DUF5-L418W (1 nM) for 48 hours.8123-112570-02

[0120] FIG. 7B shows viabilities of the same cancer cells as in FIG. 7A, that had been incubated with various concentrations of PA3MzHER2 in the presence of FP59 (1.9 nM) for 48 hours. Cell viability was assessed by an MTT assay. Cancer cells with oncogenic RAS mutations, but not cancer cells with the BRAFV600Emutation, were sensitive to RAS inhibition (PA3MzHER2 / LFn-DUF5-L418W) (A). All cancer cells tested are sensitive to protein synthesis inhibition (PA3MzHER2 / FP59) regardless of their oncogenic mutations (B).

[0121] FIGs. 8 -8C: Cytotoxicity comparison of HER2-targeting toxins versus HER2-targeting agents currently used in the clinic. FIG. 8A shows viabilities of HT1080 cells (mean ± SD) that had been incubated with various concentrations of trastuzumab, T-DM1, T-DXd, or PA3MzHER2-MMP combined with FP59 (1.9 nM) or LFn-DUF5-L418W (1 nM ), for 48 hours. FIG. 8B shows viabilities of MDA-MB-231 cells (mean ± SD) that had been incubated under the same conditions as in FIG. 8A. FIG.8C shows viabilities of MiaPaCa-2 cells (mean ± SD) that had been incubated under the same conditions as in FIG. 8A. Cell viability was assessed by an MTT assay. The toxins exhibited much higher cytotoxicity than trastuzumab and its drug conjugates to all cancer cells tested.

[0122] FIGs. 9A-9C: LFn-DUF5-L418W demonstrated potent proteolytic activity to RAS proteins. FIG. 9A shows a gel image of the Western blotting analysis for cell lysates prepared from H358 cells that had been incubated with various concentrations of PA3MzHER2-MMP / LFn-DUF5-L418W for 3 hours. FIG. 9B shows a gel image of the Western blotting analysis for cell lysates prepared from MBD-MB-231 cells that had been incubated under the same condition as in FIG. 9A. FIG. 9C shows a gel image of the Western blotting analysis for cell lysates prepared from HT1080 cells that had been incubated under the same condition as in FIG. 9A. The antibodies used were anti-pan-RAS, anti-KRAS or anti-NRAS, anti-phosphor-ERK, and anti-tubulin antibodies.

[0123] FIGs. 10A-10E: Cytotoxicity comparison of PA3MzHER2-MMP / LFn-DUF5-L418W versus small molecule KRAS inhibitors to cancer cells. FIG. 10A shows viabilities of H358 cells (mean ± SD) that had been incubated with various concentrations of LFn-DUF5-L418W (with PA3MzHER2-MMP = 5 nM), Sotorasib, Adagrasib, MRTX1133, and BI-2865 for 48 hours. FIG. 10B shows viabilities of MiaPaCa-2 cells (mean ± SD) that had been incubated under the same conditions as in FIG. 10A. FIG. 10C shows viabilities of AsPc-1 cells (mean ± SD) that had been incubated under the same conditions as in FIG. 10A.

[0124] FIG. 10D shows viabilities of MDA-MB-231 cells (mean ± SD) that had been incubated under the same conditions as in FIG. 10A. FIG. 10E shows viabilities of HT1080 cells (mean ± SD) that had been incubated under the same conditions as in FIG. 10 A. Cell8123-112570-02

[0125] viability was assessed by an MTT assay. The toxin exhibited much higher cytotoxicity than trastuzumab and its drug conjugates in all cancer cells tested.

[0126] FIGs. 11A-11B: PA3MzHER2-MMP / LFn-DUF5-L418W and PA3MzHER2-MMP / FP59 demonstrated potent antitumor activity against human fibrosarcoma HT1080 xenografts. FIG. 11A shows tumor volume of HT1080 tumors (mean ± SD) in mice that were being treated with PA3MzHER2-MMP (30 pg) combined with LFn-DUF5-L418W (10 pg) or FP59 (5 pg) twice a week at days indicated by the arrows. FIG. 11B shows body weight (mean ± SD) of these mice. 1 out of 15 tumors were eliminated after 4 toxin doses. *: p < 0.01.

[0127] FIGs. 12A-12B: PA3MzHER2-MMP / LFn-DUF5-L418W demonstrated potent antitumor activity against human lung carcinoma H358 xenografts. FIG. 12A shows tumor volume of H358 tumors (mean ± SD) in mice that were being treated with PA3MzHER2-MMP / LFn-DUF5-L418W (30 pg / 10 pg) twice a week at days as indicated by the arrows. FIG. 12B shows body weight (mean ± SD) of these mice. *: p < 0.001.

[0128] FIGs. 13A-13B: Estimation of the therapeutic index of PA3MzHER2-MMP / LFn-DUF5-L418W. FIG. 13A shows tumor volume of HT1080 tumors (mean ± SD) in mice that were being treated with low doses (10 pg / 3.3 pg) or high doses (90 pg / 30 pg) of PA3MzHER2-MMP / LFn-DUF5-L418W (I.P.) at days as indicated by the arrows. FIG. 13B shows body weight (mean ± SD) of these mice. While both doses demonstrated potent comparable antitumor activity (A), no signs of toxicity was observed even in mice treated with the high doses (B). The therapeutic index of the toxin is estimated to be at least 9 (90 pg / 30 pg -? 10 pg / 3.3 pg = 9). The toxin eliminated 10 out of 11 tumors after 3 doses.

[0129] FIGs. 14A-14B: Antitumor and side effects of LFn-DUF5 / IC3-PA. FIG. 14A shows tumor volume (mean ± SE) of LLC(KrasG12C) tumor-bearing mice that were being treated (I.P., red arrows) with LFn-DUF5 / IC3-PA (0.4 pg / 20 pg), or PBS. FIG. 14B shows the amount of abdominal fluid (mean ± SE) withdrawn from intraperitoneal cavities of the four mice with abdominal edema after 3 doses of LFn-DUF5 / IC3-PA.

[0130] FIG. 15: In vitro activity assay of LFn-DUF5. HCT116 cell lysates (50 pL) were incubated with LFn-DUF5 containing LB culture supernatant from LFn-DUF5 expressing BH500 strains, at various dilutions, for 30 min. Samples were then assessed by Western blotting using an anti-pan-RAS or anti-RAPl antibody. Even 0.01 pL of LFn-DUF5 containing supernatant cleaved most RAS and RAP1 proteins within 30 min.

[0131] FIG. 16: DUF5 crystal structure.8123-112570-02

[0132] FIG. 17: A comparison of the substrate sequences around the DUF5 cleavage site in human RAP1 and human RAS proteins. The substrate sequences of RAP1A, RAP1B, HRAS, KRAS, and NRAS shown in the figure are set forth as SEQ ID NOs: 34, 35, 36, 37, and 38, respectively.

[0133] FIG. 18: LFn-DUF5 L418W is highly selective towards RAS over RAPE Human lung carcinoma H358 cells were incubated with various concentrations of LFn-DUF5 or LFn-DUF5 L418W in the presence of PA3MzHER2-MMP (1000 ng / mL) for 3 hours. Then, cell lysates were prepared and analyzed by Western blotting using an anti -pan RAS, anti -RAP 1, anti-RhoA, anti-Racl, or anti-CDC42 antibody. Notably, LFn-DUF5 L418W exhibited proteolytic activity to all RAS proteins, but not to RAP1 or other related small GTPase family members (RAC1, RhoA, and CDC42).

[0134] FIG. 19: Reduced edema-causing activity of LFn-DUF5 L418W. Mice were treated (I.P.) with 0.2 pg / 20 pg of LFn-DUF5 / IC3-PA or LFn-DUF5 L418W / IC3-PA every other day for three times. The amount of abdominal fluid (mean ± SE) withdrawn from intraperitoneal cavities of these mice is shown.

[0135] FIGs. 20A-20B: Antitumor and minimal side effects of PA3MzHER2-MMP / LFn-DUF5 L418W. Human pancreatic AsPc-1 tumor-bearing mice were treated with PA3MzHER2-MMP / LFn-DUF5 L418W (30 pg / 10 pg) at days indicated by the arrows. After 3 weeks of treatment, all tumors disappeared. Although 6 / 9 tumors grew back when treatment was paused, these tumors remained responsive to the following treatment and were eliminated (FIG. 20A). No undesired toxicity was observed throughout the course of treatment (FIG. 20B).

[0136] FIGs. 21A-21B: Potent antitumor activity of PA3MzHER2-MMP / LFn-DUF5-L418W. Human pancreatic carcinoma AsPc-1 (KRASG12D / G12D) tumor-bearing mice were treated with i) PBS; ii) RMC-6236 (25 mg / kg, oral gavage, 5 days per week), then switched to PA3MzHER2-MMP / LFn-DUF5-L418W at day 24 (30 pg / 10 pg, intravenously (I.V.) at days indicated by the downward-pointing arrows); iii) PA3MzHER2-MMP / LFn-DUF5-L418W (30 pg / 10 pg, intraperitoneally (I.P.), three times per week); or iv) PA3MzHER2-MMP / LFn-DUF5-L418W (30 pg / 10 pg, l.V. at days indicated by the upward-pointing arrows). FIG. 21 A is a plot that shows tumor volume over the days after the first treatment. FIG. 21B are waterfall plots of tumor volume change (%) for the mice in each treatment group in FIG. 21 A. No apparent adverse effects were observed in any treatment group. Tumor size is presented as mean ± SE. As shown, complete tumor regression was achieved in all mice treated with PA3MzHER2-MMP / LFn-DUF5-L418W by day 24, and8123-112570-02

[0137] tumors that had developed resistance to RMC-6236 responded strongly to PA3MzHER2-MMP / LFn-DUF5-L418W.

[0138] DETAILED DESCRIPTION

[0139] I. Terms

[0140] Unless otherwise noted, technical terms are used according to conventional usage. Definitions of many common terms in molecular biology may be found in Krebs el al. (eds.), Lewin’s genes XII, published by Jones & Bartlett Learning, 2017. As used herein, the singular forms “a,” “an,” and “the,” refer to both the singular as well as plural, unless the context clearly indicates otherwise. As used herein, the term “comprises” means “includes.” It is further to be understood that any and all base sizes or amino acid sizes, and all molecular weight or molecular mass values, given for nucleic acids or polypeptides are approximate, and are provided for descriptive purposes, unless otherwise indicated. Although many methods and materials similar or equivalent to those described herein can be used, particular suitable methods and materials are described herein. In case of conflict, the present specification, including explanations of terms, will control. In addition, the materials, methods, and examples are illustrative only and not intended to be limiting.

[0141] To facilitate review of the various aspects, the following explanations of terms are provided.

[0142] About: Unless context indicated otherwise, “about” refers to plus or minus 5% of a reference value. For example, “about” 100 refers to 95 to 105.

[0143] Administration: The introduction of a composition into a subject by a chosen route. Administration can be local or systemic. For example, if the chosen route is intravenous, the composition is administered by introducing the composition into a vein of the subject.

[0144] Exemplary routes of administration include, but are not limited to, oral, injection (such as subcutaneous, intramuscular, intradermal, intraperitoneal, and intravenous), sublingual, rectal, transdermal (for example, topical), intranasal, vaginal, and inhalation routes.

[0145] Affibody: A small protein that binds to a target protein or peptide (e.g., EGFR, HER2, HER3, etc.) with specificity and high affinity, and is therefore a member of the family of antibody mimetics. In some aspects, an affibody is based on the small (7 kDa) protein Z, an engineered IgG-binding domain of the protein A found on the cell surface

[0146] of Staphylococcus aureus, and is usually selected from phage display libraries obtained by randomizing 13 residues exposed on the first two a-helices of the three-helix bundle. In some examples, an affibody comprises three alpha helices with 58-60 amino acids, and8123-112570-02

[0147] optionally lacks disulfide bridges. In some examples, affibodies used herein include at least 80% sequence identity to any of SEQ ID NOs: 5-7, or include or consist of any of SEQ ID NOs: 5-7, and specifically bind to HER2, HER3, or EGFR.

[0148] Anthrax toxin: A three -protein exotoxin secreted by virulent strains of Bacillus anthracis, the causative agent of anthrax. Anthrax toxin is composed of a cell-binding protein, known as protective antigen (PA), and two enzyme components, known as edema factor (EF) and lethal factor (LF). These three protein components act together to exert their physiological effects. As used herein, anthrax toxin may refer to any one, any two, or all three of the components. Anthrax toxin may be considered as AB toxin, in that each component, alone, is nontoxic. Toxic symptoms are only observed when PA and LF or EF are present together. The co-presence of PA and EF causes edema, and the co-presence of PA and LF is lethal.

[0149] PA mediates attachment to host cells by binding host cell receptors, such as tumor endothelial marker 8 (TEM8, also called ANTXR1) and capillary morphogenesis protein 2 (CMG2, also called ANTXR2). This process starts when the 83-kDa form of PA, called PA83, binds to a receptor. Following host cell surface attachment, PA is cleaved by furin to generate the PA63 form, which constitutes the mature form of the protein that oligomerizes and forms a pore to translocate the enzymatic toxin components EF and / or LF into the host cytosol.

[0150] As used herein, PA includes naturally occurring and engineered PA. In some aspects, PA is engineered to reduce or abolish its activity towards its natural toxin receptors, for example, by including one or more of D683A, L685E, and Y688K mutations (numbering according to SEQ ID NO: 1). CMG2 is the major PA receptor on cancer cells and tumor stromal cells, while healthy tissues, including kidneys and the brain, express high levels of TEM8. Therefore, in some aspects, PA is engineered to preferentially or exclusively bind CMG2 (or reduce or abolish its activity towards TEM8), for example, by including one or more of I656Q, Y681 A, and L687A mutations (numbering according to SEQ ID NO: 1), thereby sparing toxicity to healthy tissues that express the TEM8 receptor.

[0151] In some aspects, PA is engineered herein to increase its specificity toward tumor cells, by, for example, replacing the furin cleavage site to MMP cleavage site or uPA cleavage site; fusing PA to a tumor targeting component (e.g., EGFR, HER2, or HER3 affibody or antibody); mutating PA so that it is not recognized by its native receptors (e.g., the 3M mutations or the I656Q mutation); and / or mutating a pair of PAs so that an effector protein (e.g., LF) can only bind to the hetero-oligomer formed by both of the PAs, but not to the8123-112570-02

[0152] homo-oligomer formed by only one of the PAs (e.g., the ICPA mutations). PA may also be used to refer to the fusion protein formed by PA and the tumor targeting component. In some aspects, PA includes at least 80% sequence identity to any of SEQ ID NOs: 1-2 and 10-28, or include or consist of any of SEQ ID NOs: 1-2 and 10-28.

[0153] LF acts as a protease that cleaves the N-terminal of most dual specificity mitogen-activated protein kinase kinases (MAPKKs, also called MEKs): cleavage occurs within the N-terminal proline-rich region preceding the kinase domain, thus disrupting a sequence involved in directing specific protein-protein interactions necessary for the assembly of signaling complexes. LF binds to PA63 through its N-terminal domain (-250 residues). After binding, PA undergoes endocytosis, and in the acidic environment of the endosome, it forms a pore that translocates LF into the cytosol. LFn refers to the domain of LF that binds to PA. In some aspects, LFn includes at least 80% sequence identity to SEQ ID NO: 31. In some examples, LFn includes or consists of SEQ ID NO: 31.

[0154] Antibody: A protein, e.g., immunoglobulin, antigen-binding fragment, or derivative thereof, that specifically binds to an antigen, such as a protein, such as EGFR, HER2, HER3, etc. The term antibody is used herein in the broadest sense and encompasses various antibody structures, including but not limited to monoclonal antibodies, polyclonal antibodies, multispecific antibodies (e.g., bispecific antibodies), and antibody fragments, so long as they exhibit the desired antigen-binding activity. Non-limiting examples of antibodies include, for example, intact immunoglobulins and variants and fragments thereof that retain binding affinity for the antigen. Examples of antibody fragments include but are not limited to Fv, Fab, Fab', Fab'-SH, F(ab')2; diabodies; linear antibodies; single -chain antibody molecules (e.g. scFv); and multispecific antibodies formed from antibody fragments. Antibody fragments include antigen binding fragments either produced by the modification of whole antibodies or those synthesized de novo using recombinant DNA methodologies (see, e.g., Kontermann and Dubel (Ed), Antibody Engineering, Vols. 1-2, 2ndEd., Springer Press, 2010). Light and heavy chain variable regions each has a “framework” region interrupted by three hypervariable regions, also called “complementarity-determining regions” or “CDRs” (see, e.g., Kabat et al., Sequences of Proteins of Immunological Interest, U.S. Department of Health and Human Services, 1991). The framework region of an antibody, that is the combined framework regions of the constituent light and heavy chains, serves to position and align the CDRs in three-dimensional space. The CDRs are primarily responsible for binding to an epitope of an antigen.8123-112570-02

[0155] Conservative substitution: A substitution of one amino acid residue in a protein sequence for a different amino acid residue having similar biochemical properties. Typically, conservative substitutions have little to no impact on the activity of a resulting polypeptide. For example, a HER2 affibody variant including one or more conservative substitutions (for example no less than 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10, and / or no more than 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20, 25, 30, 35, 40, 45, or 50 substitutions, for example 2 to 50, 2 to 30, 2 to 20, 2 to 15, 2 to 10, 2 to 5 substitutions) retains the structure and activity of the original protein. A polypeptide can be produced to contain one or more conservative substitutions by manipulating the nucleotide sequence that encodes that polypeptide using, for example, standard procedures such as site-directed mutagenesis or PCR.

[0156] Examples of conservative substitutions are shown below.

[0157] Original Residue Conservative Substitutions

[0158] Ala Ser

[0159] Arg Lys

[0160] Asn Gin, His

[0161] Asp Glu

[0162] Cys Ser

[0163] Gin Asn

[0164] Glu Asp

[0165] His Asn; Gin

[0166] He Leu, Vai

[0167] Leu He; Vai

[0168] Lys Arg; Gin; Glu

[0169] Met Leu; He

[0170] Phe Met; Leu; Tyr

[0171] Ser Thr

[0172] Thr Ser

[0173] Trp Tyr

[0174] Tyr Trp; Phe

[0175] Vai He; Leu

[0176] Conservative substitutions generally maintain (a) the structure of the polypeptide backbone in the area of the substitution, for example, as a sheet or helical conformation,8123-112570-02

[0177] (b) the charge or hydrophobicity of the molecule at the target site, or (c) the bulk of the side chain.

[0178] The substitutions which in general are expected to produce the greatest changes in protein properties will be non-conservative, for instance changes in which (a) a hydrophilic residue, for example, seryl or threonyl, is substituted by a hydrophobic residue, for example, leucyl, isoleucyl, phenylalanyl, valyl or alanyl; (b) a cysteine or proline is substituted by any other residue; (c) a residue having an electropositive side chain, for example, lysyl, arginyl, or histadyl, is substituted by an electronegative residue, for example, glutamyl or aspartyl; or (d) a residue having a bulky side chain, for example, phenylalanine, is substituted by one not having a side chain, for example, glycine.

[0179] In some aspects, the proteins provided herein, including the effector proteins (e.g., DUF5) and the delivery proteins (e.g., PA3MzHER2-Ll) can include one or more (e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or more) substitutions, such as conservative substitutions.

[0180] Control: A reference standard, which may include concurrent control, historical control, and / or established benchmark. In some examples, control refers to a reference standard used to evaluate the effect of a drug. The control may be (i) a concurrent control obtained from subjects or samples tested under similar experimental conditions as the treatment group except for receiving the treatment (e.g., drug), (ii) a historical control based on previously collected data, or (iii) an established standard recognized in the field, such as a predefined clinical metrics or score. The control may be obtained from subjects or samples that do not receive the treatment, receive a placebo, or receive a standard treatment with known efficacy.

[0181] DUF5: Domain of unknown function 5 (DUF5), a specific endopeptidase that cleaves within the Switch 1 region (e.g., between Y32 and D33) of RAS and RAP1. As used herein, DUF5 includes any DUF5 found in Vibrio vulnificus, and any orthologs thereof, such as those found in Aeromonas hydrophila, Vibrio splendidus, Xenorhabdus nematophila, Yersinia kristensii, Photorhabdus luminescens, and Photorhabdus asymbiotica, and any naturally occurring or engineered variants thereof. Native DUF5 may be present as an effector domain of multifunctional-autoprocessing repeats-in-toxin (MARTX), such as in Vibrio vulnificus, or as a stand-alone protein such as in Photorhabdus asymbiotica. MARTX toxins are a family of large bacterial protein toxins with conserved repeat regions at the N and C termini that are predicted to transfer effector domains located between the repeats across the eukaryotic cell plasma membrane. These domains are then excised from the holotoxin by autoprocessing and released to the eukaryotic cell cytosol.8123-112570-02

[0182] DUF5 can inactivate both wild-type and mutant RAS (KRAS. NRAS, HRAS) and RAP1 proteins. An exemplary wild-type DUF 5 amino acid sequence from Vibrio vulnificus is provided in SEQ ID NO: 29, which is amino acid 3581-4089 of MARTX (renumbered to 1-509 herein). Another exemplary DUF5 amino acid sequence from Aeromonas hydrophila is provided in GENBANK® Accession No. WP_011705266.1 (aa 3069-3570), April 17, 2025. incorporated by reference herein.

[0183] Provided herein are RAS-selective DUF5. that specifically targets RAS over RAPE In some examples, the RAS-selective DUF5 comprises at least 60% sequence identity to SEQ ID NO: 30, and the L418W substitution of SEQ ID NO: 30.

[0184] EGFR and EGFR binding domain: Epidermal growth factor receptor (EGFR), also known as receptor tyrosine-protein kinase ErbB-1, is a cell surface receptor that belongs to the ErbB family of receptor tyrosine kinases. It binds to epidermal growth factor (EGF) and related ligands, initiating a cascade of signaling pathways involved in cell proliferation, survival, migration, and differentiation. When EGFR is activated by ligand binding, it undergoes dimerization (either homodimerization or heterodimerization with other ErbB family members), leading to autophosphorylation of tyrosine residues in its intracellular domain. This triggers downstream signaling pathways, such as the MAPK and PI3K / AKT pathways, which regulate various cellular processes. Exemplary EGFR protein sequences are provided in GENBANK® Accession Nos. NP_005219.2. (Homo sapiens) and NP_997538.1 (Mus musculus), March 30, 2025, incorporated by reference herein. Exemplary EGFR mRNA sequences are provided in GENBANK® Accession Nos. NM_005228.5 (Homo sapiens) and NM_207655.2(Mn.s' musculus), March 30, 2025, incorporated by reference herein.

[0185] An EGFR binding domain is a protein domain that specifically binds EGFR, including EGFR-specific affibodies and antibodies.

[0186] HER2 and HER2 binding domain: Human epidermal growth factor receptor 2 (HER2), also known as receptor tyrosine-protein kinase ErbB-2, is a cell surface receptor that belongs to the ErbB family of receptor tyrosine kinases. Unlike other members of the family, HER2 does not directly bind a ligand. HER2 activation results from heterodimerization with another ErbB member or by homodimerization when HER2 concentration is high. HER2 plays a crucial role in cell proliferation, survival, and differentiation. Aberrant HER2 signaling is implicated in various cancer, particularly in breast and gastric cancers, where HER2 overexpression or amplification correlates with aggressive tumor behavior and poor prognosis. Exemplary HER2 protein sequences are provided in GENBANK® Accession8123-112570-02

[0187] Nos. NP_004439.2 (Homo sapiens) and NP_001003817.1 (Mas musculus), March 30, 2025, incorporated by reference herein. Exemplary HER2 mRNA sequences are provided in GENBANK® Accession Nos. NM_004448.4 (Homo sapiens) and NM_001003817.1 (Mus musculus), March 30, 2025, incorporated by reference herein.

[0188] A HER2 binding domain is a protein domain that specifically binds HER2, including HER2-specific affibodies and antibodies.

[0189] HER3 and HER3 binding domain: Human epidermal growth factor receptor 3 (HER3), also known as receptor tyrosine-protein kinase ErbB-3, is a cell surface receptor that belongs to the ErbB family of receptor tyrosine kinases. HER3 has been shown to bind the ligands heregulin and NRG-2. Ligand binding causes a change in conformation that allows for dimerization, phosphorylation, and activation of signal transduction. HER3 can heterodimerize with any of the other three ErbB family members. Unlike the other family members which are activated through autophosphorylation upon ligand binding, HER3 was found to be kinase impaired. The HER2-HER3 dimer is considered the most active of the possible ErbB dimers. This heterodimer conformation allows the signaling complex to activate multiple pathways including the MAPK, PI3K / Akt, and PLCy. Exemplary HER3 protein sequences are provided in GENBANK® Accession Nos. NP_001973.2 (Homo sapiens) and NP_034283.1 (Mus musculus), March 30, 2025, incorporated by reference herein. Exemplary HER3 mRNA sequences are provided in GENBANK® Accession Nos. NM_001982.4 (Homo sapiens) and NM_010153.2 (Mus musculus), March 30, 2025, incorporated by reference herein.

[0190] A HER3 binding domain is a protein domain that specifically binds HER3, including HER3-specific affibodies and antibodies.

[0191] Heterologous: Originating from a different genetic source or species. Isolated or purified: An isolated or purified biological substance (such as a nucleic acid, protein, protein complex, or particle) has been substantially separated, produced apart from, or purified away from other components in a preparation, or other biological components (e.g., nuclei acids or proteins) in the organism in which the substance occurs. An isolated or purified biological substance can be obtained through isolation or purification from samples obtained from an organism; recombinant expression or production in host cells followed by purification, and chemical synthesis followed by purification. The term “isolated” or “purified” does not require absolute purity; rather, it is intended as a relative term, for example, referring to being more enriched with (or having a higher concentration of) the substance, compared to a crude preparation or a natural environment from which the substance is purified. In some aspects, a8123-112570-02

[0192] biological substance is purified if the substance represents at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, at least 99.5%, or greater, of the total moles of non-solvent substances in a preparation.

[0193] Matrix metalloproteinase (MMP) cleavage site: An amino acid sequence that is cleavable by a matrix metalloproteinase (MMP). MMPs are a family of calcium-dependent zinc-containing endopeptidases, including MMP-1, MMP-2, MMP-3, MMP-7, MMP-8, MMP-9, MMP-10, MMP-11, MMP-12, MMP-13, MMP-14, MMP-15, MMP-16, MMP-17, etc., which are responsible fortissue remodeling and degradation of extracellular matrix (ECM), including collagens, elastin, gelatin, matrix glycoproteins, and proteoglycan.

[0194] Elevated levels of MMPs are observed in different types of cancers and correlate with the enhanced proliferation and growth of tumors. In some examples, an MMP cleavage site can be cleaved by MMP-2, MMP-9, and / or MMP-14. Exemplary MMP cleavage site sequences include GPLGMLSQ (SEQ ID NO: 3), GPQGIWGQ (SEQ ID NO: 34), GPQGIAGQ (SEQ ID NO: 35), VPMSMRGG (SEQ ID NO: 36), QPQGLAK (SEQ ID NO: 37), GPLGLSGK(SEQ ID NO: 38), PLGLA (SEQ ID NO: 39), etc.

[0195] Operably linked: A first nucleic acid is operably linked to a second nucleic acid when the first nucleic acid is placed in a functional relationship with the second nucleic acid. In some examples, operably linked DNA sequences are contiguous and, where necessary to join two protein coding regions, in the same reading frame. In some examples, operably linked nucleic acids include a first nucleic acid contiguous with the 5' or 3' end of a second nucleic acid. In some examples, one or more regulatory elements (e.g., promoters, enhancers, etc.) are operably linked to a nucleic acid sequence when they control the expression of the sequence.

[0196] Pharmaceutically acceptable carrier: The pharmaceutically acceptable carriers (vehicles) useful in this disclosure are conventional. Remington: The Science and Practice of Pharmacy, 23rdEd., New York: Academic Press, 2020, describes compositions and formulations suitable for pharmaceutical delivery of one or more therapeutic compositions.

[0197] In general, the nature of the carrier will depend on the particular mode of administration being employed. For instance, parenteral formulations usually comprise injectable fluids that include pharmaceutically and physiologically acceptable fluids such as water, physiological saline, balanced salt solutions, aqueous dextrose, glycerol or the like as a vehicle. For solid compositions (for example, powder, pill, tablet, or capsule forms), conventional non-toxic solid carriers can include, for example, pharmaceutical grades of mannitol, lactose, starch, or magnesium stearate. In addition to biologically-neutral carriers,8123-112570-02

[0198] pharmaceutical compositions to be administered can contain minor amounts of non-toxic auxiliary substances, such as wetting or emulsifying agents, preservatives, and pH buffering agents and the like, for example sodium acetate or sorbitan monolaurate.

[0199] Peptide or protein: A polymer formed by two or more amino acids, regardless of length or post-translational modification (e.g., glycosylation or phosphorylation). Peptide, polypeptide, and protein are used interchangeably herein.

[0200] Protein complex: A stable or transient assembly of two or more proteins that interact through non-covalent or covalent bonds, including homomeric complexes (formed by identical proteins) and heteromeric complexes (formed by different proteins).

[0201] RAS: Rat sarcoma virus (RAS) proteins, including KRAS, NRAS, and HRAS proteins. RAS proteins represent a subgroup of the large family of small GTPases, which is membrane localized. KRAS, NRAS, and HRAS are the three RAS genes found in humans. Like other small GTPases, RAS proteins function as molecular switches that cycle between an inactive, GDP-bound and an active, GTP-bound state to regulate cell proliferation, differentiation, migration, and apoptosis. Overactive RAS signaling can ultimately lead to cancer. RAS proteins contain six beta strands and five alpha helices, forming two domains: a G domain of 166 amino acids that binds guanosine nucleotides, and a C-terminal membrane targeting region (CAAX-COOH, also known as CAAX box) that is lipid-modified. As used herein, RAS includes any normal or mutated form, such as those found in humans, such as hyperactivated RAS, e.g., persistently GTP-bound and chronically activated independent of external stimuli. In some examples, the RAS mutants have a mutation at residues G12 and G13 in the P-loop and at Q61 in the Switch II region. In some examples, the RAS mutants are recognizable and cleavable by DUF5-L418W. In some examples, the RAS mutants do not have a mutation at the site that is recognized and cleaved by DUF5-L418W.

[0202] RAS-selective: Referring to a protease, e.g., DUF5, that preferentially targets and cleaves RAS (e.g., normal or mutated KRAS, NRAS, and / or HRAS) relative to RAP1 (e.g., RAP1 A and / or RAP1B). In some aspects, RAS-selective means that the protease cleaves RAS with a catalytic efficiency that is greater than the catalytic efficiency for cleaving RAP1, under the same defined conditions. In some examples, the catalytic efficiency for cleaving RAS is at least about 5 times, e.g., at least about 10 times, at least about 15 times, at least about 20 times, at least about 25 times, at least about 30 times, at least about 35 times, at least about 40 times, at least about 45 times, at least about 50 times, at least about 75 times, at least about 100 times, at least about 150 times, at least about 200 times, at least about 500 times, or8123-112570-02

[0203] at least about 1000 times greater than the catalytic efficiency for cleaving RAP1. In some examples, cleavage of RAP1 is substantially reduced so that it is undetectable.

[0204] A protease’s (e.g., DUF5’s) catalytic efficiency can be determined by measuring the formation of one or more cleaved products generated upon proteolysis of a suitable substrate (e.g., RAS or RAP1) under controlled conditions. In general, a protease is contacted with a substrate comprising a cleavage site recognized by the protease in a reaction buffer under conditions permissive for enzymatic activity. The reaction is allowed to proceed for a defined period of time, and the formation of cleaved product is measured at one or more time points. Cleaved product formation may be quantified using a variety of analytical methods, including but not limited to fluorescence-based detection (e.g., Anorogenic substrates or Auorophore release), absorbance or colorimetric assays (e.g., chromogenic substrates), chromatographic techniques such as high-performance liquid chromatography (HPLC) or ultra-performance liquid chromatography (UPLC), mass spectrometry-based approaches (e.g., LC-MS or MALDI-MS), electrophoretic methods (e.g., SDS-PAGE, capillary electrophoresis, or microAuidic systems) with densitometric or signal-based quantification, and immunological assays (e.g., ELISA, western blotting, or antibody-based detection of cleavage products).

[0205] RAP1: RAS-related protein-1 (RAP1). RAP1 is a small GTPase that cycles between the GDP- and GTP-bound state. There are two isoforms of the RAP1 protein, each encoded by a separate gene, RAP1A and RAP1B. RAP1 is involved in cell adhesion and cell junction formation. In the GTP-bound state, RAP1 is biologically active and is capable of binding with high affinity to several downstream signaling molecules, including Raf-1, B-Raf, and RalGDS. In some aspects, RAP1 refers to RAP1 that functions normally, such as those found in human.

[0206] Recombinant: Of or resulting from new combinations of genetic material. A recombinant protein is a protein produced by the use of recombinant DNA technology, which involves the combination of genetic material from different sources to create a new (non-naturally occurring) DNA molecule, which is then introduced into a host organism (such as bacteria, yeast, or mammalian cells) to produce the desired protein.

[0207] Sequence identity: The identity between two or more nucleic acid sequences, or two or more amino acid sequences, is expressed in terms of the percentage identity between the sequences. Sequence identity can be measured in terms of percentage identity; the higher the percentage, the more identical the sequences. Homologs, orthologs, or variants of a protein or nucleic acid will possess a relatively high degree of sequence identity when aligned using8123-112570-02

[0208] standard methods. Similar functions can be inferred based on homology, which can in turn be inferred from sequence identity.

[0209] Any suitable method may be used to align sequences for comparison. Non-limiting examples of programs and alignment algorithms are described in: Smith and Waterman, Adv. Appl. Math. 2(4):482-489, 1981; Needleman and Wunsch, J. Mol. Biol. 48(3):443-453, 1970; Pearson and Lipman, Proc. Natl. Acad. Sci. U.S.A. 85(8):2444-2448, 1988; Higgins and Sharp, Gene, 73(l):237-244, 1988; Higgins and Sharp, Bioinformatics, 5(2): 151-3. 1989; Coipet, Nucleic Acids Res. 16(22):10881-10890, 1988; Huang et al. Bioinformatics, 8(2): 155-165, 1992; and Pearson, Methods Mol. Biol. 24:307-331, 1994., Altschul et al,, J. Mol. Biol, 215(3):403-410, 1990, presents a detailed consideration of sequence alignment methods and homology calculations. The NCBI Basic Local Alignment Search Tool (BLAST) (Altschul etal., .J. Mol, Biol. 215(3):403-410, 1990) is available from several sources, including the National Center for Biological Information and on the Internet, for use in connection with the sequence analysis programs blastp, blastn, blastx, tblastn, and tblastx. Blastn is used to compare nucleic acid sequences, while blastp is used to compare amino acid sequences. Additional information can be found at the NCBI web site.

[0210] Generally, once two sequences are aligned, the number of matches is determined by counting the number of positions where an identical nucleotide or amino acid residue is present in both sequences. The percent sequence identity between the two sequences is determined by dividing the number of matches either by the length of the sequence set forth in the identified sequence, or by an articulated length (such as 100 consecutive nucleotides or amino acid residues from a sequence set forth in an identified sequence), followed by multiplying the resulting value by 100.

[0211] Homologs and variants of a protein are typically characterized by possession of at least about 75%, for example, at least about 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% sequence identity counted over the full-length alignment with the amino acid sequence of interest.

[0212] Subject: Living multi-cellular vertebrate organisms, a category that includes human and non-human mammals, such as non-human primates, rats, mice, dogs, cats, horses, sheep, cows, rabbits, and pigs. In some examples, a subject is a human. In some examples, a subject is a veterinary subject.

[0213] Therapeutically effective amount: A quantity of an agent that, when administered to a subject (e.g., alone, or in combination with one or more other agents), is sufficient to achieve a desired biological or therapeutic effect (such as inhibiting or treating a tumor).8123-112570-02

[0214] When administered to a subject, a dosage will generally be used that will achieve target tissue concentrations that has been shown to achieve a desired in vitro effect. Ideally, a therapeutically effective amount provides a therapeutic effect without causing a substantial cytotoxic effect in the subject.

[0215] A therapeutically effective amount of an effector protein as described herein that is administered to a human or veterinary subject will vary depending upon a number of factors associated with that subject, for example the overall health of the subject. A therapeutically effective amount can be determined by varying the dosage and measuring the resulting therapeutic response, such as reduction in tumor burden, or indication of a tumor response. The disclosed agents can be administered in a single dose, or in several doses, as needed to obtain the desired response. However, the therapeutically effective amount can be dependent on the source applied, the subject being treated, the severity and type of the condition being treated, and the manner of administration.

[0216] Treating: Refers to applying a therapeutic intervention (e.g., administration of DUF5-L418W together with a delivery protein) that produces or is expected to produce a desired effect to a subject in need thereof. The desired effect includes ameliorating one or more symptoms of a disease or cancer, inhibiting or slowing the development of a disease or cancer (e.g., stable disease, or improving progression-free survival or disease-free survival), reducing the severity of the disease or cancer (e.g., partial response), completely treating or eliminating the disease or cancer (e.g., complete response), improving survival, etc.

[0217] Tumor, Neoplasia, and Cancer: A neoplasm is an abnormal growth of tissue or cells that results from excessive cell division. Neoplastic growth can produce a tumor. A neoplasm that does not metastasize or invade surrounding tissue is referred to as “benign.” A neoplasm that does metastasize or invade the surrounding tissue is referred to as “malignant.” Cancer is neoplastic growth that is malignant or has the potential to become malignant. The amount of a tumor in an individual is the “tumor burden” which can be measured as the number, volume, or weight of the tumor.

[0218] Non-limiting examples of hematological cancers include leukemias, including acute leukemias (such as 1 lq23-positive acute leukemia, acute lymphocytic leukemia, acute myelocytic leukemia, acute myelogenous leukemia and myeloblastic, promyelocytic, myelomonocytic, monocytic and erythroleukemia), chronic leukemias (such as chronic myelocytic (granulocytic) leukemia, chronic myelogenous leukemia, and chronic lymphocytic leukemia), polycythemia vera, lymphoma, Hodgkin's disease, non-Hodgkin's lymphoma (indolent and high grade forms), multiple myeloma, Waldenstrom's8123-112570-02

[0219] macroglobulinemia, heavy chain disease, myelodysplastic syndrome, hairy cell leukemia and myelodysplasia. In specific non-limiting examples, the lymphoid malignancy can be adult T cell leukemia, cutaneous T cell lymphoma, anaplastic large cell lymphoma, Hodgkin’s lymphoma, or a diffuse large B cell lymphoma.

[0220] Non-limiting examples of solid cancers, such as sarcomas and carcinomas, include fibrosarcoma, myxosarcoma, liposarcoma, chondrosarcoma, osteogenic sarcoma, and other sarcomas, synovioma, mesothelioma, Ewing's tumor, leiomyosarcoma, rhabdomyosarcoma, colon carcinoma, lymphoid malignancy, pancreatic cancer, breast cancer (including basal breast carcinoma, ductal carcinoma and lobular breast carcinoma), lung cancers, ovarian cancer, prostate cancer, hepatocellular carcinoma, squamous cell carcinoma, basal cell carcinoma, adenocarcinoma, sweat gland carcinoma, medullary thyroid carcinoma, papillary thyroid carcinoma, pheochromocytomas sebaceous gland carcinoma, papillary carcinoma, papillary adenocarcinomas, medullary carcinoma, bronchogenic carcinoma, renal cell carcinoma, hepatoma, bile duct carcinoma, choriocarcinoma, Wilms' tumor, cervical cancer, testicular tumor, seminoma, bladder carcinoma, and CNS tumors (such as a glioma, astrocytoma, medulloblastoma, craniopharyrgioma, ependymoma, pinealoma, hemangioblastoma, acoustic neuroma, oligodendroglioma, meningioma, melanoma, neuroblastoma and retinoblastoma), and those provided in Table 2. In several examples, a tumor is breast, ovarian, pancreatic, gastric or esophageal cancer.

[0221] Urokinase-type plasminogen activator (uPA) cleavage site: An amino acid sequence that is cleavable by uPA (also known as urokinase). uPA is a serine protease present in humans and other animals, e.g., in urine, blood, and the extracellular matrix of many tissues. The primary physiological substrate of uPA is plasminogen, which is an inactive form (zymogen) of the serine protease plasmin. Activation of plasmin triggers a proteolytic cascade that, depending on the physiological environment, participates in thrombolysis or extracellular matrix degradation. Elevated uPA expression level is found to be correlated with tumor malignancy. uPA breakdown of the extracellular matrix is crucial for initiating the angiogenesis which is associated with cancer growth. uPA level is elevated in breast cancer tissue, which correlates with poor prognosis in breast cancer patients.

[0222] Exemplary uPA cleavage site sequences include SGRSA (SEQ ID NO: 40), PGSGRSA (SEQ ID NO: 4), GGSGRSANAILE (SEQ ID NO: 41), KKSPGRVVGGSVAH (SEQ ID NO: 42), CPGRVVGGC (SEQ ID NO: 43), etc.

[0223] Vector: An entity containing a nucleic acid molecule of interest and permitting its expression and / or integration in a host cell. A vector may also be capable of replicating in a8123-112570-02

[0224] host cell (e.g., along with or independent of the host genome replication during cell division), for example, by including a nucleic acid sequence (such as an origin of replication) that permits its replication. An integrating vector is capable of integrating itself or the nucleic acid molecule of interest it carries into a host nucleic acid. An expression vector is a vector that contains necessary regulatory sequences to allow transcription and translation of the nucleic acid molecule of interest (e.g., one or more genes encoding a protein). Vectors include, but are not limited to, nucleic acid molecules that are single-stranded, doublestranded, or partially double-stranded; nucleic acid molecules that comprise one or more free ends, or no free ends (e.g., circular); nucleic acid molecules that include DNA, RNA, other varieties of polynucleotides known in the art, or any combination thereof. One type of vector is a "plasmid," which refers to a circular double stranded DNA loop into which additional DNA segments can be inserted, such as by standard molecular cloning techniques. Another type of vector is a viral vector, wherein virally derived DNA or RNA sequences are present in the vector for packaging into a vims (e.g., retroviruses, replication defective retroviruses, adenoviruses, herpes simplex viruses, baculoviruses, replication defective adenoviruses, and adeno-associated viruses).

[0225] IL Overview of Several Aspects

[0226] HER2 is a member of the EGFR family of receptor tyrosine kinases. HER2 overexpression or gene amplification has been identified as a key driver in breast cancer (20%), and in a variety of other solid tumors. Although HER2 has no known ligand, it forms heterodimers with other members of the EGFR family, in particular EGFR (HER1, ERBB1), typically owing to HER2 overexpression, resulting in autophosphorylation of the tyrosine residues within the cytoplasmic domains of the heterodimers, initiating a variety of signaling pathways leading to cellular proliferation and tumorigenesis even without engagement of EGF. Therefore, in addition to EGFR, HER2 is a validated target for anticancer therapies, and an anti-HER2 antibody plus chemotherapy are used as first line treatment of HER2-positive metastatic breast cancer.

[0227] Trastuzumab is an anti-HER2 monoclonal antibody, developed in 1990. It binds HER2 on the surface of cancer cells and interferes with HER2 signaling via several mechanisms: inhibition of its dimerization with EGFR; triggering receptor internalization and / or degradation; inhibition of the ERK and the PI3K-AKT signaling pathways; and mediating antibody-dependent cellular cytotoxicity. However, trastuzumab and other HER2-targeted antibodies (e.g., pertuzumab) are only effective for breast cancers harboring HER28123-112570-02

[0228] overexpression or gene amplification (HER2-high). Clinically, HER2 -positive (HER2-high) tumors are defined as tumors with score 3+ in HER2 expression detected by immunohistochemistry (IHC), and that are positive in HER2 gene amplification detected by in situ hybridization (ISH). In a large randomized trial, trastuzumab failed to provide any clinical benefit to HER2-low (IHC of 1+ or 2+ with no gene amplification) breast cancer patients (representing >50% of all breast cancers patients).

[0229] To improve the potency for targeting HER2-low breast cancers, anti-HER2 antibodydrug conjugates (ADCs), such as trastuzumab deruxtecan (T-DXd) have been developed. T-DXd is composed of trastuzumab conjugated with up to eight molecules of deruxtecan, a cytotoxic topoisomerase I inhibitor. Upon antibody -mediated internalization, deruxtecan is released into the cytosol to induce DNA damage, leading to apoptotic cell death. In a Phase III clinical trial, T-DXd was shown to double the progress-free survival rates of patients with HER2-low metastatic breast cancers when compared with a non-targeted chemotherapy chosen by the physicians (capeci tabine, gemcitabine, eribulin, paclitaxel, or nab-paclitaxel). Therefore, T-DXd has been approved by the FDA to treat patients with advanced HER2-low breast cancers. However, the efficacy of T-DXd is limited by the amount of payload it can carry (<8 deruxtecan molecules per trastuzumab). In addition, due to its toxicity to HER2-low normal cells, the side effect profile of T-DXd is comparable with that of chemotherapy. In particular, a significant portion of patients developed unacceptable interstitial lung disease. Thus, there is a critical unmet need to develop safer and more potent therapies targeting HER2-expressing (either HER2-high or HER2-low) cancers.

[0230] Through interactions with mammalian host cells, many bacterial pathogens have evolved to employ potent and specific protein toxins to cause host pathology via disrupting the RAS-RAF-MEK-ERK, NF-KB, or other pathways. These toxins usually have multiple functional domains and thus can be structurally modified to achieve high tumor specificity. Anthrax lethal toxin (LT) is such a toxin. LT is a typical A-B type bacterial toxin consisting of two polypeptides: a cellular receptor-binding and delivering component termed protective antigen (PA), and an enzymatic moiety denoted as lethal factor (LF). PA and LF are individually nontoxic. To gain entry into mammalian cells, PA binds to capillary morphogenesis protein-2 (CMG2) or tumor endothelium marker-8 (TEM8) on the cell surface. This binding results in a proteolytic activation of PA by a cell surface furin protease, releasing PA20 fragment, yielding the active PA63 oligomer. The PA oligomer then binds and translocates LF into the cytosol of target cells to exert cytotoxic effects. LF is a zinc metalloproteinase that selectively inactivates mitogen-activated protein kinase kinases8123-112570-02

[0231] (MEKs), leading to disruption of the ERK pathway that is essential for cell proliferation among many other cellular functions.

[0232] In addition to LF, other reengineered effector proteins can also be delivered into the cytosol of target cells by PA to disrupt pathways that are essential for cancer cells. This can be achieved by fusing of the N-terminal PA binding domain of LF (LFn) with a payload “passenger” peptide to be delivered. Three groups of PA-dependent effector proteins for targeting cancer cells with different oncogenic mutations have been generated herein (Table 1): 1) MEK inactivators (LF variants thereof); 2) pan-RAS inactivators (LFn-DUF5 and variants thereof); and 3) protein synthesis inhibitor (FP59).

[0233] Table 1: Reengineered tumor targeting anthrax toxins

[0234] Delivery Receptor used Activated by

[0235] component

[0236] PA (WT) TEM8 and CMG2 Furin protease

[0237] PA3MzHER2 HER2 Furin protease

[0238] PA3MzHER2- HER2 MMP-2, MMP-9, MMP-14, MMP and other MMPs

[0239] Effector Target Cancers targeted

[0240] component

[0241] LF (WT) MEKs (MEK1 / 2 / 3 / 4 / 6) Oncogenic BRAF- and

[0242] MEK-driven

[0243] LF-W271A MEK1 / 2-ERK Oncogenic BRAF- and

[0244] MEK-driven

[0245] LFn-DUF5- All RAS proteins Oncogenic KRAS-, NRAS-, L418W HRAS-driven

[0246] FP59 eEF2 (protein synthesis) All cancers (regardless of

[0247]

[0248] oncogenes)

[0249] One of the frequent oncogenic mutations in human cancers is the BRAFV600Emutation (occurs in 8% of all human cancers and 70% of melanomas). Cancer cells with the BRAFV600Emutation utilize the MEK-ERK pathway for survival. Thus, the MEK inactivators LF and the MEKl / 2-selective variant LF-W271A are used to treat cancers with BRAF mutations (FIG. 1).

[0250] LFn-DUF5 is a fusion of LFn and DUF5, a toxin effector domain (amino acids Glu3581-Gln4089) of multifunctional-autoprocessing repeats-in-toxin (MARTX) from Vibrio vulnificus. DUF5 is a potent and specific endopeptidase that cleaves and inactivates oncogenic rat sarcoma virus (RAS) proteins (including KRAS, NRAS, and HRAS), and RAS-related protein-1 (RAP1). Thus, it is also known as RAS / RAP1 -specific endopeptidase8123-112570-02

[0251] (RRSP). LFn-DUF5-L418W, provided herein, is a RAS-selective LFn-DUF5 variant that does not affect RAP1 (Table 1). This RAS-selective pan-RAS inactivator is demonstrated herein to be effective in treating cancers with oncogenic RAS mutations, which occur in approximately 30% of all human cancers and 90% of pancreatic cancers.

[0252] FP59 is a fusion protein of LFn and the catalytic domain of Pseudomonas aeruginosa exotoxin A that kills all cells (independent of their oncogenic mutations) by ADP-ribosylation of eEF2 (thus inhibiting protein synthesis). Thus, FP59 can be used to target cancers with unidentified oncogenic mutations.

[0253] These PA-dependent effector proteins are potent enzymes that function at sub-stoichiometric levels, with usually only a few molecules per cell being sufficient to inactivate their targets. Thus, also provided are engineered anthrax PA proteins, that allow for extremely high tumor specificity in delivering these cytotoxic effectors directly into cancer cells.

[0254] To redirect the PA’s receptor binding specificity to HER2, zHER2, a high affinity HER2 affibody, was fused to the C-terminus of PA. Affibodies are a class of small (58 amino acids, 6 kD in size), engineered protein binders derived from a structure domain of staphylococcal protein A. Affibodies have a three-helix bundle structure, mimicking antibodies’ antigen binding domains. Specific affibody molecules that bind a desired target protein can be identified from libraries using phage display. zHER2 is one of the best characterized affibodies, with extremely high affinity (Kd = 22 pM) to HER2. zHER2 was fused to the C-terminus of PA3M, a PA variant that does not bind to its natural toxin receptors due to alterations of three key binding residues (D683A, L685E, and Y688K), resulting in PA3MzHER2. PA3MzHER2 demonstrates high affinity to HER2, allowing specific delivery of the effector proteins to the HER2 -expressing cancer cells.

[0255] To further improve the safety of PA3MzHER2, its tumor specificity was further improved by engineering its activation to be further dependent on tumor-associated proteases. As such, the activation of PA will require two independent cancer markers. The unique requirement for PA proteolytic activation on the target cell surface provides a way to reengineer PA to be activated by tumor-associated proteases rather than the usual, ubiquitously expressed furin protease. PA variants that are activated by tumor-associated proteases such as matrix metalloproteinases (MMPs) are generated (Table 1). MMPs are tightly regulated, expressed at extremely low level in normal tissues, but overexpressed in all human cancers. The tumor-specificity of PA3MzHER2 is further enhanced by replacing the furin cleavage sequence164RKKR167with a MMP substrate sequence (e.g., GPLGMLSQ, SEQ ID NO: 3),8123-112570-02

[0256] yielding PA3MzHER2-MMP (FIG. 1). PA3MzHER2-MMP demonstrates not only potent antitumor activity in the presence of a selected effector protein, but also improved safety.

[0257] Therefore, provided herein are highly tumor-selective anthrax toxin allowing the targeting of solid tumors beyond breast cancer. Cancers with various mutations (e.g., oncogenic RAS (KRAS, NRAS, HRAS) mutations) are targeted through delivery of the novel effectors provided herein (e.g., RAS-selective pan-RAS inactivators) to the cytosol of cancer cells, by the novel, highly tumor-selective delivery proteins provided herein (e.g., PA3MzHER2-MMP). This approach yielded a higher antitumor potency and also a higher tumor-specificity compared to other known HER2-targeting agents and small molecule KRAS inhibitors.

[0258] Modifications targeting two different cancer markers can turn pathogenic bacterial toxins into potent and safe antitumor agents. It is demonstrated herein that PA3MzHER2-MMP / LFn-DUF5-L418W has potent antitumor activity, even eliminating very aggressive human tumor xenografts in mice. The engineered toxin exhibits an excellent safety feature that mice can be safely treated for at least ten weeks without any adverse effects, achieving therapeutic index > 9.

[0259] Since the effector proteins are enzymes, functioning at sub-stoichiometric levels, only a few molecules per cell are required to inactivate the target. It is demonstrated herein that the pan-RAS inactivator LFn-DUF5-L418W has much higher potency than the HER2-targeting ADCs (T-DXd and T-DM1), as well as the FDA-approved small molecule KRASG12C-allele specific inhibitors Sotorasib and Adagrasib, the KRASG12D-allele specific inhibitor MRTX1133, and the pan-KRAS inhibitor BI-2865 (which is under preclinical development). Since LFn-DUF5-L418W can irreversibly (because it is proteolytic) inactivate all RAS proteins, pathway re-activation, a common mechanism of drug resistance to small molecule inhibitors of the RAS-RAF-MEK-ERK pathway, is unlikely to occur with the engineered toxin. The different effector proteins (Table 1) can also be used in sequence or in combination to overcome potential drug resistance.

[0260] PA3MzHER2-MMP has a high affinity for HER2, allowing targeting cancers beyond breast cancers that have a wide range of HER2 expression (low to high). The high flexibility of the PA-based protein delivery system allows for further optimization of its tumorspecificity. A high affinity EGFR- or HER3-specific affibody (58 amino acids in size) can be used to direct the toxins to selectively target EGFR- or HER3-expressing cancer cells.

[0261] Using a modified Bacillus anthracis (BA) expression and purification system, the tumor targeting protein toxins can be readily made in large amounts free of8123-112570-02

[0262] lipopolysaccharide (LPS) contamination. The engineered proteins can be secreted into culture medium in their active form and in high yields (50-100 mg per liter), and can be readily purified through simple chromatographic steps.

[0263] III. Effector Proteins

[0264] Provided herein are DUF5 proteins that selectively cleave RAS proteins over RAP 1 proteins. As demonstrated herein, RAS-selective DUF5 retains potent anti-tumor activities without the side effects associated with DUF5. The RAS-selective DUF5 can be used alone or with any suitable delivery system. In some aspects, the RAS-selective DUF5 cleaves RAS at a catalytic efficiency that is at least about 5 times, e.g., at least about 10 times, at least about 15 times, at least about 20 times, at least about 25 times, at least about 30 times, at least about 35 times, at least about 40 times, at least about 45 times, at least about 50 times, at least about 75 times, at least about 100 times, at least about 150 times, at least about 200 times, at least about 500 times, or at least about 1000 times greater than the catalytic efficiency of cleaving RAP1. In some aspects, the DUF5 proteins are recombinant, purified, or isolated.

[0265] In some aspects, DUF5 provided herein comprises one or more of the following substitutions: 408A, 408K, 409R, 418R, 418F, 418C, and 418W, wherein the amino acid numbering is according to a reference DUF5 protein sequence set forth as SEQ ID NO: 29. In some examples, the substitutions are R408A, R408K, P409R, L418R, L418F, L418C, and / or L418W. In some examples, the DUF5 comprises the amino acids surrounding the substitution position, e.g., ±1, 2, 3, 4, 5, 6, or 7 amino acids upstream and / or downstream of position 408, 409, and / or 418 of SEQ ID NO: 29. In some examples, the DUF5 comprises an amino acid sequence that has one or more of the substitutions, and is at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, or at least about 99.8% identical to SEQ ID NO: 29. In some examples, the DUF5 comprises one or more (e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or more) substitutions, such as conservative substitutions, based on SEQ ID NO: 29, in addition to 408 A / K, 409R, and / or 418R / F / C / W.

[0266] Provided herein are recombinant DUF5 proteins comprising a L418W substitution, wherein the amino acid numbering is according to SEQ ID NO: 29. In some aspects, the recombinant DUF5 comprises an amino acid sequence that is at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at8123-112570-02

[0267] least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, or at least about 99.8% identical to SEQ ID NO: 30, and has the L418W substitution. In some examples, the DUF5 comprises one or more (e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or more) substitutions, such as conservative substitutions, based on SEQ ID NO: 30, and has the L418W substitution. In some aspects, the recombinant DUF5 comprises or consists of SEQ ID NO: 30. In some aspects, the DUF5 cleaves RAS (e.g., normal or mutated KRAS, NRAS, and / or HRAS) or inhibits RAS (e.g., normal or mutated KRAS, NRAS, and / or HRAS) activities at an efficiency that is at least about 5 times, e.g., at least about 10 times, at least about 15 times, at least about 20 times, at least about 25 times, at least about 30 times, at least about 35 times, at least about 40 times, at least about 45 times, at least about 50 times, at least about 75 times, at least about 100 times, at least about 150 times, at least about 200 times, at least about 500 times, or at least about 1000 times greater than the efficiency of cleaving RAP1 (e.g., RAP1A and / or RAP1B) or inhibiting RAP1 (e.g., RAP1A and / or RAP1B) activities. In some aspects, the DUF5 does not cleave or inhibit RAP1, e.g., at low to even very high concentrations of DUF5 and / or RAPE

[0268] Provided herein are recombinant DUF5 proteins comprising a R408A or R408K substitution, wherein the amino acid numbering is according to SEQ ID NO: 29. In some aspects, the recombinant DUF5 comprises an amino acid sequence that is at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, or at least about 99.8% identical to SEQ ID NO: 29, and has the R408A or R408K substitution. In some aspects, the DUF5 cleaves RAS (e.g., normal or mutated KRAS, NRAS, and / or HRAS) or inhibits RAS (e.g., normal or mutated KRAS, NRAS, and / or HRAS) activities at an efficiency that is at least about 5 times, e.g., at least about 10 times, at least about 15 times, at least about 20 times, at least about 25 times, at least about 30 times, at least about 35 times, at least about 40 times, at least about 45 times, at least about 50 times, at least about 75 times, at least about 100 times, at least about 150 times, at least about 200 times, at least about 500 times, or at least about 1000 times greater than the efficiency of cleaving RAP1 (e.g., RAP1 A and / or RAP1B) or inhibiting RAP1 (e.g., RAP1 A and / or RAP1B) activities. In some aspects, the DUF5 does not cleave or inhibit RAP1, e.g., at low to even very high concentrations of DUF5 and / or RAPE8123-112570-02

[0269] Provided herein are recombinant DUF5 proteins comprising a P409R substitution, wherein the amino acid numbering is according to SEQ ID NO: 29. In some aspects, the recombinant DUF5 comprises an amino acid sequence that is at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, or at least about 99.8% identical to SEQ ID NO: 29, and has the P409R substitution. In some aspects, the DUF5 cleaves RAS (e.g., normal or mutated KRAS, NRAS, and / or HRAS) or inhibits RAS (e.g., normal or mutated KRAS, NRAS, and / or HRAS) activities at an efficiency that is at least about 5 times, e.g., at least about 10 times, at least about 15 times, at least about 20 times, at least about 25 times, at least about 30 times, at least about 35 times, at least about 40 times, at least about 45 times, at least about 50 times, at least about 75 times, at least 100 times, at least about 150 times, at least about 200 times, at least about 500 times, or at least 1000 times greater than the efficiency of cleaving RAP1 (e.g., RAP1 A and / or RAP1B) or inhibiting RAP1 (e.g., RAP1A and / or RAP1B) activities. In some aspects, the DUF5 does not cleave or inhibit RAPE e.g., at low to even very high concentrations of DUF5 and / or RAPE

[0270] Provided herein are recombinant DUF5 proteins comprising a L418R, L418F, or L418C substitution, wherein the amino acid numbering is according to SEQ ID NO: 29. In some aspects, the recombinant DUF5 comprises an amino acid sequence that is at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, or at least about 99.8% identical to SEQ ID NO: 29, and has the L418R, L418F, or L418C substitution. In some aspects, the DUF5 cleaves RAS (e.g., normal or mutated KRAS, NRAS, and / or HRAS) or inhibits RAS (e.g., normal or mutated KRAS, NRAS, and / or HRAS) activities at an efficiency that is at least about 5 times, e.g., at least about 10 times, at least about 15 times, at least about 20 times, at least about 25 times, at least about 30 times, at least about 35 times, at least about 40 times, at least about 45 times, at least about 50 times, at least about 75 times, at least 100 times, at least about 150 times, at least about 200 times, at least about 500 times, or at least 1000 times greater than the efficiency of cleaving RAP1 (e.g., RAP1 A and / or RAP1B) or inhibiting RAP1 (e.g., RAP1 A and / or RAP1B) activities. In some aspects, the DUF5 does not cleave or inhibit RAPE e.g., at low to even very high concentrations of DUF5 and / or RAPE8123-112570-02

[0271] Also provided are fusion proteins, comprising the DUF5 disclosed herein fused to a heterologous protein. In some aspects, the heterologous protein is a domain of LF that binds to PA (LFn). In some examples, the LFn comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, or at least about 99.8% identical to SEQ ID NO: 31. In some examples, the LFn comprises or consists of SEQ ID NO: 31.

[0272] In particular aspects, the fusion protein comprises an amino acid sequence that is at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, or at least about 99.8% identical to SEQ ID NO: 33, and comprises the L418W substitution. In some examples, the fusion protein comprises one or more (e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or more) substitutions, such as conservative substitutions, based on SEQ ID NO: 33, and comprises the L418W substitution. In particular examples, the fusion protein comprises or consists of SEQ ID NO: 33. In particular aspects, the fusion protein comprises an amino acid sequence that is at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, or at least about 99.8% identical to SEQ ID NO: 32, and comprises any of the R408A, R408K, P409R, L418R, L418F, and L418C substitutions. In some examples, the fusion protein comprises one or more (e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or more) substitutions, such as conservative substitutions, based on SEQ ID NO: 32, in addition to 408A / K, 409R, and / or 418R / F / C / W.

[0273] Any other effector proteins, that have an antitumor effect and can be recognized by the delivery proteins provided herein are also contemplated. In some aspects, the effector protein is fused to LFn so that it can bind and be delivered by PA. In some examples, the LFn comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, or at least about 99.8% identical to SEQ ID NO: 31. In some examples, the LFn comprises or consists of SEQ ID NO: 31. Exemplary effector proteins are provided in Table 1.8123-112570-02

[0274] Also provided are nucleic acids, e.g., recombinant DNA or RNA, that encode any of the effector and fusion proteins provided herein. In some aspects, the nucleic acids are optimized for host cell expression. Also provided are vectors comprising the recombinant nucleic acid molecule.

[0275] IV. Delivery Proteins

[0276] Delivery proteins are used herein to deliver the effector proteins into a cell, e.g., a cancer cell. The delivery proteins provided herein are PA-based and tumor specific.

[0277] PA is engineered herein to increase its specificity toward tumor cells, by, for example, replacing the furin cleavage site to MMP cleavage site or uPA cleavage site; fusing PA to a tumor targeting component (e.g., EGFR, HER2, or HER3 affibody or antibody); mutating PA so that it is not recognized by its native receptors (e.g., the 3M mutations); and / or mutating a pair of PAs so that an effector protein (e.g., LF) can only bind to the hetero-oligomer formed by both of the PAs, but not to the homo-oligomer formed by only one of the PAs (e.g., the ICPA mutations). PA may also be used to refer to the fusion protein formed by PA and the tumor targeting component. In some aspects, PA comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to any of SEQ ID NOs: 1-2, 10-28, and 61-63. In some examples, PA comprises or consists of any of SEQ ID NOs: 1-2, 10-28, and 61-63. In some aspects, PA comprises one or more (e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or more) substitutions, such as conservative substitutions, based on any of SEQ ID NOs: 1-2, 10-28. and 61-63.

[0278] Provided herein are recombinant PAs, comprising a mutation to block binding to PA receptor (e.g., TEM8 and / or CMG2), and a mutation to change the PA furin cleavage site to an MMP cleavage site or a uPA cleavage site. In some aspects, the mutation to block binding to PA receptor comprises 683A, 685E, and 688K substitutions (e.g., D683A, L685E, and Y688K), wherein the amino acid numbering is according to a reference PA protein sequence set forth as SEQ ID NO: 1. In some aspects, the MMP cleavage site comprises the amino acid sequence set forth as GPLGMLSQ (SEQ ID NO: 3) and the uPA cleavage site comprises the amino acid sequence set forth as PGSGRSA (SEQ ID NO: 4).

[0279] In some aspects, the recombinant PA comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about8123-112570-02

[0280] 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99% identical to SEQ ID NO: 2, and comprises the 683 A, 685E, and 688K substitutions. In some examples, the recombinant PA comprises or consists of SEQ ID NO: 2.

[0281] In some aspects, the recombinant PA comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to aa 1-739 of SEQ ID NO: 11 or SEQ ID NO: 61, and comprises the MMP cleavage site GPLGMLSQ (SEQ ID NO: 3), and optionally comprises the 683A, 685E, and 688K substitutions. In some examples, the recombinant PA comprises or consists of aa 1-739 of SEQ ID NO: 11 or SEQ ID NO: 61. In some examples, the recombinant PA comprises D at the 512 position (with numbering according to a furin-activated PA sequence, e.g., SEQ ID NO: 1 or 2).In some aspects, the recombinant PA comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to aa 1-738 of SEQ ID NO: 12, and comprises the uPA cleavage site PGSGRSA (SEQ ID NO: 4), and optionally comprises the 683A, 685E, and 688K substitutions. In some examples, the recombinant PA comprises or consists of aa 1-738 of SEQ ID NO: 12.

[0282] In some aspects, the recombinant PA comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to aa 1-739 of SEQ ID NO: 17, and comprises the MMP cleavage site GPLGMLSQ (SEQ ID NO: 3), and the 207R (e.g., 1207R) substitution, and optionally comprises the 683A, 685E, and 688K substitutions. In some examples, the recombinant PA comprises or consists of aa 1-739 of SEQ ID NO: 17.

[0283] In some aspects, the recombinant PA comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least8123-112570-02

[0284] about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to aa 1-738 of SEQ ID NO: 20, and comprises the uPA cleavage site PGSGRSA (SEQ ID NO: 4), and the 200A (e.g., R200A) substitution, and optionally comprises the 683A, 685E, and 688K substitutions. In some examples, the recombinant PA comprises or consists of aa 1-738 of SEQ ID NO: 20.

[0285] In some aspects, the recombinant PA comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to aa 1-739 of SEQ ID NO: 23, and comprises the MMP cleavage site GPLGMLSQ (SEQ ID NO: 3), and 245G and 252N (e.g., K245G and R252N) substitutions, and optionally comprises the 683A, 685E, and 688K substitutions. In some examples, the recombinant PA comprises or consists of aa 1-739 of SEQ ID NO: 23.

[0286] In some aspects, the recombinant PA comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to aa 1-738 of SEQ ID NO: 26, and comprises the uPA cleavage site PGSGRSA (SEQ ID NO: 4), and the 512K (e.g., D512K) substitution, and optionally comprises the 683A, 685E, and 688K substitutions In some examples, the recombinant PA comprises or consists of aa 1-738 of SEQ ID NO: 26.

[0287] Also provided are fusion proteins of the PAs and a tumor targeting domain. In some aspects, the tumor targeting domain is a HER2 binding domain, a HER3 binding domain, or an EGFR binding domain. In some examples, the tumor targeting domain comprises a target specific affibody (e.g., a HER2, HER3, or EGFR specific affibody). In some examples, the tumor targeting domain comprises a target-specific antibody (e.g., a HER2, HER3, or EGFR specific antibody), including any fragment-based variants such as Fv, Fab, Fab', Fab'-SH, F(ab')2, scFv, etc.

[0288] In some aspects, the HER2 binding domain comprises a HER2 specific affibody comprising an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about8123-112570-02

[0289] 97%, at least about 98%, or at least about 99% identical to SEQ ID NO: 5. In some examples, the HER2 binding domain comprises SEQ ID NO: 5.

[0290] In some aspects, the HER3 binding domain comprises a HER3 specific affibody comprising an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99% identical to SEQ ID NO: 6. In some examples, the HER3 binding domain comprises or consists of SEQ ID NO: 6.

[0291] In some aspects, the EGFR binding domain comprises an EGFR specific affibody comprising an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99% identical to SEQ ID NO: 7. In some examples, the EGFR binding domain comprises or consists of SEQ ID NO: 7.

[0292] In some aspects, the PA is linked to the tumor targeting domain via a GS linker. In some examples, the GS linker comprises or consists of SEQ ID NO: 8. In some examples, the C-terminus or N-terminus of PA is linked to the tumor targeting domain, e.g., through a GS linker.

[0293] In some aspects, the fusion protein of PA and HER2 binding domain comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 11, SEQ ID NO: 61, SEQ ID NO: 12, SEQ ID NO: 17, SEQ ID NO: 20, SEQ ID NO: 23, or SEQ ID NO: 26. In some examples, the fusion protein comprises or consists of SEQ ID NO: 11, SEQ ID NO: 61, SEQ ID NO: 12, SEQ ID NO: 17, SEQ ID NO: 20, SEQ ID NO: 23, or SEQ ID NO: 26.

[0294] In some aspects, the fusion protein of PA and HER2 binding domain comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 11 or SEQ ID NO: 61, and comprises the MMP cleavage site GPLGMLSQ (SEQ ID NO: 3), and optionally comprises the 683A, 685E, and 688K8123-112570-02

[0295] substitutions. In some examples, the fusion protein comprises D at the 512 position (with numbering according to a furin-activated PA sequence, e.g., SEQ ID NO: 1 or 2).

[0296] In some aspects, the fusion protein of PA and HER2 binding domain comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 12, and comprises the uPA cleavage site PGSGRSA (SEQ ID NO: 4), and optionally comprises the 683A, 685E, and 688K substitutions.

[0297] In some aspects, the fusion protein of PA and HER2 binding domain comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 17, and comprises the MMP cleavage site GPLGMLSQ (SEQ ID NO: 3), and the 207R (e.g., I207R) substitution, and optionally comprises the 683A, 685E, and 688K substitutions.

[0298] In some aspects, the fusion protein of PA and HER2 comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 20, and comprises the uPA cleavage site PGSGRSA (SEQ ID NO: 4), and the 200A (e.g., R200A) substitution, and optionally comprises the 683 A, 685E, and 688K substitutions.

[0299] In some aspects, the fusion protein of PA and HER2 comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 23, and comprises the MMP cleavage site GPLGMLSQ (SEQ ID NO: 3), and 245G and 252N (e.g., K245G and R252N) substitutions, and optionally comprises the 683A, 685E, and 688K substitutions.8123-112570-02

[0300] In some aspects, the fusion protein of PA and HER2 comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 26, and comprises the uPA cleavage site PGSGRSA (SEQ ID NO: 4), and the 512K (e.g., D512K) substitution, and optionally comprises the 683 A, 685E, and 688K substitutions.

[0301] In some aspects, the fusion protein of PA and HER3 binding domain comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 13, SEQ ID NO: 62, SEQ ID NO: 14, SEQ ID NO: 18, SEQ ID NO: 21, SEQ ID NO: 24, or SEQ ID NO: 27. In some examples, the fusion protein comprises or consists of SEQ ID NO: 13, SEQ ID NO: 62, SEQ ID NO: 14, SEQ ID NO: 18, SEQ ID NO: 21, SEQ ID NO: 24, or SEQ ID NO: 27.

[0302] In some aspects, the fusion protein of PA and HER3 binding domain comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 13 or SEQ ID NO: 62, and comprises the MMP cleavage site GPLGMLSQ (SEQ ID NO: 3), and optionally comprises the 683A, 685E, and 688K substitutions. In some examples, the fusion protein comprises D at the 512 position (with numbering according to a furin-activated PA sequence, e.g., SEQ ID NO: 1 or 2).

[0303] In some aspects, the fusion protein of PA and HER3 binding domain comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 14, and comprises the uPA cleavage site PGSGRSA (SEQ ID NO: 4), and optionally comprises the 683A, 685E, and 688K substitutions.8123-112570-02

[0304] In some aspects, the fusion protein of PA and HER3 binding domain comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 18, and comprises the MMP cleavage site GPLGMLSQ (SEQ ID NO: 3), and the 207R (e.g., I207R) substitution, and optionally comprises the 683A, 685E, and 688K substitutions.

[0305] In some aspects, the fusion protein of PA and HER3 comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 21, and comprises the uPA cleavage site PGSGRSA (SEQ ID NO: 4), and the 200A (e.g., R200A) substitution, and optionally comprises the 683 A, 685E, and 688K substitutions.

[0306] In some aspects, the fusion protein of PA and HER3 comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 24, and comprises the MMP cleavage site GPLGMLSQ (SEQ ID NO: 3), and 245G and 252N (e.g., K245G and R252N) substitutions, and optionally comprises the 683A.

[0307] 685E, and 688K substitutions.

[0308] In some aspects, the fusion protein of PA and HER3 comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 27, and comprises the uPA cleavage site PGSGRSA (SEQ ID NO: 4), and the 512K (e.g., D512K) substitution, and optionally comprises the 683 A, 685E, and 688K substitutions.

[0309] In some aspects, the fusion protein of PA and EGFR binding domain comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at8123-112570-02

[0310] least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 15, SEQ ID NO: 63, SEQ ID NO: 16, SEQ ID NO: 19, SEQ ID NO: 22, SESQ ID NO: 25, or SEQ ID NO: 28. In some examples, the fusion protein comprises or consists of SEQ ID NO: 15, SEQ ID NO: 63, SEQ ID NO: 16, SEQ ID NO: 19, SEQ ID NO: 22, SESQ ID NO: 25, or SEQ ID NO: 28.

[0311] In some aspects, the fusion protein of PA and EGFR binding domain comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 15 or SEQ ID NO: 63, and comprises the MMP cleavage site GPLGMLSQ (SEQ ID NO: 3), and optionally comprises the 683A, 685E, and 688K substitutions. In some examples, the fusion protein comprises D at the 512 position (with numbering according to a furin-activated PA sequence, e.g., SEQ ID NO: 1 or 2).

[0312] In some aspects, the fusion protein of PA and EGFR binding domain comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 16, and comprises the uPA cleavage site PGSGRSA (SEQ ID NO: 4), and optionally comprises the 683A, 685E, and 688K substitutions.

[0313] In some aspects, the fusion protein of PA and EGFR binding domain comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 19, and comprises the MMP cleavage site GPLGMLSQ (SEQ ID NO: 3), and the 207R (e.g., I207R) substitution, and optionally comprises the 683A, 685E, and 688K substitutions.

[0314] In some aspects, the fusion protein of PA and EGFR comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about8123-112570-02

[0315] 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 22, and comprises the uPA cleavage site PGSGRSA (SEQ ID NO: 4), and the 200A (e.g., R200A) substitution, and optionally comprises the 683A, 685E, and 688K substitutions.

[0316] In some aspects, the fusion protein of PA and EGFR comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 25, and comprises the MMP cleavage site GPLGMLSQ (SEQ ID NO: 3), and 245G and 252N (e.g., K245G and R252N) substitutions, and optionally comprises the 683A, 685E, and 688K substitutions.

[0317] In some aspects, the fusion protein of PA and EGFR comprises an amino acid sequence that is at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 28, and comprises the uPA cleavage site PGSGRSA (SEQ ID NO: 4), and the 512K (e.g., D512K) substitution, and optionally comprises the 683 A, 685E, and 688K substitutions.

[0318] V. Protein Complexes

[0319] Provided are protein complexes formed by an effector protein disclosed herein, and a PA delivery protein disclosed herein. For examples, any of the DUF5-LFn (e.g., RAS-selective DUF5-LFn) disclosed herein can form a complex with any of the PAs disclosed herein that can bind LFn. In other examples, any of the DUF5-LFn (e.g., RAS-selective DUF5-LFn) disclosed herein can form a complex with any of the ICPA pairs disclosed herein.

[0320] In some aspects, the recombinant PA comprises a mutation to block binding to PA receptor (e.g., D683A, L685E, and Y688K substitutions), and a mutation to change the PA furin cleavage site to an MMP cleavage site or a uPA cleavage site. In some aspects, the recombinant PA is fused to a HER2 binding domain, a HER3 binding domain, or an EGFR binding domain. In some examples, the HER2 binding domain comprises a HER2 specific8123-112570-02

[0321] affibody comprising the amino acid sequence set forth as SEQ ID NO: 5; the HER3 binding domain comprises a HER3 specific affibody comprising the amino acid sequence set forth as SEQ ID NO: 6; or the EGFR binding domain comprises an EGFR specific affibody comprising the amino acid sequence set forth as SEQ ID NO: 7. In some examples, the recombinant PA fused to a HER2 binding domain comprises the amino acid sequence set forth as SEQ ID NO: 11, SEQ ID NO: 61, or SEQ ID NO: 12; the recombinant PA fused to a HER3 binding domain comprises the amino acid sequence set forth as SEQ ID NO: 13, SEQ ID NO: 62, or SEQ ID NO: 14; or the recombinant PA fused to an EGFR binding domain comprises the amino acid sequence set forth as SEQ ID NO: 15, SEQ ID NO: 63, or SEQ ID NO: 16.

[0322] In some aspects, the recombinant PA comprises i) a first recombinant PA comprising a mutation to change the PA furin cleavage site to a MMP cleavage site, and a I207R substitution; and ii) a second recombinant PA comprising a mutation to change the PA furin cleavage site to a uPA cleavage site, and a R200A substitution, wherein the amino acid numbering is according to a reference PA protein sequence set forth as SEQ ID NO: 1. In some examples, the first and second recombinant PAs further comprise a mutation to block binding to PA receptor (e.g., D683A, L685E, and Y688K substitutions). In some examples, the first recombinant PA is fused to a HER2 binding domain, a HER3 binding domain, or an EGFR binding domain; and / or the second recombinant PA is fused to a HER2 binding domain, a HER3 binding domain, or an EGFR binding domain.

[0323] In some examples, the first recombinant PA is i) fused to a HER2 binding domain and comprises the amino acid sequence set forth as SEQ ID NO: 17; ii) fused to a HER3 binding domain and comprises the amino acid sequence set forth as SEQ ID NO: 18; or iii) fused to an EGFR binding domain comprises the amino acid sequence set forth as SEQ ID NO: 19; and the second recombinant PA is i) fused to a HER2 binding domain and comprises the amino acid sequence set forth as SEQ ID NO: 20; ii) fused to a HER3 binding domain and comprises the amino acid sequence set forth as SEQ ID NO: 21; or iii) fused to an EGFR binding domain comprises the amino acid sequence set forth as SEQ ID NO: 22.

[0324] In some aspects, the recombinant PA comprises i) a first recombinant PA comprising a mutation to change the PA furin cleavage site to a MMP cleavage site, and K245G and R252N substitutions; and ii) a second recombinant PA comprising a mutation to change the PA furin cleavage site to a uPA cleavage site, and a D512K substitution, wherein the amino acid numbering is according to a reference PA protein sequence set forth as SEQ ID NO: 1. In some examples, the first and second recombinant PAs further comprise a mutation to block8123-112570-02

[0325] binding to PA receptor (e.g., D683A, L685E, and Y688K substitutions). In some examples, the first recombinant PA is fused to a HER2 binding domain, a HER3 binding domain, or an EGFR binding domain; and / or the second recombinant PA is fused to a HER2 binding domain, a HER3 binding domain, or an EGFR binding domain.

[0326] In some examples, the first recombinant PA is i) fused to a HER2 binding domain and comprises the amino acid sequence set forth as SEQ ID NO: 23; ii) fused to a HER3 binding domain and comprises the amino acid sequence set forth as SEQ ID NO: 24; or iii) fused to an EGFR binding domain comprises the amino acid sequence set forth as SEQ ID NO: 25; and the second recombinant PA is i) fused to a HER2 binding domain and comprises the amino acid sequence set forth as SEQ ID NO: 26; ii) fused to a HER3 binding domain and comprises the amino acid sequence set forth as SEQ ID NO: 27; or iii) fused to an EGFR binding domain comprises the amino acid sequence set forth as SEQ ID NO: 28.

[0327] In one specific example, a protein complex provided herein comprises i) a fusion protein of PA and a HER2 affibody; and ii) a fusion protein of DUF5 and LFn, wherein: the fusion protein of i) comprises an amino acid sequence that is at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 11 or SEQ ID NO: 61, and comprises SEQ ID NO: 3 and the D683A, L685E, and Y688K substitutions; and

[0328] the fusion protein of ii) comprises an amino acid sequence that is at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 33, and comprises the L418W substitution.

[0329] VI. Polynucleotides and Expression

[0330] Nucleic acid molecules (for example, cDNA or RNA molecules) encoding the amino acid sequences of effector proteins, delivery proteins, and protein complexes as described herein are provided. Nucleic acids encoding these molecules can readily be produced using the amino acid sequences provided herein, sequences available in the art, and the genetic code. In several implementations, the nucleic acid molecules can be expressed in a host cell (such as a mammalian cell) to produce a disclosed antibody or antigen binding fragment.8123-112570-02

[0331] The genetic code can be used to construct a variety of functionally equivalent nucleic acid sequences, such as nucleic acids which differ in sequence, but which encode the same effector protein or delivery protein.

[0332] Nucleic acid molecules encoding the effector proteins, delivery proteins, and protein complexes can be prepared by any suitable method including, for example, cloning of appropriate sequences or by direct chemical synthesis by standard methods. Chemical synthesis produces a single stranded oligonucleotide. This can be converted into double stranded DNA by hybridization with a complementary sequence or by polymerization with a DNA polymerase using the single strand as a template.

[0333] Exemplary nucleic acids can be prepared by cloning techniques. Examples of appropriate cloning and sequencing techniques can be found, for example, in Green and Sambrook (Molecular Cloning: A Laboratory Manual, 4thed., New York: Cold Spring Harbor Laboratory Press, 2012) and Ausubel et al. (Eds.) (Current Protocols in Molecular Biology, New York: John Wiley and Sons, including supplements, 2017).

[0334] Nucleic acids can also be prepared by amplification methods. Amplification methods include the polymerase chain reaction (PCR), the ligase chain reaction (LCR), the transcription-based amplification system (TAS), and the self-sustained sequence replication system (3 SR).

[0335] The nucleic acid molecules can be expressed in a recombinantly engineered cell such as bacteria, plant, yeast, insect and mammalian cells. The effector proteins, delivery proteins, and protein complexes can be expressed as individual proteins and optionally expressed simultaneously or as a fusion protein to generate a protein complex containing the effector and delivery proteins.

[0336] One or more DNA sequences encoding the effector proteins, delivery proteins, and protein complexes can be expressed in vitro by DNA transfer into a suitable host cell. The cell may be prokaryotic or eukaryotic. Numerous expression systems available for expression of proteins including Bacillus anthracis (e.g., BH500 strain), E. coli, other bacterial hosts, yeast, and various higher eukaryotic cells such as the COS, CHO, HeLa and myeloma cell lines, can be used to express the disclosed antibodies and antigen binding fragments. In a particular example, the host cell is BH500 cell. Methods of stable transfer, meaning that the foreign DNA is continuously maintained in the host, are known in the art. Hybridomas expressing the antibodies of interest are also encompassed by this disclosure.

[0337] The expression of nucleic acids encoding the effector proteins, delivery proteins, and protein complexes described herein can be achieved by operably linking the DNA or cDNA8123-112570-02

[0338] to a promoter (which is either constitutive or inducible), followed by incorporation into an expression cassette. The promoter can be any promoter of interest, including a cytomegalovirus promoter and a human T cell lymphotrophic virus promoter (HTLV)-l. Optionally, an enhancer, such as a cytomegalovirus enhancer, is included in the construct. The cassettes can be suitable for replication and integration in either prokaryotes or eukaryotes. Typical expression cassettes contain specific sequences useful for regulation of the expression of the DNA encoding the protein. For example, the expression cassettes can include appropriate promoters, enhancers, transcription and translation terminators, initiation sequences, a start codon (i.e., ATG) in front of a protein-encoding gene, splicing signals for introns, sequences for the maintenance of the correct reading frame of that gene to permit proper translation of mRNA, and stop codons. The vector can encode a selectable marker, such as a marker encoding dmg resistance (for example, ampicillin or tetracycline resistance).

[0339] To obtain high level expression of a cloned gene, it is desirable to construct expression cassettes which contain, for example, a strong promoter to direct transcription, a ribosome binding site for translational initiation (e.g., internal ribosomal binding sequences), and a transcription / translation terminator. For E. coli, this can include a promoter such as the T7, trp, lac, or lambda promoters, a ribosome binding site, and preferably a transcription termination signal. For eukaryotic cells, the control sequences can include a promoter and / or an enhancer derived from, for example, an immunoglobulin gene, HTLV, SV40 or cytomegalovirus, and a polyadenylation sequence, and can further include splice donor and / or acceptor sequences (for example, CMV and / or HTLV splice acceptor and donor sequences). The cassettes can be transferred into the chosen host cell by well-known methods such as transformation or electroporation for E. coli and calcium phosphate treatment, electroporation or lipofection for mammalian cells. Cells transformed by the cassettes can be selected by resistance to antibiotics conferred by genes contained in the cassettes, such as the amp, GPt, neo, and hyg genes.

[0340] Modifications can be made to a nucleic acid encoding a polypeptide described herein without diminishing its biological activity. Some modifications can be made to facilitate the cloning, expression, or incorporation of the targeting molecule into a fusion protein. Such modifications include, for example, termination codons, sequences to create conveniently located restriction sites, and sequences to add a methionine at the amino terminus to provide an initiation site, or additional amino acids (such as poly His) to aid in purification steps.

[0341] Once expressed, the effector proteins, delivery proteins, and protein complexes thereof, can be purified according to standard procedures in the art, including ammonium8123-112570-02

[0342] sulfate precipitation, affinity columns, column chromatography, and the like (see, generally, Simpson et al. (Eds.), Basic methods in Protein Purification and Analysis: A Laboratory Manual, New York: Cold Spring Harbor Laboratory Press, 2009). The effector proteins, delivery proteins, and protein complexes thereof need not be 100% pure. Once purified, partially or to homogeneity as desired, if to be used prophylatically, the polypeptides should be substantially free of endotoxin.

[0343] VI. Pharmaceutical Compositions and Kits

[0344] Pharmaceutical compositions comprising an effector protein (e.g., RAS-selective DUF5-LFn), a delivery protein, a protein complex, a nucleic acid molecule, and / or a vector as disclosed herein and a pharmaceutically acceptable carrier are also provided. Such compositions can be administered to subjects by a variety of administration modes, for example, intramuscular, subcutaneous, intravenous, intra-arterial, intra-articular, intraperitoneal, or parenteral routes. Methods for preparing administrable compositions are described in more detail in such publications as Remington: The Science and Practice of Pharmacy, 23rdEd., New York: Academic Press, 2020.

[0345] Thus, an effector protein, a delivery protein, a protein complex, a nucleic acid molecule, and / or a vector described herein can be formulated with pharmaceutically acceptable carriers to help retain biological activity while also promoting increased stability during storage within an acceptable temperature range. Potential carriers include, but are not limited to, physiologically balanced culture medium, phosphate buffer saline solution, water, emulsions (e.g., oil / water or water / oil emulsions), various types of wetting agents, cryoprotective additives or stabilizers such as proteins, peptides or hydrolysates (e.g., albumin, gelatin), sugars (e.g., sucrose, lactose, sorbitol), amino acids (e.g., sodium glutamate), or other protective agents. The resulting aqueous solutions may be packaged for use as is or lyophilized. Lyophilized preparations are combined with a sterile solution prior to administration for either single or multiple dosing.

[0346] Formulated compositions, especially liquid formulations, may contain a bacteriostat to prevent or minimize degradation during storage, including but not limited to effective concentrations (usually ^1% w / v) of benzyl alcohol, phenol, m-cresol, chlorobutanol, methylparaben, and / or propylparaben. A bacteriostat may be contraindicated for some patients; therefore, a lyophilized formulation may be reconstituted in a solution either containing or not containing such a component.8123-112570-02

[0347] The pharmaceutical compositions of the disclosure can contain as pharmaceutically acceptable vehicles substances as required to approximate physiological conditions, such as pH adjusting and buffering agents, tonicity adjusting agents, wetting agents and the like, for example, sodium acetate, sodium lactate, sodium chloride, potassium chloride, calcium chloride, sorbitan monolaurate, and triethanolamine oleate.

[0348] In some particular aspects, the pharmaceutical composition comprises the recombinant DUF5 protein comprising a L418W substitution, or a nucleic acid molecule encoding the recombinant DUF5 protein.

[0349] In some particular aspects, the pharmaceutical composition comprises a protein complex formed by the recombinant DUF5-LFn comprising a L418W substitution, and a PA, or ICPA pairs disclosed herein; or one or more nucleic acid molecules encoding the protein complex.

[0350] In some particular aspects, the pharmaceutical composition comprises any of the fusion proteins comprising or consisting of any of the amino acid sequences set forth as SEQ ID NO: 11, SEQ ID NO: 61, SEQ ID NO: 12, SEQ ID NO: 17, SEQ ID NO: 20, SEQ ID NO: 23, SEQ ID NO: 26, SEQ ID NO: 13, SEQ ID NO: 62, SEQ ID NO: 14, SEQ ID NO: 18, SEQ ID NO: 21, SEQ ID NO: 24, or SEQ ID NO: 27, SEQ ID NO: 15, SEQ ID NO: 63, SEQ ID NO: 16, SEQ ID NO: 19, SEQ ID NO: 22, SESQ ID NO: 25, and SEQ ID NO: 28; or a nucleic acid molecule encoding the fusion protein.

[0351] In one specific example, a pharmaceutical composition provided herein comprises i) a fusion protein of PA and a HER2 affibody; and ii) a fusion protein of DUF5 and LFn, wherein:

[0352] the fusion protein of i) comprises an amino acid sequence that is at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 11 or SEQ ID NO: 61, and comprises SEQ ID NO: 3 and the D683A, L685E, and Y688K substitutions: and

[0353] the fusion protein of ii) comprises an amino acid sequence that is at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 33, and comprises the L418W substitution.8123-112570-02

[0354] Also provided are kits, comprising one, two or more of the pharmaceutical compositions disclosed herein. In some aspects, the two or more pharmaceutical compositions are contained in different containers or different compartments of a container.

[0355] In some particular aspects, the kit comprises a pharmaceutical composition comprising the recombinant DUF5-LFn comprising a L418W substitution, or a nucleic acid molecule encoding the recombinant DUF5-LFn; and a pharmaceutical composition comprising a PA protein comprising or consisting of any of SEQ ID NOs: 10-28, or a nucleic acid molecule encoding the recombinant PA protein.

[0356] In some particular aspects, the kit comprises a pharmaceutical composition comprising the recombinant DUF5-LFn comprising a L418W substitution (or a nucleic acid molecule encoding the recombinant DUF5-LFn); and one or two pharmaceutical compositions comprising i) a PA protein comprising or consisting of SEQ ID NO: 17, SEQ ID NO: 18, or SEQ ID NO: 19 (or the nucleic acid molecule encoding the recombinant fusion protein), and ii) a PA protein comprising or consisting of SEQ ID NO: 20, SEQ ID NO: 21, or SEQ ID NO: 22 (or the nucleic acid molecule encoding the recombinant fusion protein, together or separately.

[0357] In some particular aspects, the kit comprises a pharmaceutical composition comprising the recombinant DUF5-LFn comprising a L418W substitution (or a nucleic acid molecule encoding the recombinant DUF5-LFn); and one or two pharmaceutical compositions comprising i) a PA protein comprising or consisting of SEQ ID NO: 23, SEQ ID NO: 24, or SEQ ID NO: 25 (or the nucleic acid molecule encoding the recombinant fusion protein), and ii) a PA protein comprising or consisting of SEQ ID NO: 26, SEQ ID NO: 27, or SEQ ID NO: 28 (or the nucleic acid molecule encoding the recombinant fusion protein, together or separately.

[0358] In one specific example, a kit provided herein comprises i) a fusion protein of PA and a HER2 affibody; and ii) a fusion protein of DUF5 and LFn, wherein:

[0359] the fusion protein of i) comprises an amino acid sequence that is at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 11 or SEQ ID NO: 61, and comprises SEQ ID NO: 3 and the D683A, L685E, and Y688K substitutions; and

[0360] the fusion protein of ii) comprises an amino acid sequence that is at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about8123-112570-02

[0361] 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, at least about 99.5%, at least about 99.8%, or at least about 99.9% identical to SEQ ID NO: 33, and comprises the L418W substitution.

[0362] VII. Method of Treatment

[0363] Provided are methods of treating cancer in a subject (such as human or a veterinary subject), comprising administering a therapeutically effective amount of an effector protein (e.g., RAS-selective DUF5 or DUF5-LFn), a delivery protein (e.g., an engineered PA or ICPA pairs), a protein complex, a nucleic acid molecule, a vector, or a pharmaceutical composition as disclosed herein to the subject.

[0364] In these applications, the therapeutically effective amount of the effector protein (e.g., RAS-selective DUF5 or DUF5-LFn), delivery protein (e.g., an engineered PA or ICPA pairs), protein complex, nucleic acid molecule, vector, or pharmaceutical composition is administered to the subject in an amount sufficient to inhibit growth, replication, or metastasis of cancer cells, or to inhibit a sign or a symptom of the cancer. Thus, in some examples, the disclosed method reduces the size, weight and / or volume of a primary tumor and / or a metastasis by at least 20%, at least 25%, at least 50%, at least 75%, at least 80%, at least 90%, at last 95%, at least 99%, or even 100% compared to a control or compared to pretreatment condition of the subject. In some examples, the disclosed methods reduce the number of metastases by at least 20%, at least 25%, at least 50%, at least 75%, at least 80%, at least 90%, at last 95%, at least 99%, or even 100% compared to a control or compared to pre -treatment condition of the subject.

[0365] These methods can be used to treat cancer that overexpresses or has elevated levels of HER2, HER3, EGFR, MMP, and / or uPA / uPAR, such as HER2 and MMP, HER2 and uPA / uPAR, HER3 and MMP, HER3 and uPA / uPAR, EGFR and MMP, and EGFR and uPA / uPAR. When ICPA proteins are used, even higher levels of tumor specificity can be achieved, as three of four tumor markers selected from MMP, uPA / uPAR, EGFR, HER2, and HER3 can be targeted. The methods can be used to treat cancer that has mutated or overactive RAS (e.g., KRAS, NRAS, and / or HRAS) activity, while minimizing side effects caused by targeting RAPE

[0366] The methods can also be used to treat cancer that has low levels of HER2 expression (“HER2-low'’), such as HER2-low breast cancer. In some aspects, the level of HER2 expression is determined by an immunohistochemistry (IHC) test and / or fluorescence in situ hybridization (FISH) test. Cancer with a HER2 IHC score of 1+, or 2+ with a negative FISH8123-112570-02

[0367] result, is generally considered HER2-low cancer. If the IHC result is 0, the cancer is considered HER2-negative. If the IHC result is 0 with HER2+ staining in no more than 10% of tumor cells, the cancer is considered HER2-ultralow. If the IHC is 2+ and the FISH is positive, or if the IHC is 3+, the cancer is considered HER2-positive. As demonstrated herein, the methods are effective for HER2-low and HER2 -positive cancers.

[0368] In some aspects, the methods comprise local or systemic administration of the effector protein and the delivery protein. The effector protein and delivery proteins can be mixed together before administration to a subject, or can be administered separately to the subject without mixing beforehand. In some aspects, the agents are delivered to the subject through intra-arterial administration, intravenous administration, topical administration, subcutaneous administration, or intramuscular administration.

[0369] Administration of the effector protein (e.g., RAS-selective DUF5 or DUF5-LFn), delivery protein (e.g., an engineered PA or ICPA pairs), protein complex, nucleic acid molecule, vector, or pharmaceutical composition disclosed herein can also be accompanied by administration of other anti-cancer agents or therapeutic treatments (such as surgical resection of a tumor). Any suitable anti-cancer agent can be administered in combination with the effector protein (e.g., RAS-selective DUF5 or DUF5-LFn), delivery protein (e.g., an engineered PA or ICPA pairs), protein complex, nucleic acid molecule, vector, or pharmaceutical composition disclosed herein (for example sequentially or concurrently). Exemplary anti-cancer agents include, but are not limited to, chemotherapeutic agents, such as, for example, mitotic inhibitors, alkylating agents, anti -metabolites, intercalating antibiotics, growth factor inhibitors, cell cycle inhibitors, enzymes, topoisomerase inhibitors, anti-survival agents, biological response modifiers, anti-hormones (e.g. anti-androgens) and anti-angiogenesis agents. Other anti-cancer treatments include radiation therapy and other antibodies that specifically target cancer cells.

[0370] The method of treatment may be used, for example, to treat solid as well as hematological cancers. In some particular aspects, the method is used to treat breast cancer, ovarian cancer, pancreatic cancer, lung cancer, bladder cancer, gastric cancer, esophageal cancer, colorectal cancer, melanoma, or a hematologic cancer.8123-112570-02

[0371] EXAMPLES

[0372] Overview

[0373] The native effector LF is a zinc-dependent metalloproteinase that cleaves and inactivates multiple mitogen-activated protein kinase kinases (MEKs), resulting in the inactivation of the three key mitogen-activated protein kinase (MAPK) pathways: ERK (through cleavage of MEK1 / 2), p38 (through cleaving MEK3 / 6), and Jun N-terminus kinase (JNK) (through cleaving MEK4). In addition to the native effector, other effector proteins can also be delivered to target cells through the anthrax toxin protein delivery system, by fusion to the LF’s N-terminal PA-binding domain (LFn). The multifunctionalautoprocessing repeats-in-toxin (MARTX, amino acids 1-5206) effector domain DUF5 (domain of unknown function 5, amino acids 3581-4089 of MARTX, renumbered 1-509 herein) from Vibrio vulnificus was identified to be a specific endopeptidase that cleaves within the RAS Switch 1 region (between Y32 and D33), thereby inactivating both wild-type and mutant RAS (KRAS, NRAS, HRAS) proteins. The fusion of DUF5 and diphtheria toxin (RRSP-DTB) could reduce the growth of human tumor xenografts in a mouse model. Since diphtheria toxin receptor is widely expressed in human cells (whereas intrinsically, a mouse lacks active diphtheria toxin receptor), RRSP-DTB lacks tumor specificity in human.

[0374] Here, a LFn-DUF5 fusion was generated, so it can be delivered into tumors using tumor-selective PA variants. However, LFn-DUF5 also cleaves and inactivates RAS-related protein-1 (RAP1), which has an important role in normal physiology by regulating intercellular tight junction, thereby maintaining vascular integrity. Thus, to avoid the potential undesired side effects of targeting RAP1, novel DUF5 variants that selectively cleave RAS while sparing RAP1 were identified herein (e.g., DUF5-L418W). These novel RAS-selective DUF5 effectors can be used on their own or with any delivery systems to selectively target RAS proteins without affecting RAP1 proteins.

[0375] In some examples, these DUF5 variants are fused to LFn, and combined with the highly tumor-selective PA-based delivery system provided herein (e.g., PA3MzHERl / 2 / 3-MMP / uPA, the ICPA delivery system, etc.), form a potent, highly tumor-selective RAS inactivators, allowing safe targeting of cancers covering all distinct RAS alterations, as well as wildtype KRAS amplification.

[0376] A HER2-specific PA variant, termed PA3MzHER2, that binds specifically to HER2 with a sub-nanomolar affinity was generated. This was achieved by fusing a HER2-specific affibody (zHER2, 58 amino acids in size) to the C-terminus of PA. PA3MzHER2, in the8123-112570-02

[0377] presence of an effector protein, can efficiently kill a variety of cancer cells even those classified as having low HER2 expression. A PA3MzHER2 variant was further generated by replacing the furin activation site with a matrix metalloproteinase (MMP) substrate sequence, resulting in PA3MzHER2-MMP that relies on two different cancer markers (HER2 and MMP) for activation, thus achieving high tumor specificity.

[0378] PA3MzHER2 and PA3MzHER2-MMP, along with different effector proteins, including a novel pan-RAS inactivator LFn-DUF5-L418W, were characterized in vitro with respect to their abilities in targeting both HER2-high and HER2-low cancer cells with various oncogenic mutations, with a particular focus on oncogenic RAS mutations. As demonstrated herein, the engineered toxins are effective across a wide range of HER2-expressing (high or low) cancer cells with various clinically observed oncogenic RAS mutations, and achieve superior effects in both antitumor potency and specificity compared to HER2-taregting antibodies and drag conjugates thereof, as well as small molecule KRAS inhibitors. It is shown herein that PA3MzHER2 plus LFn-DUF5-L418W or FP59 were quite toxic to mice. Remarkably, PA3MzHER2-MMP plus LFn-DUF5-L418W or FP59 were well tolerated in mice, allowing for long-term treatment.

[0379] The engineered toxin system is versatile in that different effector proteins can be used to target cancers with different oncogenic mutations: LFn-DUF5-L418W (pan-RAS inactivator) for cancers with any oncogenic RAS mutations; LF-W271 A (MEK inactivator) for cancers with BRAF or MEK mutations; FP59 (potent protein synthesis inhibitor) for cancers with broader or numerous oncogenic mutations. Since cancer cells with oncogenic RAS mutations or BRAF mutations are addicted to these oncogenic proteins for survival, effectors LFn-DUF5-L418W and LF-W271A are expected to provide additional tumor specificity comparing to FP59.

[0380] Oncogenic RAS proteins have long been considered as “undraggable” until the recent FDA-approval of the allele-specific covalent inhibitors Sotorasib and Adagrasib for treating non-small cell lung cancer (NSCLC) with the KRASG12Cmutation. However, treated patients inevitably developed drag resistance, mostly due to other secondary KRAS mutations (G12D, G12R, G12V, G13D, or Q61H) that are not targeted by these inhibitors. Since the novel pan-RAS inactivator LFn-DUF5-L418W is able to inactivate all these RAS proteins, the dragresistance due to secondary KRAS mutations is unlikely to occur with this approach.

[0381] In vivo toxicity and antitumor studies of PA3MzHER2-MMP plus LFn-DUF5-L418W were performed. The maximum tolerated dose and potential toxicity of the engineered toxin were determined in mice. The antitumor efficacy was evaluated in various mouse tumor8123-112570-02

[0382] models, focusing on cancers with various oncogenic RAS mutations. It is shown herein that the engineered toxin can eliminate aggressive human fibrosarcoma HT1080 (NRASQ61K) and lung H358 (KRASG12C) tumor xenografts in mice.

[0383] The HER2-only targeted toxin (with PA3MzHER2 as the delivery component) caused unacceptable in vivo toxicity, indicating that targeting HER2 alone is insufficient to achieve high tumor specificity. In contrast, the HER2- and MMP-dual targeted toxin (with PA3MzHER2-MMP as the delivery component) exhibited extremely high tumor specify, with a therapeutic index > 9. Strikingly, the dual cancer marker-targeted toxin can even eliminate most of the HT1080 (NRASQ61K) tumors and NCI-H358 (KRASG12C) tumors without obvious toxicity. Because mice tolerated long-term, high doses of PA3MzHER2-MMP plus LFn-DUF5-L418W (90 pg / 30 pg), the approach appears unlikely to have any significant safety concerns.

[0384] Long-term therapies with the toxin may induce neutralizing antibodies that block its activity in immunocompetent subjects. However, this can be safely and effectively overcome by B-cell depletion using a pentostatin / cyclophosphamide (PC) regimen. Additionally, tolerogenic nanoparticles comprised of biodegradable polymers that encapsulate rapamycin are capable of inhibiting anti-drug antibody productions in rodents, allowing long-term therapies.

[0385] Although the FDA-approved KRASG12Cinhibitors Sotorasib and Adagrasib demonstrate efficacy to NSCLC patients with the KRASG12Cmutation, all patients inevitably developed drag resistance. The mechanisms of drag-resistance are mostly due to a secondary KRAS mutation (2°KRAS) (such as G12D, G12R, G12V, G13D, or Q61H), 2°NRAS (such as NRASQ61K, NRASG13R), or pathway re-activation due to the release of ERK-dependent negative feedback inhibition, resulting in activation of the WT KRAS, NRAS, and HRAS. These common drag resistance mechanisms are unlikely to occur in cancer cells treated with the engineered toxin provided herein, because of the potent and irreversible inactivation (due to proteolytic activity) of all RAS proteins. Other drug resistance mechanisms, such as ERK activation via acquired BRAF and MEK activating mutations, or PI3K pathway activation (due to 2° PTEN inactivation or P1K3CA activation mutations) may be overcome by including an additional LF-W271A (MEK-specific inactivator), or PI3K inhibitor (e.g., BEZ235) to the treatment.

[0386] One common drug-resistance mechanism to T-DXd that cancer cells develop is lowering HER2 expression, while simultaneously upregulating HER3 expression. A HER3-targeting toxin is provided herein, wherein a high affinity HER3 affibody (zHER3-6899) is8123-112570-02

[0387] fused to the C-terminus of PA3M-MMP to yield PA3MzHER3-MMP. Therefore, the potential drug-resistance due to HER2-downregulation HER3 -upregulation can be readily overcome by using PA3MzHER3-MMP. Thus, the anthrax protein delivery system is quite versatile: through sequential or combinatorial usage of different effector proteins, and through redirecting the binding specificities, many drug-resistance mechanisms can be overcome.

[0388] Example 1

[0389] Generation and purification of HER-2 targeting pa variants B. anthracis can express and secrete very high levels of protein toxin components. Based on this, a highly efficient anthrax protein expression system was developed, which utilizes a non-virulent B. anthracis host strain (BH500), and a PA promoter and signal peptide-based expression plasmid pYS5 to express various recombinant proteins. These recombinant proteins are usually secreted into the culture medium at > 50 mg per liter in their active form and can be readily purified free of lipopolysaccharide (LPS) contamination. The expression plasmids of PA3MzHER2 and PA3MzHER2-MMP were transformed into BH500 host strain. The recombinant proteins secreted into culture medium were collected and purified in amounts > 100 mg.

[0390] Example 2

[0391] HER2 as a general marker for cancer targeting

[0392] HER2 overexpression or gene amplification is the key oncogenic driver in approximately 20% of breast cancers and, to a lesser extent, various other types of cancers. These HER2 -driven cancers can be treated with HER2-neutralizing antibodies (trastuzumab and pertuzumab). It is hypothesized that other cancer cells would also express HER2 at lower levels that might however be sufficient to be targeted by our high affinity HER2-targeting toxins.

[0393] To survey HER2 expression in cancer cells, a set of 21 human cancer cell lines was analyzed. Real-time PCR analysis showed that all these cancer cell lines expressed HER2, albeit at various levels (FIG. 2A). Human breast cancer lines BT474 and SKBR3 are among the high HER2-expressing cell lines. When compared to HER2 levels seen in human fibrosarcoma HT1080 cells (low HER2 -expressing cells), the high HER2-expressing cells express HER2 at a > 100-fold higher level (FIG. 2A). Western blotting analyses confirmed these cancer cell lines expressed markedly different levels of HER2 (FIG. 2B).8123-112570-02

[0394] Example 3

[0395] HER2-dependent cytotoxicity of HER2-targeting PA variants To examine the HER2 -targeting activity of the novel PA variants, PA3MzHER2 (SEQ ID NO: 10) and PA3MzHER2-MMP (SEQ ID NO: 11 with 512D), the above cancer cell lines were treated with PA3MzHER2 and PA3MzHER2-MMP in the presence of FP59. FP59 is a broad effector that kills cells (independent of oncogenic mutations) in a strict PA-dependent manner, serving as a convenient molecular tool to test the delivery activity of PA variants. As shown in FIGs. 3A and 3B, the high HER2 cells (e.g., AU565, BT474, and SKBR3) were highly susceptible to the novel HER2-targeting toxins, and the low HER2 cells (e.g., HT1080 and MDA-MB-231) were also sensitive to the killing by these toxins.

[0396] To verify the toxins’ HER2-dependency on cytotoxicity, HER2-knocked out HT1080 and MCF7 cells were generated using a CRISPR gene editing approach. Lack of HER2 expression by HT1080 (HER2-KO) cells and MCF7 (HER2-KO) cells was confirmed by anti-HER2 Western blotting analysis (FIG. 2B). Remarkably, while HT1080-wildtype (WT) and MCF7-WT cells were susceptible to PA3MzHER2 plus FP59, the corresponding HER2-KO cells were completely resistant to the toxin (FIG. 3C), demonstrating that the novel HER2-targeting toxins are strictly dependent on HER2 expression (even at low levels) on the target cell.

[0397] Example 4

[0398] HER2 binding affinity of HER2-targeting PA variants The potent and specific cytotoxicity of the novel HER2-targeting toxins to low HER2 cancer cells (FIGs. 3A-3C) suggests that the novel toxins likely have a high binding affinity for HER2.

[0399] To assess the binding affinity of PA3MzHER2-MMP for HER2 (SEQ ID NO: 11 with 512D), a nontoxic PA3MzHER2-MMP variant was generated by further introducing the K245G and R252N alterations. The resulted PA3MzHER2-MMP-GN (SEQ ID NO: 23) cannot form an active oligomer after proteolytic activation, but can serve as a nontoxic HER2-binding competitor of PA3MzHER2-MMP. The apparent affinity for HER2 was evaluated in a Schild plot analysis (Zuo, Zehua et al. “A potent tumor-selective ERK pathway inactivator with high therapeutic index.” PNAS nexus vol. 1,3 pgacl04. 1 Jul. 2022, doi:10.1093 / pnasnexus / pgacl04; and Liu, Shihui et al. “Characterization of the interaction between anthrax toxin and its cellular receptors.” Cellular microbiology vol. 9,4 (2007): 977-8123-112570-02

[0400] 87. doi:10.1111 / j.1462-5822.2006.00845. x). MDA-MB-231 cells (low HER2) were incubated with various concentrations of PA3MzHER2-MMP plus FP59 (constant at 1.9 nM) in the presence of different fixed concentrations of PA3MzHER2-MMP-GN. As the concentration of PA3MzHER2-MMP-GN increases, the cytotoxicity dose-response curve shifts rightward (FIG. 4A).

[0401] The EC50 (concentration required to kill 50% of the cells) of PA3MzHER2-MMP in the presence of various fixed concentrations of PA3MzHER2-MMP-GN were determined (FIG. 4B). Prism program was used to perform a non-liner regression to fit the EC50 to the equation, Y= -Log(X+10LogKd)-P, to determine the Kd of PA3MzHER2-MMP-GN to HER2-expressing cells (where Y=-Log[EC50] (nM), X= [PA3MzHER2-MMP-GN] (nM), P is a constant, see http: / / www.graphpad.com / curvefit / schild.htm for detailed discussion) (FIG.

[0402] 4B). The apparent affinity of PA3MzHER2-MMP-GN (and by implication, PA3MzHER2-MMP) to HER2-expressing cells was in the sub-nanomolar range (Kd = 0.21 nM). A similar binding affinity was obtained using SKOV3 cells (high HER2) (data not shown). The method used here to determine affinity constants directly accesses the binding affinity in their native environments.

[0403] Therefore, the novel HER2 -targeting toxins bind to cell surface HER2 with a high affinity, consistent with their high cytotoxicity to even low HER2 cancer cells (FIGs. 3A-3C). These results demonstrate that although HER2 may not be oncogenic in low HER2 cancer cells, it serves as a valid target for the novel engineered toxins that bind HER2 with a high affinity.

[0404] Example 5

[0405] Greatly attenuated in vivo toxicity of PA3MzHER2-MMP To evaluate the tumor specificity of PA3MzHER2-MMP (SEQ ID NO: 11 with 512D) and PA3MzHER2 (SEQ ID NO: 10), human noncancerous lung epithelial BEAS-2B cells were treated with PA3MzHER2 and PA3MzHER2-MMP in the presence of FP59. BEAS-2B cells were susceptible to PA3MzHER2 (FIG.5A), indicating that PA3MzHER2 is cytotoxic to certain normal cells that express low levels of HER2. In contrast, BEAS-2B cells were resistant to PA3MzHER2-MMP (FIG. 5A), demonstrating that the HER2- and MMP-dual targeted toxin has much lower cytotoxicity to normal cells.

[0406] To further assess their potential in vivo toxicity, mice were injected (I.P. route) with multiple doses of PA3MzHER2 (30 pg) with either FP59 (5 pg) or LFn-DUF5-L418W (SEQ ID NO: 33) (10 pg), or of PA3MzHER2-MMP (30 pg) with either FP59 (5 pg) or LFn-8123-112570-02

[0407] DUF5-L418W (10 pg), every other day. Strikingly, while all the mice in the PA3MzHER2 group succumbed to the challenge after two to four doses, all the mice in the PA3MzHER2-MMP group survived the challenge, even after nine-doses of administration without any signs of malaise (FIG. 5B). These results demonstrate that the strategy of requiring two distinct cancer markers for engagement and activation greatly (or entirely) diminishes the undesirable in vivo toxicity.

[0408] Example 6

[0409] HER2 specificity among other EGFR family members To further provide evidence that PA3MzHER2 (SEQ ID NO: 10) and PA3MzHER2-MMP (SEQ ID NO: 11 with 512D) solely use HER2 among other EGFR family members as the receptor for cell surface binding, EGFR-, HER2-, HER3-, and HER4-knockout HT1080 and MCF7 cells will be generated using CRISPR genome editing. It is expected that while HT1080 HER2-KO and MCF7 HER2-KO cells are resistant to the HER2-targeting toxins, the EGFR-KO, HER3-KO and HER4-KO cells will remain sensitive to the toxins. The effects of the HER2-blocking antibodies, such as trastuzumab and pertuzumab, on the sensitivity of HT1080 cells and MCF7 cells to the HER2-targeting toxins will also be examined in cytotoxicity assays. It is expected that pre-incubation of these cells with the HER2-blocking antibodies will reduce sensitivity of the cells to the HER2-tageting toxins. These experiments will further verify the absolute HER2 specificity of the engineered toxins in their cytotoxic action.

[0410] Example 7

[0411] Requirement of MMP for PA3MzHER2-MMP activation In addition to HER2, the cytotoxicity of PA3MzHER2-MMP to cancer cells is expected to rely on the presence of MMP activity. HT1080 cells were treated with PA3MzHER2 (SEQ ID NO: 10) or PA3MzHER2-MMP (SEQ ID NO: 11 with 512D) plus FP59, in the presence of various concentrations of a specific MMP inhibitor GM6001.

[0412] Notably, GM6001 diminished the cytotoxicity of PA3MzHER2-MMP but not PA3MzHER2 in a dose-dependent manner (FIG. 6). Thus, PA3MzHER2-MMP requires the simultaneous presence of two distinct cancer markers, HER2 and MMP, to exert its anticancer cytotoxicity.

[0413] To further verify the requirement of MMP for proteolytic activation of PA3MzHER2-MMP, an in vitro cleavage assay will be performed using recombinant MMPs, such as MMP-2, MMP -9, and MMP-14 (commercially available from several sources), which are expected8123-112570-02

[0414] to proteolytically activate PA3MzHER2-MMP. PA3MzHER2-MMP and PA3MzHER2 will be incubated with recombinant MMP-2, MMP-9, soluble form of MMP-14, and soluble form of furin for various lengths of time. Then proteolytic processing of the PA proteins will be examined by Western blotting against PA using an anti-PA serum (Liu, Shihui, and Stephen H Leppla. “Cell surface tumor endothelium marker 8 cytoplasmic tail-independent anthrax toxin binding, proteolytic processing, oligomer formation, and internalization.” The Journal of biological chemistry vol. 278,7 (2003): 5227-34. doi:10.1074 / jbc.M210321200). It is expected that PA3MzHER2-MMP can be proteolytically processed into the 63 kD and 20 kD PA fragments by all these MMPs but not by furin, and that PA3MzHER2 can only be processed by furin but not the MMPs.

[0415] MMP-2, MMP-9, and MMP-14 knockout HT1080 cells will also be generated. These MMPs are often described as major players in cancer development. It is expected that these MMP-KO cells will have reduced sensitivity to PA3MzHER2-MMP but not to PA3MzHER2 in the presence FP59.

[0416] Example 8

[0417] In vitro cytotoxicity of HER2-targeting toxins to cancer cells with various oncogenic mutations

[0418] The cytotoxicity of the toxins will be further investigated in an extensive set of cancer cell lines, that harbor various cancer-driver mutations (e.g., in KRAS. NRAS, HRAS, BRAF. and PIK3CA) and are of different tissue origins (Table 2).

[0419] Table 2: Cancer cell lines (ATCC cell lines by gene mutation database) Cancer cell line Oncogenes Tumor suppressor genes SW620 KRAS p.G12V APC p.Q1338* (nonsense) (Colorectal carcinoma) (homo) TP53 p.R273H

[0420] TP53 p.P309S

[0421] HCT116 KRAS p.G13D CDKN2A p.R24fs*20 (Colorectal carcinoma) PIK3CA p.H1047R

[0422] Colo205 BRAF p.V600E APC p.T1556fs*3

[0423] (Colorectal carcinoma) TP53 p.Y103_Llll>L

[0424] HT29 BRAF p.V600E APC p.T1556fs*3

[0425] (Colorectal carcinoma) PIK3CA p.P449T TP53 p.R273H

[0426] SMAD4 pQ311*

[0427] RKO BRAF p.V600E Unknown

[0428] (Colorectal carcinoma) PIK3CA p.H1047R

[0429] AsPc-1 KRAS p.G12D TP53 p.C135fs*35

[0430] (Pancreatic carcinoma) (homo) CDKN2A p.L78fs*41

[0431]

[0432] Capan-2 KRAS p.G12V (het) Unknown8123-112570-02

[0433] (Pancreatic carcinoma)

[0434] MiaPaCa-2 KRAS p.G12C TP53 p.R248W (homo) (Pancreatic carcinoma) (homo) CDKN2A p.0? (homo) MDA-MB-231 KRAS p.G13D TP53 p.R280K (homo) (breast adenocarcinoma) BRAF p.G464V

[0435] AU565 Unknown TP53 p.R175H (homo) (breast adenocarcinoma)

[0436] BT474 (breast ductal) PIK3CAp.KlllN TP53 p.E285K

[0437] (het)

[0438] SK-BR-3 ERBB2 amplification Unknown

[0439] (breast adenocarcinoma)

[0440] SK-OV-3 ERBB2 amplification CDKN2A p.0? (homo) (Ovary adenocarcinoma) PIK3CA p.H1047R TP53 p.S90fs*33 (homo)

[0441] (het)

[0442] RL-95-2 HRAS p.Q61H(het) TP53 p.V73fs*50

[0443] (Uterus carcinoma) TP53 p.V218del

[0444] PTEN p.N323fs*2

[0445] PTEN p.N323fs*21

[0446] NCI-H358 (NSCLC) KRAS p.G12C (het) Unknown

[0447] NCI-H460 KRAS p.Q61H CDKN2A p.0?

[0448] (lung large cell carcinoma) (homo)

[0449] PIK3CA p.E545K

[0450] (het)

[0451] NCI-H1155 KRAS p.Q61H TP53 p.R273H (homo) (lung large cell carcinoma) (homo) PTEN p.R233*(homo) A549 (NSCLC) KRAS p.G12S (homo) CDKN2A p.0?

[0452] Calu-1 (NSCLC) KRAS p.G12C (het) Unknown

[0453] NCI-H2087 (NSCLC) NRAS p.Q61K(het) TP53 p.V157F (homo)

[0454] BRAF p. L597V

[0455] (het)

[0456] NCI-H2347 (NSCLC) NRAS p.Q61R Unknown

[0457] (homo)

[0458] NCI-H2030 (NSCLC) KRAS p.G12C CDKN2A p.0?

[0459] (homo)

[0460] CHP-212 (Neuroblastoma) NRAS p.Q61R Unknown

[0461] (homo)

[0462] HT1080 (Fibrosarcoma) NRAS p.Q61K(het) CDKN2A p.0? (homo) SNU-387 NRAS p.Q61K(het) TP53 p.K164*(homo) CDKN2A (hepatocellular carcinoma) p.0? (homo)

[0463] A2058 BRAF p.V600E TP53 p.V274F (Melanoma) PTENp.L112Q

[0464] PTEN p.V175fs*3

[0465] SK-MEL-2 (Melanoma) NRAS p.Q61K TP53 p.G245S (het)

[0466] (homo)

[0467] HT144 (Melanoma) BRAF p.V600E CDKN2A p.0?

[0468] PTEN p.?

[0469] C32 (Melanoma) BRAF p.V600E (het) CDKN2A p.0? (homo)

[0470]

[0471] CDKN2A p.*157S (homo)8123-112570-02

[0472] These cell lines are expected to express HER2 at various levels. These cells will be treated with various concentrations of PA3MzHER2-MMP (SEQ ID NO: 11 with 512D) and PA3MzHER2 (SEQ ID NO: 10) in the presence of various types of effector proteins, e.g.., LFn-DUF5-L418W (SEQ ID NO: 33) (RAS inhibition), LF-W271A (MEK inhibition), and FP59 (protein synthesis inhibition) (Table 1). It is expected that cancer cells with an oncogenic RAS mutation will be sensitive to the RAS inhibition, whereas cancer cells with a BRAFV600Emutation (initiating MEK-ERK signaling independent of RAS) will be insensitive to the RAS inhibition. Conversely, cancer cells with the BRAFV600Emutation will be more sensitive than the cells with oncogenic RAS mutations to the MEK inhibitor.

[0473] As shown in FIG. 7A, cancer cells with an oncogenic RAS mutation were sensitive to PA3MzHER2-MMP plus LFn-DUF5-L418W, but cancer cells with the BRAFV600Emutation were resistant to PA3MzHER2-MMP plus LFn-DUF5-L418W. As shown in FIG. 7B, regardless of their oncogenic mutations, all cancer cells tested were sensitive to PA3MzHER2-MMP plus FP59 (protein synthesis inhibition).

[0474] Example 9

[0475] In vitro cytotoxicity of HER2-targeting toxins compared to clinically used HER2- targeting agents

[0476] Since the effector proteins of the HER2 -targeting toxins are enzymes, functioning at sub-stoichiometric levels, it is expected that PA3MzHER2-MMP (SEQ ID NO: 11 with 512D) plus LFn-DUF5-L418W (SEQ ID NO: 33) (pan-RAS inactivator) or FP59 (protein synthesis inhibitor) will have much higher potency than trastuzumab and its drug conjugates (T-DXd and T-DM1) that are currently used in the clinic.

[0477] A side-by-side comparison of their cytotoxicity to various cancer cells with low HER2 expression was performed, including HT1080, MDA-MB-231, and MiaPaCa-2 cancer cells. Notably, in all cases, the HER2-targeting toxins exhibited much higher cytotoxicity than trastuzumab and its drug conjugates (FIGs. 8A-8C). The cytotoxicity to cancer cells with high HER2 expression (e.g., SK-OV3 and SK-BR3 cells) will be studied.8123-112570-02

[0478] Example 10

[0479] In vitro cytotoxicity of PA3MzHER2-MMP / LFn-DUF5-L418W compared to small molecule KRAS inhibitors

[0480] LFn-DUF5-L418W (SEQ ID NO: 33) is a RAS-specific endopeptidase that can efficiently cleave oncogenic RAS proteins after being delivered by PA3MzHER2-MMP (SEQ ID NO: 11 with 512D) into the cytosol of cancer cells (FIGs. 9A-9C). Based on its enzymatic nature, the toxin would have much higher potency than the current KRAS small molecule inhibitors.

[0481] The efficacy of LFn-DUF5-L418W plus PA3MzHER2-MMP was compared with the efficacies of KRASG12Callele-specific inhibitors Sotorasib and Adagrasib (FDA-approved), the KRASG12Dinhibitor MRTX1133, and the pan-KRAS inhibitor BI-2865 (both under preclinical evaluation), in various cancer cells including NCI-H358(KRASG12C), MiaPaCa-2 (KRASG12C), ASPC-1 (KRASG12D), MDA-MB-231 (KRASG13D), and HT1080 (NRASQ61K) cancer cells.

[0482] As expected, the HER2 -targeting toxin exhibited potent cytotoxicity to all tested human cancer cells with ECso < 0.1 nM. In contrast, the ECso of Adagasib or Sotorasib to KRASG12Ccancer cells was around 10 nM, and MRTX1133 and BI-2865 only showed variable and relatively modest potency (FIGs. 10A-10E). These results demonstrate that the HER2 -targeting toxin has potency to all cancer cells that rely on oncogenic RAS for survival irrespective of the specific mutations, and is superior to both the allele-specific and pan-KRAS small molecule inhibitors.

[0483] Example 11

[0484] In vivo antitumor activity of HER2-targeting toxins Human fibrosarcoma HT1080 (NRASQ61K) and human lung H358 (KRASG12C) tumor xenografts were established by subcutaneous injection of 5xl06HT1080 cells and 5xl06H358 cells, respectively, to mice. When tumors were fully established (> 100 mm3), the tumor-bearing mice were treated with PBS, or PA3MzHER2-MMP (SEQ ID NO: 11 with 512D) (30 pg) combined with LFn-DUF5-L418W (SEQ ID NO: 33) (10 pg) or FP59 (5 pg) via an intraperitoneally (I.P.) delivery route.

[0485] Notably, the novel toxin exhibited potent antitumor activity in both tumor models, and the mice tolerated toxin treatment for at least 6 weeks (twice per week) without displaying any obvious toxicity (FIGs. 11 A-l IB and 12A-12B). Strikingly, most of the HT1080 tumors (12 out of 15) were eliminated after four doses of the toxin treatment, while all the mice in8123-112570-02

[0486] PBS-treated group reached euthanasia criteria due to tumor burden (FIGs. HA and 1 IB). Similarly, H358 tumors were also responsive to the toxin treatment, and all the tumors remained minimum in size or were eliminated two weeks after the first treatment (FIGs. 12A and 12B). Therefore, the tumor-selective toxin demonstrated potent antitumor activity to tumors with RAS mutations both in vitro and in vivo.

[0487] Example 12

[0488] Remarkably high therapeutic index of PA3MzHER2-MMP / LFn-DUF5-L418W The excellent safety feature and high antitumor efficacy suggest that the therapeutic index (maximum tolerated dose / minimum effective antitumor dose) of PA3MzHER2-MMP / LFn-DUF5-L418W (SEQ ID NO: 11 with 512D / SEQ ID NO: 33) is high. To estimate its therapeutic index, PA3MzHER2-MMP / LFn-DUF5-L418W was injected in mice with HT1080 tumor xenografts at a low dose (10 pg / 3.3 pg) and at a high dose (90 pg / 30 pg) for 10 weeks (I.P., twice a week). Amazingly, the toxic demonstrated potent, comparable antitumor activity at both doses, and no signs of toxicity was observed even in mice treated with the high dose of the toxin (FIGs. 13 A and 13B). Therefore, the therapeutic index of the toxin is estimated to be at least 9 (90 pg / 30 pg 4- 10 pg / 3.3 pg = 9).

[0489] Example 13

[0490] Maximum tolerated dose of PA3MzHER2-MMP / LFn-DUF5-L418W Dose-escalation studies will be performed to determine the maximum tolerated dose (MTD6) of PA3MzHER2-MMP (SEQ ID NO: 11 with 512D) plus LFn-DUF5-L418W (SEQ ID NO: 33) or FP59 (3:1 ratio in weight) for six I.P. injections (at days 1, 3, 5, 7, 9, and 11) within two weeks. MTD6 is the maximum dosing that does not cause any death or obvious signs of malaise in a group of five mice receiving all six injections. Since mice were able to tolerate PA3MzHER2-MMP / LFn-DUF5-L418W with dosing of 90 pg / 30 pg, the studies will start with six injections of 100 pg / 33.3 pg of PA3MzHER2-MMP / LFn-DUF5-L418W using C57BL / 6J mice (Jackson Laboratory). Several rounds of dose-escalating will be performed until the MTD6 or 150 pg / 50 pg is reached. If no toxicity is observed at 150 pg / 50 pg, no further dose-escalation will be performed.

[0491] To identify potential tissue damage caused by the toxin administered beyond its MTD dose, the challenged C57BL / 6J mice will be sacrificed, and the major tissues or organs will be collected and processed for H&E staining for histology analyses, including brain, heart, kidney, lung, liver, GI tract, spleen, adrenal gland, femur, skin, mammary gland, salivary8123-112570-02

[0492] gland, thyroid gland, muscle, pancreas, gall bladder, bladder, ovaries, and uterus. Blood will also be collected and processed for a complete blood count and a limited blood-chemistry panel to examine alterations in alkaline phosphatase, alanine aminotransferase, aspartate aminotransferase, creatinine kinase, and lactate dehydrogenase levels. These studies will provide insight into the potential dose-limiting toxicity caused by RAS signaling inhibition.

[0493] Example 14

[0494] Bio-distribution of HER2- targeting toxins

[0495] Since the toxin functions by binding to and entering target cells to cleave RAS proteins, it is expected that it will quickly disappear from the circulation. Tissue distribution and potential off-target toxicities of the toxin will be determined in HT1080 tumor-bearing mice as used above. When HT1080 tumors reach -500 mm3, the tumor-bearing mice will be injected a single dose (I.P.) of PBS, PA3MzHER2-MMP / LFn-DUF5-L418W (SEQ ID NO: 11 with 512D / SEQ ID NO: 33) (30 pg / 10 pg), or PA3MzHER2 / LFn-DUF5-L418W (SEQ ID NO: 10 / SEQ ID NO: 33) (30 pg / 10 pg) (n = 12, male and female mice per group). Mice will be euthanized by CO2 inhalation at 6 h, 24 h, and 72 h (n = 4 for each time point).

[0496] Tumors and major organs (brain, heart, kidney, lung, liver, GI tract, and spleen) will be immediately removed. Tissue lysates will be analyzed by Western blotting using anti-pan-RAS, -KRAS, -NRAS, -HRAS, or phospho-ERK antibodies. These tissue lysates will also be analyzed by Western blotting using anti-PA and anti-LFn antibodies. It is expected that the RAS cleavage and the ERK pathway disruption mediated by the MMP- and HER2-dual targeted toxin will mostly occur in tumors and not in normal tissues, whereas the activity mediated by the HER2-only targeted toxin will be seen in tumor tissues and, to a lesser extent, certain normal tissues. These studies will also inform whether the ERK pathway disruption by a single toxin administration will last beyond the period of 72 hours, based on which dosing intervals can be further determined.

[0497] Example 15

[0498] Further studies of in vivo antitumor activity of PA3MzHER2-MMP / LFn-DUF5-L418W Because of the availability of many well-defined human cancer cell lines, human tumor xenografts are convenient and informative models for testing the responsiveness of cancer cells with various genetic mutations. The anti-tumor activity of PA3MzHER2-MMP / LFn-DUF5-L418W (SEQ ID NO: 11 with 512D / SEQ ID NO: 33) will be ftirther8123-112570-02

[0499] examined in various human cancer xenografts in C57BL / 6J athymic nude (Foxnlnu / nu) or NOD / SCID / gamma (NSG) mice (The Jackson Laboratory).

[0500] Tumors will be established from cancer cells having oncogenic RAS mutations, e.g., human pancreatic carcinomas AsPc-l(KRASG12D), Capan-2 (KRASG12V), and MiaPaCa-2 (KRASG12C); human breast carcinoma MDA-MB-231(KRASG13D); and human lung carcinoma NCI-H1155 (KRASQ61H) (Liu, Shihui et al. “Solid tumor therapy by selectively targeting stromal endothelial cells.” Proceedings of the National Academy of Sciences of the United States of America vol. 113,28 (2016): E4079-87. doi:10.1073 / pnas.1600982113; Liu, Shihui et al. “Matrix metalloproteinase-activated anthrax lethal toxin demonstrates high potency in targeting tumor vasculature.” The Journal of biological chemistry vol. 283,1 (2008): 529-540. doi:10.1074 / jbc.M707419200: Schafer, Jeffrey M et al. “Efficient targeting of head and neck squamous cell carcinoma by systemic administration of a dual uPA and MMP-activated engineered anthrax toxin.” PloS one vol. 6,5 (2011): e20532. doi:10.1371 / joumal.pone.0020532). To establish human tumor xenografts, 5xl06cells per mouse will be injected intradermally into athymic nude or NSG mice. When tumors reach -100 mm3in size, the tumor-bearing mice will be randomized into groups (10 mice per group) and injected (I.P.) with PBS or various doses of the toxin (< 1 / 2 of the MTD) twice a week for eight weeks. To benchmark antitumor activity of the toxin against small molecule KRAS inhibitors, in some experiments, tumor-bearing mice will also be treated with Sotorasib (10 mg / kg body weight), Adagrasib (10 mg / kg), MRTX1133 (10 mg / kg), or BI-2865 (90 mg / kg) by daily oral gavage for eight weeks.

[0501] It is expected that the toxin will be safe and have potent antitumor activity in all these tumor models, whereas the KRAS allele-specific inhibitors Sotorasib (KRASG12C), Adagrasib (KRASG12C), and MRTX1133 (KRASG12D) will only be active to MiaPaCa-2 (KRASG12C) or AsPc-l(KRASG12D) tumors. BI-2865 (pan-KRAS inhibitor) may have efficacies to a wider range of these tumors. Due to the enzymatic nature, consistent with the results demonstrated in in vitro studies (Example 10) and in vivo studies of human fibrosarcoma HT1080 (NRASQ61K) and human lung H358 (KRASG12C) tumor xenografts (Example 11), the toxin is expected to have higher potency than all small molecule inhibitors in all xenograft models tested.

[0502] Tumor size and bodyweight will be monitored daily. Tumors will be measured with digital calipers (FV Fowler); tumor volumes will be estimated using the formula: tumor volume (mm3) = i (length x width x height in mm). To study the tumor response to treatment, tumors from the control group and the toxin treated groups (n = 4) will be8123-112570-02

[0503] dissected, fixed in 10% formalin and embedded in paraffin. Tumor samples will be analyzed by a variety of parameters including tumor histology by H&E staining, apoptosis by TUNEL staining, tumor cell proliferation by Ki-67 staining, and levels of tumor angiogenesis by CD31 staining (Liu, Shihui et al. “Solid tumor therapy by selectively targeting stromal endothelial cells.” Proceedings of the National Academy of Sciences of the United States of America vol. 113,28 (2016): E4079-87. doi: 10.1073 / pnas.1600982113; Liu, Shihui et al. “Matrix metalloproteinase-activated anthrax lethal toxin demonstrates high potency in targeting tumor vasculature.” The Journal of biological chemistry vol. 283,1 (2008): 529-540. doi:10.1074 / jbc.M707419200; Liu, Shihui et al. “Potent antitumor activity of a urokinase-activated engineered anthrax toxin.” Proceedings of the National Academy of Sciences of the United States of America vol. 100,2 (2003): 657-62. doi: 10.1073 / pnas.0236849100; Peters, Diane E et al. “Comparative toxicity and efficacy of engineered anthrax lethal toxin variants with broad anti-tumor activities.” Toxicology and applied pharmacology vol. 279,2 (2014): 220-9. doi:10.1016 / j.taap.2014.06.010).

[0504] Example 16

[0505] In vivo antitumor activity of PA3MzHER2-MMP / FP59 in human breast cancer xenografts

[0506] Human breast carcinoma SK-BR3 and BT474 cells that express high levels of HER2 will be used to establish orthotopic tumors. 10-12 weeks old female NSG mice will be injected unilaterally with 2.5xl06cells in 200 pL of 50=50 matrigel / collagen I into the fourth mammary fat pad at the base of the nipple. When tumors reach ~50 mm3in size, the tumorbearing mice will be randomized into groups (10 mice per group) and injected (I.P.) with PBS or PA3MzHER2-MMP (SEQ ID NO: 11 with 512D) / FP59 (30 pg / 5 pg) twice a week for eight weeks. Since both SK-BR3 and BT474 tumors are HER2 (gene amplification)-driven tumors, and no oncogenic RAS or BRAF mutations are identified, FP59 will be used as the effector. Because these tumors are expected to be sensitive to trastuzumab, T-DXd, and T-DM1, side-by-side comparisons of the antitumor efficacies of the toxin and these FDA-approved drugs will be performed. Thus, tumor-bearing mice will also be treated with T-DXd (10 mg / Kg) or T-DM1 (15 mg / Kg) intravenously once per 2 weeks.8123-112570-02

[0507] Example 17

[0508] In vivo antitumor activity of PA3MzHER2-MMP / LFn-DUF5-L418W in patient-derived xenograft (PDX) models

[0509] PDX tumors from pancreatic cancer patients will be analyzed, which more faithfully recapitulates the heterogeneity, histology, and molecular alterations of patient tumors, and are particularly useful for developing novel targeted therapies based on genetic profiles.

[0510] Pancreatic and colorectal PDX models will be analyzed, because approximately 90% and 40%, respectively, of these human cancers contain oncogenic KRAS mutations. Frozen PDX tumors will be implanted into 4-6-week-old anesthetized (NOD / SCID gamma) mice. Once tumors reach 1 cm3in size, they will be dissected and cut into 10-15 pieces, each of which will be implanted into a different NSG mouse. When tumors reach -100 mm3, the PDX-bearing mice will be grouped (n=10 per group, both male and female) and treated with PBS or various doses of PA3MzHER2-MMP / LFn-DUF5-L418W (SEQ ID NO: 11 with 512D / SEQ ID NO: 33) twice a week for eight weeks.

[0511] Example 18

[0512] In vivo antitumor activity of PA3MzHER2-MMP / LFn-DUF5-L418W in genetically engineered mouse tumor model

[0513] Certain genetically engineered mice can develop spontaneous, autochthonous tumors that more faithfully mimic the genetic and biological evolution of human cancers. KPC (KrasLSL G12D; p53LoxP; Pdxl (pancreas-specific promoter)-CreER triple mutant) is a well-established model of tamoxifen-inducible pancreatic ductal adenocarcinoma (PDAC). KPC mice are commercially available from the Jackson Laboratory (#032429). Oral gavage of tamoxifen to breast-feeding mothers can induce expression of oncogenic KrasG12Dand deletion of p53 in pups, leading to their development of PDAC by 12 weeks of age with a high penetration rate (> 90%). Tumor-bearing mice will succumb to tumor burden within 24 weeks of age. We will use this PDAC tumor model to examine the antitumor activity of the toxin. The tumor-bearing mice (12 weeks of age, 10 mice per group, both male and female) will be randomized into groups treated with PBS or various doses of PA3MzHER2-MMP / LFn-DUF5-L418W (SEQ ID NO: 11 with 512D / SEQ ID NO: 33) and survival monitored. Tumor-bearing mice will be euthanized and tumors dissected after 8 weeks of treatment and tumor burden (size and number) will be determined and compared among groups. In this immunocompetent setting, long-term therapies with the toxin may induce neutralizing antibodies that may, over time, block therapeutic activity. A8123-112570-02

[0514] pentostatin / cyclophosphamide (PC) regimen can safely block the neutralizing antibody production against therapeutic toxins. Therefore, in a separate study, PC will be coadministered with PA3MzHER2-MMP / LFn-DUF5-L418W to evaluate the long-term (> 10 weeks) therapeutic effects of the toxin.

[0515] Example 19

[0516] Antitumor and side effects of DUF5 in mice

[0517] To test the antitumor activity of LFn-DUF5 / IC3-PA, Lewis lung carcinoma (LLC) KrasGI2C) syngeneic tumors were established in C57BL / 6J mice (both male and female) by subcutaneous injection of IxlO6LLC (KrasG12C) cells. When LLC tumors reached > 100 mm3, the tumor-bearing mice were treated with PBS or LFn-DUF5 / IC3-PA.

[0518] IC3-PA comprises PA-L1-I207R / I656Q and PA-U2-R200A / I656Q, which are activated by MMPs and uPA, respectively. Upon proteolytic activation, PA-L1-I207R / I656Q and PA-U2-R200A / I656Q form oligomers and gain the capacity to bind LF or LF variants. Each active LF-binding site is formed by two subsites (1207 and R200) on two adjacent hetero PA protomers PA-L1-1207R / 1656Q and PA-U2-R200A / 1656Q, each providing one subsite, i.e. R200 and 1207, respectively. IC3-PA requires simultaneous presence of the three tumor markers MMPs, urokinase, and CMG2 for its cytotoxic action. IC3-PA selectively targets CMG2 receptor over TEM8 receptor, because the I656Q mutation reduces the ability of the PA to bind TEM8. IC3-PA is disclosed in Zuo, Zehua et al., “A potent tumor-selective ERK pathway inactivator with high therapeutic index,” PNAS nexus vol. 1,3 pgacl04. 1 Jul.

[0519] 2022, doi:10.1093 / pnasnexus / pgacl04, the entire disclosure of which is incorporated by reference herein.

[0520] Although LFn-DUF5 / IC3-PA exhibited potent antitumor activity in this tumor model (FIG. 14A), a fraction of the tumor-bearing mice (4 out 9) displayed abdominal edema following repetitive intraperitoneal administration (FIG. 14B). This was likely due to DUF5’s undesired activity towards RAPL since RAP1 has an established role in maintaining vascular integrity by regulating intercellular tight junction.

[0521] Example 20

[0522] Identification, and antitumor and minimal side effects of RAS-selective DUF5 variants Based on the known important function of RAP1 in regulating intercellular tight junction and its role in maintaining vascular integrity, it is reasoned that the undesirable8123-112570-02

[0523] edema or vascular leakage caused by DUF5 was due to its activity toward RAP1.

[0524] Experiments were thus performed to identify DUF5 variants that selectively target RAS without affecting RAPE

[0525] B. anthracis BH500 host strain was used to express and secrete high levels of LFn-DUF5 variant proteins into culture supernatants in their active form. LFn-DUF5’s proteolytic activity to RAS and RAP1 proteins can be detected using culture supernatants of BH500 host strains, allowing high throughput functional analysis. To test this, various dilutions of BH500 (LFn-DUF5) culture supernatant (containing secreted LFn-DUF5) were incubated with HCT116 cell lysates prepared by RIPA lysis buffer. Remarkably, both RAS and RAP1 proteins were efficiently cleaved, as shown by Western blotting using anti-pan RAS or anti-RAP1 antibody (FIG. 15). This efficient protein expression and testing system was then used to screen relative RAS versus RAP1 proteolytic activity of LFn-DUF5 variants in a high throughput manner.

[0526] DUF5 has 509 amino acids (numbered aa 1-509, corresponding to aa 3581-4089 of MARTX), forming an N-temiinal lobe (N lobe, aa 1-179) and a C-terminal lobe (C lobe, aa 180-509) (FIG. 16). The N lobe is a finger-like structure, responsible for attaching to the inner membrane of the cell, thereby co-localizing with its substrates RAS and RAP1, both of which function at the cytoplasmic face of the plasma membrane. The C lobe, responsible for substrate recognition and catalytic activity, contains two tightly associated Cl (aa 180-275) and C2 domains (aa 276-509). To identify LFn-DUF5 variants with selectivity to RAS over RAP1, a comprehensive alanine-scan mutagenesis screen was performed, covering all residues of the C lobe.

[0527] 330 LFn-DUF5 Ala variants covering the entire Cl (aa 180-275) and C2 (aa 276-509) domains were generated, by changing each of the residues to Ala, or to Ser when the naturally occurring amino acid is Ala. The expression plasmids of these LFn-DUF5 Ala variants were constructed using the In-Fusion technology strategy, verified by Sanger DNA sequencing, and transformed into BH500 allowing secretion of the variant proteins into the supernatant. The LFn-DUF5 variant protein containing supernatants were then analyzed for proteolytic activity towards RAS and RAP1 proteins as described above.

[0528] The Ala-scan mutagenesis identified residues in both the Cl and C2 domains that are key to DUF5’s proteolytic activity to both RAS and RAP1, including the catalytic residues — the two Glu-His pairs (Glu320-His322, Glu^-His450) located in C2. The Ala-scan also identified a disordered loop in DUF5 structure (amino acids Leu401-Gin421), which is adjacent to the catalytic residues (Glu320-His322, and Glu3:,0-His450), and is also functionally important8123-112570-02

[0529] to DUF5. One of the variants in this loop, DUF5 R408A, is of particular interest, because while it retains proteolytic activity toward RAS, its activity toward RAP1 is greatly reduced (Table 3). It is hypothesized that this loop, adjacent to the catalytic site, may play a role in substrate recognition.

[0530] Table 3: Ala-scan and Arg-scan mutagenesis to identify LFn-DUF5 variants selective for RAS over RAP1

[0531] Ala-scan variant Arg-scan variant Amino acids Leu

[0532] V401-Arg ariants421

[0533] relative activity relative activity Ala or Arg-scan

[0534] (RAS vs. RAP1) (RAS vs. RAP1) LFN-DUF5 LDANSSARPNVQGTEHGLMYR

[0535] active to both active to both (WT) (SEQ ID NO: 39)

[0536] LFn-DUF5 XDANSSARPNVQGTEHGLMYR

[0537] active to both No activity (L401) (SEQ ID NO: 40)

[0538] LFn-DUF5 LXANSSARPNVQGTEHGLMYR

[0539] active to both No activity (D402) (SEQ ID NO: 41)

[0540] LFn-DUF5 LDXNSSARPNVQGTEHGLMYR

[0541] active to both active to both (A403) (SEQ ID NO: 42)

[0542] LFn-DUF5 LDAXSSARPNVQGTEHGLMYR

[0543] active to both active to both (N404) (SEQ ID NO: 43)

[0544] LFn-DUF5 LDANXSARPNVQGTEHGLMYR

[0545] active to both active to both (S405) (SEQ ID NO: 44)

[0546] LFn-DUF5 LDANSXARPNVQGTEHGLMYR

[0547] active to both No activity (S406) (SEQ ID NO: 45)

[0548] LFn-DUF5 LDANSSXRPNVQGTEHGLMYR

[0549] active to both Partial to both (A407) (SEQ ID NO: 46)

[0550] LFn-DUF5 LD ANS S AXPN VQGTEHGLM YR

[0551] RAS > RAP1 RAS > RAP1 (R408) (SEQ ID NO: 47)

[0552] LFn-DUF5 LDANSSARXNVQGTEHGLMYR

[0553] active to both RAS > RAP1 (P409) (SEQ ID NO: 48)

[0554] LFn-DUF5 LDANSSARPXVQGTEHGLMYR

[0555] active to both active to both (N410) (SEQ ID NO: 49)

[0556] LFn-DUF5 LDANSSARPNXQGTEHGLMYR

[0557] active to both Partial to both (V411) (SEQ ID NO: 50)

[0558] LFn-DUF5 LDANSSARPNVXGTEHGLMYR

[0559] active to both active to both (Q412) (SEQ ID NO: 51)

[0560] LFn-DUF5 LDANSSARPNVQXTEHGLMYR

[0561] No activity active to both (G413) (SEQ ID NO: 52)

[0562] LFn-DUF5 LDANSSARPNVQGXEHGLMYR

[0563] active to both Partial to both (T414) (SEQ ID NO: 53)

[0564] LFn-DUF5 LDANSSARPNVQGTXHGLMYR

[0565] active to both active to both (E415) (SEQ ID NO: 54)

[0566] LFn-DUF5 LDANSSARPNVQGTEXGLMYR

[0567] Partial to both Partial to both (H416) (SEQ ID NO: 55)

[0568] LFn-DUF5 LDANSSARPNVQGTEHXLMYR

[0569] No activity Partial to both

[0570]

[0571] (G417) (SEQ ID NO: 56)8123-112570-02

[0572] LFn-DUF5 LD ANS S ARPNVQGTEHGXMYR

[0573] No activity RAS » RAP1 (L418) (SEQ ID NO: 57)

[0574] LFn-DUF5 LDANSSARPNVQGTEHGLXYR

[0575] active to both active to both (M419) (SEQ ID NO: 58)

[0576] LFn-DUF5 LDANSSARPNVQGTEHGLMXR

[0577] No activity No activity (Y420) (SEQ ID NO: 59)

[0578] LFn-DUF5 LD ANS S ARPNVQGTEHGLMYX

[0579] No activity Partial to both

[0580]

[0581] (R421) (SEQ ID NO: 60)

[0582] The substrate sequences of RAP 1 and RAS share high similarity around the cleavage site by DUF5 (between Y32 and D33), except that residue 31 is a positively charged Lys in RAP1, and a negatively charged Glu in RAS (FIG. 17). To identify DUF5 variants that distinguish RAP1 and RAS, a negatively-charged Arg-scan mutagenesis for the loop of Leu401-Arg421was performed, by changing each of the residues to Arg, or to Lys when the naturally occurring amino acid is Arg. This Arg-scan mutagenesis identified DUF5 (L418R) as an additional RAS-selective variant (Table 3). To further optimize its RAS versus RAP1 selectivity, L418 was changed to other 17 amino acids (Table 4), and DUF5 L418W was identified as highly RAS-selective (FIG. 18).

[0583] Table 4. Optimization of RAS-selective DUF5 variants by changing L418 to other 17 amino acids

[0584] DUF5 variant Relative activity

[0585] (RAS vs. RAP1)

[0586] 1 L418F RAS>RAP1

[0587] 2 L418I active to both

[0588] 3 L418M active to both

[0589] 4 L418V active to both

[0590] 5 L418S Inactive

[0591] 6 L418P Inactive

[0592] 7 L418T Partial to both

[0593] 8 L418H instable

[0594] 9 L418G Inactive

[0595] 10 L418D Inactive

[0596] 11 L418K

[0597] 12 L418N Inactive

[0598] 13 L418E Inactive

[0599] 14 L418C RAS » RAP1

[0600] 15 L418W RAS » RAP1

[0601] 16 L418G Inactive

[0602]

[0603] 17 L418Y Partial to both8123-112570-02

[0604] LFn-DUF5 L418W protein was purified for further analyses. To verify whether LFn-DUF5 L418W has reduced edema-causing activity, LFn-DUF5 L418W / IC3-PA or LFn-DUF5(WT) / IC3-PA was injected (I.P.) into mice. Surprisingly, the mice challenged with LFn-DUF5 L418W / IC3-PA developed much lower levels of abdominal edema compared to the mice injected with LFn-DUF5(WT) / IC3-PA (FIG. 19). Therefore, as demonstrated herein, DUF5 L418W is a RAS-selective variant with much reduced edema-causing activity compared to wildtype DUF5. DUF5 L418W also has potent antitumor activity. In vivo toxicity and antitumor activity of LFn-DUF5 L418W / PA3MzHER2-MMP were evaluated. In one study, human pancreatic carcinoma AsPc-1 (KRASG12D) tumor-bearing mice were treated with PA3MzHER2-MMP / LFn-DUF5 L418W (30 pg / 10 pg) twice a week for 16 weeks. While exhibiting potent antitumor activity, the toxin did not show obvious toxicity to mice (FIGs. 20A-20B).

[0605] Example 21

[0606] Potent antitumor activity of reengineered toxin

[0607] Human pancreatic carcinoma AsPc-1 (KRASG12D / G12D) tumor-bearing mice were treated with i) PBS; ii) RMC-6236 (25 mg / kg, oral gavage, 5 days per week), then switched to PA3MzHER2-MMP / LFn-DUF5-L418W (SEQ ID NO: 11 with 512D / SEQ ID NO: 33) at day 24 (30 pg / 10 pg, intravenously (I.V.) at days indicated by the downward-pointing arrows); iii) PA3MzHER2-MMP / LFn-DUF5-L418W (30 pg / 10 pg, intraperitoneally (I.P.), three times per week); or iv) PA3MzHER2-MMP / LFn-DUF5-L418W (30 pg / 10 pg, I.V. at days indicated by the upward-pointing arrows).

[0608] Complete tumor regression was achieved in all mice treated by

[0609] PA3MzHER2-MMP / LFn-DUF5-L418W (I.P. or I.V.) by day 24 (FIGs. 21A-21B), demonstrating strong antitumor efficacy of the reengineered toxin. Although RMC-6236 initially showed antitumor activity, the tumors gradually became less responsive to the drug, as evidenced by the increase in tumor volume starting around day 10 after the first treatment (FIGs. 21 A-21B). Remarkably, when the RMC-6236 treatment was switched to PA3MzHER2-MMP / LFn-DUF5-L418W (I.V.) at day 24, the tumors responded quickly and robustly, as evidenced by the significant and continuous decrease in tumor volume following the switch (FIGs. 21 A-21B). This shows that the toxin approach is effective even for drug-resistant tumors. No apparent adverse effects were observed in any treatment group.

Claims

8123-112570-02We claim:

1. A recombinant DUF5 protein comprising a L418W substitution, wherein the amino acid numbering is according to a reference DUF5 protein sequence set forth as SEQ ID NO: 29 [WT DUF5].

2. The recombinant DUF5 protein of claim 1, comprising an amino acid sequence at least 90% identical to SEQ ID NO: 30 [DUF5-L418W no signal peptide], wherein the recombinant DUF5 protein has the L418W substitution.

3. The recombinant DUF5 protein of claim 1, comprising an amino acid sequence at least 95% identical to SEQ ID NO: 30 [DUF5-L418W no signal peptide], wherein the recombinant DUF5 protein has the L418W substitution.

4. The recombinant DUF5 protein of claim 1, comprising the amino acid sequence set forth as SEQ ID NO: 30 [DUF5-L418W no signal peptide].

5. The recombinant DUF5 protein of any one of the prior claims, wherein the recombinant DUF5 protein is rat sarcoma virus (RAS)-selective.

6. The recombinant DUF5 protein of claim 5, wherein the RAS proteins comprise KRAS, NRAS, and HRAS.

7. The recombinant DUF5 protein of any one of the prior claims, fused to a heterologous protein.

8. The recombinant DUF5 protein of claim 7, wherein the heterologous protein is a domain of Anthrax lethal factor that binds to Anthrax protective antigen (LFn).

9. The recombinant DUF5 protein of claim 8, wherein the LFn comprises an amino acid sequence at least 90% identical to SEQ ID NO: 31 [LFn no signal peptide],10. The recombinant DUF5 protein of claim 8, wherein the LFn comprises an amino acid sequence at least 95% identical to SEQ ID NO: 31 [LFn no signal peptide].8123-112570-0211. The recombinant DUF5 protein of claim 8, wherein the LFn comprises the amino acid sequence set forth as SEQ ID NO: 31 [LFn no signal peptide].

12. The recombinant DUF5 protein of claim 8, comprising the amino acid sequence set forth as SEQ ID NO: 33 [LFn-DUF5 L418W no signal peptide].

13. A protein complex comprising the recombinant DUF5 protein of any one of claims 8-12, and a recombinant Anthrax protective antigen (PA).

14. The protein complex of claim 13, wherein the recombinant PA comprises: a mutation to block binding to PA receptor; anda mutation to change the PA furin cleavage site to a matrix metalloproteinase (MMP) cleavage site or a urokinase-type plasminogen activator (uPA) cleavage site.

15. The protein complex of claim 14, wherein the mutation to block binding to PA receptor comprises D683A, L685E, and Y688K substitutions, wherein the amino acid numbering is according to a reference PA protein sequence set forth as SEQ ID NO: 1 [WT PA],16. The protein complex of claim 14 or 15, wherein the MMP cleavage site comprises the amino acid sequence set forth as GPLGMLSQ (SEQ ID NO: 3) and the uPA cleavage site comprises the amino acid sequence set forth as PGSGRSA (SEQ ID NO: 4).

17. The protein complex of any one of claims 13-16, wherein the recombinant PA is fused to a HER2 binding domain, a HER3 binding domain, or an EGFR binding domain.

18. The protein complex of claim 17, whereinthe HER2 binding domain comprises a HER2 specific affibody comprising the amino acid sequence set forth as SEQ ID NO: 5 [zHER2:343]:the HER3 binding domain comprises a HER3 specific affibody comprising the amino acid sequence set forth as SEQ ID NO: 6 [zHER3:8699]; orthe EGFR binding domain comprises an EGFR specific affibody comprising the amino acid sequence set forth as SEQ ID NO: 7 [zEGFR2377].8123-112570-0219. The protein complex of claim 17 or 18, whereinthe recombinant PA fused to a HER2 binding domain comprises an amino acid sequence that is at least 99% identical to SEQ ID NO: 11, SEQ ID NO: 61 or SEQ ID NO: 12 [PA3MzHER2-Ll, PA3MzHER2-Ll, PA3MzHER2-U2, no signal peptide];the recombinant PA fused to a HER3 binding domain comprises an amino acid sequence that is at least 99% identical to SEQ ID NO: 13, SEQ ID NO: 62, or SEQ ID NO: 14 [PA3MzHER3-Ll, PA3MzHER3-Ll, PA3MzHER3-U2, no signal peptide]; orthe recombinant PA fused to an EGFR binding domain comprises the amino acid sequence that is at least 99% identical to SEQ ID NO: 15, SEQ ID NO: 63, or SEQ ID NO: 16 [PA3MzEGFR-Ll, PA3MzHER3-Ll, PA3MzEFGR-U2, no signal peptide].

20. The protein complex of claim 13, comprising:i) a first recombinant PA comprising a mutation to change the PA furin cleavage site to a MMP cleavage site, and a I207R substitution, and a second recombinant PA comprising a mutation to change the PA furin cleavage site to a uPA cleavage site, and a R200A substitution; orii) a first recombinant PA comprising a mutation to change the PA furin cleavage site to a MMP cleavage site, and K245G and R252N substitutions, and a second recombinant PA comprising a mutation to change the PA furin cleavage site to a uPA cleavage site, and a D512K substitution,wherein the amino acid numbering is according to a reference PA protein sequence set forth as SEQ ID NO: 1 [WT PA],21. The protein complex of claim 20, wherein the first and second recombinant PAs further comprise a mutation to block binding to PA receptor.

22. The protein complex of claim 21, wherein the mutation to block binding to PA receptor comprises D683A, L685E, and Y688K substitutions.

23. The protein complex of any one of claims 20-22, wherein the MMP cleavage site comprises the amino acid sequence set forth as GPLGMLSQ (SEQ ID NO: 3), and the uPA cleavage site comprises the amino acid sequence set forth as PGSGRSA (SEQ ID NO: 4).8123-112570-0224. The protein complex of any one of claims 20-23, whereinthe first recombinant PA is fused to a HER2 binding domain, a HER3 binding domain, or an EGFR binding domain; and / orthe second recombinant PA is fused to a HER2 binding domain, a HER3 binding domain, or an EGFR binding domain.

25. The protein complex of claim 24, whereinthe HER2 binding domain comprises a HER2 specific affibody comprising the amino acid sequence set forth as SEQ ID NO: 5 [zHER2:343];the HER3 binding domain comprises a HER3 specific affibody comprising the amino acid sequence set forth as SEQ ID NO: 6 [zHER3:8699]; orthe EGFR binding domain comprises an EGFR specific affibody comprising the amino acid sequence set forth as SEQ ID NO: 7 [zEGFR2377].

26. The protein complex of any one of claims 20-25, whereinthe first recombinant PA isi) fused to a HER2 binding domain and comprises the amino acid sequence set forth as SEQ ID NO: 17 [PA3MzHER2-Ll I207R, no signal peptide];ii) fused to a HER3 binding domain and comprises the amino acid sequence set forth as SEQ ID NO: 18 [PA3MzHER3-Ll I207R, no signal peptide]; oriii) fused to an EGFR binding domain comprises the amino acid sequence set forth as SEQ ID NO: 19 [PA3MzEGFR-Ll I207R, no signal peptide]; andthe second recombinant PA isi) fused to a HER2 binding domain and comprises the amino acid sequence set forth as SEQ ID NO: 20 [PA3MzHER2-U2 R200A, no signal peptide];ii) fused to a HER3 binding domain and comprises the amino acid sequence set forth as SEQ ID NO: 21 [PA3MzHER3-U2 R200A, no signal peptide]; oriii) fused to an EGFR binding domain comprises the amino acid sequence set forth as SEQ ID NO: 22 |PA3MzEGFR-U2 R200A, no signal peptide].

27. The protein complex of any one of claims 20-25, whereinthe first recombinant PA isi) fused to a HER2 binding domain and comprises the amino acid sequence set forth as SEQ ID NO: 23 [PA3MzHER2-Ll GN, no signal peptide];8123-112570-02ii) fused to a HER3 binding domain and comprises the amino acid sequence set forth as SEQ ID NO: 24 [PA3MzHER3-Ll GN, no signal peptide]; oriii) fused to an EGFR binding domain comprises the amino acid sequence set forth as SEQ ID NO: 25 [PA3MzEGFR-Ll GN, no signal peptide]; andthe second recombinant PA isi) fused to a HER2 binding domain and comprises the amino acid sequence set forth as SEQ ID NO: 26 [PA3MzHER2-U2 D512K, no signal peptide];ii) fused to a HER3 binding domain and comprises the amino acid sequence set forth as SEQ ID NO: 27 [PA3MzHER3-U2 D512K, no signal peptide]; oriii) fused to an EGFR binding domain comprises the amino acid sequence set forth as SEQ ID NO: 28 [PA3MzEGFR-U2 D512K, no signal peptide].

28. A recombinant fusion protein comprising recombinant Anthrax PA and a HER2 affibody, comprising an amino acid sequence that is at least 99% identical to SEQ ID NO: 11, SEQ ID NO: 12, SEQ ID NO: 17, SEQ ID NO: 20, SEQ ID NO: 23, SEQ ID NO: 26, or SEQ ID NO: 61.

29. A recombinant fusion protein comprising recombinant Anthrax PA and a HER3 affibody, comprising an amino acid sequence that is at least 99% identical to SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 18, SEQ ID NO: 21, SEQ ID NO: 24, SEQ ID NO: 27, or SEQ ID NO: 62.

30. A recombinant fusion protein comprising recombinant Anthrax PA and an EGFR affibody, comprising an amino acid sequence that is at least 99% identical to SEQ ID NO: 15, SEQ ID NO: 16, SEQ ID NO: 19, SEQ ID NO: 22, SEQ ID NO: 25, SEQ ID NO: 28, or SEQ ID NO: 63.

31. A recombinant nucleic acid molecule encoding the recombinant DUF5 protein, the protein complex, or the recombinant fusion protein of any one of claims 1-30.

32. A vector comprising the recombinant nucleic acid molecule of claim 31.8123-112570-0233. A pharmaceutical composition, comprising the recombinant DUF5 protein, the protein complex, the recombinant fusion protein, the nucleic acid molecule, or the vector, of any one of claims 1-32.

34. A pharmaceutical composition, comprising the recombinant DUF5 protein of any one of claims 1-12 or a nucleic acid molecule encoding the recombinant DUF5 protein.

35. A pharmaceutical composition, comprising the protein complex of any one of claims 11-27 or one or more nucleic acid molecules encoding the protein complex.

36. A pharmaceutical composition, comprising the fusion protein of any one of claims 28-30 or a nucleic acid molecule encoding the fusion protein.

37. A kit, comprising:the recombinant DUF5 protein of any one of claims 8-12 or a nucleic acid molecule encoding the recombinant DUF5 protein, andthe recombinant fusion protein of any one of claims 28-30 or a nucleic acid molecule encoding the recombinant fusion protein.

38. The kit of claim 37, comprising:the recombinant DUF5 protein or the nucleic acid molecule encoding the recombinant DUF5 protein, andi) the recombinant fusion protein comprising SEQ ID NO: 17, SEQ ID NO: 18, or SEQ ID NO: 19 or the nucleic acid molecule encoding the recombinant fusion protein, and the recombinant fusion protein comprising SEQ ID NO: 20, SEQ ID NO: 21, or SEQ ID NO: 22 or the nucleic acid molecule encoding the recombinant fusion protein; orii) the recombinant fusion protein comprising SEQ ID NO: 23, SEQ ID NO: 24, or SEQ ID NO: 25 or the nucleic acid molecule encoding the recombinant fusion protein, and the recombinant fusion protein comprising SEQ ID NO: 26, SEQ ID NO: 27, or SEQ ID NO: 28 or the nucleic acid molecule encoding the recombinant fusion protein.

39. A method, comprising:8123-112570-02producing a recombinant DUF5 protein by expressing a nucleic acid molecule encoding the recombinant DUF5 protein of any one of claims 1-12 in a host cell and purifying the recombinant DUF5 protein;producing a protein complex comprising a recombinant DUF5 protein and an Anthrax PA protein by expressing one or more nucleic acid molecules encoding the recombinant DUF5 protein and the Anthrax PA protein of the protein complex of any one of claims 13-27 in a host cell and purifying the protein complex;producing a recombinant fusion protein comprising recombinant Anthrax PA and a HER2 affibody by expressing a nucleic acid molecule encoding the fusion protein of claim 28 in a host cell and purifying the fusion protein;producing a recombinant fusion protein comprising recombinant Anthrax PA and a HER3 affibody by expressing a nucleic acid molecule encoding the fusion protein of claim 29 in a host cell and purifying the fusion protein; orproducing a recombinant fusion protein comprising recombinant Anthrax PA and an EGFR affibody by expressing a nucleic acid molecule encoding the fusion protein of claim 30 in a host cell and purifying the fusion protein.

40. A method of treating cancer in a subject, comprising administering to the subject a therapeutically effective amount of:the pharmaceutical composition of claim 34 and the pharmaceutical composition of claim 36; orthe pharmaceutical composition of claim 35.

41. A protein complex, comprising i) a fusion protein of Anthrax protective antigen (PA) and a HER2 affibody; and ii) a fusion protein of DUF5 and a domain of Anthrax lethal factor that binds to Anthrax PA (LFn), wherein:the fusion protein of i) comprises an amino acid sequence that is at least 90% identical to SEQ ID NO: 11 or SEQ ID NO: 61, and comprises SEQ ID NO: 3 and the D683A, L685E, and Y688K substitutions; andthe fusion protein of ii) comprises an amino acid sequence that is at least 90% identical to SEQ ID NO: 33, and comprises the L418W substitution.8123-112570-0242. A pharmaceutical composition, comprising i) a fusion protein of Anthrax protective antigen (PA) and a HER2 affibody: and ii) a fusion protein of DUF5 and a domain of Anthrax lethal factor that binds to Anthrax PA (LFn), wherein:the fusion protein of i) comprises an amino acid sequence that is at least 90% identical to SEQ ID NO: 11 or SEQ ID NO: 61, and comprises SEQ ID NO: 3 and the D683A, L685E, and Y688K substitutions; andthe fusion protein of ii) comprises an amino acid sequence that is at least 90% identical to SEQ ID NO: 33, and comprises the L418W substitution.

43. A kit, comprising i) a fusion protein of Anthrax protective antigen (PA) and a HER2 affibody; and ii) a fusion protein of DUF5 and a domain of Anthrax lethal factor that binds to Anthrax PA (LFn), wherein:the fusion protein of i) comprises an amino acid sequence that is at least 90% identical to SEQ ID NO: 11 or SEQ ID NO: 61, and comprises SEQ ID NO: 3 and the D683A, L685E, and Y688K substitutions;the fusion protein of ii) comprises an amino acid sequence that is at least 90% identical to SEQ ID NO: 33, and comprises the L418W substitution; andthe fusion protein of i) and the fusion protein of ii) are contained in separate containers or separate compartments of a container.

Images