Method for site-specific chemical modification of nano antibody and application thereof

Abstract

The invention discloses a method for site-specific chemical modification of a nano antibody and application of the method, and belongs to the technical field of chemical modification of nano antibodies. The invention aims to provide more modification sites and strategies for the nano antibody. According to the method, on the basis of NHS ester reaction, natural lysine is completely or selectively mutated into arginine by rationally designing an amino acid sequence of a nano antibody, so that accurate control of modification sites is realized. According to one method, after all lysine is mutated, NHS ester only reacts with alpha-amino at the N end of the nano antibody, so that specific modification of the N end is realized; according to the other method, cysteine is introduced to the N end of the nano antibody to react with a 2-cyanobenzothiazole derivative so as to close the N-end alpha-amino group, meanwhile, single lysine is reserved or introduced to serve as a specific modification site, other lysine mutates, and precise functionalization of a nano antibody skeleton or a CDR region is achieved. And the method has wide potential in basic research, imaging and treatment development.

Description

Technical Field

[0001] This invention belongs to the field of nanobody chemical modification technology, and relates to a method for site-specific chemical modification of nanobodies and its application. Background Technology

[0002] Nanobodies (VHHs), which are antibodies composed solely of a single variable region fragment of heavy chain antibodies (HCAbs), are emerging as a new generation of antibody forms in basic research, diagnostics, and biomedical applications. Compared to traditional antibodies (approximately 150 kDa), nanobodies offer several advantages, including small size (13–15 kDa), high stability and specificity, good tissue penetration, low immunogenicity, and cost-effective expression through bacterial systems. To expand their applications, researchers have developed site-specific chemical modification strategies to endow nanobodies with new functions, such as fluorophores, DNA, PEG chains, anticancer drugs, cell-penetrating peptides, small molecule ligands, photosensitizing groups, and radioactive reagents, thereby enabling the construction of functionalized nanobodies to meet a wider range of application needs.

[0003] Despite growing research interest, current site-specific modification methods for nanobodies remain limited in labeling flexibility and are often plagued by insufficient labeling efficiency. For example, C-terminal modification via expression protein linkage (EPL)—introducing an inteptide tag at the C-terminus of the nanobody and then linking it to a reagent containing an N-terminal cysteine—sometimes suffers from low conversion rates, is limited to C-terminal labeling, and may experience yield reductions due to hydrolysis of thioester intermediates. N-terminal cysteine ​​can be modified via natural chemical linkage (NCL) using active thioesters or via cyanobenzothiazole (CBT), but these methods are limited to the N-terminal site and typically require multi-step organic synthesis to prepare the functionalized thioester or CBT reagent. Chemoenzymatic methods utilize post-translational modified enzymes, such as mTGase for labeling C-terminal Q tags (LLQS) with amines; Sortase for linking C-terminal LPSTG tags with Gn-functionalized probes; and Snoopligase for linking SnoopTagJr with DogTag. However, these methods may suffer from incomplete conversion and typically only introduce a single label at the N-terminus or C-terminus, thus limiting the flexibility of modification. Therefore, there is an urgent need to develop a robust, multifunctional, and easy-to-operate site-specific nanobody functionalization strategy. Summary of the Invention

[0004] The purpose of this invention is to provide nanobodies with more modification sites and strategies.

[0005] This invention provides a multi-modified nanobody, which retains any lysine in any segment of CDR1-CDR3 or FR1-FR4 of the nanobody, and mutates the remaining lysine to arginine. Functional groups are attached to the retained lysine sites to mutate all lysine in the nanobody to arginine before N-terminal modification.

[0006] Further specifying, if the retained lysine is in the CDR region of the nanobody, the retained lysine site is the binding site between the nanobody and the antigen; if there is no lysine in the CDR region, then the arginine at the binding site between the nanobody and the antigen is mutated to lysine. If the retained lysine is in the FR region, the retained lysine site is the site furthest from the binding site between the nanobody and the antigen. If there is no lysine in the FR region, the arginine, which is furthest from the binding site between the nanobody and the antigen, is mutated to lysine.

[0007] Furthermore, the N-terminus of the nanobody is modified with NHS ester.

[0008] Further specifying, the functional group is a photocage group.

[0009] Further, the C-terminus of the nanobody is modified with VcMMAE or TMP.

[0010] This invention provides a bivalent specific nanobody, in which one nanobody retains any lysine residue from any segment of CDR1-CDR3 of the nanobody, and a photocage group is modified at the site of the retained lysine residue. In the other nanobody, all arginine residues are mutated to lysine residues, and glycine-glycine-tryptophan (GGS) links the two nanobodies.

[0011] This invention provides a bispecific modified nanobody, which retains any lysine in any segment of FR1-FR4 of the nanobody, and mutates the remaining lysine to arginine, performs CBT modification at the N-terminus, and attaches a functional group to the site of the retained lysine.

[0012] Furthermore, the retained lysine site is the site furthest from the binding site between the nanobody and the antigen. If there is no lysine in the FR region, the arginine at the binding site between the nanobody and the antigen is mutated to lysine.

[0013] Further specifying, the functional group is Nvoc, Bhc, DEACM, TMR or PEG5k.

[0014] This invention provides an application of the above-mentioned nanobody in immunoimaging for non-diagnostic and non-therapeutic purposes.

[0015] This invention provides the application of the above-mentioned nanobody in the preparation of tumor-targeted therapeutic drugs.

[0016] This invention provides an application of the above-mentioned nanobody in the nucleation of branched microtubules in Xenopus laevis egg extract for non-diagnostic and non-therapeutic purposes.

[0017] Beneficial Effects: We propose a precise chemogenetic programming strategy for nanobodies based on the "CDR-scaffold autonomy" principle. This method is robust and versatile, enabling the modification of functional tags at the N-terminus, scaffold region, and / or CDR loop. We demonstrate that surface lysine-to-arginine (K2R) mutations preserve the folded structure of nanobodies—verified through AI 3D structural simulations of over one hundred reported nanobodies—and maintain their binding activity, confirmed by live-cell colocalization, pull-down experiments, and BLI analysis. Notably, we resolved the crystal structure of the GBPR(R34K):EGFP complex and compared it with the native GBP:EGFP complex, confirming the preservation of the 3D structure and the conservation of antigen-binding interactions. This framework establishes a molecular basis for the precise chemical modification of nanobodies using NHS chemistry (with optional N-terminal CBT linkage).

[0018] We demonstrated various N-terminal modifications of nanobodies, incorporating different functional tags, including fluorophores, bioorthogonal groups, and biotin. A simple one-step bioconjugation process enabled the generation of Cy5-conjugated EGFR nanobodies for live-cell and in vivo immunofluorescence imaging. To achieve bispecific and multispecific modifications, we established a scaffold labeling pipeline: first, N-terminal cysteine ​​residues were blocked via CBT, followed by lysine labeling of the scaffold using NHS esters. This pipeline enabled the construction of multifunctional PEGylated NDCs exhibiting selective cytotoxicity against EGFR-positive cancer cells and tumors, and successfully prepared pentafunctionalized nanobodies—potentially the highest level of site-specific functionalization reported to date.

[0019] Furthermore, we achieved precise photocage formation on the CDR ring, generating photoactivated nanobodies including a visible-light-sensitive bispecific dimer that enables chemo-optical genetic regulation of branched microtubule nucleation in Xenopus laevis egg extract. We anticipate that this chemogenetic platform will be a powerful complement to the toolkit of limited site-specific nanobodies and even protein functionalization, possessing broad potential in basic research, imaging, and therapeutic development. Attached Figure Description

[0020] Figure 1A site-specific labeling strategy for nanobodies based on chemogenetics: modifications on the N-terminus, scaffold, and CDRs, following the principle of "CDR-scaffold autonomy"; Figure 2 The NHS ester and small molecule probes used in this study; Figure 3 To verify the retention of the binding ability of the nanobody K2R after live cell colocalization and RMSD statistical analysis; Figure 4 To validate the "CDR-scaffold autonomy" principle for crystal structure and affinity assays; a. Crystal structure of the GBPR(R34K): EGFP complex, resolved at 2.4 Å resolution; b. The antibody-antigen binding interface between the GBPR(R34K): EGFP complex and the GBP:EGFP complex (PDB: 3OGO) remains conserved; c. Biomembrane layer interference (BLI) measurements of GBP, GBPR, and GBPR(R34K) binding to EGFP; d. Summary of Kon, Koff, and Kd values; Figure 5 The N-terminus of the nanobody specifically achieves various chemical modifications; a. Schematic diagram of different molecules modified at the N-terminus of the nanobody Gly-GBPR; b. N-terminal DBCO-labeled Gly-GBPR; (i) Bioconjugation scheme for N-terminal labeling of Gly-GBPR using commercially available DBCO-PEG4-NHS, (ii) SDS-PAGE and in-gel fluorescence analysis of DBCO-labeled nanobody (excitation wavelength 475–480 nm, for FAM), and further labeling with 3 times molar excess of 5-FAM-azide to confirm the introduction of the DBCO group, (iii) secondary labeling with 4 times equivalent of Cy5-NHS to verify complete modification of the nanobody, (iv) MS² characterization of DBCO-GBPR after trypsin digestion to confirm N-terminal modification; c. Schematic diagram of GBPR structure with N-terminal FAM labeling using FAM-NHS (left), SDS-PAGE and in-gel fluorescence analysis (middle), and MS² confirmation of N-terminal labeling (right). Figure 6 Various nanobodies achieved N-terminal specific biotinylation modification; a. Schematic diagram of biotin-GBPR structure labeled with biotin-NHS (left), SDS-PAGE analysis with WB detection (middle), and pulldown experiment using Strep-Tactin magnetic beads to demonstrate its maintained binding activity (right); b. Site-specific N-terminal biotinylation of Gly-RBPR to obtain biotin-RBPR; c. Site-specific N-terminal biotinylation of Gly-NbREGFR to obtain biotin-NbREGFR; Figure 7 For site-specific N-terminal labeling and its application in immunoimaging; a. Preparation and characterization of Cy5-NbREGFR: (i) Schematic diagram of the structure of Cy5-labeled EGFR nanobody (Cy5-NbREGFR) in the far-red region, (ii) its SDS-PAGE and in-gel fluorescence analysis (left), and secondary labeling using 4-fold equivalent DEAC-NHS (excitation wavelength 475–480 nm) to confirm complete N-terminal glycine modification (right), (iii) its MS² spectrum characterization (right); b. Live-cell immunofluorescence detection: left, schematic diagram; middle, representative confocal microscopy image (scale bar: 50 μm); right, quantitative analysis of relative fluorescence intensity of different cells (n = 8 cells; one-sided independent Student's t test); c. After establishing a tumor xenograft model by inoculating HeLa cells under the armpit of BALB / c nude mice, 50 µg of Cy5-NbREGFR or the negative control conjugate Cy5-GBPR was injected intravenously for in vivo fluorescence imaging. Tumor sites are indicated by yellow arrows. Images were acquired at 0, 0.25, 0.5, 1, 2, and 4 hours post-injection; d. Left, mean fluorescence irradiance of tumors [p / s / cm² / sr] / [µW / cm²] using Cy5-NbREGFR and Cy5-GBPR imaging (both groups were n = 3 experiments; mean ± standard deviation); Right, calculated tumor to background (T / B) ratio; Figure 8 Scaffold labeling enables bispecific modification of different nanobodies; a. Schematic diagram of the bioconjugation strategy; b. SDS-PAGE and in-gel fluorescence analysis confirmed the successful DEAC scaffold labeling of three nanobodies; c. MS² characterization results show the N-terminal labeling of HO-CBT and the K42 site of GBPR (K42); d. MS² characterization results show the N-terminal labeling of HO-CBT and the K67 site of NbREGFR (K67); e. MS² characterization results show the N-terminal labeling of HO-CBT and the K89 site of NbRPD-L1 (K89). Figure 9To achieve bispecific modification of nanobodies with different molecules for scaffold labeling; a. Site-specific labeling bioconjugation scheme for Cys-GBPR(K42), which can selectively link TMR or PEG5kDa chains; b. SDS-PAGE and in-gel fluorescence characterization of TMR-labeled Cys-GBPR(K42); c. MS² characterization results showing HO-CBT labeling at the N-terminus and TMR labeling at the K42 site of Cys-GBPR(K42); d. SDS-PAGE and in-gel fluorescence characterization of PEG5kDa-labeled Cys-GBPR(K42); Figure 10 Scaffold labeling enables triple functionalization of nanobodies; a. Bioconjugation scheme using bifunctional biotin-CBT reagent for triple specific labeling; b. Synthetic route of biotin-CBT(1); c. Biotin-luci-GBP R SDS-PAGE and in-gel fluorescence characterization of (K42-PEG5K); d. SDS-PAGE and in-gel fluorescence characterization of H2N-CBT-modified nanobody α-amino and NHS-PEG5kDa-modified backbone lysine unique lysine; e. SDS-PAGE and in-gel fluorescence characterization of N3-CBT-modified bispecific nanobody α-amino and NHS-DEACM-modified CDR region lysine unique lysine unique lysine.

[0021] Figure 11Scaffold labeling enables the generation and multifunctional biochemical characterization of quadruple-functionalized NDCs; a. A one-pot bioconjugation scheme for quadruple-specific functionalization, generating PEGylated NDCs: Biotin-luci-NbREGFR(K67-PEG5K)-VcMMAE, i.e., NDC-PEG5K; b. SDS-PAGE and in-gel fluorescence characterization of NbREGFR(K67-PEG5K); c. SDS-PAGE and in-gel fluorescence analysis showed that 8-fold equivalent VcMMAE-NHS achieved complete labeling, generating NDC-PEG5K; d. MS² characterization results showed that Biotin-CBT specifically labeled the N-terminus of NDC-PEG5K; e. MS² characterization results showed that VcMMAE specifically labeled the C-terminus of NDC-PEG5K; f. Pull-down experiments showed that NDC-PEG5K maintained comparable binding affinity to endogenous EGFR compared to Biotin-luci-NbEGFR; g. Representative confocal microscopy images showed that endocytosed NDC-PEG5K (1 μM, 2 h, PBS wash 2 × 5 min) was visualized by fluorescein fluorescence (excitation 405 nm) and co-localized with mCherry-Rab7a (excitation 561 nm) labeled late endosomes; h. The in vivo half-life of NDC-PEG5K and non-PEGylated NDC was determined by ELISA; Figure 12 NDC-PEG5K exhibited potent cytotoxicity against EGFR-positive cells and tumors; a. CCK8 cytotoxicity assays showed that NDC-PEG5K selectively inhibited EGFR-positive HeLa and MDA-MB231 cells, but had no inhibitory effect on EGFR-negative MCF-7 cells; in contrast, NbREGFR(K67-PEG5K) showed no cytotoxicity, while free VcMMAE showed non-selective toxicity; b. Schematic diagram of in vivo experiments in a HeLa tumor-bearing mouse model, where mice were administered PBS, 1.5 mg / kg NDC (i.e., Biotin-luci-NbREGFR(K67)-VcMMAE), or 1.5 mg / kg NDC-PEG5K via tail vein every other day for a total of seven administrations; c. Tumor volume curves for the NDC-PEG5K, NDC, and PBS groups; d. Changes in mouse body weight over time; e. Photographs of tumor resection on day 14 after drug or PBS treatment; Figure 13Site-specific pentavalent functionalization of nanobodies; a. Bioconjugation scheme of the pentavalent functionalized nanobodies: Biotin-luci-NbREGFR(K67-DEAC)-Cys(disulfoCy5)-TMP, comprising C-terminal TMP ①, N-terminal biotin ②, fluorescein ③, disulfonated Cy5 ④, and the scaffold DEAC ⑤; b. SDS-PAGE and multichannel in-gel fluorescence characterization of the pentavalent functionalized nanobodies: excitation wavelength 310 nm for fluorescein / CBT (blue), 475–480 nm for DEAC (green), and 670–685 nm for disulfonated Cy5 (red); c. MS² spectra show the specific linkage of biotin-CBT at the N-terminal cysteine ​​(c), the modification of DEAC at the scaffold K67 (d), and the modification of TMP at the C-terminus (e); the modification of disulfonated Cy5-Mal was not detected due to its negative charge; Figure 14 Site-specific CDR labeling generates DEACM photocage nanobodies; a. Bioconjugation scheme of GBP photocage at the CDR1_R34K site; b. Schematic structure of Luci-GBPR(K34-DEACM) (left) and its MOE molecular modeling image, showing the DEACM photocage located at the antibody-antigen binding interface (right); c. Left: SDS-PAGE and in-gel fluorescence analysis of photocage-treated Luci-GBPR(K34-DEACM); Right: Secondary labeling with 4-fold equivalent Cy5-NHS confirms complete modification of the CDR1_R34K residue in Luci-GBPR(K34-DEACM); d. MS² characterization confirms HO-CBT modification of the N-terminal cysteine ​​(left) and DEACM photocage modification at the CDR1_R34K site (right); e. UV-Vis absorption spectrum of Luci-GBPR(K34-DEACM); f. Natural and non-reducing gel electrophoresis showed that Luci-GBPR (K34-DEACM) was photoactivated under visible light (405 nm, 8 W, 5 cm, 2 ℃); Figure 15MeNv photocage nanobodies were generated for site-specific CDR labeling; a. Schematic structure of Luci-GBPR(K34-MeNv) (left) and MOE molecular modeling image showing MeNv located at the antibody-antigen binding interface (right); b. Left: SDS-PAGE and in-gel fluorescence analysis of photocage Luci-GBPR(K34-MeNv); Right: Secondary labeling with 4-fold equivalent Cy5-NHS confirmed complete modification of CDR1_R34K residues in Luci-GBPR(K34-MeNV); c. MS² characterization confirmed HO-CBT modification of N-terminal cysteine ​​(left) and MeNv photocage modification at the CDR1_R34K site (right); d. UV-Vis absorption spectrum of Luci-GBPR(K34-MeNv); e. Native and non-reducing gel electrophoresis showed that Luci-GBPR(K34-MeNv) exhibited good absorption at visible light (365 nm, 800 nm). Photoactivation occurs under irradiation at W, 5 cm, 2 ℃. Figure 16 Site-specific CDR labeling generates Bhc photocage nanobodies; a. Schematic structure of Luci-GBPR(K34-Bhc) (left) and MOE molecular modeling image showing Bhc located at the antibody-antigen binding interface (right); b. Left: SDS-PAGE and in-gel fluorescence analysis of photocage Luci-GBPR(K34-Bhc); Right: Secondary labeling with 4-fold equivalent Cy5-NHS confirms complete modification of CDR1_R34K residues in Luci-GBPR(K34-Bhc); c. MS² characterization confirms HO-CBT modification of N-terminal cysteine ​​(left) and photocage modification of Bhc at the CDR1_R34K site (right); d. UV-Vis absorption spectrum of Luci-GBPR(K34-Bhc); e. Native and non-reducing gel electrophoresis show that Luci-GBPR(K34-Bhc) exhibits good absorption at visible light (365 nm, 800 nm). Photoactivation occurs under irradiation at W, 5 cm, 2 ℃. Figure 17Site-specific CDR labeling generates Nvoc photocage nanobodies; a. Schematic structure of Luci-GBPR(K34-Nvoc) (left) and MOE molecular modeling image showing Nvoc located at the antibody-antigen binding interface (right); b. Left: SDS-PAGE and in-gel fluorescence analysis of photocage Luci-GBPR(K34-Nvoc); Right: Secondary labeling with 4-fold equivalent Cy5-NHS confirms complete modification of CDR1_R34K residues in Luci-GBPR(K34-Nvoc); c. MS² characterization confirms HO-CBT modification of N-terminal cysteine ​​(left) and Nvoc photocage modification at the CDR1_R34K site (right); d. UV-Vis absorption spectrum of Luci-GBPR(K34-Nvoc); e. Native and non-reducing gel electrophoresis show that Luci-GBPR(K34-Nvoc) exhibits good absorption at visible light (365 nm, 8 W, 500 nm). Photoactivation occurs under irradiation at 2000 cm and 2 ℃. Figure 18 A summary of key photophysical parameters and representative characteristics of photoactivated nanobodies; Figure 19 CDR labeling generates photoactivated bispecific nanobodies (pBsNb); a. Schematic diagram of the bioconjugation strategy of pBsNb; b. SDS-PAGE and in-gel fluorescence analysis of the obtained construct (left), and the complete modification of the CDR1_R34K residue was confirmed by secondary labeling with 4 times equivalent Cy5-NHS (right); c. MS² spectrum showing the N-terminal HO-CBT labeling (left) and DEACM conjugation at the CDR1_R34K site (right). Figure 20Photocage-encapsulated bispecific nanobody pBsNb was used for the chemo-optical genetic regulation of branched microtubule nucleation. a. Pull-down experiments showed that pBsNb was activated under 405 nm illumination, triggering dimerization between EGFP and mCherry-AlfaTag; b. Schematic diagram of conformational disruption due to xTPX2 photoinactivation; c. Total internal reflection fluorescence (TIRF) microscopy results showed the nucleation activity of xTPX2 in Xenopus oocyte cytoplasm extract at 20, 23, and 26 minutes after immune depletion (ID), after supplementation with photoTPX2 after ID, and after addition of Luci-GBPR(R34K)-NbRAlfa bispecific nanobody (BsNb). EB1-mScarlet (green) labeled the positive end of microtubules, and Alexa647-labeled porcine brain tubulin (red) was used to visualize the microtubules. Scale bar: 20 μm; d. Statistical results of microtubule numbers in each c plot (n = 3 fields of view for each condition); e. TIRF results show that xTPX2 immunization depletion followed by photoTPX2 supplementation and the addition of photocage-encapsulated dual nanobody Luci-GBPR(K34-DEACM)-NbRAlfa followed by 405 nm light irradiation (8 W, 5 cm, 16℃, 1 min) inhibited branched microtubule nucleation in Xenopus oocyte cytoplasm extract; f. Statistical results of microtubule numbers at different time points in the e plot (n = 3 fields of view); Scale bar: 10 μm. Detailed Implementation

[0022] Example 1. Design of a chemical genetics pipeline: Nanobodies for site-specific modification 1. Nanobodies (Nbs or V) H Hs) consists of a well-folded scaffold region and three complementarity-determining regions (CDRs)—CDR1, CDR2, and the typically longer CDR3—which are grafted onto the scaffold to form a specific spatial configuration that supports high-affinity and high-specificity antigen binding. Figure 1 a). Therefore, if the folding and geometry of the scaffold remain intact after engineering modifications, CDRs should be able to maintain their functional conformation and antigen-binding capacity. Based on this principle, we envision replacing all surface lysines (or all but one) with their hydrophilic analogue, arginine. Arginine plays a role highly similar to lysine in scaffold folding through hydrophilic interactions with environmental water molecules, thereby obtaining nanobodies containing only one N-terminal amino group or only one lysine on the scaffold or CDRs, achieving site-specific chemical labeling. Figure 1(b) When performing backbone-specific modification on nanobodies, lysine residues on the scaffold are retained. The specific position on the scaffold where lysine residues are retained is determined based on the structure of the nanobody-antigen complex in the PDB (Protein Data Bank). Lysine residues that are far from the interaction interface are retained, which is beneficial because the chemical molecules added during modification do not affect the binding of the nanobody to the antigen. For example, GBPR (K42) retains lysine residues on scaffold 2, NbREGFR (K67) retains lysine residues on scaffold 3, and NbRPDL1 (K89) also retains lysine residues on scaffold 3.

[0023] 2. The purpose of backbone (scaffold) modification is to modify chemical groups at positions far from the CDR region, such as modifying PEG with a large molecular weight, so as to minimize its impact on the binding of the antigen to the CDR region of the nanobody. Therefore, when the backbone is specifically modified, there are two steps: (1) According to the structure of the nanobody and antigen complex, the original nanobody selects lysine that is far from the CDR region. The orientation of the side chain of this lysine is not on the side of the CDR region, and it is farthest from the CDRs. The distance between the amino group of this lysine side chain and the antigen interaction interface can be measured using pymol, generally in Å. Select the antigen binding site and lysine that meets the conditions. (2) The position of the lysine to be retained has been determined in the first step. Then, other lysines on the sequence are mutated into arginine, including other unselected lysines on the CDR and backbone. Therefore, when modifying the backbone, whether or not the CDR region contains lysine is irrelevant to the retention of lysine in the backbone. If the CDR region contains lysine, it needs to be mutated into arginine in the second step.

[0024] 3. There is only one case involving bivalent nanobodies: one nanobodies has one arginine residue in the CDR1 region mutated to lysine, and the other nanobodies have all lysine residues mutated to arginine. The above situation does not exist. The bivalent nanobodies are GBPR(R34K)-NbRAlfa, involving two nanobodies, GBP and NbAlfa: (1) First, all lysine residues of GBP were mutated to arginine. Cell-level experiments and affinity assays showed that the lysine-to-arginine mutation did not affect its ability to bind antigens (K2R). Based on this, according to the complex structure and the interaction site between the GBP nanobodies and the antigen, the original arginine residue at position 34 of GBP (CDR1) was selected to be mutated to lysine (R34K). Similarly, cell-level experiments and affinity assays showed that the R34K mutation did not affect binding, so the GBPR(R34K) mutant was obtained. Subsequently, a photocage molecule was added to the K site of GBPR(R34K) (the photocage molecule breaks when exposed to light, lysine is exposed, and the binding ability is restored; when not exposed to light, the presence of the photocage molecule hinders GBPR(R34K) from binding to the antigen). The binding of GBPR(R34K) to the antigen can be regulated by light. All lysines of NbAlfa were mutated to arginine to obtain NbRAlfa. Cell-level experiments verified that the binding ability to the antigen was still maintained. (2) In order to regulate the formation of the ternary complex, a photocage-based bivalent nanobody was designed. First, Cys-GBPR(R34K)-NbRAlfa was obtained by fusion expression in E. coli. The first amino acid of the fusion protein is cysteine. Then, the C-terminus of GBPR(R34K) is connected to the N-terminus of NbRAlfa through three amino acids of GGS to obtain the fused bivalent nanobody. This bivalent nanobody has only one lysine in the CDR1 region of the first nanobody, which is used to modify the photocage molecule.

[0025] Here, we refer to this lysine-to-arginine mutation as K2R engineering, and use the capital letter "R" to denote engineered nanobodies (such as GBP). R 、Nb R EGFR). Since the N-terminal cysteine ​​residue can be quantitatively blocked via the CBT ligation reaction, the remaining single lysine residue can be site-specifically modified using the widely adopted N-hydroxysuccinimide (NHS) chemistry. This forms the basis of our chemogenetic pipeline, enabling site-specific labeling at the N-terminus, on scaffolds, or on CDRs while maintaining antigen-binding function. Figure 1 c). The NHS esters and small molecule probes used, such as Figure 2 As shown.

[0026] Example 2. Validating “CDR-scaffold autonomy”: To assess whether antigen-binding function was preserved after K2R (lysine to arginine) engineering, we conducted live-cell colocalization experiments.

[0027] (1) The specific method involves fusing nanobodies (modified with K2R) with fluorescent proteins (EGFP or mCherry) and targeting mitochondria or plasma membranes, while simultaneously co-expressing their corresponding antigens (labeled with another orthogonal fluorescent protein, such as mCherry or EGFP). Figure 3 The four nanobodies of ad were first modified with K2R to obtain Nb R The specific method of modification is to replace all arginine codons in the nanobody gene sequence with lysine codons at the gene level, then construct a plasmid for transfection into eukaryotic cells, and co-transfect the K2R-modified nanobody with a fluorescent protein fusion plasmid and the antigen with another fluorescent protein fusion plasmid into the cells. The co-localization of the antigen and the two fluorescent proteins carried by the nanobody is observed by confocal microscopy.

[0028] Figure 3 The first step, where all lysine residues are mutated to arginine, is to verify that the binding affinity of the nanobody is still maintained after K2R, thus proving the feasibility of the K2R principle. The second step involves mutating arginine in the CDR region to lysine. Since the CDR regions of the nanobodies involved in this invention do not contain lysine, it is not a mutation back, but rather the original arginine in the CDR region is mutated to lysine.

[0029] The data here not only verifies the feasibility of the K2R principle, but also lays the groundwork for the functional modification of different sites on nanobodies.

[0030] (2) RBPR is a version of RBP modified by K2R. The structure of RBPR was then predicted using Alphafold3, and the antigen binding site was selected. Arginine at position 109 of the CDR3 region (the original RBP has arginine at position 109) was mutated to lysine, i.e., RBPR (R109K). NbRAlfa and NbRRho were both modified by K2R and their structures were predicted by Alphafold3. NbRAlfa (R59K) is NbRAlfa with arginine at position 59 of the CDR2 region (the original NbAlfa has arginine at position 59) mutated to lysine. NbRRho (R107K) is NbRRho with arginine at position 107 of the CDR3 region (the original NbRho has arginine at position 107) mutated to lysine.

[0031] (3) The selection principle is to predict the modified structure or the structure of the interacting complex using Alphafold3, and to mutate the arginine at the interaction interface or the arginine closest to the interaction interface in CDR1-3 to lysine. Using PyMOL software to analyze the structure of the nanobody and antigen complex, the interacting amino acids can be displayed. Arginine is selected to be mutated to lysine. If there are multiple arginines in the CDR region, they can be mutated to lysine one by one, and cell-level tests are performed to select the most suitable mutation site. If there is no arginine in the CDR region, the arginine closest to the interaction interface is selected to be mutated to lysine. The indicator is to measure the distance between arginine and the interaction interface and select the closest arginine. (Since the CDR regions of the nanobodies involved lack lysine, arginine was mutated to lysine. If the original nanobodies' interaction interfaces contain lysine, then K2R is unnecessary; only the lysine at the interaction interface needs to be retained, while other scaffold regions undergo K2R modification.) In this invention, retaining one lysine is essential for backbone modification molecules; retaining one lysine or mutating arginine to lysine is essential for CDR region modification photocage molecules; however, for N-terminal modification molecules, only all lysine in the nanobodies needs to be mutated to arginine, without retaining lysine.

[0032] (4) If there are multiple segments, it is necessary to combine the complex structure information in the original nanobody PDB and select the arginine at the interaction interface of the CDRs region for mutation. If they are all at the interaction interface and their contribution is the same, they can be mutated one by one to verify the mutation effect at the cell level.

[0033] (5) Use PyMOL to visualize the amino acids at the interaction interface. For example, CDRs contain two arginine residues that form salt bridges and other interaction bonds with the amino acid residues of the antigen, indicating that the two arginine residues contribute almost equally to the binding of the antigen. However, only one arginine residue in the CDR region needs to be mutated to lysine for modification. Therefore, each arginine residue needs to be mutated individually. Combine cell-level experiments and the effects of the modified molecules to screen for suitable arginine mutation sites. Since the purpose of introducing lysine residues into the CDR region is to modify photocage molecules for regulation, only lysine residues at the interaction site can be selected for modification. However, the physicochemical properties of the photocage molecule itself may have different effects on the modification effect of lysine residues. Therefore, each lysine residue needs to be screened individually. If the interaction interface of the CDR region contains lysine residues, there is no need to mutate arginine residues to lysine residues. It is only necessary to retain the lysine residues in the CDR region and modify them.

[0034] (6) GBP RThe amino acid sequence of GBP before modification: VQLVESGGALVQPGGSLRLSCAASGFPVNRYSMRWYRQAPGKEREWVAGMSSAGDRSSYEDSVKGRFTISRDDARNTVYLQMNSLKPEDTAVYYCNVNVGFEYWGQGTQVTVSS (SEQ ID NO.1); Modified GBP R Amino acid sequence: VQLVESGGALVQPGGSLRLSCAASGFPVNRYSMRWYRQAPGREREWVAGMSSAGDRSSYEDSVRGRFTISRDDARNTVYLQMNSLRPEDTAVYYCNVNVGFEYWGQGTQVTVSS (SEQ IDNO.2); Modified GBP R (R34K) Amino acid sequence: VQLVESGGALVQPGGSLRLSCAASGFPVNRYSMKWYRQAPGREREWVAGMSSAGDRSSYEDSVRGRFTISRDDARNTVYLQMNSLRPEDTAVYYCNVNVGFEYWGQGTQVTVSS (SEQID NO.3); (7) Method: The gene sequence of the above nanobody fusion mCherry was cloned into the pCDNA3.1 vector by gene cloning, and the insertion site was located between the CMV promoter and the SV40 polyA multiple cloning site.

[0035] The fused GBP-mCherry sequence is as follows: (The linear GBP sequence is followed by the mCherry sequence, with the remaining amino acids being either the initiating methionine or linking amino acids.) M VQLVESGGALVQPGGSLRLSCAASGFPVNRYSMRWYRQAPGKEREWVAGMSSAGDRSSYEDSVKGRF TISRDDARNTVYLQMNSLKPEDTAVYYCNVNVGFEYWGQGTQVTVSSPPVATMVSKGEEDNMAIIKEFMRFKVHMEGSVNGHEFEIEGEGEGRPYEGTQTAKLKVTKGGPLPFAWDILSPQFMYGSKAYVKHPADIPDYLKLSFPEGFKWERVMNFEDGGVVTVTQDSS LQDGEFIYKVKLRGTNFPSDGPVMQKKTMGWEASSERMYPEDGALKGEIKQRLKLKDGGHYDAEVKTTYKAKKPVQLPGAYNVNIKLDITSHNEDYTIVEQYERAEGRHSTGGMDELYK (SEQ ID NO.4); GBP R and GBP R The (R34K) fusion plasmid can be prepared simply by replacing the GBP sequence mentioned above.

[0036] GBP R The sequence is: VQLVESGGALVQPGGSLRLSCAASGFPVNRYSMRWYRQAPGREREWVAGMSSAGDRSSYEDSVRGRFTISRDDARNTVYLQMNSLRPEDTAVYYCNVNVGFEYWGQGTQVTVSS (SEQ ID NO.5); GBP R The (R34K) sequence is: VQLVESGGALVQPGGSLRLSCAASGFPVNRYSMKWYRQAPGREREWVAGMSSAGDRSSYEDSVRGRFTISRDDARNTVYLQMNSLRPEDTAVYYCNVNVGFEYWGQGTQVTVSS (SEQ ID NO. 6); GBP / GBP R GBP R (R34K)-mCherry and EGFP-mito were co-transfected into HeLa cells. EGFP-mito was cloned by fusing the two genes and inserting them into the multiple cloning site between the CMV Promoter and SV40polyA in the pCDNA3.1 vector. The EGFP-mito sequence is as follows: (the straight line represents EGFP; ESGDASGSGSGSRAQASNSKLIAKSAEDEKAKEEPGNHRIVILAMLAIGVFSLGALIKIIQLRKNN is the mito sequence, which can be localized to the outer mitochondrial membrane after cell expression; the rest are linking amino acids) MVSKGEELFTGVVPILVELDGDVNGHKFSVSGEGEGDATYGKLTLK FICTTGKLPVPWPTLVTTLTYGVQCFSRYPDHMKQHDFFKSAMPEGYVQERTIFFKDDGNYKTRAEVKFEGDTLVN RIELKGIDFKEDGNILGHKLEYNYNSHNVYIMADKQKNGIKVNFKIRHNIEDGSVQLADHYQQNTPIGDGPVLLPD NHYLSTQSALSKDPNEKRDHMVLLEFVTAAGITLGMDELYK GGESGDASGGSGSRAQASNSKLIAKSAEDEKAKEEPGNHRIVILAMLAIGVFSLGALIKIIQLRKNN (SEQ ID NO.7); (8) Nb R EGFR was later used to specifically modify the N-terminus of the nanobody, involving NbEGFR. This nanobody was based on the premise that the K2R nanobody did not affect nanobody binding. All lysine residues in the EGFR nanobody were mutated to arginine, without requiring further verification of binding at the cell level through transfection experiments. NbEGFR was directly introduced into the nanobody using prokaryotic E. coli. R EGFR is expressed. First, His8-mCherry-TEV-Nb is expressed. R The EGFR gene sequence was cloned into the pET28a prokaryotic expression vector, and the gene was inserted between the T7 promoter and T7 terminator. His8 served as a protein purification tag, mCherry as a fluorescent tag, and the TEV sequence (ENLYFQ↓G) provided a site for TEV protease cleavage. The fusion protein His6-mCherry-TEV'-Nb was expressed in E. coli. R EGFR was digested with TEV enzyme to obtain Nb, where the first amino acid at the N-terminus is Gly. R EGFR. The sequence is: (HHHHHHHHH is the His8 sequence; the straight line is the mCherry sequence; ENLYFQG is the TEV enzyme recognition sequence; AAQVRLEESGGGSVQTGGSLRLTCAASGRTSRSYGMGWFRQAPGREREFVSGISWRGDSTGYADSVRGRFTISRDNARNTVDLQMNSLRPEDTAIYYCAAAAGSAWYGTLYEYDYWGQGTQVTVSS is Nb R EGFR sequence; the rest are initiating methionine or linking amino acids) MHHHHHHHHLE MVSKGEEDNMAIIKEFMRFKVHMEGSVNGHEFEIEGEGEGRPYEGTQTAKLKVTKGG PLPFAWDILSPQFMYGSKAYVKHPADIPDYLKLSFPEGFKWERVMNFEDGGVVTVTQDSSLQDGEFIYKVKLRGTN FPSDGPVMQKKTMGWEASSERMYPEDGALKGEIKQRLKLKDGGHYDAEVKTTYKAKKPVQLPGAYNVNIKLDITSH NEDYTIVEQYERAEGRHSTGGMDELYK GGSENLYFQGGSGSMAAQVRLEESGGGSVQTGGSLRLTCAASGRTSRSYGMGWFRQAPGREREFVSGISWRGDSTGYADSVRGRFTISRDNARNTVDLQMNSLRPEDTAIYYCAAAAGSAWYGTLYEYDYWGQGTQVTVSS (SEQ ID NO. 8); (9) Obtain the fluorescent fusion plasmid as shown above. By cloning the two gene segments into the pCDNA3.1 vector, the insertion site is located between the CMV promoter and SV40 polyA.

[0037] RBP-EGFP sequences: (The straight line represents the EGFP sequence, and AQVQLVESGGGLVQAGGSLRLSCATSGFTFSDYAMGWFRQAPGKEREFVAAISWSGHVTDYADSVKGRFTISRDNVKNTVYLQMNSLKPEDTAVYSCAAAKSGTWWYQRSENDFGSWGQGTQVTVSKEAI represents the RBP sequence) MVSKGEELFTGVVPILVELDGDVNGHKFSVSGEGEGDATYGKLTLKFIC TTGKLPVPWPTLVTTLTYGVQCFSRYPDHMKQHDFFKSAMPEGYVQERTIFFKDDGNYKTRAEVKFEGDTLVNRIE LKGIDFKEDGNILGHKLEYNYNSHNVYIMADKQKNGIKVNFKIRHNIEDGSVQLADHYQQNTPIGDGPVLLPDNHY LSTQSALSKDPNEKRDHMVLLEFVTAAGITLGMDELYK SGLRSMAQVQLVESGGGLVQAGGSLRLSCATSGFTFSDYAMGWFRQAPGKEREFVAAISWSGHVTDYADSVKGRFTISRDNVKNTVYLQMNSLKPEDTAVYSCAAAKSGTWWYQRSENDFGSWGQGTQVTVSKEAI (SEQ ID NO. 9); RBP R and RBP R For the fusion plasmid of (R109K) and fluorescent protein, only the RBP sequence needs to be replaced, while the rest remains the same. R The sequence: AQVQLVESGGGLVQAGGSLRLSCATSGFTFSDYAMGWFRQAPGREREFVAAISWSGHVTDYADSVRGRFTISRDNVRNTVYLQMNSLRPEDTAVYSCAAARSGTWWYQRSENDFGSWGQGTQVTVS (SEQ IDNO.10); RBP R (R109K) Sequence: AQVQLVESGGGLVQAGGSLRLSCATSGFTFSDYAMGWFRQAPGREREFVAAISWSGHVTDYADSVRGRFTISRDNVRNTVYLQMNSLRPEDTAVYSCAAARSGTWWYQKSENDFGSWGQGTQVTVS (SEQ ID NO. 11); RBP / RBP R / RBP R(R109K)-EGFP and mCherry-mito were co-transfected into HeLa cells. The mCherry-mito sequence is as follows: (the straight line represents the mCherry sequence; LILAMLAIGVFSLGAFIKIIQLRKNN is the mito localization sequence, which can target the outer mitochondrial membrane after expression; the others are linking amino acids.) MVSKGEEDNMAIIKEFMRFKVHMEGSVNGHEF EIEGEGEGRPYEGTQTAKLKVTKGGPLPFAWDILSPQFMYGSKAYVKHPADIPDYLKLSFPEGFKWERVMNFEDGG VVTVTQDSSLQDGEFIYKVKLRGTNFPSDGPVMQKKTMGWEASSERMYPEDGALKGEIKQRLKLKDGGHYDAEVKT TYKAKKPVQLPGAYNVNIKLDITSHNEDYTIVEQYERAEGRHSTGGMDELYK SGLRSRAQASNSAVDGTLILAMLAIGVFSLGAFIKIIQLRKNN (SEQ ID NO. 12); (10) The cloning method of NbAlfa-mCherry fusion protein and GBP-mCherry is the same as above, except that the nanobody sequence is changed. The sequence of NbAlfa is: SGEVQLQESGGGLVQPGGSLRLSCTASGVTISALNAMAMGWYRQAPGERRVMVAAVSERGNAMYRESVQGRFTVTRDFTNKMVSLQMDNLKPEDTAVYYCHVLEDRVDSFHDYWGQGTQVTVSS (SEQ ID NO.13). Nb R Alfa: SGEVQLQESGGGLVQPGGSLRLSCTASGVTISALNAMAMGWYRQAPGERRVMVAAVSERGNAMYRESVQGRFTVTRDFTNRMVSLQMDNLRPEDTAVYYCHVLEDRVDSFHDYWGQGTQVTVS (SEQ ID NO. 14); Nb R Alfa(R59K):SGEVQLQESGGGLVQPGGSLRLSCTASGVTISALNAMAMGWYRQAPGERRVMVAAVSEKGNAMYRESVQGRFTVTRDFTNRMVSLQMDNLRPEDTAVYYCHVLEDRVDSFHDYWGQGTQVTVS (SEQ IDNO.15); NbAlfa / Nb R Alfa / Nb RAlfa(R59K)-mCherry and AlfaTag-EGFP-mito were co-transfected into HeLa cells. The three gene segments AlfaTag-EGFP-mito were inserted together into the polyclonal restriction site between CMV Promoter and SV40PolyA in pCDNA3.1. The amino acid sequence is: (PSRLEEELRRRLTEP is the AlfaTag sequence; the straight line is the EGFP sequence;) ESGDASGSGSGSRAQASNSKLIAKSAEDEKAKEEPGNHRIVILAMLAIGVFSLGALIKIIQLRKNN is a mito sequence, the rest are starting amino acids (methionine) or linking amino acids. MPSRLEEELRRRLTEPGSPVAT MVSKGEEL FTGVVPILVELDGDVNGHKFSVSGEGEGDATYGKLTLKFICTTGKLPVPWPTLVTTLTYGVQCFSRYPDHMKQHDF FKSAMPEGYVQERTIFFKDDGNYKTRAEVKFEGDTLVNRIELKGIDFKEDGNILGHKLEYNYNSHNVYIMADKQKN GIKVNFKIRHNIEDGSVQLADHYQQNTPIGDGPVLLPDNHYLSTQSALSKDPNEKRDHMVLLEFVTAAGITLGMDE LYK GGESGDASGGSGSRAQASNSKLIAKSAEDEKAKEEPGNHRIVILAMLAIGVFSLGALIKIIQLRKNN (SEQID NO.16); (11) The NbRho-mCherry cloning method is the same as above, except that the nanobody gene is replaced. The NbRho sequence is: EVQLQASGGGFVQPGGSLRLSCAASGDTWWSSAMGWFRQAPGKEREFVSAISFYPTEYTYYADSKGRFTISRDNSKNTVYLQMNSLRAEDTATYYCAWIAWGPWMRTSWYWGQGTQVTVSS (SEQ ID NO.17). Nb R Rho sequences: EVQLQASGGGFVQPGGSLRLSCAASGDTWWSSAMGWFRQAPGREREFVSAISFYPTEYTYYADSRGRFTISRDNSRNTVYLQMNSLRAEDTATYYCAWIAWGPWMRTSWYWGQGTQVTVSS (SEQ ID NO. 18); Nb R The sequence of Rho(R107K): EVQLQASGGGFVQPGGSLRLSCAASGDTWWSSAMGWFRQAPGREREFVSAISFYPTEYTYYADSRGRFTISRDNSRNTVYLQMNSLRAEDTATYYCAWIAWGPWMKTSWYWGQGTQVTVSS (SEQ ID NO. 19); NbRho / Nb R Rho / Nb R Rho(R107K)-mCherry and EGFP-RhoA-CAAX (cloning method as above, all three gene segments are inserted into the same vector and site as above) were co-transfected into HeLa cells. Their amino acid sequences are: (the straight line is EGFP). MAAIRKKLVIVGDGACGKTCLLIVFSKDQFPEVYVPTVFENYVADIEVDGKQVELALWDTAGQEDYDRLRPLSYPDTDVILMCFSIDSPDSLENIPEKWTPEVKHFCPNVPIILVGNKKDLRNDEHTRRELAKMKQEPVKPEEGRDMANRIGAFGYMECSAKTKDGVREVFEMATRAALQARRGKKKSGCLVL is RhoA; KMSKDGKKKKKSKTKCVIM is CAAX, which can be localized to the inner cell membrane after cell expression; the rest are linked amino acids). MVSKGEELFTGVVPILVELDGDVNGHKFSVSGEGEGDATYGKLTLKFICTTGKLPVPWPTLVTTLTYG VQCFSRYPDHMKQHDFFKSAMPEGYVQERTIFFKDDGNYKTRAEVKFEGDTLVNRIELKGIDFKEDGNILGHKLEY NYNSHNVYIMADKQKNGIKVNFKIRHNIEDGSVQLADHYQQNTPIGDGPVLLPDNHYLSTQSALSKDPNEKRDHMV LLEFVTAAGITLGMDELYK SGLRSRGMAAIRKKLVIVGDGACGKTCLLIVFSKDQFPEVYVPTVFENYVADIEVDGKQVELALWDTAGQEDYDRLRPLSYPDTDVILMCFSIDSPDSLENIPEKWTPEVKHFCPNVPII LVGNKKDLRNDEHTRRELAKMKQEPVKPEEGRDMANRIGAFGYMECSAKTKDGVREFEMATRAALQARRGKKKSGCLVLSNSAVDGTAGPGSGGSGKMSKDGKKKKKKSKTKCVIM (SEQ ID NO.20); The results showed that mCherry-bound nanobodies (RBP), AlfaTag nanobodies (NbAlfa), and RhoA nanobodies (NbRho) maintained their complete binding activity after K2R engineering, even with the introduction of an additional lysine residue in the CDR region. Figure 3ac, upper part). Pearson correlation coefficient (PCC) analysis further confirmed strong colocalization between different channels (ac, upper part). Figure 3 (ac, bottom left).

[0038] RBP R It is a modified version of RBP using K2R, and then Alphafold3 is used to predict RBP. R The structure was modified by selecting arginine at position 109 of the CDR3 region (the original RBP has arginine at position 109) and mutating it to lysine, thus creating the RBP. R (R109K).

[0039] Nb R Alfa and Nb R Rho structures were all modified using K2R and predicted using Alphafold3; Nb R Alfa (R59K) is Nb R The arginine at position 59 of the CDR2 region of Alfa (the original NbAlfa had arginine at position 59) is mutated to lysine, Nb R Rho(R107K) is Nb R The arginine residue at position 107 of the CDR3 in Rho (the original NbRho has arginine at position 107) is mutated to lysine. The selection principle is to predict the modified structure or the structure of the interacting complex through Alphafold3.

[0040] The single-site mutation to lysine involved here is to verify the feasibility of the K2R principle. First, all lysine residues of the nanobody were mutated to arginine, and then, based on this, the original arginine residues in the CDR region were mutated to lysine. At the cellular level, it was demonstrated that neither mutation affected nanobody binding. Here is... Figure 3 and Figure 4 The experimental data in these two figures are to support the claim that mutating all lysine to arginine in nanobodies does not affect the structure and function of the nanobodies, because the physicochemical properties of these two amino acids are almost identical, both being basic amino acids; at the same time, for the CDRs region, which is the most critical site for binding with the antigen, mutating the original arginine to lysine does not affect the binding, further demonstrating that the K2R principle is feasible for nanobodies. Figure 4 Taking the GBP nanobody as an example, a complete mutation of lysine to arginine was performed on its sequence. Simultaneously, the original arginine position 34 of its CDR1 sequence was mutated to lysine (according to structural and literature reports, arginine position 34 of GBP is involved in antigen binding). R(R34K) was analyzed, and the structure of its complex with the antigen EGFP and its affinity for the antigen were determined, both of which showed no significant changes. These data validate the feasibility of the K2R principle at the cellular level, and further validate it at the structural level and through affinity data.

[0041] Arginine synthesis of nanobodies involves directly mutating the lysine codon to the arginine codon at the gene level. Figure 3 The article discusses the selection of arginine to lysine in the CDRs region. This is not a reversion mutation; the purpose here is to verify that lysine in nanobodies can be mutated to arginine, and that arginine can also be mutated to lysine without altering the nanobody binding affinity. Therefore, arginine in the CDRs region is used as an example to demonstrate this mutation. The arginine site selection is based on the interacting amino acids between the nanobody and the antigen (which can be displayed using PyMOL software) and the interaction sites reported in the literature. Since the CDR regions of the nanobodies discussed in this article do not involve lysine, there is no reversion mutation. If a nanobody's CDR region happens to have one lysine that participates in antigen interaction, there is also no reversion mutation. This is because either all lysines are mutated to arginine, or the lysine in the CDRs region is retained and the rest are mutated to arginine, these two methods are independent. The former only requires mutating the lysine codon to the arginine codon, while the latter mutates all codons except the interacting lysine codon to the arginine codon. The two methods are independent of each other.

[0042] Structural prediction of the K2R mutant using AlphaFold 3 showed that its structure was highly similar to the reported crystal structures, with a root mean square deviation (RMSD) value of less than 0.5 Å. Figure 3 (ac, bottom right). It is noteworthy that the widely used GFP-binding protein (GBP) retains its binding ability even after K2R engineering and further introduction of the R34K mutation into CDR1. Figure 3 d).

[0043] To generalize this principle, we analyzed over 100 nanobody crystal structures from the Protein Data Bank, predicted their K2R mutant structures using AlphaFold 3, and calculated RMSD values. The results showed that over 94% of the mutants had RMSDs below 0.5 Å, with all results below 0.65 Å. This further reinforces our proposed structure-function principle, namely CDR-scaffold autonomy—as long as the overall folding remains intact, CDRs can retain antigen-binding function after scaffold engineering. Figure 3 e).

[0044] Example 2. Crystallographic and Biochemical Verification: GBPR Binding Mode and Affinity Maintenance of EGFP 1. Obtain GBP R (R34K): GBP R (R34K) is a mutation of arginine at position 34 of CDR1 to lysine, GBP R The lysine in the GBP scaffold was mutated to arginine. The CDR region does not contain lysine, and the amino acid at position 34 was arginine before the modification.

[0045] 2. The original nanobody is GBP. Based on its structure with the antigen complex and literature reports, its arginine at position 34 participates in the interaction with the antigen and is located in CDR1. No other arginines in the entire CDR region participate in the interaction, and there is no lysine. To regulate the antigen-binding activity of the nanobody's CDR region, it is necessary to modify and label the CDR region with photocage molecules. Since lysine and arginine have similar physicochemical properties, the proposed approach is to mutate the arginine selected in CDR1 to lysine, and then modify it with photocage molecules using the amino-specific reaction between NHS ester and lysine.

[0046] 3. To verify that arginine in nanobodies can be mutated to lysine, and lysine can be mutated to arginine, neither of which affects the binding of the nanobodies, therefore... Figure 3-4 The experiment. First, on Figure 3 Four nanobodies were subjected to lysine mutations to replace all their lysine residues with arginine. Cell-level transfection experiments showed that the nanobodies still maintained their binding to the antigen, indicating that lysine in nanobodies can be mutated to arginine. Subsequently, for the argininated nanobodies, the original arginine residues in the CDR region of the nanobodies were mutated to lysine (not a reversion mutation). Cell-level transfection experiments showed that this did not affect the binding of the nanobodies, thus indicating that arginine in nanobodies can be mutated to lysine.

[0047] 4. To further verify the results, a GBP nanobody was selected as a general example. The binding force after mutation was detected by biomembrane interferometry and the structure of the complex was resolved by X-ray crystal structure analysis. Both remained consistent with the original nanobody.

[0048] 5. In addition to cell-level transfection experiments, in vitro experiments with GBP R (R34K) was expressed and purified from E. coli. His8-TEV-GBP RThe three (R34K) modules were cloned sequentially into the pET28a vector and inserted into the multiple cloning site between the T7 promoter and the T7 terminator. The amino acid sequence is as follows: (HHHHHHHH is the His8 protein purification tag, which can bind to a nickel ion chromatography column; ENLYFQG is the sequence for TEV protease recognition and digestion; the straight line is the nanobody sequence; the rest are starting methionine or linking amino acids) MHHHHHHHHLEENLYFQGGS VQLVESGGALVQPGGSLRLS CAASGFPVNRYSMKWYRQAPGREREWVAGMSSAGDRSSYEDSVRGRFTISRDDARNTVYLQMNSLRPEDTAVYYCN VNVGFEYWGQGTQVTVSS (SEQ ID NO.21); After expression in E. coli and in vitro purification, the plasmid was obtained as His8-TEV-GBP. R The (R34K) fusion protein was then digested with TEV protease and purified again by nickel ion chromatography to obtain pure GBP. R (R34K) protein, the sequence is: GGSVQLVESGGALVQPGGSLRLSCAASGFPVNRYSMKWYRQAPGREREWVAGMSSAGDRSSYEDSVRGRFTISRDARNTVYLQMNSLRPEDTAVYYCNVNVGFEYWGQGTQVTVSS (SEQ ID NO. 22); 6. Obtain GBP R (R34K):EGFP complex: GBP was purified from prokaryotic E. coli. R The proteins GBPR (R34K) and EGFP were incubated at a molar ratio of 1:1.2 at 4°C for 1–2 hours. The nanobody GBPR (R34K) and the antigen EGFP autonomously bound together in the presence of each other, without interference or linkage, forming an antigen-antibody complex through non-covalent interactions. The complex was then purified by molecular sieve chromatography.

[0049] To further characterize the K2R-engineered GBP nanobodies, and given the widespread use of EGFP, we purified GBP... R (R34K):EGFP complex, and its crystal structure was resolved. Figure 4 a). The results showed that the three-dimensional structure of GBPR(R34K) was almost identical to that of wild-type GBP, with a root mean square deviation (RMSD) of 0.29, and its antigen-binding interface remained completely conserved. Figure 4 (b) It is worth noting that K34 is located within the binding interface, making it an ideal site for photocaging, thereby generating photoactivated GBPs.

[0050] Subsequently, we used the biofilm layer interferometry (BLI) method to determine GBP and GBP. R and GBP R (R34K) binding affinity to EGFP ( Figure 4 c). All mutants maintained high affinity for EGFP, reaching sub-nanomolar levels, consistent with previously reported high affinity (Kd = 1.2 nM). Figure 4 d). These results, encompassing crystallographic, structural, and biochemical analyses, further support the CDR-scaffold autonomy principle. Figure 4 In the middle section, colocalization experiments in live HeLa cells showed that antibody-antigen binding function was still preserved after scaffold K2R engineering and the introduction of lysine in CDRs. Top: Representative microscopic images (mito: mitochondrial targeting sequence; CAAX: plasma membrane targeting sequence; scale bar: 10 μm); Bottom left: Pearson correlation coefficient (PCC) between different channels (n = 10 cells; independent two-sided Student's t test); Bottom right: Comparison of the original nanobody crystal structure with the structures of the NbR and NbR(K) mutants predicted by AlphaFold3.

[0051] Example 3. Chemogenetic N-terminal Modification of Nanobodies Based on the above findings, we developed a simple nanobody functionalization strategy that utilizes commercially available NHS esters to perform site-specific N-terminal modification of nanobodies to mount various functional groups, including bioorthogonal groups, fluorophores, and affinity tags. Figure 5 a) Nanobodies with N-terminal glycine residues can be conveniently generated by cleaving the N-terminal TEV sequence (ENLYFQ↓G) using a cysteine-containing TEV protease. We prepared GBP nanobodies with N-terminal glycine residues. R (Cly-GBP) R The NHS ester was reacted with DBCO-PEG4-NHS (10 equivalents) in PBS buffer at pH 8.0 and incubated overnight at 2 °C. Excess NHS ester was desorbed via Tris... Quenching with HCl (pH 8.0, final concentration 20 mM, on ice, 1 h) followed by size exclusion chromatography (SEC) to obtain purified N-terminated DBCO-GBP. R ( Figure 5 b).

[0052] We further demonstrated that this labeling strategy can be widely used to mount various functional groups, is compatible with different nanobodies, and retains antigen-binding function after modification. Using the same experimental protocol, FAM-NHS (3-fold equivalent) was successfully labeled with Gly-GBP. RThis generated N-terminal FAM-modified GBP. R ( Figure 5 c).

[0053] Similarly, biotin-PEG 45 -NHS (3-times equivalent) can efficiently prepare N-terminal biotinylated GBP R ( Figure 6 a, left and middle). Pull-down experiments using Strep-Tactin beads confirmed the biotinylated GBP. R It still retains its ability to bind to EGFP. Figure 6 a, right).

[0054] To verify the generality of this method, we applied it to other K2R-engineered nanobodies, including RBPs bound to mCherry. R and Nb targeting EGFR R EGFR (both nanobody sequences have all lysine residues mutated to arginine) Figure 6 bc). SDS-PAGE results showed single biotinylation, and Western blot (α-biotin) further confirmed the identity and specificity of the biotin tag.

[0055] Example 4. N-terminal Cy5-modified Nb for live cell and in vivo immunoimaging R Preparation of EGFR Cy5-Nb R EGFR: Nb R EGFR is a nanobody example demonstrating its function, based on the successful establishment of the K2R principle mentioned earlier. The entire nanobody contains only a single amino group at the N-terminus of the protein. This N-terminal amino group is then labeled using Cy5-NHS active ester molecules. The NHS ester specifically reacts with the amino group and does not react with other groups. Cy5 is thus labeled at the N-terminus, resulting in Cy5-Nb. R EGFR.

[0056] To achieve far-infrared immunofluorescence imaging, we employed an established NHS-based modification strategy to prepare N-terminal Cy5-labeled EGFR nanobodies (Cy5-Nb). R EGFR). The conjugate was characterized by SDS-PAGE and in-gel fluorescence analysis, followed by secondary labeling with DEAC-NHS to confirm complete N-terminal glycine modification, and further verified by tandem mass spectrometry. Figure 7 a).

[0057] Cy5-Nb REGFR was applied to several live cell lines—including HeLa, MDA-MB-231, HepG2, and MCF-7—and incubated for 30 minutes at a final concentration of 1 μM, followed by washing with PBS (2 × 5 min). Confocal microscopy results showed strong EGFR labeling signal in HeLa cells, moderate signal in MDA-MB-231 and HepG2, and almost no signal in MCF-7, consistent with the known EGFR expression profiles of these cell lines. Figure 7 b, middle and right parts).

[0058] To evaluate the in vivo imaging capability of Cy5-conjugated nanobodies, we administered 5 × 10⁵ nanobodies subcutaneously to the axilla of BALB / c nude mice. 6 A xenograft mouse model was established using HeLa cells. Subsequently, mice were administered Cy5-Nb via tail vein injection. R EGFR or Cy5-GBP R (Control). Whole-body fluorescence imaging ( Figure 7 c), and the quantification of fluorescence intensity at the tumor site ( Figure 7 d, left) and the ratio of tumor signal to background signal ( Figure 7 d, right), both confirmed Cy5-Nb R Selective accumulation of EGFR in EGFR-positive tumors. These results demonstrate that our N-terminal labeling strategy enables the simple generation of functionalized nanobodies suitable for live-cell immunoassay and in vivo tumor imaging.

[0059] Example 5. Bispecific nanobody modification achieved by chemical genetic scaffold labeling To extend labeling beyond the N-terminus, we developed a pipeline for site-specific modification of the scaffold region, a challenge posed by conventional chemoenzymatic and EPL / NCL methods. Even when the nanobody contains only one scaffold lysine, the unmodified N-terminal amino group can interfere with the reactivity of the NHS ester. To address this issue, we devised a one-pot, two-step strategy: i) quantitative blocking of the N-terminal cysteine ​​of the K2R-engineered nanobody via CBT linking; ii) selective labeling of a single scaffold lysine (…). Figure 8 a).

[0060] The N-terminal cysteine ​​residue was generated by cleaving an engineered ENLYFQ↓C sequence with a TEV protease, followed by modification with commercially available HO-CBT (4 equivalents) under mild conditions (PBS, 2 °C, overnight). This reaction not only quantitatively blocked the N-terminal cysteine ​​residue but also generated a fluorescein group, thus enabling fluorescence detection.

[0061] Cys-GBP R(K42): Cys-GBP R (K42): First, at the gene level, the 42nd lysine of its scaffold sequence is preserved (because it is furthest from the interaction interface and does not bind to the antigen on the same side), while the codons of other lysines are mutated to arginine codons, resulting in GBP. R (K42), then construct His8-mCherry-TEV'-GBP R (K42) The plasmid expressing the fusion protein was inserted sequentially into the pET28a plasmid vector, between the multiple cloning restriction sites between the T7 promoter and T7 terminator. The sequences are: (HHHHHHHH is the His 8 amino acid sequence, used as a protein purification tag; the straight line is the mCherry sequence; ENLYFQC is the TEV protease recognition tag sequence, ENLYFQ↓C, TEV enzyme recognizes this sequence and breaks it before C after digestion; VQLVESGGALVQPGGSLRLSCAASGFPVNRYSMRWYRQAPGKEREWVAGMSSAGDRSSYEDSVRGRFTISRDDARNTVYLQMNSLRPEDTAVYYCNVNVGFEYWGQGTQVTVSS is GBP R (K42) sequence; the rest are initiating methionine or linking amino acids) MHHHHHHHHLE MVSKGEEDNMAIIKEFMRFKVHMEGSVNGHE FEIEGEGEGRPYEGTQTAKLKVTKGGPLPFAWDILSPQFMYGSKAYVKHPADIPDYLKLSFPEGFKWERVMNFEDG GVVTVTQDSSLQDGEFIYKVKLRGTNFPSDGPVMQKKTMGWEASSERMYPEDGALKGEIKQRLKLKDGGHYDAEVK TTYKAKKPVQLPGAYNVNIKLDITSHNEDYTIVEQYERAEGRHTGGMDELYK KASENLYFQCGGSGSTGMVQLVESGGALVQPGGSLRLSCAASGFPVNRYSMRWYRQAPGKEREWVAGMSSAGDRSSYEDSVRGRFTISRDDARNTVYLQMNSLRPEDTAVYYCNVNVGFEYWGQGTQVTVSS (SEQ ID NO. 23);.

[0062] The cloned plasmid was transformed into E. coli for expression, yielding His8-mCherry-TEV'-GBP. R The (K42) fusion protein was purified using a nickel ion chromatography column, followed by TEV protease digestion and purification again using a nickel column to obtain Cys-GBP. R The (K42) sequence is: CGGSGSTGMVQLVESGGALVQPGGSLRLSCAASGFPVNRYSMRWYRQAPGKEREWVAGMSSAGDRSSYEDSVRGRFTISRDDARNTVYLQMNSLRPEDTAVYYCNVNVGFEYWGQGTQVTVSS (SEQ ID NO. 24); Cys-Nb R EGFR(K67): Cys-Nb R EGFR(K67): This nanobody is similar to the Cys-GBP mentioned above. R (K42) Cloning and expression are the same, except that GBP is used. R The (K42) sequence is replaced with Nb R The EGFR(K67) sequence was ultimately used to obtain Cys-Nb. R The sequence of EGFR(K67) is: CGGSGSMAAQVRLEESGGGSVQTGGSLRLTCAASGRTSRSYGMGWFRQAPGREREFVSGISWRGDSTGYADSVKGRFTISRDNARNTVDLQMNSLRPEDTAIYYCAAAAGSAWYGTLYEYDYWGQGTQVTVSS (SEQ IDNO.25); Cys-Nb R PD-L1(K89): TEV'-Nb R The PD-L1(K89) module gene was sequentially inserted into the PTXB1 vector, at the multiple cloning site between the T7 promoter and Intein, to obtain TEV'-Nb. R The PD-L1(K89)-Intein-His6 fusion protein expression plasmid has the following amino acid sequence: (ENLYFQC is the sequence of TEV'; the straight line is Nb). R PD-L1(K89) sequence; CITGDALVALPEGESVRIADIVPGARPNSDNAIDLKVLDRHGNPVLADRLFHSGEHPVYTVRTVEGLRVTGTANHPLLCLVDVAGVPTLLWKLIDEIKPGDYAVIQRSAFSVDCAGFARGKPEFAPTTYTVGVPGLVRFLEAHHRDPDAQAIADELTDGRFYYAKVASVTDAGVQPVYSLRVDTADHA is the Intein sequence; HHHHHH is the His6 sequence; the rest are the starting methionine and linking amino acids) MGGSENLYFQCGGSGSTG QVQLQESGGGLVQPGGSLRLSCAASGRMSSRRCMAWFRQAPGRERERVAR LLTTSGSTYLADSVRGRFTISQNNARSTVYLQMNSLKPEDTAMYYCAADSFEDPTCTLVTSSGAFQYWGQGTQVTV SS GGGSYPYDVPDYALEGGGCITGDALVALPEGESVRIADIVPGARPNSDNAIDLKVLDRHGNPVLADRLFHSGEHPVYTVRTVEGLRVTGTANHPLLCLVDVAGVPTLLWKLIDEIKPG DYAVIQRSAFSVDCAGFARGKPEFAPTTYTVGVPGLVRFLEAHHRDPDAQAIADELTDGRFYYAKVASVTDAGVQPVYSLRVDTADHAFITNGFVSHATGLTGLNSGGHHHHHH (SEQ ID NO.26); TEV'-Nb was expressed using prokaryotic Escherichia coli and purified using nickel ion column chromatography. R PD-L1(K89)-Intein-His6 protein was first expressed via an expression ligation (EPL) reaction. In a pH 8.0 buffer, 0.55 M of 4-mercaptophenylacetic acid and 1 M of sodium 2-mercaptoethanesulfonate were added, along with 2 mM Cys. The reaction was carried out at 4°C for 48 h. The protein was then purified again using a nickel column to obtain TEV'-Nb. R PD-L1(K89) protein was then digested with TEV protease to obtain Cys-Nb. R PD-L1 (K89), the sequence is: CGGSGSTGQVQLQESGGGLVQPGGSLRLSCAASGRMSSRRCMAWFRQAPGRERERVARLLTTSGSTYLADSVRGRFTISQNNARSTVYLQMNSLKPEDTAMYYCAADSFEDPTCTLVTSSGAFQYWGQGTQVTVSSGGGSYPYDVPDYALEGGGC (SEQ ID NO. 27); To demonstrate this labeling strategy, we generated three nanobody precursors with N-terminal cysteine ​​residues: Cys-GBP R (K42), Cys-Nb R EGFR (K67) and the K2R-engineered PD-L1 nanobody Cys-Nb R PD-L1(K89).

[0063] (2) K42 is located at FR2, GBP R(K42) indicates that the nanobody retains only the original lysine at position 42, with all other lysines mutated to arginine, resulting in the entire nanobody containing only one lysine. K67 and K89 are both located in the FR3 region. (Same as GBP) R (K42) contains only one lysine, thus retaining the only lysine on the scaffold.

[0064] Based on the original nanobody structure, the lysine sites that needed to be retained in the backbone were selected, while the amino acids at other sites in the nanobody sequence were all mutated to arginine, not reverted. Figure 3-4 It has been demonstrated that mutations between lysine and arginine have no effect on nanobody binding. Therefore, it is only necessary to select the lysine that needs to be retained and mutate the others to arginine. Then, the retained lysine is specifically functionalized.

[0065] These nanobodies were labeled in PBS buffer (pH 8.0) using a one-pot method. First, they were reacted overnight at 2 °C with HO-CBT (2-cyano-6-hydroxybenzothiazole) (4 equivalents), followed by reaction at 2 °C with diethylaminocoumarin-NHS (DEAC-NHS, 4 equivalents) for 4 hours. Finally, they were labeled with Tris... Quenching was performed with HCl (20 mM, 1 h). SDS-PAGE and in-gel fluorescence analysis confirmed successful dual labeling: the DEAC group (excitation wavelength 475–480 nm) was attached to the lysine residue on the scaffold, while the fluorescein group (excitation wavelength 310 nm) was formed at the N-terminus. Figure 8 b). These DEAC-labeled nanobody conjugates were digested with trypsin or α-Lytic protease, and the resulting peptides were analyzed by nanoLC-MS². Figure 8 The obtained MS² spectrum confirmed the N-terminal labeling of HO-CBT and the modification of the DEAC group on the expected scaffold lysine.

[0066] We further evaluated whether other groups (such as the orange-red emitting tetramethylrhodamine (TMR) fluorophore or polyethylene glycol (PEG) chains) could be linked to the scaffold lysine ( Figure 9 a). The results showed that 5-TMR (excitation wavelength 520–525 nm) was also successfully linked to Cys-GBP. R (K42), confirmed by SDS-PAGE and intragel fluorescence analysis. Figure 9 b), MS² spectra further confirmed the labeling of 5-TMR at the K42 residue, while the N-terminal cysteine ​​residue was blocked by HO-CBT ( Figure 9 c).

[0067] PEGylation is known to prolong the in vivo half-life of nanobodies, improve pharmacokinetic characteristics, and reduce immunogenicity. To demonstrate PEGylation on a scaffold, we used HO-PEG5kDa-NHS-labeled Cys-GBP. R (K42). SDS-PAGE results showed successful ligation of a single PEG5kDa chain. Figure 9 d, left), while intragel fluorescence analysis (excitation wavelength 365 nm) confirmed the formation of the N-terminal fluorescein group ( Figure 9 d, right). These results highlight the versatility of our stent labeling strategy, which is applicable to stents including GBP. R (K42), Nb R EGFR (K67) and Nb R Various nanobodies, including PD-L1 (K89), are available, and different groups, such as fluorophores and PEG chains, can be linked.

[0068] Example 6. Triple functionalization of nanobodies via chemical genetic scaffold labeling Encouraged by the above results, we further explored multispecific functionalization, a more challenging goal in antibody engineering. For clarity, each installed function is labeled ① / ② / ③ / ④ / ⑤ in sequence. We first designed a triple functionalization scheme ( Figure 10 a) Utilizing the bifunctional reagent Biotin-CBT. This molecule was synthesized via a one-step amide coupling reaction between commercially available Biotin-LC-COOH and H2N-CBT. Figure 10 (b) During the CBT linking process, both the biotin affinity tag (①) and the fluorescein group (②) can be modified simultaneously. Subsequently, a third functional group, such as a PEG chain (③), can be further mounted on the scaffold by NHS chemistry.

[0069] Biotin-luci-GBP R (K42-PEG5K) nanobody is Cys-GBP R (K42), the method of obtaining the sequence and the above questions have been answered in detail.

[0070] No need to filter, because Figure 3-4 Data has shown that mutations between lysine and arginine do not affect the binding of nanobodies (i.e., the K2R principle). Based on this principle, and the structure of the GBP and antigen complex, the side chain of lysine at position 42 is furthest from the CDR region (which can be observed with the naked eye using Pymol software). Therefore, lysine at position 42 on the nanobody is selected and retained, while all other lysines are mutated to arginine.

[0071] Tagging process: (1) Tagging Cys-GBP using CBT-BiotinR (K42) yields Biotin-luci-GBP R (K42)(CBT reacts with an amino acid at the N-terminus of the protein to generate luciferin, hence the name Luciferin (abbreviated as Luci)); (2) Biotin-luci-GBP R (K42) was labeled with PEG5K-NHS ester, and this polymer could be specifically labeled onto the lysine residue at position 42 to obtain Biotin-luci-GBP. R (K42-PEG5K) was purified to pure Biotin-luci-GBP by molecular sieve chromatography. R (K42-PEG5K).

[0072] In this way, we successfully constructed the trifunctional nanobody Biotin-luci-GBP. R (K42-PEG5K), and confirmed by SDS-PAGE and in-gel fluorescence characterization. Figure 10 c). Figure 10 d represents the SDS-PAGE and in-gel fluorescence characterization of H2N-CBT modified α-amino and NHS-PEG5kDa modified lysine as the backbone of the nanobody. Figure 10 e represents the SDS-PAGE and in-gel fluorescence characterization of the N3-CBT-modified bispecific nanobody α-amino and the NHS-DEACM-modified CDR region's only lysine residue.

[0073] Example 7. Construction of a multifunctional PEGylated nanobody drug conjugate (NDC) with selective cytotoxicity for targeted therapy of EGFR-positive cancer cells and tumors. To achieve targeted cancer therapy, we developed a strategy to generate a quadruple-specific functionalized PEGylated nanobody drug conjugate (NDC-PEG5K) in a one-pot workflow by integrating a cysteine-maleimide coupling reaction into a chemogenetic marker platform. C-terminal cysteine ​​can be conveniently introduced via an EPL (expression protein linking) reaction between cysteine ​​and the nanobody-integral peptide fusion protein. Therefore, we prepared the precursor Cys-Nb. R EGFR(K67)-Cys, the process of which includes: i) cleavage of the C-terminal intima-peptide tag via EPL-mediated cleavage; ii) cleavage of the N-terminal TEV' tag via TEV protease (ENLYFQ↓C).

[0074] This precursor underwent stepwise modification to achieve four different functions: Biotin-CBT labeling at the N-terminal cysteine ​​residue to link biotin (①), simultaneously generating a fluorescein group (②); HO-PEG5kDa-NHS was added in a one-pot reaction to link a third functional group, the PEG chain, onto the scaffold (③); finally, the cytotoxic load VcMMAE-Mal (④) was coupled to the C-terminal cysteine ​​residue ( ). Figure 11 a).

[0075] SDS-PAGE and in-gel fluorescence analysis confirmed the presence of Biotin-luci-Nb. R Successful preparation and purification of EGFR(K67-PEG5K) intermediate ( Figure 11 b). To verify the integrity of the drug conjugation, we performed a secondary labeling experiment using an excess of Cy5-Mal (10 equivalents). The results showed that 8 equivalents of VcMMAE-Mal were sufficient to completely saturate the C-terminal cysteine ​​( Figure 11 c). Subsequently, trypsin digestion and MS² analysis confirmed the expected marker sites for each functional group ( Figure 11 de).

[0076] This quadruple-functionalized NDC integrates four different features: i) a potent cytotoxic payload (VcMMAE) for targeted cell killing; ii) a biotin tag for affinity binding and immune detection; iii) a fluorescein group for fluorescence-based tracking; and iv) a PEG chain for improved pharmacokinetics.

[0077] Therefore, we demonstrated that the biotin tag can achieve pull-down experiments, confirming that NDC-PEG5K still maintains high binding affinity compared to unmodified EGFR nanobodies. Figure 11 f). The fluorescein group facilitated visualization of intracellular transport, revealing its strong co-localization with late endosomes (f). Figure 11 g). The linkage of the PEG5kDa chain significantly increased the in vivo half-life of NDC ( t 1 / 2 ELISA assays revealed that PEG 5kDa prolonged the duration of NDC drug use from 2.15 ± 0.06 hours to 7.87 ± 0.59 hours, approximately a fourfold increase. Figure 11 h).

[0078] Example 8. Construction of quadruple functionalized NDCs: Selective cytotoxicity against EGFR-positive cancer cells and tumors CCK8 assays showed that NDC-PEG5K exhibited selective cytotoxicity against EGFR-positive HeLa and MDA-MB-231 cells, while no cytotoxicity was observed in EGFR-negative MCF-7 cells. Figure 12 a). In contrast, nanobody conjugates lacking VcMMAE did not exhibit cytotoxicity, while free VcMMAE showed reduced, non-selective toxicity in all cell lines ( Figure 12 a, right).

[0079] Furthermore, in the HeLa tumor xenograft mouse model ( Figure 12 (b) We confirmed that NDC-PEG5K is significantly more effective than NDC in inhibiting tumor growth in vivo, highlighting the beneficial role of the PEG5kDa chain in NDC-PEG5K therapeutic applications. Figure 12 c). The mice maintained stable body weight, indicating no systemic toxicity. Figure 12 d), and the analysis of the resected tumor further verified that NDC-PEG5K has stronger antitumor activity than NDC ( Figure 12 e).

[0080] Therefore, these results demonstrate that we have successfully constructed a multifunctional and efficient NDC drug using a scaffold labeling strategy, capable of targeting EGFR-positive cancer cells and tumors.

[0081] (1) Biotin-luci-NbREGFR(K67)-VcMMAE: Cys- Nb R EGFR(K67)-Cys is made by retaining lysine at position 67 of the FR3 region and mutating the other lysines through the K2R mutation, so that it contains only one lysine.

[0082] (2) The first step of modification is the specific reaction of Biotin-CBT with the N-terminus Cys to obtain Biotin-CBT-Nb. R EGFR(K67)-Cys, because CBT reacts with the N-terminal Cys to become a fluorescein, is named Biotin-luci-Nb. R EGFR(K67)-Cys; the second modification step involves reacting the product from the first step with NHS-PEG5K. NHS (succinimide ester) specifically reacts with the primary amino group of lysine 67 to yield Biotin-luci-Nb. R The third modification step involves reacting the product modified in the second step with Mal-VcMMAE. Mal (maleimide) reacts with the C-terminal Cys, ultimately yielding Biotin-luci-Nb. R EGFR(K67-PEG5K)-VcMMAE.

[0083] (3) Biotin-luci-Nb R EGFR(K67-PEG5K)-VcMMAE. This is a nanobody designed using a K2R-based modification method, allowing for multi-step specific modifications. VcMMAE is one of its functionalizations, representing a drug molecule. Specific modification of the nanobody with this drug molecule enables targeted delivery to a specific region. Biotin, as a tag, is used in immunoblotting experiments to identify binding to the antigen. Luci, which fluoresces under excitation light, is used to indicate and track the nanobody drug molecule.

[0084] (4) Based on the K2R principle, select the 67th lysine that needs to be retained. The selection requirements have been answered above.

[0085] Cys-Nb R EGFR(K67)-Cys: TEV'-Nb R Nb R The EGFR(K67) module gene was sequentially inserted into the PTXB1 vector, at the multiple cloning site between the T7 promoter and Intein, to obtain TEV'-Nb. R The fusion protein expression plasmid of EGFR(K67)-Intein-His6 has the following amino acid sequence: (ENLYFQC is the sequence of TEV'; the straight line is Nb) R PD-L1 (K89) sequence; CITGDALVALPEGESVRIADIVPGARPNSDNAIDLKVLDRHGNPVLADRLFHSGEHPVYTVRTVEGLRVTGTANHPLLCLVDVAGVPTLLWKLIDEIKPGDYAVIQRSAFSVDCAGFARGKPEFAPTTYTVGVPGLVRFLEAHHRDPDAQAIADELTDGRFYYAKVASVTDAGVQPVYSLRVDTADHA is the Intein sequence; HHHHHH is the His6 sequence; the rest are start codons and linking amino acids) MGGSENLYFQCGSGSM AAQVRLEESGGGSVQTGGSLRLTCAASGRTSRSYGMGWFRQAPGREREFVSG ISWRGDSTGYADSVKGRFTISRDNARNTVDLQMNSLRPEDTAIYYCAAAAGSAWYGTLYEYDYWGQGTQVTVSSGGGSYPYDVPDYALEGGGCITGDALVALPEGESVRIADIVPGARPNSDNAIDLKVLDRHGNPVLADRLFHSGEHPVYTVRTVEGLRVTGTANHPLLCLVDVAGVPTLLWKLIDEIKPG DYAVIQRSAFSVDCAGFARGKPEFAPTTYTVGVPGLVRFLEAHHRDPDAQAIADELTDGRFYYAKVASVTDAGVQPVYSLRVDTADHAFITNGFVSHATGLTGLNSGGHHHHHH (SEQ ID NO.28); (5) The expression was performed using prokaryotic Escherichia coli, and the product was purified by nickel ion column chromatography to obtain TEV'-Nb. R EGFR(K67)-Intein-His6 protein was first expressed via an expression ligation (EPL) reaction. In a pH 8.0 buffer, 0.55 M of 4-mercaptophenylacetic acid and 1 M of 2-mercaptoethanesulfonate were added, along with 2 mM Cys. The reaction was carried out at 4°C for 48 h. The protein was then purified again using a nickel column to obtain TEV'-Nb. R EGFR(K67)-Cys protein was then digested with TEV protease to obtain Cys-Nb. R EGFR(K67)-Cys, the sequence is: CGSGSMAAQVRLEESGGGSVQTGGSLRLTCAASGRTSRSYGMGWFRQAPGREREFVSGISWRGDSTGYADSVKGRFTISRDNARNTVDLQMNSLRPEDTAIYYCAAAAGSAWYGTLYEYDYWGQGTQVTVSSGGGSYPYDVPDYALEGGGC (SEQ ID NO. 29); (6) Cys-Nb R EGFR(K67)-Cys first reacts with CBT-Biotin to yield Biotin-luci-Nb R EGFR(K67)-Cys; (2) The product from the first step was further reacted with PEG5K-NHS ester to obtain Biotin-luci-Nb R EGFR(K67-PEG5K)-Cys was purified by molecular sieve chromatography to obtain a pure modified product; (3) the third modification was Biotin-luci-Nb R The C-terminal Cys of EGFR(K67-PEG5K)-Cys reacts with Mal-VcMMAE to give the final product Biotin-luci-Nb.R EGFR(K67-PEG5K)-VcMMAE.

[0086] Example 9. Construction of a five-fold functionalized nanobody conjugate To push the boundaries of site-specific nanobody engineering, we developed a strategy for achieving fivefold functionalization through C-terminal integration of an EPL reaction. In this process, Nb R The C-terminus of EGFR(K67) is linked to the small molecule ligand trimethoprim (TMP ①) via EPL to generate the precursor Cys-Nb. R EGFR(K67)-Cys-TMP. Based on our established one-pot workflow, four additional functions were sequentially installed: Biotin-CBT was used to modify the N-terminal cysteine ​​residue (②) and simultaneously generate a fluorescein group (③); disulfonicated Cy5 was linked to the C-terminal cysteine ​​residue via maleimide coupling (④); DEAC-NHS was used to modify the lysine residue of the scaffold (⑤). Figure 13 a).

[0087] Cys-Nb R EGFR(K67)-Cys-TMP: This protein is associated with Cys-Nb R The preparation of EGFR(K67)-Cys is similar, except that in the EPL process, Cys-TMP molecules are added at a final concentration of 2 mM instead of Cys. The rest of the preparation process is the same, using the same precursor protein TEV'-Nb. R EGFR(K67)-Intein-His 6。

[0088] SDS-PAGE and in-gel fluorescence characterization confirmed the successful introduction of the five groups. Figure 13 b), and tandem MS² analysis after trypsin digestion further validated the expected labeling sites of each functional group ( Figure 13 ce).

[0089] Therefore, this scaffold labeling method enables one-pot preparation of five-fold functionalized nanobodies, representing one of the most diverse and precisely engineered antibody formats reported to date.

[0090] Example 10. Chemogenetic CDR labeling for photoactivated nanobodies Targeted engineering of complementarity-determining regions (CDRs) opens up a variety of application opportunities, including the regulation and maturation of binding affinity, ligand sensitivity, and photoactivation of nanobodies. However, current site-specific chemical modification methods for CDRs remain limited. Here, we demonstrate the potential for CDR functionalization by constructing photocage-like nanobodies, making them a tool for photocontrolled antigen modulation.

[0091] Given the wide range of applications of GFP and its variants in biology, we designed photoactivated GBPs using a two-step process: first, we blocked the N-terminal cysteine ​​residue using HO-CBT, and then photocage modification was performed at the CDRs using NHS esters. Figure 14 a). A key advantage of this strategy is its high versatility, enabling the introduction of multiple types of photocages into the CDR region without the need for more complex non-natural amino acid (UAA) introductions.

[0092] Therefore, we developed a GBP R (CDR1_R34K): Crystal structure analysis of the EGFP complex ( Figure 14 a) shows that K34 is located at the antibody-antigen binding interface and forms a key hydrogen bond with EGFP. We hypothesize that modifying DEACM at the K34 site will disrupt binding through steric hindrance and interference with key interactions. Figure 14 b). Photocage-NHS esters (such as DEACM-NHS) can be synthesized in one step using bis(succinimide) carbonate (DSC). Employing a two-step bioconjugation strategy, we successfully generated visible light-sensitive GBP. R (K34-DEACM) photocage nanobody was confirmed by SDS-PAGE and in-gel fluorescence analysis. Figure 14 c, left). Secondary labeling with excess Cy5-NHS confirmed complete modification of the CDR1_R34K residues. Figure 14 c, right). Trypsin digestion combined with MS² analysis further validated the N-terminal HO-CBT site-specific modification and the DEACM modification of CDR1_R34K (c). Figure 14 d). Photophysical characterization shows that luci-GBP R (K34-DEACM) carrying a DEACM photocage exhibits absorption in the 350–460 nm range (λ). max = 393 nm), covering the visible light region, indicating its visible light photolysis ( Figure 14 e). Natural gel electrophoresis confirmed 405 nm visible light-induced EGFP binding, forming GBP. R (R34K):EGFP complex ( Figure 14f). Therefore, we successfully generated a DEACM photocage-coated GBP nanobody with DEACM Lys UAA residues introduced at the CDR1_K34 site and demonstrated highly efficient visible light photoactivation.

[0093] Building on this success, we further constructed the MeNv photocage nanobody luci-GBP. R (K34-MeNv) was prepared using a similar two-step process. Figure 15 Photophysical characterization and natural gel electrophoresis confirmed its highly efficient photoactivation under ultraviolet light (ac). Figure 15 de).

[0094] Subsequently, we developed the challenging Bhc optical cage-encapsulated GBP luci-GBP. R (K34-Bhc) introduces a Bhc photocage sensitive to two-photon (2P) signals. Figure 16 a). SDS-PAGE, in-gel fluorescence, and MS² characterization verified the successful preparation of Bhc photocage nanobodies. Figure 16 bc), which exhibits extremely fast photoactivation, requiring only two seconds ( Figure 16 de).

[0095] To further highlight the diversity of CDR photocage strategies, we also prepared Nvoc photocage-encapsulated GBP luci-GBP. R (K34-Nvoc) introduces the widely used Nvoc photocage, which exhibits the lowest photosensitivity. Figure 17 ).

[0096] Key photophysical parameters and representative characteristics of four photoantibodies Figure 18 The results are summarized below. In conclusion, these results establish that chemogenetic strategies can serve as a robust method for CDR photocage formation to generate diverse photoactivated nanobodies. This approach can introduce not only bulky, visible-light-sensitive DEACM photocages that are difficult to achieve through genetic code expansion, but also two-photon-sensitive Bhc photocages, offering potential for optogenetic applications in deep tissues.

[0097] Example 11. Chemogenetic Engineering of Photoactivated Bispecific Nanobodies (pBsNb) for Optogenetic Control of Biological Processes Building on the above successes, we engineered a photoactivated bispecific nanobody (pBsNb), Luci-GBP. R (K34-DEACM)-Nb R Alfa (two nanobodies purified through fusion expression, linked together by the amino acid GGS). 1. His8-mCherry-TEV'-GBP R (R34K)-Nb R The genes of the five Alfa modules were cloned sequentially into the multiple cloning site between the T7 promoter and T7 Terminator of the pET28a plasmid vector. The amino acid sequence is: (HHHHHHHH is His8; the straight line is the mCherry tag; ENLYFQC is the TEV protease recognition and digestion sequence; VQLVESGGALVQPGGSLRLSCAASGFPVNRYSMKWYRQAPGREREWVAGMSSAGDRSSYEDSVRGRFTISRDDARNTVYLQMNSLRPEDTAVYYCNVNVGFEYWGQGTQVTVSS is GBP) R (R34K); SGEVQLQESGGGLVQPGGSLRLSCTASGVTISALNAMAMGWYRQAPGERRVMVAAVSERGNAMYRESVQGRFTVTRDFTNRMVSLQMDNLRPEDTAVYYCHVLEDRVDSFHDYWGQGTQVTVSS is Nb R Alfa; the rest are initiating methionine or linking amino acids) MHHHHHHHHLE MVSKGEEDNMAIIKEFMRFKVHMEGSVNGHEFEIEGEGEGRPYE GTQTAKLKVTKGGPLPFAWDILSPQFMYGSKAYVKHPADIPDYLKLSFPEGFKWERVMNFEDGGVVTVTQDSSLQD GEFIYKVKLRGTNFPSDGPVMQKKTMGWEASSERMYPEDGALKGEIKQRLKLKDGGHYDAEVKTTYKAKKPVQLPG AYNVNIKLDITSHNEDYTIVEQYERAEGRHTGGMDELYKK ASENLYFQCGGSGSM VQLVESGGALVQPGGSLRLS CAASGFPVNRYSMKWYRQAPGREREWVAGMSSAGDRSSYEDSVRGRFTISRDDARNTVYLQMNSLRPEDTAVYYCN VNVGFEYWGQGTQVTVSS GGSSGEVQLQESGGGLVQPGGSLRLSCTASGVTISALNAMAMGWYRQAPGERRVMVAAVSERGNAMYRESVQGRFTVTRDFTNRMVSLQMDNLRPEDTAVYYCHVLEDRVDSFHDYWGQGTQVTVSS (SEQ IDNO.30); The above plasmid was expressed in *E. coli* and purified using a nickel ion chromatography column to obtain His8-mCherry-TEV'-GBP. R (R34K)-Nb R Alfa fusion protein was then digested with TEV protease, followed by purification using a nickel ion chromatography column to obtain the bivalent nanobody Cys-GBP, whose first amino acid is cysteine. R (R34K)-Nb RAlfa, the sequence is: CGGSGSMVQLVESGGALVQPGGSLRLSCAASGFPVNRYSMKWYRQAPGREREWVAGMSSAGDRSSYEDSVRGRFTISRDDARNTVYLQMNSLRPEDTAVYYCNVNVGFEYWGQGTQVTVSS GGSSGEVQLQESGGGLVQPGGSLRLSCTASGVTISALNAMAMGWYRQAPGERRVMVAAVSERGNAMYRESVQGRFTVTRDFTNRMVSLQMDNLRPEDTAVYYCHVLEDRVDSFHDYWGQGTQVTVSS (SEQ ID NO.31).

[0098] 2. Cys-GBP R (R34K)-Nb R Alfa labeling: (1) First, HO-CBT was used for labeling, which specifically reacts with the first amino acid Cys of the bivalent nanobody to obtain Luci-GBP. R (R34K)-Nb R Alfa; (2) The product labeled in the first step was purified by ultrafiltration to obtain a pure product, and then lysine was specifically labeled using DEACM-NHS ester. The bivalent nanobody contained only one lysine, and the product Luci-GBP was obtained. R (K34-DEACM)-Nb R Alfa.

[0099] It is a GBP optically caged by DEACM R It is composed of K2R-engineered AlfaTag nanobodies. This construct enables the light-controlled regulation of protein functions through photoinduced proximity interactions.

[0100] pBsNb can be prepared via a one-pot, two-step method: using Cys-GBP R (R34K)-Nb R Alfa was used as a precursor, and the N-terminus was first blocked with CBT, followed by DEACM-NHS labeling in PBS at pH 8.0. Figure 19 a). SDS-PAGE and in-gel fluorescence analysis (DEACM excitation wavelength 365–370 nm) confirmed successful preparation. Figure 19 b, left). Subsequent secondary labeling with excess Cy5-NHS verified complete modification of the CDR1_R34K site. Figure 19 b, right). Trypsin digestion combined with MS² analysis further confirmed the N-terminal HO-CBT marker and the DEACM modification of GBP at the CDR1_R34K site (b). Figure 19 c).

[0101] Example 12. Regulation of branched microtubule nucleation using photocage-cageed dual nanoparticle antibody pBsNb Functionally, pBsNb can mediate light-dependent dimerization: under 405 nm illumination, mCherry-AlfaTag dimers with eDHFR-EGFR, a process verified by pull-down experiments using mCherry-AlfaTag-coated magnetic beads. Figure 20 a). It is worth noting that 405 nm visible light activation is more advantageous than ultraviolet light because 405 nm laser diodes are routinely configured in standard confocal laser scanning microscopes.

[0102] Photocage-encapsulated dual nanobody pBsNb was used to regulate branched microtubule nucleation, a crucial process in spindle assembly and cell division, achieved through photocontrolled regulation of the nucleation factor xTPX2 (a target protein of Xenopus kinesin-like protein 2). Branched microtubule nucleation, the generation of daughter microtubules from existing mother microtubules, is best visualized in Xenopus laevis egg extracts due to its superior biochemical accessibility.

[0103] To achieve optical control, we fused xTPX2 with N-terminal EGFP and C-terminal AlfaTag to generate EGFP-xTPX2-AlfaTag (photoTPX2), which forms a functional complex with pBsNb ( Figure 20 b, above). Under 405 nm illumination, DEACM photolysis in pBsNb disrupts the native conformation of photoTPX2, effectively deactivating the photoTPX2:pBsNb complex (photoCTPX2). Figure 20 b, down).

[0104] Branched microtubule nucleation was completely eliminated by immunodepletion of xTPX2 in Xenopus laevis egg extract; nucleation was restored after the addition of photoTPX2, indicating functional reconstruction of xTPX2. Figure 20 (cd). However, photoTPX2:Luci-GBP R (R34K)-Nb R The addition of the Alfa complex to Xenopus oocyte cytoplasm extract containing immune-depleted xTPX2 failed to restore nucleation, demonstrating that Luci-GBP R (R34K)-Nb R Alfa, the photolytic form of pBsNb, effectively deactivates photoTPX2 in this process. Figure 20 cd).

[0105] Finally, Xenopus oocyte cytoplasmic extract supplemented with photoTPX2 and pBsNb (endogenous xTPX2 had been immunodepleted) was exposed to 405 nm light. Total internal reflection fluorescence (TIRF) microscopy revealed strong microtubule nucleation and a typical “branched fan” structure under dark conditions, suggesting that photoCTPX2 was fully active. Figure 20 e, left; Figure 20 f, top). In contrast, the illuminated group showed very little branching nucleation, with only sparse individual microtubules (f, top). Figure 20 e, right; Figure 20 e, below), indicating that nucleation is effectively suppressed under photoactivation.

[0106] These results establish that pBsNb can serve as a powerful tool for photo-controlled xTPX2 for optogenetic regulation of branched microtubule nucleation, providing a valuable platform for elucidating biochemical mechanisms in cell-free systems.

Claims

1. A method for site-specific chemical modification of nanobodies, characterized in that, One lysine residue is retained in the amino acid sequence of the nanobody, and all others are mutated to arginine. The N-terminal α-amino group of the nanobody or the only lysine residue in the amino acid sequence is reacted with N-hydroxysuccinimide ester to attach a functional group.

2. A method for site-specific chemical modification of nanobodies, characterized in that, After mutating all natural lysine in the amino acid sequence of the nanobody to arginine, the α-amino group at the N-terminus of the nanobody was subjected to site-specific chemical modification using NHS ester.

3. The method according to claim 2, characterized in that, The chemical modification groups are affinity tags, fluorescent molecules, and bioorthogonal groups; the nanobodies are green fluorescent protein (GFP) nanobodies, EGFR nanobodies, PD-L1 nanobodies, fluorescent protein mCherry nanobodies, or Alfa tag nanobodies.

4. The method according to claim 1, characterized in that, The specific steps are as follows: Introducing a cysteine ​​residue at the N-terminus and reacting it with a 2-cyanobenzothiazole derivative to block the α-amino group at the N-terminus, while retaining any lysine residue in the nanobody backbone region and mutating all the remaining lysine residues to arginine, thereby achieving site-specific chemical modification of the only remaining lysine residue on the nanobody backbone using NHS ester. A cysteine ​​residue needs to be introduced at the N-terminus to react with the CBT derivative to block the α-amino group at the N-terminus. At the same time, all lysine residues in the nanobody backbone region are mutated to arginine, and a lysine residue is introduced into the CDR region by mutation, so as to achieve site-specific chemical modification of the only lysine residue in the CDR region using NHS ester.

5. The method according to claim 4, characterized in that, The CBT derivatives are 6-amino-2-cyanobenzothiazole, 6-hydroxy-2-cyanobenzothiazole, 6-biotin-2-cyanobenzothiazole, or 6-azido-2-cyanobenzothiazole. The nanobody backbone has only one lysine residue, which undergoes site-specific chemical modification. The chemical modification groups are PEG chains and fluorescent molecules. The single lysine residue in the CDR region is specifically modified using NHS ester, with the modification group being a photocage molecule, i.e., a photoresponsive protecting group that can be removed under light. The chemical modification groups include a biotinylate tag for affinity, FAM and Cy5 groups for fluorescent molecules, and DBCO for bioorthogonal groups. The C-terminus of the backbone-site-specifically modified nanobody is modified with VcMMAE or TMP.

6. The chemical group according to claim 5, characterized in that, The PEG chain is a PEG with a molecular weight of 5 kDa. 5000 The fluorescent molecules are either 7-diethylaminocoumarin-3-carboxylic acid (DEAC) or 5-carboxytetramethylrhodamine (TMR); the nanobodies are either green fluorescent protein (GFP) nanobodies, EGFR nanobodies, PD-L1 nanobodies, fluorescent protein mCherry nanobodies, or Alfa tag nanobodies.

7. The method according to claim 5, characterized in that, Photocage molecules are 1-(3,4-methylenedioxy-6-nitrophenyl)-ethoxycarbonyl, [7-(diethylamino)coumarin-4-yl]methyl, (4,5-dimethoxy-2-nitrophenyl)methanol or 6-bromo-7-hydroxy-4-(hydroxymethyl)-2H-chromen-2-one.

8. A photocage-like bispecific nanobody, characterized in that, The first nanobody requires the introduction of a cysteine ​​residue at its N-terminus to react with a 2-cyanobenzothiazole derivative to block the α-amino group at the N-terminus. Simultaneously, lysine residues in all backbone regions of both nanobodies are mutated to arginine residues. A lysine residue is introduced into the CDR region of the second nanobody. This allows for site-specific chemical modification of the unique lysine residue on the bispecific nanobody using NHS esters to create a photocage molecule. The two nanobodies are then linked together by a gene-encoded linker.

9. The photocage-encapsulated bispecific nanobody according to claim 8, characterized in that, Photocage-encapsulated bispecific nanobody is Luci-GBP R (K34-DEACM)-Nb R Alfa nanobodies are green fluorescent protein (GFP) nanobodies, EGFR nanobodies, PD-L1 nanobodies, fluorescent protein mCherry nanobodies, or Alfa tag nanobodies.

10. The application of the method according to any one of claims 1-7 or the photocage-encapsulated bispecific nanobody according to claim 8 or 9 in immunoimaging for non-diagnostic and non-therapeutic purposes, in the preparation of tumor-targeted therapeutic drugs, or in the regulation of branched microtubule nucleation in Xenopus laevis egg extract for non-diagnostic and non-therapeutic purposes.

Images