Engineered bacteria for bladder cancer

Abstract

The present invention relates to compositions comprising one or more bacteria engineered to express a payload that recruits immune cells. The present invention also relates to a kit comprising said compositions. The compositions of the invention find utility in the field of medicine, such as for treating 5 cancer including bladder cancer.

Description

[0001] Engineered Bacteria for Bladder Cancer

[0002] Field of the Invention

[0003] The present invention relates to bacteria with therapeutic applications. In particular, the bacteria of the invention may find utility in the treatment of bladder cancer.

[0004] SEQUENCE LISTING

[0005] The instant application contains a Sequence Listing which has been submitted herewith and is hereby incorporated by reference in its entirety. Said .xml copy, created on February 27, 2025, is named 008681454, and is 20,219 bytes in size.

[0006] Background

[0007] Cancer is a leading cause of death globally, accounting for an estimate of 10 million deaths in 2020 [1], Among different cancers, bladder cancer has been historically overlooked, with relatively fewer new therapies developed for patients [2], Correspondingly, there has not been an improvement in survival rates for patients in the last three decades. Currently, the clinical gold standard for treating non-muscle invasive bladder cancer (NMIBC) involves a standard surgical procedure known as transurethral resection of the bladder tumour (TURBT), followed by the intravesical instillation of Bacillus Calmette- Guerin (BCG) [3, 4], BCG works by binding to the cancerous urothelial cells in the bladder wall and is internalized by these cells. The internalization process triggers the release of a wide range of immunostimulatory cytokines such as TNF-a, IFN-y, IL-2, IL-6, IL-8, and IL-12 [5-7] and initiates an immune response that can eliminate the bladder carcinoma cells [7],

[0008] Most solid human tumours display one of the three distinct immune phenotypes: immune desert, immune excluded, or immune inflamed

[0015] , In an immune desert tumour, T cells are absent from the tumour microenvironment (TME). Immune excluded tumours are characterised by having T cells accumulate in the tumour periphery but do not infiltrate the TME. Lastly, immune inflamed tumours are characterised by T cells infiltrating the TME but not killing the carcinoma cells. In a metastatic bladder carcinoma cohort, it was reported that most of the tumours exhibited the immune excluded phenotype (47%), with the remaining tumours being immune inflamed (27%) and immune desert (27%)

[0016] , Most bladder tumours belong to the immune-excluded phenotype.

[0009] Despite widespread use of BCG, there are a number of disadvantages associated with this therapy. For example, while approximately 80% of patients with high-risk NMIBC respond to BCG therapy, up to 90% experience recurrence within five years [8], Additionally, a significant number of patients develop refractoriness to BCG after multiple instillations [9], It has been reported that 8% of BCG treatments had to be interrupted because of adverse events

[0010] , The lack of clinical response in patients are in part due 8880361

[0010] 2 to suboptimal binding of BCG to fibronectin

[0011] and vigorous phagocytosis of BCG by leukocytes

[0012] , Bladder tumours may differ in the type and degree of immune cell infiltration, secreted factor profile in the tumour microenvironment (TME) and tissue architecture

[0013] , It has been reported that despite being in its weakened state, attenuated BCG can become active and cause multi-system disease in treated patients

[0024] , Finally, bladder carcinoma patients have been increasingly affected by a shortage of BCG in recent years as pharmaceutical companies have ceased production of BCG, citing high production costs, limited scalability, lengthy incubation periods, and the drug’s generic status

[0014] ,

[0011] Recent investigations into alternative treatments include the development of alternative BCG strains, delivering immune checkpoint inhibitors alongside BCG and using gene therapy and intravesical biologic therapies.

[0012] With regards to BCG alternative strains, the attenuated Salmonella enterica typhi Ty21 a strain [21 , 22] and Salmonella enterica typhi ZH9 strain

[0023] are undergoing testing to determine whether these strains are effective alternatives to conventional BCG.

[0013] More recently, the intravesical instillation of BCG and immune checkpoint inhibitors (e.g., Atezolizumab, Pembrolizumab, Durvalumab, Nivolumab, and Avelumab) has been approved as therapy for NMIBC and muscle-invasive bladder cancer (MIBC)

[0025] , However, immune checkpoint inhibitors are best suited to treat immune inflamed tumours and not immune excluded or immune desert tumours.

[0014] Viral gene therapies have been developed for instillation in the bladder for sustained delivery of immunomodulators

[0026] , A prominent example recently approved by the US FDA is Adstildadrin (nadofaragene firadenovec), a non-replicating adenoviral gene therapy that is used to deliver interferon alpha 2b to treat high-risk BCG unresponsive NMIBC patients with carcinoma in situ (CIS) with or without papillary tumours

[0052] , The viral vector enters the cells of the bladder wall and causes these cells to transiently secrete interferon alpha 2b. The transient nature of this expression necessitates repeated patient administration every three months. Another intravesical viral gene therapy undergoing clinical trials is cretostimogene grenadenorepvec (CG0070), which utilizes a recombinant oncolytic adenovirus to deliver granulocyte-macrophage colony stimulating factor (GM-CSF)

[0053] , This viral vector preferentially infects and replicates within tumour cells. This transient expression requires more frequent patient administration, with initial treatments every three weeks or once a week for three weeks.

[0015] Other modalities that have been developed include the intravesical ‘TAR-210’ delivery system for treating high-risk BCG-unresponsive NMIBC with FGFR3 alterations

[0054] , This drug-eluting device slowly releases erdafitinib into the bladder after instillation and is currently undergoing clinical trials. Another therapy currently at the clinical trial stage is the Alphal H, a tumoricidal synthetic peptide-lipid complex

[0055] , The intravesical delivery of this complex triggers a dose-dependent cell death response by perturbing the membrane and inhibiting cancer-related genes in cancer cells. Finally, N-803 (Anktiva),a mutant IL-15- based immunostimulatory fusion protein complex (IL15RaFc) has been approved by the US FDA

[0056] , N- 803 is instilled together with BCG into the patient’s bladder to treat BCG unresponsive NMIBC carcinoma in situ and papillary disease. 8880361

[0016] 3

[0017] Shen et al, 2024 reported the discovery of commensal strain Lactobacillus plantarum with an intrinsic cancer-binding mechanism. L. plantarum binding to nasoparyngeal carcinoma (NPC) was shown to be mediated by oligopeptide-binding (OppA) protein recognition of heparan sulfate on the surface of these cancer cells

[0050] , This native interaction was used to deliver prodrugs activated by tumour-associated biosignals to release chemotherapy compounds near NPCs.

[0018] The present invention has been devised in light of the above considerations.

[0019] Summary of the Invention

[0020] The present inventors made the surprising finding that bacteria engineered to deliver cytokines can be used to increase the infiltration of immune cells at tumour sites to kill or enhance the killing of tumour cells.

[0021] In a first aspect, the present invention provides bacteria that are engineered to express a payload comprising a cytokine, which activates and / or attracts an immune cell by binding to an immune cell receptor. In some embodiments, the cytokine is secreted from the bacteria. The secretion of these cytokines can generate a concentration gradient to attract and / or activate immune cells to the bacteria. In some embodiments, the cytokine may be expressed on a surface of the bacteria, e.g., attached to a receptor or surface membrane protein. In some embodiments, the immune cell is (a) a lymphocyte. In preferred embodiments, the lymphocyte is a T cell or a natural killer (NK cell). In further preferred embodiments, the T cell is a CD4+ cell, a CD8+ cell, or a natural killer T cell (NKT cell). In other embodiments, the immune cell is (b) an antigen presenting cell (APC), preferably a dendritic cell (DC). In some embodiments, the bacteria are commensal organisms. These are organisms that reside on the surface of the human body or at mucosa without harming human health. In some embodiments, the bacteria are gram-positive, e.g., of the genus Lactobacillus. In preferred embodiments, the bacteria are of the genus Lactobacillus. In some embodiments, these bacteria are of the species: (a) Lactobacillus plantarum, preferably of the strain WCFS1 , or (b) Lactobacillus salivarius, preferably of the strain DSM20555, or (c) Lactobacillus reuteri, preferably of the strain DSM20016. In some embodiments, the cytokine is any one of CXCL9, CXCL10, CCL4, IL-15, CX3CL1 , IL-21 , CXCL11 , IFN-a, TRAIL, CD40L, GM-CSF, IFN-y, TNF-a, CCL3, CCL5, or any combination thereof. In preferred embodiments, the cytokine is any one of CXCL9, CXCL10, CX3CL1 , IL-15, CCL4, or any combination thereof. In some embodiments, the cytokine is a chemokine. Secretion of these chemokines can generate a concentration gradient to attract particular immune cells to the site of secretion, i.e., to the engineered bacteria. In some embodiments, the bacteria is engineered to deliver two or more cytokines, e.g. including CXCL9, CXCL10, CX3CL1 and / or IL-15. In some embodiments, the two or more cytokines are: (a) CXCL9 and CXCL10, or (b) CX3CL1 and IL-15. In some embodiments, the bacteria are engineered to express CCL4, e.g. in combination with CXCL9, CXCL10, CX3CL1 and / or IL-15.

[0022] In some embodiments, the bacteria further comprise an adhesion molecule that is: (a) overexpressed from the genome of the bacteria as compared to non-engineered bacteria, and / or (b) expressed from a 8880361

[0023] 4 nucleic acid introduced into the bacteria, and / or (c) exogenously coated on a surface of the bacteria. In preferred embodiments, the adhesion molecule is an oligopeptide binding protein (OppA). In some embodiments, the adhesion molecule comprises a membrane anchor with an amino acid sequence having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to any one of SEQ ID NOs: 1-10. In some embodiments, the membrane anchor is cleavable, e.g., by a signal peptidase. In some embodiments, the adhesion molecule comprises a surface adhesin with an amino acid sequence having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to any one of SEQ ID NOs: 11-17. In some embodiments, the adhesion molecule comprises a membrane anchor and a surface adhesin. In some embodiments, the membrane anchor and the surface adhesin are heterologous to each other. In some embodiments, the membrane anchor consists of a secretion signal peptide. In preferred embodiments, the secretion signal peptide is a lipoprotein secretion signal peptide, e.g., Lp_0783. In some embodiments, the membrane anchor comprises a secretion signal peptide and a cleavage site. The signal peptide may comprise the N-terminal of the membrane anchor sequence and the cleavage site may comprise the C-terminal of the membrane anchor. In some embodiments, the cleavage site is cleaved. In other embodiments, the cleavage site is not cleaved. This may be for example because the membrane anchor comprises a signal peptide which is not accessible to cleavage enzymes. In some embodiments, the cleavage site is cleaved In preferred embodiments, the cleavage site is cleavable by an enzyme, preferably signal peptidase II (SPase II). In some embodiments, the membrane anchor is partially cleaved such that a portion of the membrane anchor is retained in the adhesion molecule. Accordingly, in some embodiments, the adhesion molecule comprises a partial membrane anchor sequence, and a surface adhesin when anchored to a surface of the bacteria. The adhesion molecule may be anchored to the surface of a bacteria by utilising the lipoprotein anchoring machinery of the engineered bacteria, e.g., using a diacylglycerol transferase enzyme. Therefore, the membrane anchor may form a transmembrane anchor. In some embodiments, the membrane anchor is a transmembrane anchor.

[0024] In a second aspect, the present invention provides a composition according to the first aspect. In some embodiments, the composition comprises a first bacteria expressing CXCL9 and a second bacteria expressing CXCL10, according to the bacteria of the first aspect. In other embodiments, the first bacteria express CXCL10 and a second bacteria express CXCL9, according to the bacteria of the first aspect. In other embodiments, the composition comprises a first bacteria expressing CX3CL1 and a second bacteria expressing a payload comprising IL-15. In other embodiments, the first bacteria express IL-15 and a second bacteria express CX3CL1 , according to the bacteria of the first aspect. In some embodiments, the composition is for use as a medicament. In some embodiments, the composition is used in a method of treatment.

[0025] In a third aspect, the present invention provides bacteria according to the first aspect or a composition according to the second aspect for use in the treatment of cancer in a subject. In preferred embodiments, the cancer is bladder cancer. For example, the bladder cancer may be a non-muscle invasive bladder 8880361

[0026] 5 cancer (NMIBC) or muscle invasive bladder cancer (MIBC). When the cancer is a bladder cancer, the treatment may comprise the step of administering the bacteria to the subject via intravesical instillation, e.g. using a catheter. The secretion of cytokines in the bladder can generate a concentration gradient to attract and / or activate immune cells to the bladder cancer. In some embodiments, the cancer exhibits an immune excluded and / or immune desert phenotype. In some embodiments, the treatment is administered in place of Bacillus Calmette-Guerin (BCG) therapy or alongside BCG therapy. In some embodiments, the treatment is administered as a follow up therapy suitable for NMIBC patients who have experienced BCG failure.

[0027] In a fourth aspect, the present invention proves a kit comprising a first container containing a first composition according to the second aspect and a second container containing a second composition according to the second aspect, in which the payload expressed by the bacteria of the first composition comprises a different cytokine to the cytokine comprised within the payload that is expressed by the bacteria of the second composition. For example, the first bacteria may express CXCL9, and the second bacteria may express CXCL10. In other examples, the first bacteria may express CX3CL1 , and the second bacteria may express IL-15.

[0028] The inventors also found that the commensal strain Lactobacillus plantarum exhibited a native affinity for a variety of bladder carcinoma cells through its cell surface protein OppA. The inventors have demonstrated that overexpressing OppA can enhance bacterial binding to bladder carcinoma cells. These overexpressed OppA genes produce an OppA polypeptide. The OppA polypeptide may be endogenous or exogenous to the bacteria. The OppA polypeptide may be a homologous or heterologous polypeptide that is expressed from a nucleic acid introduced into the bacteria. In some embodiments, the nucleic acid encodes an OppA polypeptide that has the same sequence as an OppA polypeptide encoded in its genome. OppA fusion constructs were developed with enhanced bladder carcinoma cell binding capabilities. These fusion constructs comprise a membrane anchor and an OppA surface adhesin. The membrane anchor and OppA surface adhesin may be ‘homologous’ to one another (i.e., derived from the same source molecule) or ‘heterologous’ to one another (i.e., derived from different source molecules) and / or synthetically engineered.

[0029] Therefore, in another aspect, the invention provides bacteria engineered to express or overexpress an adhesion molecule as described herein. In preferred embodiments, the adhesion molecule is an OppA polypeptide or encodes an OppA polypeptide. The OppA polypeptide may be endogenous or exogenous to the bacteria. The OppA polypeptide may be a homologous or heterologous polypeptide that is expressed from a nucleic acid introduced into the bacteria. In some embodiments, the nucleic acid encodes an OppA polypeptide that has the same sequence as an OppA polypeptide that is natively expressed by the bacteria from its genome. In some embodiments, the OppA polypeptide may be exogenously coated on a surface of the bacteria as described herein.

[0030] In preferred embodiments, the adhesion molecule is a fusion construct. The fusion construct may comprise a membrane anchor and a surface adhesin as described herein. In some embodiments, the 8880361

[0031] 6 membrane anchor and surface adhesin are homologous to one another. In preferred embodiments, the membrane anchor and surface adhesin are heterologous to one another.

[0032] In some embodiments, the fusion construct comprises a membrane anchor comprising an amino acid sequence having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to any sequence listed in Table 1 . In some embodiments, the fusion construct comprises a surface adhesin comprising an amino acid sequence having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to any sequence listed in Table 2. For example, in preferred embodiments, the fusion molecule comprises the membrane anchor of SEQ ID NO: 1 and the surface adhesin of any one of SEQ ID NOs: 11-17. In other embodiments, the fusion molecule comprises the membrane anchor of SEQ ID NO: 2 and the surface adhesin of SEQ ID NOs: 11-17. In some embodiments, the fusion molecule comprises the membrane anchor of SEQ ID NO: 3 and the surface adhesin of any one of SEQ ID NOs: 11-17. In some embodiments, the fusion molecule comprises the membrane anchor of SEQ ID NO: 4 and the surface adhesin of any one of SEQ ID NO: 11 - 17. In some embodiments, the fusion molecule comprises the membrane anchor of SEQ ID NO: 5 and the surface adhesin of any one of SEQ ID NOs: 11-17. In some embodiments, the fusion molecule comprises the membrane anchor of SEQ ID NO: 6 and the surface adhesin of any one of SEQ ID NOs: 11-17. In some embodiments, the fusion molecule comprises the membrane anchor of SEQ ID NO: 7 and the surface adhesin of any one of SEQ ID NOs: 11-17. In some embodiments, the fusion molecule comprises the membrane anchor of SEQ ID NO: 8 and the surface adhesin of any one of SEQ ID NOs: 11-17. In some embodiments, the fusion molecule comprises the membrane anchor of SEQ ID NO: 9 and the surface adhesin of any one of SEQ ID NOs: 11-17. In some embodiments, the fusion molecule comprises the membrane anchor of SEQ ID NO: 10 and the surface adhesin of any one of SEQ ID NOs: 11-17.

[0033] Table 1. Exemplary membrane anchor sequences 8880361

[0034] 7

[0035] Table 2. Exemplary surface adhesin sequences 8880361

[0036] 8 8880361

[0037] 9

[0038] In some embodiments, the bacteria are engineered to:

[0039] (a) express a payload comprising a cytokine, wherein the payload activates and / or attracts an immune cell by binding to an immune cell receptor, and, (b) express or overexpress an adhesion molecule.

[0040] The invention includes the combination of the aspects and preferred features described except where such a combination is clearly impermissible or expressly avoided.

[0041] In one aspect, the invention provides a composition comprising one or more bacteria, wherein the one or more bacteria are engineered to express a payload, wherein the payload comprises at least two immunomodulators, wherein at least one of the immunomodulators is an attractor, wherein the attractor recruits immune cells by binding chemokine receptors. Thus, the attractor is a cytokine, typically a chemokine. The attractor recruits the immune cells by directing their migration along a concentration gradient. 8880361

[0042] 10

[0043] In some embodiments, the composition comprises one bacterium engineered to express the payload comprising at least two immunomodulators. In some embodiments, the composition comprises two or more bacteria each engineered to express at least one immunomodulator, wherein the immunomodulators expressed by the two or more bacteria together constitute the payload.

[0044] In another aspect, the invention provides a kit comprising a first container containing a first composition comprising one bacterium engineered to express at least one first immunomodulator, and a second container containing a second composition comprising one bacterium engineered to express at least one second immunomodulator, wherein the first immunomodulator is different from the second immunomodulator, and wherein at least one of the immunomodulators is an attractor, wherein the attractor recruits an immune cell by binding a chemokine receptor on the immune cell. The bacteria and immunomodulators are defined herein.

[0045] In some embodiments, the at least two immunomodulators are selected from the group consisting of a chemokine, an interferon (IFN), an interleukin (IL), a tumour necrosis factor (TNF), a haematopoietin, a transforming growth factor beta (TGF-p), a colony stimulating factor (CSF), VEGF-A, MIF, BAFF, CXCL5, CXCL1 , HGF, CXCL11 , IFN-y, IL-1a, IL-2, IL-3, IL-4, IL-7, IL-15, IL-18, IL-22, M-CSF, CCL2, CCL20, SCF, SDF-1a, CXCL9, CXCL10, CCL4, IL-15, CX3CL1 , IL-21 , IFN-a, TRAIL, CD40L, GM-CSF, TNF-a, CCL3, CCL5, and an immune checkpoint inhibitor. In some embodiments, the at least two immunomodulators do not include CCL5. In preferred embodiments, the at least two immunomodulators are selected from the group consisting of CX3CL1 , CXCL9, CXCL10, CCL4, CCL5, IL-15, IL-2 and an immune checkpoint inhibitor. In yet more preferred embodiments, the at least two immunomodulators are selected from the group consisting of CX3CL1 , CXCL9, CXCL10, CCL4, IL-15, IL-2 and an immune checkpoint inhibitor.

[0046] In some embodiments, the at least one attractor is selected from the group consisting of a chemokine, IL- 8, CXCL5, CXCL1 , CXCL11 , CCL2, CCL20, SDF-1a, CXCL9, CXCL10, CCL4, CX3CL1 , CXCL11 , CCL3 and CCL5. In some embodiments, the at least one attractor is selected from the group consisting of a chemokine, IL-8, CXCL5, CXCL1 , CXCL11 , CCL2, CCL20, SDF-1a, CXCL9, CXCL10, CCL4, CX3CL1 , CXCL11 , and CCL3. In preferred embodiments, the at least one attractor is / are selected from the group consisting of CX3CL1 , CXCL9, CXCL10, CCL4 and CCL5. In preferred embodiments, the at least one attractor is selected from the group consisting of CX3CL1 , CXCL9, CXCL10 and CCL4.

[0047] In some embodiments, the payload comprises CXCL9 and CXCL10. In some embodiments, the payload comprises CX3CL1 and IL-15. In some embodiments, the payload comprises CXCL9, CXCL10 and CCL4. In some embodiments, the payload comprises CXCL9, CXCL10 and IL-2. In some embodiments, the payload comprises CXCL9, CXCL10, CCL4 and IL-2. In some embodiments, the payload comprises CXCL9, CXCL10 and CCL5. In some embodiments, the payload comprises CXCL9, CXCL10 and IL-15. In some embodiments, the payload comprises CXCL9, CXCL10, IL-15 and IL-2. 8880361

[0048] 11

[0049] In some embodiments, the composition comprises a first bacterium engineered to express CXCL9 and a second bacterium engineered to express CXCL10. In some embodiments, the composition comprises a first bacterium engineered to express CX3CL1 and a second bacterium engineered to express IL-15.

[0050] In some embodiments, at least one of the (at least two) immunomodulators is an activator, wherein an activator enhances the functional potency, persistence, or cytotoxic activity of recruited immune cells. In some embodiments, the at least one activator is selected from the group consisting of IL-15, IL-2 and an immune checkpoint inhibitor. In some embodiments, the immune checkpoint inhibitor is an antibody, nanobody, peptide, aptamer, RNA-based inhibitor or small molecule. In preferred embodiments, the immune checkpoint inhibitor is a PD-L1 inhibitor, a PD-1 inhibitor or a CTLA-4 inhibitor.

[0051] In some embodiments, the activator is IL-2. In some embodiments, the payload comprises IL-2 and at least one additional activator selected from the group consisting of IL-15, a PD-L1 inhibitor, a PD-1 inhibitor and a CTLA-4 inhibitor. In some embodiments, the activator is a PD-L1 inhibitor. In some embodiments, the payload comprises a PD-L1 inhibitor and at least one additional activator selected from the group consisting of IL-15, IL-2, a PD-1 inhibitor and a CTLA-4 inhibitor. In some embodiments, the activator is a PD-1 inhibitor. In some embodiments, the payload comprises a PD-1 inhibitor and at least one additional activator selected from the group consisting of IL-15, IL-2, a PD-L1 inhibitor and a CTLA-4 inhibitor. In some embodiments, the activator is a CTLA-4 inhibitor. In some embodiments, the payload comprises a CTLA-4 inhibitor and at least one additional activator selected from the group consisting of IL-15, IL-2, a PD-L1 inhibitor and a PD-1 inhibitor.

[0052] In some embodiments, the immunomodulators are secreted from the bacteria.

[0053] In some embodiments, the immune cell recruited by the attractor is a lymphocyte. In some embodiments, the lymphocyte is a T cell or a natural killer cell (NK). In some embodiments, the T cell is a CD4+ cell or a CD8+ cell. In some embodiments, the immune cell recruited by the attractor is a natural killer T cell (NKT) or an antigen presenting cell (APC). In some embodiments, the APC is a dendritic cell (DC). In some embodiments, the immune cell whose functional potency, persistence, or cytotoxic activity is enhanced by the activator is a lymphocyte. In some embodiments, the lymphocyte is a T cell or a natural killer cell (NK). In some embodiments, the T cell is a CD4+ cell or a CD8+ cell. In some embodiments, the immune cell whose functional potency, persistence, or cytotoxic activity is enhanced by the activator is a natural killer T cell (NKT) or an antigen presenting cell (APC). In some embodiments, the APC is a dendritic cell (DC).

[0054] In some embodiments, the bacterium is a commensal bacterium. In some embodiments, the bacteria are commensal bacteria. In some embodiments, the bacterium is a gram-positive bacterium, optionally wherein the gram-positive bacterium is of the genus Lactobacillus. In some embodiments, the bacteria are gram-positive bacteria, optionally wherein the gram-positive bacteria are of the genus Lactobacillus. In some embodiments, the bacterium / bacteria of the genus Lactobacillus is / are of the species Lactobacillus plantarum. In some embodiments, the Lactobacillus plantarum bacterium / bacteria is / are of 8880361

[0055] 12 the strain WCFS1 . In other embodiments, the bacterium / bacteria of the genus Lactobacillus is / are of the species Lactobacillus salivarius. In some embodiments, the Lactobacillus salivarius bacterium / bacteria is / are of the strain DSM20555. In other embodiments, the bacterium / bacteria of the genus Lactobacillus is / are of the species Lactobacillus reuteri. In some embodiments, the Lactobacillus reuteri bacterium / bacteria is / are of the strain DSM20016.

[0056] In some embodiments, the one or more bacteria are engineered to express an adhesion molecule that is that is overexpressed from the genome of the bacterium / bacteria as compared to a non-engineered bacterium / bacteria and / or expressed from a nucleic acid introduced into the bacterium / bacteria and / or exogenously coated on a surface of the bacterium / bacteria. In some embodiments, the adhesion molecule is an oligopeptide binding protein (OppA). In some embodiments, the adhesion molecule comprises a membrane anchor with an amino acid sequence having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to any one of SEQ ID NOs: 1-10, optionally wherein the membrane anchor is cleavable. In other embodiments, the adhesion molecule comprises a surface adhesin with an amino acid sequence having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to any one of SEQ ID NOs: 11-17. In some embodiments, the adhesion molecule comprises a membrane anchor and a surface adhesin that are heterologous to each other. In some embodiments, the membrane anchor comprises a secretion signal peptide and a cleavage site, optionally wherein the cleavage site is cleavable by an enzyme, further optionally wherein the enzyme is a signal peptidase II (SPase II).

[0057] In one aspect, the invention provides the composition according to any one of the embodiments disclosed herein for use as a medicament. Disclosed herein is a composition according to any one of the embodiments disclosed herein for use in the treatment of cancer in a subject, optionally wherein said cancer is bladder cancer, further optionally wherein said bladder cancer is non-muscle invasive bladder cancer (NMIBC) or muscle invasive bladder cancer (MIBC). In some embodiments, the cancer exhibits an immune excluded and / or immune desert phenotype. In some embodiments, the treatment is administered in place of Bacillus Calmette-Guerin (BCG) therapy or alongside BCG therapy.

[0058] Summary of the Figures

[0059] Embodiments and experiments illustrating the principles of the invention will now be discussed with reference to the accompanying figures in which:

[0060] Figure 1. Schematic representation of the strategy to employ engineered L. plantarum WCFS to deliver different combinations of immunomodulatory proteins to bladder tumours within the bladder. 1) L. plantarum WCFS1 strains were modified to secrete an immunomodulatory protein and anchor surface adhesion protein. 2) Different combinations of engineered strains (2 assets) were assembled for delivery by intravesical installation. 3) Engineered strains present surface adhesion proteins to bind bladder 8880361

[0061] 13 tumours. 4) Sustained delivery of immunomodulatory proteins. 5) Desirable populations of lymphocytes are attracted to the site, e.g., to infiltrate tumours and kill bladder carcinoma cells.

[0062] Figure 2. Measuring relative binding efficiency of Lactobacillus strains to bladder carcinoma cell lines.

[0063] Figure 3. Schematic showing the presence of native OppA binding protein on the L. plantarum WCFS1 cell surface and the anchoring of exogenous OppA genetic variants to facilitate more robust and targeted binding to GAGs (e.g., heparan sulphate) on the surface of bladder carcinoma cells.

[0064] Figure 4. Measuring binding affinity of seven OppA genetic variants to MB49, T24, J82 and UM-UC-3 bladder carcinoma cell lines. (A) Geometric mean fluorescence intensity (MFI) bladder carcinoma cells tagged with seven OppA genetic variants. **** p < 0.00005. (B) MB49 cell count versus fluorescence intensity histogram. (C) T24 cell count versus fluorescence intensity histogram. (D) J82 cell count versus fluorescence intensity histogram. (E) UM-UC-3 cell count versus fluorescence intensity histogram.

[0065] Figure 5. Fusion construct and vector for L. plantarum WCFS1 surface anchoring protein. (A) The pTRK892 shuttle vector served as the starting point for construction. The vector backbone consisted of the pWV01 replicon and the erythromycin resistance gene (Erm). The gene of interest was placed under the control of the strong constitutive phosphoglycerate mutase (pgm) promoter from L. acidophilus NCFM. A different synthetic RBS was generated for each specific genetic construct using the RBS calculator tool developed by Salis Lab. The Term 908 terminator was used for all constructs. (B) Fusion constructs were designed for N-terminal anchoring of OppA genetic variants with the Lp_0783 secretion signal peptide and lipoprotein anchor, and inserted into the gene expression cassette. The Lp_0783 membrane anchor comprised an N-terminal signal peptide sequence and a C-terminal Spase II cleavage site. The SPase II cleavage site was at the boundary between Lp_0783 and the OppA anchor. All OppA genetic variants were tagged with the Myc-Tag for antibody detection. C) Insertion of the N-terminal anchoring constructs in the gene expression cassette.

[0066] Figure 6. Measuring surface display and binding efficiency of anchored OppA genetic variants. (A) Western blot detection of N-terminal anchored OppA expression. (B) Flow cytometry analysis is used to detect expression and surface display of Lp_0783, LBA1665, and LVIS_1959 OppA genetic variants via an N-terminal anchor. Measuring relative binding efficiency of engineered L. plantarum WCFS1 with N- terminal anchored LVIS_1959 to (C) MB49, (D) J82, (E) T24 and (F) UM-UC-3 cell lines. * p < 0.05, ** p < 0.005, *** p < 0.0005, **** p < 0.00005, ns = not significant.

[0067] Figure 7. Measuring the relative binding retention of engineered L. plantarum WCFS1 with N-terminal anchored LVIS_1959 in MB49 im-planted bladders. (A) Timeline of mouse experiment. Single instillation of engineered strains into the bladder. Bladders were har-vested one week after instillation of engineered strains. (B) CFU / g of bacteria recovered from bladder tissue after four days. (C) CFU / g of bacteria recovered from bladder tissue after seven days. * p < 0.05, ns = not significant.

[0068] Figure 8. Development of two therapeutic assets administering different chemokines and cytokines.

[0069] Engineered L. plantarum WCFS1 strains were to be administered via intravesical instillation. (A) L. 8880361

[0070] 14 plantarum WCFS1 strains delivering CXCL9 and CXCL10. (B) L. plantarum WCFS1 strains delivering CX3CL1 and IL-15.

[0071] Figure 9. Confirming chemokine secretion from L. plantarum WCFS1 strains comprising an immunomodulator gene expression cassette. (A) Cloning the immunomodulator gene expression cassettes for chemokine secretion into L. plantarum WCFS1 . (B) Expression and secretion of human CXCL9 and CXCL10. (C) Expression and secretion of mouse CXCL9 and CXCL10. (D) Expression and secretion of human CX3CL1 , mouse CX3CL1 and mouse IL-15.

[0072] Figure 10. Confirming chemokine secretion from other Lactobacillus species comprising an immunomodulator gene expression cassette. (A) Expression and secretion of CXCL9 from L. salivarius DSM20555 comprising a Lp_0373-hCXCL9 construct. (B) Expression and secretion of CXCL9 from L. reuteri DSM20016 comprising a Lp_0373-hCXCL9 construct.

[0073] Figure 11. 3D microfluidic device setup fortesting immune cell migration.

[0074] Figure 12. Experimental setup fortesting the bioactivity of L. plantarum WCFS1 secreted chemokines. (A) Loading media containing secreted chemokines from cultures on the 3D microfluidics device. (B) CXCL9 & CXCL10 chemokine mix has a summative effect on T cell migration. * p < 0.05, **** p < 0.00005, ns = not significant. (C) 3D microfluidics device configurations. The J82 cell line was cultured in the middle chamber as depicted for subsequent experiments. (D) Timeline of experiment.

[0075] Figure 13. Migration data for activated T cells when exposed to L. plantarum WCFS1 secreted chemokines, in the presence of J82 carcinoma cells. (A) Lp-hCXCL9 + Lp-hCXCL10 OD600 = 0.6, (B) Lp-hCX3CL1 OD600 = 0.6 and (C) Lp-CCL4 = 0.6. * p < 0.05, ** p < 0.005, *** p < 0.0005, **** p < 0.00005, ns = not significant.

[0076] Figure 14. Migration data for single positive CD8+ and CD4+ T cells exposed to L. plantarum WCFS1 secreted chemokines after 3 days, in the presence of J82 carcinoma cells. (A) CD8+ T cell count. (B) CD4+ T cell count. * p < 0.05, ** p < 0.005, *** p < 0.0005, **** p < 0.00005, ns = not significant.

[0077] Figure 15. Experimental setup fortesting the bioactivity of L. plantarum WCFS1 secreted CX3CL1 to attract NK cells. (A) Generating culture of Lp-hCX3CL1 in OD600 = 0.6 in R10 medium and loading media containing secreted soluble hCX3CL1 onto the 3D microfluidics device. (B) 3D microfluidics device configurations for NK-92 cell migration. (C) Timeline of the experiment.

[0078] Figure 16. Evaluating the chemoattractant capacity of secreted soluble human CX3CL1 to enhance NK- 92 cell chemotaxis and measure the corresponding cytotoxicity by migrated NK-92 cells toward J82 carcinoma cells (A) Migration data for NK-92 cells when exposed to L. plantarum WCFS1 secreted soluble hCX3CL1 , in the presence of dispersed J82 carcinoma cells. * p < 0.05, ** p < 0.005, *** p < 0.0005, ns = not significant. (B) Quantification of the NK-92 cell-mediated killing of J82 carcinoma cells represented by the ratio of the surface area of DRAQ7 to surface area of CMRA (n=3).

[0079] Figure 17. Cloning of the gene expression cassettes for both chemokine and LVIS_1959 binding protein into L. plantarum WCFS1 into the PTRK892 plasmid backbone. 8880361

[0080] 15

[0081] Figure 18. Therapeutic efficacy of engineered bacteria after the administration of 2 doses. (A) Timeline of mouse experiment. Weekly instillations of engineered L. plantarum WCFS1 strains into the bladder, over the course of 2 weeks. (B) Bladder mass, (C) Bladder volume. * p < 0.05, ** p < 0.005.

[0082] Figure 19. Testing the therapeutic efficacy of the engineered L. plantarum WCFS1 strains with and without the LVIS_1959 surface modification. Weekly instillations of engineered L. plantarum WCFS1 strains into the bladder, over the course of two weeks. (A) Bladder mass and (B) Bladder volume of harvested bladders from the MB49 tumour bearing mice. * p < 0.05, ** p < 0.005.

[0083] Figure 20. Measuring (A) CD8+ T cells, (B) CD4+ T cells, (C) NK1.1+ cells and (D) CD103+ DCs infiltration within resected bladders treated with the combination of the Lp-mCXCL9-LVIS_1959 and Lp- mCXCL10-LVIS_1959 strains. ** p < 0.005, *** p < 0.0005.

[0084] Figure 21. Measuring (A) NK1 .1+ cells and (B) CD8+ T cell infiltration within resected bladders treated with the combination of the Lp-mCX3CL1-LVIS_1959 and Lp-mlL-15-LVIS_1959 strains. * p < 0.05, ** p < 0.005.

[0085] Figure 22. Therapeutic efficacy of engineered L. plantarum WCFS1 after the administration of 4 doses. (A) Timeline of mouse experiment. Weekly instillations of engineered L. plantarum WCFS1 strains into the bladder, over the course of 4 weeks. (B) Bladder mass, (C) Bladder volume. * p < 0.05, ** p < 0.005.

[0086] Figure 23. Downstream immunomodulators upregulated in response to commercial CXCL9 and CXCL10, or CCL4, and engineered L. plantarum WCFS1 secreted CXCL9 and CXCL10, or CCL4, were quantified from the culture media in the 3D microfluidics device. (A) IL-21 , (B) CXCL11 , (C) IFN-a, (D) TRAIL, (E) CD40L, (F) GM-CSF, (G) TNFa, (H) CX3CL1 , (I) IL-15. * p < 0.05, ** p < 0.005, *** p < 0.0005, **** p < 0.00005, ns = not significant.

[0087] Figure 24. Downstream immunomodulators upregulated in response to (A) commercial CX3CL1 with NK- 92 cells and (B) engineered L. plantarum WCFS1 secreted CX3CL1 , were quantified from the culture media in the 3D microfluidics device. * p < 0.05, ** p < 0.005, *** p < 0.0005, **** p < 0.00005, ns = not significant.

[0088] Figure 25. Downstream immunomodulators upregulated in response to L. plantarum WCFS1 strains engineered to secrete mCX3CL1 and mlL-15, were quantified from the mouse bladder tumour lysates. (A) CCL3, (B) CCL5 and (C) CXCL10. * p < 0.05, ** p < 0.005, *** p < 0.0005, **** p < 0.00005, ns = not significant.

[0089] Figure 26. Expression and secretion of (A) human and (B) mouse CCL4 by L. plantarum WCFS1 .

[0090] Figure 27. Experimental set-up fortesting the bioactivity of L. plantarum WCFS1 secreted chemokines. (A) Generating cultures of Lp-hCCL4 in ODeoo = 0.3, 0.6 and 1.2. Loading media containing secreted chemokines on the 3D microfluidic device. (B) Concentration of secreted chemokines in AIM-V media + 2% human AB serum after 24h. 8880361

[0091] 16

[0092] Figure 28. Migration data for T cells when exposed to L. plantarum WCFS1 secreted chemokines, in the presence of dispersed J82 carcinoma cells. (A) Lp-hCCL4 OD600 = 0.3, (B) Lp-hCCL4 OD600 = 0.6, (C) Lp-hCCL4 OD600 = 1 .2. * p < 0.05, ** p < 0.005, *** p < 0.0005, **** p < 0.00005, ns = not significant.

[0093] Figure 29. Engineered L. plantarum WCFS1 strains, each producing and secreting a single immunomodulator, can be delivered alone or in rationally designed combinations. Bladder (A) mass and (B) volume for four doses of engineered strains and controls (n = 3-9 mice per group; mean ± s.d.). * p < 0.05, ** p < 0.005. Statistical analysis was conducted using Two-sample T-Test with unequal variance.

[0094] Figure 30. Expression and secretion of (A) PD-L1 nanobody and (B) CTLA-4 nanobody by L. plantarum WCFS1. The expected molecular weight is 15kDa.

[0095] Detailed Description of the Invention

[0096] Aspects and embodiments of the present invention will now be discussed with reference to the accompanying figures. Further aspects and embodiments will be apparent to those skilled in the art. All documents mentioned in this text are incorporated herein by reference.

[0097] The present inventors have discovered that bacteria with intrinsic or engineered affinity for cancer cells can be used to deliver immunomodulators such as cytokines to tumour sites and that the sustained expression of cytokines therefrom can be used to recruit immune cells to the TME. For example, by producing a concentration gradient of cytokines, particular immune cells can be recruited to infiltrate the TME to kill, or enhance the killing of, tumour cells.

[0098] Lactobacillus strains are naturally present in the microbiome of many tumour and surrounding healthy tissues, for example in healthy and carcinomatous nasopharyngeal tissue. The inventors have identified a commensal Lactobacillus plantarum (Lp) strain (Lactobacillus plantarum WCFS1) with surprising affinity for bladder cancer cells. Without being bound by theory, the surface of this Lp strain contains OppA adhesin proteins with strong affinity for heparan sulphate, which is abundantly expressed on many cancer cells. The presence of several of these surface adhesins on L. plantarum makes this species particularly advantageous for targeting and colonising cancerous tissue.

[0099] Lactobacillus are commensal strains that are generally regarded as safe (GRAS) by the US FDA for consumption by humans and many animals. L. plantarum is cleared from non-cancerous tissues at much faster rates than strains like Salmonella typhimurium, Listeria monocytogenes, Pseudomonas aeruginosa and Escherichia coll, thus lowering the risk of bacterial infections following administration. Lactobacillus is also naturally present in many mucosa and they can thus be administered mucosally as another means of targeted delivery to avoid the side effects of systemic administration.

[0100] The present inventors found that increasing the expression of adhesins such as OppA can enable or enhance cancer cell binding in bacterial cells. This can be achieved through various mechanisms, for example by (i) increasing endogenous OppA expression from the genome of OppA expressing bacteria, and / or (ii) introducing vectors encoding homologous or heterologous OppA adhesins into bacteria, this 8880361

[0101] 17 includes for example a vector encoding an OppA sequence that is also natively expressed by the bacteria, and / or (iii) by coating bacteria with exogenous adhesins (e.g., recombinantly expressed and purified OppA proteins).

[0102] To understand how bladder tumours may differ among bladder carcinoma patients, the immune phenotype of bladder tumours was investigated. As described herein, the commensal strain L. plantarum WCFS1 was engineered to present bacterial surface adhesins to boost its binding efficiency to glycosaminoglycans (GAGs) expressed at the surface of bladder carcinoma cells, and to secrete immunomodulatory proteins such as cytokines and chemokines, capable of enhancing the infiltration of activated tumour-infiltrating lymphocytes within the bladder tumours to kill carcinoma cells (Figure 1). Two different therapeutic approaches, comprising different engineered L. plantarum WCFS1 strains, were developed and administered to bladder carcinoma patients via intravesical instillation.

[0103] Accordingly, disclosed herein are bacteria engineered to deliver an immunomodulator to activate and / or attract immune cells. Also disclosed herein is an engineered payload for the secretion of an immunomodulator to attract immune cells by generating a chemoattractant gradient.

[0104] In one aspect, the present invention provides bacteria engineered to express a payload comprising a cytokine, wherein the payload activates and / or attracts an immune cell by binding to an immune cell receptor. In some embodiments, the immune cell is attracted to the bladder, preferably to a bladder cancer.

[0105] In another aspect, the present invention provides a bacteria engineered to express or overexpress an adhesion molecule. In preferred embodiments, the adhesion molecule is OppA. Described herein, the engineered bacteria comprising said adhesion molecule could be used for the targeted and sustained delivery of any payload known to one skilled in the art. This could include for example, a chemotherapy agent, gene therapy agent or any other therapeutic.

[0106] In a further aspect, the present invention provides bacteria engineered to:

[0107] (a) express a payload comprising a cytokine wherein the payload activates and / or attracts an immune cell by binding to an immune cell receptor, and,

[0108] (b) express or overexpress an adhesion molecule.

[0109] Described herein are compositions comprising one or more bacteria. Also described herein are kits comprising one or more containers comprising one or more compositions comprising one or more bacteria.

[0110] Bacteria

[0111] The present invention relates to any bacteria. In some embodiments, the bacteria are gram-positive bacteria. In some embodiments, the bacteria are gram-negative bacteria. In a preferred embodiment, the bacteria are commensal bacteria. Commensal bacteria include bacteria native to the microbiome of the subject (e.g. mammalian or, more specifically, human microbiome). This includes, for example, bacteria 8880361

[0112] 18 that residue on the surface of the body or at mucosa without harming the subject’s health. The bacteria may be derived from a cancer tissue or a tissue where the cancer can occur. The bacteria may be a component of the native microbiome of the tissue or cancer tissue. The tissue may be a mucosal tissue, e.g., the lining of the respiratory, gastrointestinal or genitourinary tracts. Examples of tissues from which the bacteria may be derived include but are not limited to nasal, nasopharyngeal, laryngeal, throat, tracheal, bronchial, oral, esophageal, gastric, intestinal, colorectal, bladder, urethral, vaginal, ovarian or breast tissue.

[0113] In some embodiments, the bacteria are probiotic. In preferred embodiments, the bacteria are of the genus Lactobacillus. In further embodiments, the bacteria are of the species Lactobacillus plantarum. In other embodiments, the bacteria are of the species Lactobacillus salivarius. In other embodiments, the bacteria are of the species Lactobacillus reuteri. In some embodiments, the bacteria are Lactobacillus salivarius of the strain DSM20555. In other embodiments, the bacteria are Lactobacillus reuteri of the strain DSM20016. In preferred embodiments, the bacteria are Lactobacillus plantarum of the strain WCFS1.

[0114] In some embodiments, the bacteria natively express endogenous OppA from their genome. In other embodiments, the bacteria do not natively express endogenous OppA from their genome.

[0115] As described herein, the present invention provides a live commensal Lactobacillus plantarum WCFS1 strain engineered to express and secrete immunomodulatory protein. In some embodiments, the bacteria comprise a gene expression cassette containing the DNA sequence of an immunomodulatory protein fused to a selected secretion signal peptide. This enables the intravesical delivery of a single immunomodulatory protein with an engineered bacterial strain over an extended period of time and without external maintenance.

[0116] Payload

[0117] Described herein is a payload comprising an immunomodulator. Preferably, the immunomodulator is a cytokine. Accordingly, in some aspects, the payload comprises a cytokine. In some embodiments, the payload comprises two or more cytokines. In some embodiments, the cytokine is secreted from the bacteria. In other embodiments, the cytokine is presented on a surface of the bacteria. In some embodiments, the cytokines are secreted from the bacteria to generate a chemoattractant gradient. This chemoattractant gradient can attract immune cells, increasing the infiltration of immune cells in for example TMEs.

[0118] In some embodiments, the cytokine can be a chemokine, an interferon (IFN), an interleukin (IL), a tumour necrosis factor (TNF), a haematopoietin, a transforming growth factor beta (TGF-p), or a colony stimulating factor (CSF). In some embodiments, the cytokine is any one of VEGF-A, MIF, BAFF, CXCL5, CXCL1 , HGF, CXCL11 , IFN-y, IL-1a, IL-3, IL-4, IL-7, IL-15, IL-18, IL-22, M-CSF, CCL2, CCL20, SCF, SDF-1a, or any combination thereof. In preferred embodiments, the cytokine is any one of CXCL9, CXCL10, CCL4, IL-15, CX3CL1 , IL-21 , CXCL11 , IFN-a, TRAIL, CD40L, GM-CSF, IFN-y, TNF-a, CCL3, CCL5, or any combination thereof. The cytokine binds to a receptor on an immune cell. 8880361

[0119] 19

[0120] In some embodiments, the cytokine is any cytokine that binds a cytokine receptor on an immune cell. In some embodiments, the receptor is CXCR3, CCR5, IL-15R, CX3CR1 , IL-21 R, IFNAR, DR4, DR5, DcR1 , DcR2, osteoprotegrin, CD40, GM-CSFR, IFNGR, TNFR1 , TNFR2, CCR1 , CCR3, CCR4, CCR5, or any combination thereof. In preferred embodiments, the cytokine binds to receptor CXCR3, CCR5, IL-15R, CX3CR1 , or any combination thereof.

[0121] The inventors discovered that the administration of bacteria of any aspect of the present invention as described herein caused the upregulation of anti-tumour factors. These anti-tumour factors include cytokines such as interleukins and chemokines and interferons, growth factors, enzymes such as metalloproteases, and immune receptors amongst others. In some embodiments, the bacteria of the present invention had the effect of upregulating anti-tumour factors IL-15, IL-21 , CXCL11 , IFN-a, TRAIL, CD40L, GM-CSF, IFN-y, TNF-a, VEGF-A, MIF, CX3CL1 , BAFF, CXCL5, CXCL1 , HGF, IL-1 a, IL-3, IL-4, IL-7, IL-18, IL-22, M-CSF, CCL2, CCL20, MMP-1 , SCF, SDF-1a, TNF-RII, TRAIL, VEGF-A, CCL3, CCL5, or any combination thereof, as compared to controls. Without being bound by theory, this is understood to be one of the mechanisms by which the present invention elicits anti-tumour effects in the bladder.

[0122] Accordingly, in some embodiments, the present invention provides bacteria of the present invention engineered to express a payload comprising any one of IL-15, IL-21 , CXCL11 , IFN-a, TRAIL, CD40L, GM-CSF, IFN-y, TNF-a, VEGF-A, MIF, CX3CL1 , BAFF, CXCL5, CXCL1 , HGF, IL-1 a, IL-3, IL-4, IL-7, IL- 18, IL-22, M-CSF, CCL2, CCL20, MMP-1 , SCF, SDF-1a, TNF-RII, TRAIL, VEGF-A, CCL3, CCL5, or any combination thereof.

[0123] Further described herein is a cytokine delivery system. In some embodiments, the bacteria are engineered to deliver one cytokine. In some embodiments, the bacteria are engineered to deliver two or more cytokines. In some embodiments, a first bacteria delivers a first cytokine, and a second bacteria delivers a second cytokine. In some embodiments, the first cytokine and the second cytokine are the same. In other embodiments, the first cytokine and the second cytokine are different cytokines. In a preferred embodiment, these cytokines are CXCL9 and CXCL10. In another preferred embodiment, these cytokines are CX3CL1 and IL-15.

[0124] In other embodiments, the bacteria are engineered to deliver three or more cytokines. In some embodiments, a first bacteria delivers a first cytokine, a second bacteria delivers a second cytokine and a third bacteria delivers a third cytokine. In some embodiments, the first cytokine, the second cytokine and the third cytokine are the same. In other embodiments, the first cytokine, the second cytokine and the third cytokine are different cytokines from one another. In a preferred embodiment, these cytokines are CXCL9, CXCL10 and CCL4. In another preferred embodiment, these cytokines are CX3CL1 , IL-15 and CCL4.

[0125] As described herein, the payload is expressed from a gene expression cassette. The gene expression cassette may comprise a promoter, for example, the pgm promoter. The gene expression cassette may also comprise a synthetic ribosome binding site. The gene expression cassette may also comprise a secretion signal peptide to enable the secretion of the encoded cytokine. In preferred embodiments, the 8880361

[0126] 20 secretion peptide is Lp_0373. The gene expression cassette may also comprise a Myc tag. The gene expression cassette may also comprise a His tag.

[0127] In one aspect, the present invention provides a live commensal Lactobacillus plantarum WCFS1 strain engineered to anchor and present a surface adhesion molecule to enhance its binding to GAGs present on the surface of bladder carcinoma cells. In some embodiments, the bacteria comprise a gene expression cassette containing the DNA sequence of a fusion protein that consists of a bacterial membrane anchor and selected surface adhesin. This improves the binding efficiency of the engineered bacterial strain to the bladder tumour in the bladder wall to prolong its retention, providing the following advantages:

[0128] (A) Counteracting the multiple filling, storage and voiding of urine by the bladder per day which can reduce the bacterial CFU count in the bladder tumour site.

[0129] (B) Allowing prolonged and consistent delivery of immunomodulatory proteins to retain concentration gradient within the bladder tumour microenvironment (TME) to effectively reach immune cells that are immobilised in the tumour periphery.

[0130] (C) Addresses the limitations of direct instillation of recombinant immunomodulatory proteins in the bladder, where the proteins can be easily excreted due to the filling and voiding of urine by the bladder. The quick excretion of the recombinant immunomodulatory proteins without a carrier prevents the formation of the concentration gradient to attract immune cells in the bladder TME.

[0131] (D) Bacterial delivery systems can colonise and multiply at the tumour site. This can help to prolong the delivery of the immunomodulatory proteins at the tumour site.

[0132] Attractors and Activators

[0133] Described herein is a payload comprising at least two immunomodulators, wherein at least one of the immunomodulators is an attractor. An attractor is preferably a cytokine that recruits immune cells by binding chemokine receptors and directing their migration along a gradient. In some embodiments, the attractor is a chemokine, IL-8, CXCL5, CXCL1 , CXCL11 , CCL2, CCL20, SDF-1a, CXCL9, CXCL10, CCL4, CX3CL1 , CXCL11 , CCL3, CCL5, or any combination thereof. In some embodiments, the attractor is a chemokine, IL-8, CXCL5, CXCL1 , CXCL11 , CCL2, CCL20, SDF-1a, CXCL9, CXCL10, CCL4, CX3CL1 , CXCL11 , CCL3, or any combination thereof. In preferred embodiments, the attractor is CXCL9, CXCL10, CCL4, CX3CL1 , CXCL11 , CCL3, CCL5, or any combination thereof. In preferred embodiments, the attractor is CXCL9, CXCL10, CCL4, CX3CL1 , CCL5, or any combination thereof. In preferred embodiments, the attractor is CXCL9, CXCL10, CCL4, CX3CL1 , or any combination thereof. In preferred embodiments, the payload comprises CXCL9 and CXCL10. In preferred embodiments, the payload comprises CXCL9, CXCL10 and CCL4. In preferred embodiments, the payload comprises CXCL9, CXCL10 and CCL5. In preferred embodiments, the payload comprises CXCL9, CXCL10 and IL-15. In preferred embodiments, the payload comprises CXCL9, CXCL10, IL-15 and IL-2. 8880361

[0134] 21

[0135] In some embodiments, at least one of the immunomodulators is an activator. An activator is an immunostimulatory molecule that enhances the functional potency, persistence, or cytotoxic activity of infiltrating immune cells. In some embodiments, the activator can be an interferon (IFN), an interleukin (IL), a tumour necrosis factor (TNF), a haematopoietin, a transforming growth factor beta (TGF-p), or a colony stimulating factor (CSF). In some embodiments, the activator is any one of VEGF-A, MIF, BAFF, HGF, IFN-y, IL-1a, IL-3, IL-4, IL-7, IL-15, IL-18, IL-22, M-CSF, SCF, or any combination thereof. In preferred embodiments, the activator is any one of IL-2, IL-15, IL-21 , IFN-a, TRAIL, CD40L, GM-CSF, IFN-y, TNF-a, or any combination thereof. In preferred embodiments, the activator is any one of IL-15, IL- 2, or any combination thereof. In some embodiments, the activator is an immune checkpoint inhibitor. In some embodiments, the immune checkpoint inhibitor is an antibody, nanobody, peptide, aptamer, RNA- based inhibitor or small molecule. In preferred embodiments, the immune checkpoint inhibitor is a PD-L1 inhibitor, a PD-1 inhibitor or a CTLA-4 inhibitor.

[0136] In preferred embodiments, the payload comprises CX3CL1 and IL-15. In preferred embodiments, the payload comprises CXCL9, CXCL10 and IL-2. In preferred embodiments, the payload comprises CXCL9, CXCL10, CCL4 and IL-2.

[0137] In some embodiments, the payload comprises IL-2 and at least one additional activator. In some embodiments, the payload comprises IL-15 and at least one additional activator. In some embodiments, the payload comprises CTLA-4 inhibitor and at least one additional activator. In some embodiments, the payload comprises a PD-L1 inhibitor and at least one additional activator. In some embodiments, the payload comprises a PD-1 inhibitor and at least one additional activator. In some embodiments, the payload comprises any one of the combinations of immunomodulators in Table 3. In preferred embodiments, the payload comprises any one of the combinations 1-24 of immunomodulators in Table 3.

[0138] In some embodiments, one bacterium is engineered to express the payload comprising at least two immunomodulators. In other embodiments, two or more bacteria are each engineered to express at least one immunomodulator, wherein the immunomodulators expressed by the two or more bacteria together constitute the payload. In preferred embodiments, a first bacterium is engineered to express CXCL9 and a second bacterium is engineered to express CXCL10. In preferred embodiments, a first bacterium is engineered to express CX3CL1 and a second bacterium is engineered to express IL-15.

[0139] Table 3. Combinations of immunomodulators 8880361

[0140] 22 8880361

[0141] 23 8880361

[0142] 24 8880361

[0143] 25

[0144] Adhesion molecule

[0145] Described herein is an adhesion molecule. In some embodiments, the adhesion molecule is encoded by a nucleic acid sequence. In some embodiments, the bacteria express the adhesion molecule on a surface of the bacteria that is capable of attaching to a cancer cell. The adhesion molecule may attach to an extracellular molecule on a surface of the cancer cell. The adhesion molecule may attach to a polypeptide or polysaccharide. In some embodiments, the adhesion molecule is capable of attaching to a polysaccharide. In preferred embodiments, the polysaccharide is heparan sulphate.

[0146] In some embodiments, the adhesion molecule comprises a membrane anchor and a surface adhesin. In some embodiments, the membrane anchor is a cleavable membrane anchor. In some embodiments, the membrane anchor is a secretion signal peptide. In some embodiments, the membrane anchor comprises a secretion signal peptide. In further embodiments, the membrane anchor comprises a cleavage site. In further embodiments, this cleavage site is cleavable by an enzyme. In other embodiments, the cleavage site is not cleavable. This may be for example because the membrane anchor comprises a signal peptide which is not accessible to cleavage enzymes. In some embodiments, the cleavage site is cleaved. In still further embodiments, the enzyme is a signal peptidase such as signal peptidase II (SPase II). The cleavable membrane anchor can be completely or partially cleaved from the adhesion molecule to enable anchoring of the surface adhesin to a surface of the bacteria. In preferred embodiments, the secretion signal peptide comprises an OppA lipoprotein signal peptide.

[0147] For example, the cleavable membrane anchor may comprise an N-terminal signal peptide for secretion with a SPase II cleavage site at the C-terminal end of the signal peptide. This membrane anchor can be transcribed and translated, forming part of the adhesion molecule within the cell cytoplasm. During the protein secretion process, the SPase II enzyme cleaves at its specific target site and the diacylglycerol transferase enzyme catalyses a coupling reaction between an amino acid of the surface adhesin and a phospholipid of the bacteria membrane. The utilisation of this anchoring strategy can result in the N- 8880361

[0148] 26 terminal anchoring of the chosen surface adhesin, with the C-terminal of the protein facing outwards. The membrane anchor is therefore completely or partially cleaved from the surface adhesin upon anchoring

[0051] .

[0149] Accordingly, the adhesion molecule may be anchored to a surface of the bacteria. In some embodiments, the adhesion molecule consists of a surface adhesin. In some embodiments, the adhesion molecule comprises the surface adhesin. In some embodiments, the membrane anchor is partially cleaved such that a portion of the membrane anchor sequence is retained to form part of the adhesion molecule when anchored to the bacteria surface. Accordingly, in some embodiments, the adhesion molecule comprises a partial membrane anchor sequence, and a surface adhesin when anchored to a surface of the bacteria. Therefore, the membrane anchor may form a transmembrane anchor. These adhesion molecules can be encoded by a nucleic acid comprising a membrane anchor sequence that encodes an amino acid sequence having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to any one of SEQ ID NOs: 2 or 4, prior to cleavage and membrane anchoring.

[0150] In some embodiments, the membrane anchor and the surface adhesin are homologous to one another. In other embodiments, the membrane anchor and the surface adhesin are heterologous to one another. In some embodiments, the membrane anchor and / or surface adhesin are synthetically engineered. In some embodiments, the membrane anchor and the surface adhesin may be a fusion construct. The fusion construct may confer enhanced bladder carcinoma cell binding capabilities as compared to a bacteria without the fusion construct.

[0151] In some embodiments, the membrane anchor comprises an amino acid sequence having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to any one of SEQ ID NOs: 1-10. In further embodiments, the membrane anchor is cleavable. In preferred embodiments, the membrane anchor comprises an amino acid sequence having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to SEQ ID NO: 1 .

[0152] In some embodiments, the surface adhesin is an OppA polypeptide. In some embodiments, the surface adhesin comprises an amino acid sequence having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to any one of SEQ ID NOs: 11-17.

[0153] In some embodiments, the adhesion molecule is overexpressed from the genome of the bacteria as compared to non-engineered bacteria. In some embodiments, the adhesion molecule is expressed from a nucleic acid introduced into the bacteria. In some embodiments, the adhesion molecule is exogenously coated on a surface of the bacteria. In preferred embodiments, the adhesion molecule is an oligopeptide binding protein (OppA).

[0154] In some embodiments, the OppA polypeptide is expressed on a surface of the bacteria. In one embodiment, the OppA polypeptide is an endogenous polypeptide that is expressed from the genome of 8880361

[0155] 27 the Lactobacillus bacteria. In some embodiments, the endogenous OppA polypeptide is overexpressed from the genome of the bacteria, as compared to non-engineered bacteria. In some embodiments, the OppA polypeptide is a polypeptide that is expressed from a nucleic acid introduced into the bacteria. The OppA polypeptide may be endogenous or exogenous to the bacteria. The OppA polypeptide may be a homologous or heterologous polypeptide that is expressed from a nucleic acid introduced into the bacteria. In some embodiments, the nucleic acid encodes an OppA polypeptide that has the same sequence as an OppA polypeptide that is natively expressed by the bacteria from its genome. In yet another embodiment, the OppA polypeptide is exogenously coated on a surface of the bacteria. The OppA polypeptide may be a purified or semi-purified preparation that is added to the bacteria and coated on the surface of the bacteria. The OppA polypeptide may be covalently or non-covalently attached to a surface of the bacteria. The OppA polypeptide may be attached to the cell wall or cell membrane of the bacteria. The bacteria may contain one or more OppA polypeptides. The exogenously coated OppA polypeptide may be an OppA polypeptide that has the same sequence as an OppA polypeptide that is natively expressed by the bacteria, or the exogenously coated OppA polypeptide may not have the same sequence as an OppA polypeptide that is natively expressed by the bacteria.

[0156] The bacteria may contain one or more OppA polypeptides. In preferred embodiments, the bacteria contains two or more OppA polypeptides, such as two, three, four, five, six or seven OppA polypeptides containing an amino acid sequence having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to an amino acid sequence as set forth in any one of SEQ ID NOs: 11-17.

[0157] In some embodiments, the adhesion molecule may be expressed from a nucleic acid introduced into the bacteria. This nucleic acid may be a gene expression cassette. The gene expression cassette may comprise a promoter, for example, the pgm promoter. The gene expression cassette may also comprise a synthetic ribosome binding site. The gene expression cassette may also comprise a Myc tag.

[0158] Immune cell

[0159] As described herein, the present invention can be used for the attraction of immune cells through the generation of a chemoattractant gradient. For bladder cancers in particular, most bladder tumours belong to the immune-excluded phenotype, which calls for a therapeutic strategy that can increase T cell infiltration and retention.

[0160] Accordingly, the present invention provides for bacteria to express a payload comprising a cytokine, wherein expression of the payload activates or attracts an immune cell by binding an immune cell receptor. In some embodiments, the immune cell is a leukocyte. In further embodiments, the leukocyte is a monocyte. In other embodiments, the leukocyte is a macrophage. In other embodiments, the leukocyte is a neutrophil. In other embodiments, the leukocyte is a lymphocyte. In further embodiments, the lymphocyte is a B cell. In other embodiments, the lymphocyte is a T cell. In further embodiments, the T cell is a CD8+ cell. In other embodiments, the T cell is CD4+ cell. In some embodiments, the T cell is a 8880361

[0161] 28 gamma delta (y6) T cell. In other embodiments, the T cell is a T cell receptor (TCR) engineered T cell. In other embodiments, the T cell is a chimeric antigen receptor (CAR) T cell. In other embodiments, the T cell is an NKT cell.

[0162] In other preferred embodiments, the lymphocyte is an NK cell. In some embodiments, the NK cell is a CAR NK cell.

[0163] In other embodiments, the immune cell is an antigen presenting cell (APC). In further embodiments, the APC is a dendritic cell (DC).

[0164] Compositions

[0165] In some aspects, the invention provides compositions comprising the bacteria of any aspect of the present invention. The composition may comprise one or more bacteria as described herein. In some embodiments, the composition comprises a first bacteria and a second bacteria. The first bacteria and second bacteria may express a payload as described herein.

[0166] In some embodiments, the first bacteria and the second bacteria express a payload comprising a cytokine as described herein. In some embodiments, the cytokine is any one of VEGF-A, MIF, BAFF, CXCL5, CXCL1 , HGF, CXCL11 , IFN-y, IL-1a, IL-3, IL-4, IL-7, IL-15, IL-18, IL-22, M-CSF, CCL2, CCL20, SCF, SDF-1a, or any combination thereof. In preferred embodiments, the cytokine is any one of CXCL9, CXCL10, CCL4, IL-15, CX3CL1 , IL-21 , CXCL11 , IFN-a, TRAIL, CD40L, GM-CSF, IFN-y, TNF-a, CCL3, CCL5, or any combination thereof. The cytokine may bind to a receptor on an immune cell.

[0167] In some embodiments, the cytokine is any cytokine that binds a cytokine receptor on an immune cell. In some embodiments, the receptor is CXCR3, CCR5, IL-15R, CX3CR1 , IL-21 R, IFNAR, DR4, DR5, DcR1 , DcR2, osteoprotegrin, CD40, GM-CSFR, IFNGR, TNFR1 , TNFR2, CCR1 , CCR3, CCR4, CCR5, or any combination thereof. In preferred embodiments, the cytokine binds to receptor CXCR3, CCR5, IL-15R, CX3CR1 , or any combination thereof. In some embodiments, the first bacteria and the second bacteria comprise the same payload. In further embodiments, the first bacteria and second bacteria express a payload comprising the same cytokine or cytokines. In other embodiments, the first bacteria and the second bacteria comprise different payloads. In further embodiments the first bacteria and the second bacteria express different payloads comprising different cytokines.

[0168] In preferred embodiments, the first bacteria express a payload comprising CXCL9 and the second bacteria express a payload comprising CXCL10. In other embodiments, the first bacteria express a payload comprising CXCL10 and the second bacteria express a payload comprising CXCL9. In preferred embodiments, the first bacteria express a payload comprising CX3CL1 and the second bacteria express a payload comprising IL-15. In other embodiments, the first bacteria express a payload comprising IL-15 and the second bacteria express a payload comprising CX3CL1 . 8880361

[0169] 29

[0170] Kits

[0171] In some aspects, the invention provides a kit comprising the bacteria of any aspect of the present invention. In some embodiments, the kit comprises a composition comprising one or more bacteria as described herein. In further embodiments, the kit may comprise a container comprising a composition comprising the bacteria as described herein.

[0172] In further embodiments, the kit comprises at least one container wherein the at least one container comprises at least one composition wherein the at least one composition comprises at least one bacteria as described herein. In some embodiments, the kit comprises a first container and a second container. In further embodiments, the first container comprises a first composition and the second container comprises a second composition. In some embodiments, the first composition comprises a first bacteria and the second composition comprises a second bacteria. In some embodiments, the first and second bacteria comprise the same payload. In further embodiments, the first and second bacteria express a payload comprising the same cytokine or cytokines. In some embodiments, the first and second bacteria comprise different payloads. In further embodiments, the first and second bacteria express different payloads comprising different cytokines. In other words, the payload expressed by the bacteria of the first composition comprises a different cytokine to the cytokine comprised within the payload that is expressed by the bacteria of the second composition.

[0173] In some embodiments, they cytokine is any one of VEGF-A, MIF, BAFF, CXCL5, CXCL1 , HGF, CXCL11 , IFN-y, IL-1a, IL-3, IL-4, IL-7, IL-15, IL-18, IL-22, M-CSF, CCL2, CCL20, SCF, SDF-1a, or any combination thereof. In preferred embodiments, the cytokine is any one of CXCL9, CXCL10, CCL4, IL-15, CX3CL1 , IL-21 , CXCL11 , IFN-a, TRAIL, CD40L, GM-CSF, IFN-y, TNF-a, CCL3, CCL5, or any combination thereof. The cytokine may bind to a receptor on an immune cell.

[0174] In some embodiments, the cytokine is any cytokine that binds a cytokine receptor on an immune cell. In some embodiments, the receptor is CXCR3, CCR5, IL-15R, CX3CR1 , IL-21 R, IFNAR, DR4, DR5, DcR1 , DcR2, osteoprotegrin, CD40, GM-CSFR, IFNGR, TNFR1 , TNFR2, CCR1 , CCR3, CCR4, CCR5, or any combination thereof. In preferred embodiments, the cytokine binds to receptor CXCR3, CCR5, IL-15R, CX3CR1 , or any combination thereof.

[0175] In preferred embodiments, the kit comprising the first container comprising the first composition comprising the first bacteria express a payload comprising the cytokine CXCL9 and the second container comprising the second composition comprising the second bacteria express a payload comprising the cytokine CXCL10. In other embodiments, the first bacteria express a payload comprising the cytokine CXCL10 and the second bacteria expresses a payload comprising the cytokine CXCL9.

[0176] In preferred embodiments, the kit comprising the first container comprising the first composition comprising the first bacteria express a payload comprising the cytokine CX3CL1 and the second container comprising the second composition comprising the second bacteria express a payload comprising the cytokine IL-15. In other embodiments, the first bacteria express a payload comprising the cytokine IL-15 and the second bacteria expresses a payload comprising the cytokine CX3CL1 . 8880361

[0177] 30

[0178] Methods of treatment

[0179] The bacteria of any aspect the present invention can be used as a medicament. In some embodiments, the bacteria of the present invention are for use in the treatment of cancer in a subject.

[0180] To understand how bladder tumours may differ among bladder carcinoma patients, the present inventors investigated the immune phenotype of bladder tumours. Most solid human tumours display one of the three distinct immune phenotypes: immune desert, immune excluded, or immune inflamed

[0015] , In a metastatic bladder carcinoma cohort, it was reported that most of the tumours exhibited the immune excluded phenotype (47%), with the remaining tumours being immune inflamed (27%) and immune desert (27%)

[0016] , Most bladder tumours belong to the immune-excluded phenotype, which calls for a therapeutic strategy that can increase T cell infiltration and retention. The present invention provides a method for the treatment of bladder cancer by increasing the infiltration of immune cells at tumour sites. This enables the killing of tumour cells.

[0181] The immune phenotype of bladder tumours from different patients may differ. The difference can come in the form of the type and degree of immune cell infiltration, secreted factor profile in the TME, and the tumour tissue architecture

[0013] , The present invention can deliver selective groups of cytokines expressed by the bacteria described herein to enhance the infiltration of specific populations of immune cells into bladder tumours. The combination of cytokines can be tailored to treat tumours according to their immune phenotype, which may be determined e.g. by liquid biopsy or tumour biopsy.

[0182] In contrast to intravesical instillation of BCG and immune checkpoint inhibitors, (e.g., Atezolizumab, Pembrolizumab, Durvalumab, Nivolumab, and Avelumab), the inventors consider that the present invention can be effective in treating immune excluded and / or immune desert tumours. Furthermore, engineered bacteria of the invention may comprise surface modifications to enhance tumour binding to deliver immunomodulatory proteins directly to the tumour site over an extended period, without the need for external maintenance.

[0183] Accordingly, the bacteria of the present invention may be administered mucosally, peri-tumourally or intra-tumourally. Administration may be via injection, spray, direct application (e.g., direct application to a mucosal surface or tumour tissue) or implant (e.g., an implanted device or scaffold). In preferred embodiments, the bacteria are administered by intravesical instillation, e.g.. using a catheter.

[0184] The bacteria may be administered mucosally. Mucosa as used herein can be any mucosa such as oral mucosa, buccal mucosa, nasal mucosa, pulmonary mucosa, rectal mucosa, urethral mucosa, bladder mucosa, vaginal mucosa and ocular mucosa. Mucosal administration as used herein encompasses delivery to one or more mucosa. Mucosal administration includes but is not limited to oral administration, intranasal administration and bladder instillation. In preferred embodiments, the bacteria is administered via intravesical instillation into the bladder. Oral mucosal administration includes buccal, sublingual and gingival routes of delivery. 8880361

[0185] 31

[0186] The bacteria may be administered peri-tumorally, i.e., directly contacting or in close proximity to the tumour. The bacteria may be administered intra-tumourally, i.e., directly into a tumour mass. A skilled person may determine an appropriate route of administration based on the site of the cancer, the properties of the Lactobacillus bacteria (such as the ability of the bacteria to home or adhere to a particular tissue or mucosa), the drug to be delivered, etc.

[0187] The cancer may be a carcinoma. Carcinomas are malignancies originating from epithelial cells, e.g., from epithelial cells lining the inner surfaces of the body, and are often associated with mucosal layers where native microbiomes are prevalent. Examples of carcinomas include but are not limited to malignant melanoma, non-small cell lung cancer (e.g., squamous non-small cell lung cancer and non-squamous non-small cell lung cancer), small cell lung cancer, head and neck cancer (e.g., oral cancer, nasal cancer, nasopharyngeal cancer, oropharyngeal cancer, hypopharyngeal cancer, laryngeal cancer, salivary gland cancer and tongue cancer), renal cell cancer (e.g., clear cell renal cell cancer), breast cancer, ovarian cancer (e.g., serous ovarian cancer and ovarian clear cell adenocarcinoma), uterine cancer (e.g., cervical cancer and endometrial cancer), anal cancer (e.g., anal canal cancer), colorectal cancer (e.g., MSI-H and / or dMMR positive colorectal cancer), hepatocellular cancer, esophageal cancer, gastric cancer, esophagogastric junction cancer, pancreatic cancer, urothelial cancer (e.g., bladder cancer, upper urinary tract cancer, ureteral cancer, renal pelvis cancer and urethral cancer), prostate cancer, fallopian tube cancer, primary peritoneal cancer, malignant pleural mesothelioma, gallbladder cancer, bile duct cancer, biliary tract cancer, skin cancer (e.g., uveal melanoma and Merkel cell carcinoma), testicular cancer (germ cell tumor), vaginal cancer, vulvar cancer, penile cancer, small intestine cancer, endocrine system cancer, thyroid cancer, parathyroid cancer, adrenal carcinoma, ocular retinoblastoma, neuroendocrine tumour, squamous cell carcinoma and the like.

[0188] In a preferred embodiment, cancer is a cancer of a mucosal tissue. Non-limiting examples of mucosal cancers include head and neck cancer (e.g., oral cancer, nasal cancer, nasopharyngeal cancer, oropharyngeal cancer, hypopharyngeal cancer, laryngeal cancer, salivary gland cancer and tongue cancer), lung cancer (e.g., small cell lung cancer, non-small cell lung cancer), esophageal cancer, esophagogastric junction cancer, gastric cancer, colorectal cancer, pancreatic duct cancer, uterine cancer (e.g., cervical cancer and endometrial cancer), ovarian cancer, bladder cancer, vaginal cancer, vulvar cancer, and anal cancer.

[0189] In some embodiments, the cancer is a nasopharyngeal, oral, lung, bladder, gastric, colorectal, skin, breast or ovarian cancer. In one embodiment, the cancer is a nasopharyngeal cancer, e.g., nasopharyngeal carcinoma.

[0190] In a preferred embodiment, the cancer is bladder cancer. In some embodiments, the bladder cancer is NMIBC. In other preferred embodiments, the bladder cancer is MIBC.

[0191] In some embodiments, the cancer exhibits an immune excluded phenotype. In some embodiments, the cancer exhibits an immune desert phenotype. In some embodiments, the cancer exhibits an immune inflamed phenotype. 8880361

[0192] 32

[0193] The treatment described herein can serve as an alternative therapy to BCG or a complementary therapy to BCG. Accordingly, in some embodiments, the treatment is administered in place of BCG therapy. In some embodiments, the treatment is administered alongside BCG. In some embodiments, the treatment is administered as a follow up therapy suitable for NMIBC patients who have experienced BCG failure.

[0194] The invention described herein is clinically relevant in at least two aspects.

[0195] Firstly, this therapy is suitable as a follow-up therapy for NMIBC patients who have experienced BCG failure

[0017] , The different categories of BCG failure include (1) BCG relapsing tumours - recurrence of high-grade tumour after BCG maintenance treatment, (2) progression from non-muscle-invasive to muscle-invasive tumour following BCG therapy, (3) BCG intolerance - severe side effects that prevent further BCG instillation, (4) BCG refractory tumours - the appearance of high-grade tumours during BCG maintenance therapy, and (5) BCG-unresponsive tumours - comprise of BCG refractory and early BCG relapsing tumours. The present invention provides patients with recurred or progressed tumours after BCG failure an additional treatment option besides the pre-existing follow-up options, such as intravesical chemotherapy, radiation therapy, systemic chemotherapy, systemic immunotherapy, and cystectomy.

[0196] Secondly, the engineered strains such as Lactobacillus plantarum WCFS1 can be incorporated into current clinical practices by being instilled into a NMIBC patient’s bladder following TURBT. They can be administered in place of BCG (alternative therapy) or administered together with BCG (complementary therapy). The full BCG treatment schedule can be followed when administering the engineered strains. This involves an initial induction therapy that occurs weekly for six weeks, followed by maintenance therapy, which would be over the course of one year (15 instillations) or three years (27 instillations) [11 , 18],

[0197] Therefore, this therapy can serve as a second-line treatment for patients with high-grade NMIBC who do not respond to BCG therapy, or forthose with muscle-invasive bladder cancer (MIBC) as an option before considering cystectomy — a procedure that can significantly impact quality of life. Additionally, this therapy offers an alternative for patients who might choose it over BCG therapy following TURBT.

[0198] Pharmaceutical compositions

[0199] Pharmaceutical compositions may be prepared using a pharmaceutically acceptable “carrier” composed of materials that are considered safe and effective. The term “carrier” refers to diluents, binders, lubricants and disintegrants. Those with skill in the art are familiar with such pharmaceutical carriers and methods of compounding pharmaceutical compositions using such carriers. "Pharmaceutically acceptable" refers to molecular entities and compositions that are "generally regarded as safe", e.g., that are physiologically tolerable and do not typically produce an allergic or similar untoward reaction, such as gastric upset and the like, when administered to a human. In some embodiments, this term refers to molecular entities and compositions approved by a regulatory agency of the US federal or a state government, as the GRAS list under section 204(s) and 409 of the Federal Food, Drug and Cosmetic Act, that is subject to premarket 8880361

[0200] 33 review and approval by the FDA or similar lists, the U.S. Pharmacopeia or another generally recognised pharmacopeia for use in animals, and more particularly in humans.

[0201] The pharmaceutical compositions provided herein may include one or more excipients, e.g., solvents, solubility enhancers, suspending agents, buffering agents, isotonicity agents, antioxidants or antimicrobial preservatives. When used, the excipients of the compositions will not adversely affect the stability, bioavailability, safety, and / or efficacy of the active ingredients. Thus, the skilled person will appreciate that compositions are provided wherein there is no incompatibility between any of the components of the dosage form. Excipients may be selected from the group consisting of buffering agents, solubilizing agents, tonicity agents, chelating agents, antioxidants, antimicrobial agents, and preservatives.

[0202] Pharmaceutical compositions may comprise, in additional to one of the above substances, a pharmaceutically acceptable excipient, carrier, buffer, stabiliser or other material well known to those skilled in the art. Such substances should be non-toxic and should not interfere with the efficacy of the active ingredient.

[0203] Definitions

[0204] The terms “treatment”, “treat”, or “treating” are used herein to refer to the reduction in severity of a disease or condition, the reduction in the duration of a disease; the amelioration or elimination of one or more symptoms associated with a disease or condition, or the provision of beneficial effect to a subject with a disease or condition. The term also encompasses prophylaxis of a disease or condition or its symptoms thereof. “Prophylaxis” is known in the art to mean decreasing or reducing the occurrence or severity of a particular disease outcome. For example, delaying progression of cancer in a subject.

[0205] As used herein, the term “subject” refers to a human or any non-human animal (e g, mouse, rat, rabbit, dog, cat, cattle, swine, sheep, horse or primate). In many embodiments, a subject is a human being. A subject can be a patient, which refers to a human presenting to a medical provider for diagnosis or treatment of a disease. The term “subject” is used herein interchangeably with “individual” or “patient.” In some embodiments, the subject is human. A subject can be afflicted with or is susceptible to a disease or disorder but may or may not display symptoms of the disease or disorder.

[0206] As used herein, “endogenous” refers to a molecule that is natively encoded and / or present within a host organism. For example, an OppA Lp_0783 nucleic acid or polypeptide synthesised from the genome of a Lactobacillus plantarum WCFS1 bacteria is endogenous to said host bacteria.

[0207] As used herein, “exogenous” refers to a molecule that is not natively encoded and / or present within a host organism. For example, an OppA LVIS_1959 nucleic acid or polypeptide is not natively encoded and / or present in host bacteria Lactobacillus plantarum WCFS1 . As a further example, an OppA Lp_0783 nucleic acid or polypeptide that is synthesised outside a Lactobacillus plantarum WCFS1 bacteria but introduced into the cytoplasm or onto the surface of said host bacteria is exogenous to said host bacteria. Any polypeptide produced by the host organism from a nucleic acid introduced into the host organism would be considered exogenous to said host organism. 8880361

[0208] 34

[0209] As used herein, an “immunomodulator” refers to any substance capable of stimulating or suppressing the immune response. This includes, for example, a cytokine. In the context of the present invention, the immunomodulator is preferably immunostimulatory (e.g. an immunostimulatory cytokine).

[0210] As used herein, a “commensal” organism refers to any organism that resides on the surface of a host organism or at mucosa without harming the health of the host organism. This includes, for example, Lactobacillus plantarum which is commensal to the human bladder tumour microenvironment.

[0211] As used herein, a “probiotic” refers to a live organism that provides health benefits when administered to a host organism. This includes, for example, engineered Lactobacillus plantarum as described herein for the treatment of bladder cancer in humans.

[0212] As used herein, a “payload” refers to a molecule or substance that is used to deliver a therapeutic effect. This includes, for example, a DNA vector encoding a cytokine.

[0213] As used herein, an “adhesion molecule” refers to any molecule capable of adhering to a surface or molecule. This includes, for example, an OppA polypeptide.

[0214] As used herein, a “membrane anchor” refers to a region of the adhesion molecule that facilitates the anchoring of polypeptides to the membrane of a cell or is itself anchored to the membrane of a cell. This includes, for example, the signal peptide Lp_0783SP which forms the membrane anchor prior to protease cleavage in Lactobacillus plantarum.

[0215] As used herein, a “surface adhesin” refers to the region of the adhesion molecule that is capable of adhering to a surface or molecule. This includes, for example, an OppA polypeptide.

[0216] As used herein, a “homologous” sequence or gene or polypeptide refers to any sequence or gene or polypeptide that is natively encoded and / or present in a host organism.

[0217] As used herein, a “heterologous” sequence or gene or polypeptide refers to any sequence or gene or polypeptide that is natively encoded and / or present in an organism that is not the host organism. A heterologous sequence or gene or polypeptide can include a synthetically engineered sequence or gene or polypeptide.

[0218] The features disclosed in the foregoing description, or in the following claims, or in the accompanying drawings, expressed in their specific forms or in terms of a means for performing the disclosed function, or a method or process for obtaining the disclosed results, as appropriate, may, separately, or in any combination of such features, be utilised for realising the invention in diverse forms thereof.

[0219] While the invention has been described in conjunction with the exemplary embodiments described above, many equivalent modifications and variations will be apparent to those skilled in the art when given this disclosure. Accordingly, the exemplary embodiments of the invention set forth above are considered to be illustrative and not limiting. Various changes to the described embodiments may be made without departing from the spirit and scope of the invention. 8880361

[0220] 35

[0221] For the avoidance of any doubt, any theoretical explanations provided herein are provided for the purposes of improving the understanding of a reader. The inventors do not wish to be bound by any of these theoretical explanations.

[0222] Any section headings used herein are for organizational purposes only and are not to be construed as limiting the subject matter described.

[0223] Throughout this specification, including the claims which follow, unless the context requires otherwise, the word “comprise” and “include”, and variations such as “comprises”, “comprising”, and “including” will be understood to imply the inclusion of a stated integer or step or group of integers or steps but not the exclusion of any other integer or step or group of integers or steps.

[0224] It must be noted that, as used in the specification and the appended claims, the singular forms “a,” “an,” and “the” include plural referents unless the context clearly dictates otherwise. Ranges may be expressed herein as from “about” one particular value, and / or to “about” another particular value. When such a range is expressed, another embodiment includes from the one particular value and / or to the other particular value. Similarly, when values are expressed as approximations, by the use of the antecedent “about,” it will be understood that the particular value forms another embodiment. The term “about” in relation to a numerical value is optional and means for example + / - 10%.

[0225] Numbered Statements of Invention

[0226] 1 . A bacterium that is engineered to express a payload comprising a cytokine, wherein the payload activates and / or attracts an immune cell by binding to an immune cell receptor.

[0227] 2. The bacterium according to statement 1 , wherein the cytokine is secreted from the bacterium.

[0228] 3. The bacterium according to statement 1 or 2, wherein the immune cell is:

[0229] (a) a lymphocyte, optionally wherein said lymphocyte is a T cell or a natural killer cell (NK), further optionally wherein the T cell is a CD4+ cell, a CD8+ cell, or

[0230] (b) a natural killer T cell (NKT), or an antigen presenting cell (APC), optionally wherein said APC is a dendritic cell (DC).

[0231] 4. The bacterium according to any one of statements 1-3, wherein the bacterium is a commensal bacterium.

[0232] 5. The bacterium according to any one of statements 1-4, wherein the bacterium is a gram-positive bacterium, optionally wherein the gram-positive bacterium is of the genus Lactobacillus.

[0233] 6. The bacterium according to statement 5 wherein the bacterium of the genus Lactobacillus is of the species:

[0234] (a) Lactobacillus plantarum, optionally of the strain WCFS1 , or 8880361

[0235] 36

[0236] (b) Lactobacillus salivarius, optionally of the strain DSM20555, or

[0237] (c) Lactobacillus reuteri, optionally of the strain DSM20016.

[0238] 7. The bacterium according to any one of statements 1-6, wherein the cytokine is any one of CXCL9, CXCL10, CCL4, IL-15, CX3CL1 , IL-21 , CXCL11 , IFN-a, TRAIL, CD40L, GM-CSF, IFN-y, TNF-a, CCL3, CCL5, or any combination thereof.

[0239] 8. The bacterium according to any one of statements 1-7, wherein the cytokine is a chemokine.

[0240] 9. The bacterium according to any one of statements 1-8, wherein the bacterium is engineered to deliver two or more cytokines, optionally wherein the two or more cytokines are:

[0241] (a) CXCL9 and CXCL10, or

[0242] (b) CX3CL1 and IL-15.

[0243] 10. The bacterium according to any one of statements 1-9, further comprising an adhesion molecule that is:

[0244] (a) overexpressed from the genome of the bacterium as compared to a non-engineered bacterium, and / or

[0245] (b) expressed from a nucleic acid introduced into the bacterium, and / or

[0246] (c) exogenously coated on a surface of the bacterium, optionally wherein said adhesion molecule is an oligopeptide binding protein (OppA).

[0247] 11 . The bacterium according to statement 10, wherein the adhesion molecule comprises a membrane anchor with an amino acid sequence having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to any one of SEQ ID NOs: 1-10, optionally wherein the membrane anchor is cleavable.

[0248] 12. The bacterium according to statement 10, wherein the adhesion molecule comprises a surface adhesin with an amino acid sequence having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to any one of SEQ ID NOs: 11-17.

[0249] 13. The bacterium of any one of statement 10-12, wherein the adhesion molecule comprises a membrane anchor and a surface adhesin that are heterologous to each other.

[0250] 14. The bacterium according to any one of statement 10-13, wherein the membrane anchor comprises a secretion signal peptide and a cleavage site, optionally wherein the cleavage site is cleavable by an enzyme, further optionally wherein the enzyme is a signal peptidase II (SPase II).

[0251] 15. A composition comprising one or more bacteria according to any of statements 1-14. 8880361

[0252] 37

[0253] 16. The composition according to statement 15, comprising a first bacterium and a second bacterium according to any of claims 1-14, wherein the first bacterium expresses a payload comprising CXCL9, and the second bacterium expresses a payload comprising CXCL10.

[0254] 17. The composition according to statement 15, comprising a first bacterium and a second bacterium according to any of claims 1-14, wherein the first bacterium expresses a payload comprising CX3CL1 , and the second bacterium expresses a payload comprising IL-15.

[0255] 18. The bacterium according to any one of statements 1-14 or the composition according to any one of claims 15-17, for use as a medicament.

[0256] 19. The bacterium according to any one of statements 1-14 or the composition according to any one of claims 15-17, for use in the treatment of cancer in a subject, optionally wherein said cancer is bladder cancer, further optionally wherein said bladder cancer is non-muscle invasive bladder cancer (NMIBC) or muscle invasive bladder cancer (MIBC).

[0257] 20. The bacterium for the use according to statement 19, wherein the cancer exhibits an immune excluded and / or immune desert phenotype.

[0258] 21. The bacterium for the use according to statement 19 or 20, wherein the treatment is administered in place of Bacillus Calmette-Guerin (BCG) therapy or alongside BCG therapy.

[0259] 22. The bacterium for the use according to statement 19 or 20, wherein the treatment is administered as a follow up therapy suitable for NMIBC patients who have experienced BCG failure.

[0260] 23. A kit comprising a first container containing a first composition according to statement 15 and a second container containing a second composition according to statement 15, wherein the payload expressed by the bacterium of the first composition comprises a different cytokine to the cytokine comprised within the payload that is expressed by the bacterium of the second composition.

[0261] Examples

[0262] EXAMPLE 1 - Schematic overview of engineered bacteria for the sustained delivery of immunomodulatory proteins in bladder tumours

[0263] The inventors generated four genetically engineered commensal Lactobacillus plantarum WCFS1 strains, with each strain modified to deliver a single immunomodulatory protein (Figure 1). Each strain is modified to secrete an immunomodulatory protein and anchor a surface adhesion protein that can enhance its binding to bladder carcinoma cell GAGs. Two different therapeutic combinations were assembled to deliver different sets of immunomodulators, which enhance the infiltration of specific immune cells in the bladder tumour to kill bladder carcinoma cells.

[0264] The first combination comprises two engineered strains, with one strain delivering CXCL9 and another strain delivering CXCL10. The combined administration of CXCL9 and CXCL10 is intended to enhance 8880361

[0265] 38 the infiltration of T cells into bladder tumours to kill carcinoma cells. The second combination consists of two strains, with one strain delivering CX3CL1 and the other strain delivering IL-15. The combined administration of CX3CL1 and IL-15 is aimed at attracting and activating NK cells and cytotoxic T lymphocytes (CTLs) into bladder tumours to kill carcinoma cells.

[0266] To ensure that the engineered L. plantarum WCFS1 strains can perform targeted and sustained delivery of immunomodulatory protein at the bladder tumour site, each engineered strain may be designed to contain two gene expression cassettes. The first gene expression cassette was used to continuously secrete a single immunomodulatory protein. It contains the DNA sequence of an immunomodulatory protein fused to a selected secretion signal peptide. In vitro validation experiments have shown that the secreted immunomodulatory proteins from the engineered L. plantarum WCFS1 strains were bioactive, and the instillation of the engineered strains was able to initiate tumour regression in the MB49 orthotopic syngeneic murine model for bladder carcinoma

[0267] The second gene expression cassette was used to control the expression of a selected bacterial surface adhesin that has a high affinity to glycosaminoglycans (GAGs) expressed on the surface of bladder carcinoma cells residing in bladder tumours. This cassette contains the DNA sequence of a fusion protein that consists of a bacterial membrane anchor and the surface adhesin. The presentation of the fusion protein on the surface of L. plantarum WCFS1 has been shown to boost the binding efficiency of the engineered strains to bladder carcinoma cells.

[0268] EXAMPLE 2 - Engineering bacteria with enhanced binding capabilities to bladder carcinoma glycosaminoglycans

[0269] Identifying commensal bacteria with enhanced bladder carcinoma binding

[0270] An in vitro bacteria binding assay was performed to determine which Lactobacillus strains had the highest efficiency to the human bladder carcinoma cell lines T24, J82, UM-UC-3, SW780, and the mouse bladder carcinoma cell line, MB49. The four strains that were used for this screening were Lactobacillus casei DSM20011 , L. plantarum WCFS1 , Lactobacillus reuteri DSM20016 and Lactobacillus salivarius DSM20555. Among the four Lactobacillus strains, the L. plantarum WCFS1 strain had the highest CFU count of bacteria that remained bound to each individual carcinoma cell compared to the three other Lactobacillus strains (Figure 2). This trend was observed across the different bladder carcinoma cell lines. The drastic difference in binding efficiency indicated that the L. plantarum WCFS1 strain was the preferred candidate for the bacterial delivery system.

[0271] Identifying adhesion molecules with enhanced binding to bladder carcinoma cells

[0272] To find a suitable binding protein that could be expressed and presented on the surface of L. plantarum WCFS1 to further enhance its binding efficiency to bladder carcinoma cells, a literature survey was conducted to identify adhesive proteins. Only adhesive proteins that originated from gram-positive 8880361

[0273] 39 bacteria were considered, as they had to be compatible with protein secretion and anchoring machinery in L. plantarum WCFS1 strain

[0027] (Figure 3).

[0274] A study in 2013 reported that Lactobacillus salivarius Lv72 could adhere to HeLa cell surface proteoglycans and revealed that the bacterial adhesin responsible for the binding interaction was the oligopeptide-binding protein (OppA)

[0028] , an extracellularly exposed lipoprotein that is part of the ABC transporter system involved in oligopeptide internalisation [29, 30], Further analysis found that this OppA protein interacted with glycosaminoglycans (GAGs) found on the surface of HeLa cells. These GAGs consisted of heparan sulphate, heparin (over-sulphated form of heparan sulphate), and chondroitin sulphate A, B and C. This OppA protein and its genetic variants were deemed to be suitable candidates as it has been reported that GAGs in bladder tumours have varying levels of content of hyaluronic acid, heparan sulphate, dermatan sulphate and chondroitin sulphate at different cancer stages

[0031] ,

[0275] In the 2013 study, the identity of the L. salivarius Lv72 OppA was found to be KC703973. Therefore, in silico screening strategies were used to identify 7 OppA genetic variants from the KC703973 OppA. The 7 OppA genetic variants were cloned into the pET-28b(+) expression plasmid for protein production in Escherichia coli BL21 . A Myc-tag was added to each protein C-terminus for antibody detection. Fast performance liquid chromatography (FPLC) was used to isolate the seven selected OppA candidates with the HiTrap® Heparin HP column.

[0276] A binding assay that employed flow cytometry as a measurement tool was used to determine the relative binding strengths of the isolated OppA genetic variants to the MB49, T24, J82, and UM-UC-3 bladder carcinoma cell lines. In this experiment, all 7 OppA genetic variants had a significantly higher geometric mean fluorescence intensity (MFI) compared to the PBS control (Figure 4A). The Lp_0783, LBA1665, and LVIS_1959 OppA genetic variants had a noticeably higher MFI than the other four OppA protein genetic variants, indicating a stronger binding affinity to the four bladder carcinoma cell lines. This observation was more clearly seen in the cell count versus fluorescence intensity histograms, where a greater rightward shift was present for bladder carcinoma cell lines bound to Lp_0783, LBA1665 and LVIS_1959 (Figure 4B-E). Further testing of the suitability of Lp_0783, LBA1665 and LVIS_1959 was conducted in the following sections.

[0277] Designing and expressing OppA fusion constructs

[0278] The next step involved designing fusion proteins for the anchoring of OppA genetic variants on the surface ofL. plantarum WCFS1 . As the OppA genetic variants here are known lipoproteins [29, 30], a native L. plantarum WCFS1 lipoprotein anchor was used to display the heterologous OppA genetic variants on the bacteria surface. The lipoprotein anchor from the OppA genetic variants consist of a N- terminal signal peptide for secretion with a signal peptidase II (SPase II) cleavage site at the C-terminal end of the signal peptide (Figure 5). The cleavage boundary of this lipoprotein anchor consists of an alanine in position -1 and a cysteine at position +1 (AJ.C), with the rest of the surface presented protein coming after the cleavage site

[0032] , During the protein secretion process, the SPase II enzyme cleaves at its specific target site and the diacylglycerol transferase enzyme catalyses a coupling reaction between the cysteine at position +1 and a phospholipid of the membrane

[0033] , The utilisation of this lipoprotein 8880361

[0279] 40 anchoring strategy will result in the N-terminal anchoring of the chosen heterologous protein, with the C- terminal of the protein facing outwards.

[0280] Among the three short-listed OppA lipoproteins, Lp_0783 was native to the L. plantarum WCFS1 . Therefore, the Lp_0783 secretion signal peptide was used as the anchor. The LBA1665 and LVIS_1959 OppA variants were fused to the Lp_0783 secretion signal peptide. The SPase II cleavage boundary between the Lp_0783 signal peptide and the rest of the OppA protein sequence was retained to ensure that the protein could be cleaved by the SPase II enzyme and anchored to the cell membrane. The Lp_0783 OppA genetic variant did not require further design.

[0281] To introduce the OppA fusion DNA constructs into the L. plantarum WCFS1 strain, the PTRK892 shuttle vector was used. This vector was chosen because it contained the broad host range pWV01 replicon that could be used in most Lactobacillus strain

[0034] (Figure 5). The gene expression cassette found in this vector came with the phosphoglycerate mutase (pgm) promoter from Lactobacillus acidophilus NCFM, the p-glucuronidase (GusA3) reporter gene from Lactobacillus gasseri ADH and the Term 908 terminator [34-36], The fusion proteins for anchoring the OppA genetic variants were cloned between the pgm promoter and Term 908 terminator, replacing the GusA3 reporter gene (Figure 6B). Using the RBS calculator by Salis Lab, a unique RBS was designed for each gene expression cassette, replacing the original RBS in the original gene expression cassette. The translation initiation rate used was 100,000.

[0282] To detect the expression of the OppA fusion DNA constructs, western blot assay was used. The assay showed enhanced expression of all constructs, with particularly increased expression of N-terminal expression of the N-terminal anchored LVIS_1959 (Figure 6A). To confirm the surface localization of the N-terminal anchored OppA genetic variants on the surface of the L. plantarum WCFS1 strain, the engineered cells were stained with the Myc-Tag fluorescence tagged antibody and ran through a flow cytometer (Figure 6B). The flow cytometry results readings showed a net increase in fluorescence signal for all constructs compared to the Lp-empty control, with particularly increased expression of the Lp- LVIS_1959 construct.

[0283] LVIS_1959 was selected as the preferred fusion construct. Having confirmed the expression and presentation of LVIS_1959 on the surface of L. plantarum WCFS1 , the next step involved testing if the L. plantarum WCFS1 surface modification led to enhanced binding efficiency of the L. plantarum WCFS1 strain to bladder carcinoma cell lines. Using the same binding assay described for Figure 2, it was revealed that the L. plantarum WCFS1 strain presenting the N-terminal anchored LVIS_1959 resulted in a significantly higher CFU count per cell than the empty vector control, for the MB49, J82, T24 and UM-UC- 3 bladder carcinoma cell lines (Figure 6C-F).

[0284] Taking these results into consideration, L. plantarum WCFS1 strain with N-terminal anchored LVIS_1959 was further tested for its enhanced retention in the MB49 orthotopic syngeneic murine model for bladder carcinoma. The tumour model was established by implanting the MB49 carcinoma cell line in the bladder of each mouse. After allowing the tumour to grow within the bladder for a week, the Lp-LVIS_1959-N- terminal anchored strain and Lp-empty control strain, were instilled into the bladder (Figure 7A). Two different groups of mice were euthanized after four days and seven days to track how much bacteria were 8880361

[0285] 41 retained over time. The harvested bladders were sliced into small pieces and pounded using a pestle to recover the bound bacteria. The resulting tissues were reconstituted in PBS and plated on agar plates for CFU counting.

[0286] It was found that the CFU / g of bacteria recovered from MB49 implanted bladders treated with the Lp- LVIS_1959-N-terminal anchored strain was not significantly higher than the Lp-empty control strain after four days (Figure 7B). However, the significant difference in CFU / g of recovered bacteria was observed after seven days.

[0287] EXAMPLE 3 - Engineering bacteria to deliver immunomodulators to enhance the regression of bladder carcinoma by recruiting immune cells

[0288] Identifying cytokines of interest

[0289] After establishing the suitable anchor and adhesive protein for boosting the binding efficiency of the engineered L. plantarum WCFS1 strain to bladder carcinoma cells, the next step was to identify and search for suitable immunomodulators that could be delivered to the bladder TME to enhance the infiltration of immune cells and elicit anti-tumour responses.

[0290] In silico gene expression profiling was performed on a public human bladder carcinoma dataset to look for chemokines and cytokines with diminished expression levels in the bladder tumours compared to healthy bladder tissues. Healthy bladder tissues were used as the reference here with the understanding that an active immunosurveillance system is present in the normal bladder wall. This has been verified in several studies where biopsies from small cohorts of healthy control patients with non-bladder pathologies have shown that the normal immune landscape of the bladder consists of CD8+T cells, and DCs in the urothelium, and CD8+T cells, CD4+T cells, macrophages and DCs in the bladder lamina propria

[0037] , The therapeutic strategy involved replenishing chemokines that recruited immune cells lacking in bladder carcinoma tissues

[0203] ,

[0291] From this list of chemokines and cytokines, two different therapeutic strategies were developed. The first therapeutic strategy involved the combined administration of CXCL9 and CXCL10 (Figure 8A), and the second strategy consisted of the combined administration of soluble CX3CL1 and IL-15 (Figure 8B). The L. plantarum WCFS1 strain was engineered to individually secrete each therapeutic protein. Two different therapeutic assets were assembled from the four genetically engineered L. plantarum WCFS1 strains. The first asset consisted of one strain delivering CXCL9 and another strain delivering CXCL10. The second asset consisted of one strain delivering CX3CL1 and the other strain delivering IL-15. Both therapeutic assets were to be administered via intravesical instillation.

[0292] Expressing cytokines in Lactobacillus

[0293] Having established a working gene expression cassette for L. plantarum WCFS1 to express and anchor OppA genetic variants, the same components were used to assemble a gene expression cassette to 8880361

[0294] 42 secrete human and mouse CXCL9, CXCL10, CX3CL1 , and mouse IL-15. The gene cassette was cloned into the PTRK892 plasmid backbone (Figure 9A).

[0295] After successfully cloning the gene constructs, the engineered L. plantarum WCFS1 cultures were grown overnight and the secreted proteins in the MRS media were extracted using trichloroacetic acid (TCA) precipitation. Western blot was subsequently used to detect protein expression and secretion of chemokines from engineered L. plantarum WCFS1 strains (Figure 9B-D).

[0296] In order to confirm the Lactobacillus genus was capable of expressing and secreting immunomodulators, a gene expression cassette comprising Lp_0373-hCXCL9 under the pgm promoter was cloned into L. salivarius strain DSM20555 and L. reuteri strain DSM20016 using the same methodology as described above. Western blot confirmed the protein expression and section of this chemokines from L. salivarius (Figure 10A) and L. reuteri (Figure 10B), confirming this technology can be applied to any Lactobacillus species.

[0297] Cytokine induced immune cell migration

[0298] A commercially available 3D microfluidic device was used to test if the L. plantarum WCFS1 secreted chemokines were able to attract immune cells. This device contained three chambers. The middle chamber was used to culture carcinoma cells within a solidified collagen gel. One of the side channels was used to introduce chemokines to the device. The chemokines could diffuse from one end of the collagen gel to the other end, forming a concentration gradient of chemokines across the gel. Immune cells were introduced at the opposite channel, which was separated from the chemokine channel by the middle collagen gel. This setup allowed the immune cells to be exposed to the soluble factors secreted by carcinoma cells and the chemokine gradient within the collagen gel. The number of migrated immune cells in the collagen gel was taken as indicators to judge the potency of the chemokines (Figure 11). The cells were stained with fluorescence dyes to identify the different cells in the collagen gel. The J82 carcinoma cells were stained with the CMRA dye, the immune cells (T cells or NK-92 cells) were stained with the BMQC dye and the cell-impermeable nuclear dye DRAQ7 was introduced to detect doublestranded DNA of the permeabilized dead cells. In the migration assay experiments testing hCCL4 and the combination of hCXCL9 and hCXCLI 0, IL-2 (1 OOlU / ml) was supplemented to the AIM-V + 2% AB serum culture media used throughout the 3D microfluidic device to ensure T cell survival.

[0299] In the first migration assay experiment, activated T cells were used. Secreted chemokines were generated by seeding bacteria (Lp-hCXCL9, Lp-hCXCL10 and Lp-hCX3CL1) in AIM-V + 2% human AB serum cultures at the ODeoo = 0.6. Lp-empty was also cultured at ODeoo = 0.6 to serve as a control.

[0300] An equal volume of medium containing secreted hCXCL9 and hCXCLI 0 was combined to generate a medium mix and loaded on the 3D microfluidic device (Figure 12A). This was because it was found that the combined use of CXCL9 and CXCL10 was found to stimulate significantly greater T cell migration compared to the individual administration of each chemokine (Figure 12B). The medium containing secreted hCX3CL1 was tested individually. The J82 bladder carcinoma cell line was cultured in collagen 8880361

[0301] 43 gel within the middle chamber for all subsequent experiments. The device configuration and the experimental timeline can be referred to in Figure 12C-D.

[0302] It was found that the combined administration of secreted chemokines of hCXCL9 and hCXCLI O cultures significantly increased activated T cell migration compared to the blank and empty vector controls (Figure 13A) at the 48h and 72h time-points. A similar observation was made for the secreted hCX3CL1 culture (Figure 13B) and hCCL4 (Figure 13C). These results collectively showed that the L. plantarum WCFS1 strain could produce and secrete bioactive and functional chemokines.

[0303] Flow cytometry was used to select for CD8+ and CD4+ T cells. Migration assays were performed on these subpopulations confirming secreted chemokines hCXCL9 and hCXCLIO, and hCCL4 caused significant migration of CD8+ T cells (Figure 14A) and CD4+ T cells (Figure 14B) compared to empty vector controls.

[0304] The same experimental setup as Figure 13 was used to evaluate the chemoattractant capacity of secreted soluble human CX3CL1 to enhance NK cell chemotaxis and measure the corresponding cytotoxicity by migrated NK cells toward J82 carcinoma cells. Some minor modifications were introduced to the migration assay, namely the culture medium used (AIM-V to R10 media) and the effector cell (activated T cells to NK-92 cells). The R10 medium containing secreted hCX3CL1 was introduced to the device and the J82 bladder carcinoma cell line was cultured in collagen gel within the middle chamber (Figure 15A). 140IU / ml of commercially available IL-15 was added to the R10 medium that was used throughout the whole device, making it a constant variable. The device configurations and experimental timeline can be referred to in (Figure 15B-C).

[0305] It was found that the secreted soluble hCX3CL1 was able to significantly increase NK-92 cell migration when compared to the blank and empty vector controls after 48h of incubation (Figure 16A). For this experiment, there was also a small statistically detected increase in NK-92 migration cell counts for secreted soluble hCX3CL1 between the 24h and 48h incubation time.

[0306] Cytokine induced immune cell migration causes carcinoma cell death

[0307] Further analysis was performed on the regions in the middle collagen region of the device to detect and quantify the NK-92 killing of J82 carcinoma cells. When the J82 carcinoma cells were killed, the CMRA dye that was used to stain the J82 cells faded and the cell-impermeable nuclear dye DRAQ7 in the medium rapidly stained the double stranded DNA of the permeabilized dead cells. This change in the surface area of the different dyes in the resulting confocal microscopy images allowed to track J82 carcinoma cell death based on the surface of CMRA (J82 carcinoma cells) and DRAQ7 (cell death). To measure the surface area of the different dyes, the Imaris Surfaces model was used to mask the CMRA channel and DRAQ7 channel separately. The masked surface area for each channel was then summed for all the images used for analysis. As six images were used for counting migrated NK-92 cells, the same images were used for summing the masked surface areas. The ratio of the surface area of the DRAQ7 channel versus the surface area of the CMRA channel for each device was then calculated and used as an indirect quantification of J82 cell death. 8880361

[0308] 44

[0309] This analysis revealed that the devices with secreted soluble hCX3CL1 had a significantly higher ratio of surface area of DRAQ7 to surface area of CMRA compared to the empty vector control and blank control, signifying more significant J82 cell death (Figure 16B). It was also found that there was no significant difference between the empty vector control and blank control devices. These results indicated a link between the enhancement of NK-92 migration by CX3CL1 and the ratio of surface area of DRAQ7 to surface area of CMRA, the parameter used to measure J82 cell death in the in vitro experimental setup of Figure 15A. These results suggest that Lactobacillus secreted CX3CL1 not only induces immune cell migration but immune cell induced death of bladder carcinoma cells.

[0310] EXAMPLE 4 - Testing the therapeutic efficacy of the engineered bacteria in the MB49 orthotopic bladder tumour mouse model

[0311] Lactobacillus expressing engineered OppA and cytokine payloads regress bladder tumours

[0312] After confirming the bioactivity of L. plantarum WCFS1 secreted chemokines, the next step involved testing the therapeutic efficacy of the engineered bacteria in the MB49 orthotopic bladder tumour mouse model.

[0313] A final gene assembly was performed, transforming Lactoabillcus with an engineered construct comprising a fusion OppA and an immunomodulator for secretion (Figure 17A). The following four different engineered bacterial strains were generated: Lp-mCXCL9-LVIS_1959, Lp-mCXCL10- LVIS_1959, Lp-mCX3CL1-LVIS_1959 and Lp-mlL-15-LVIS_1959. This was done by cloning the gene expression cassette for secreting each specific chemokine / cytokine and the gene expression cassette for the N-terminal anchored LVIS_1959 binding protein into the PTRK892 plasmid backbone.

[0314] The tumour model was established by implanting the MB49 bladder carcinoma cell line in the bladder of each mouse. After allowing the tumour to grow within the bladder for a week, seven treatment groups were instilled into the bladder weekly, over the course of two weeks (Figure 18A). The seven treatment groups consisted of the (1) PBS negative control group, (2) Lp-LVIS_1959 negative control group, (3) combination of Lp-mCXCL9-LVIS_1959 and Lp-mCXCL10-LVIS_1959, (4) Lp-mCX3CL1-LVIS_1959, (5) Lp-mlL-15-LVIS_1959, (6) combination of Lp-mCX3CL1-LVIS_1959 and Lp-mlL-15-LVIS_1959, and (7) BCG positive control. On day 27, the mice were euthanized, and the bladders were harvested for further analysis. Healthy mouse bladders were also harvested to serve as a reference indicator for the ideal bladder mass and volume.

[0315] It was found that the bladder masses of mice instilled with the BCG positive control, Lp-mCX3CL1- LVIS_1959 monotherapy and the combination therapy of Lp-mCX3CL1-LVIS_1959 and Lp-mlL-15- LVIS_1959 were significantly smaller than the Lp-LVIS_1959 control group (Figure 18B). Additionally, the bladder masses from mice treated with the combination therapy of Lp-mCX3CL1-LVIS_1959 and Lp-mlL- 15-LVIS_1959 were found to be significantly lower than the BCG treated mice. There was no significant difference in bladder mass between the healthy bladders and the bladders treated with the combination therapy of Lp-mCX3CL1 -LVIS_1959 and Lp-mlL-15-LVIS_1959. 8880361

[0316] 45

[0317] For the bladder volume measurements, there was an overall significant decrease for treatment groups 3- 7 compared to the Lp-LVIS_1959 control group (Figure 18C). The bladder volume of bladders treated with BCG and the combination therapy of Lp-mCX3CL1-LVIS_1959 and Lp-mlL-15-LVIS_1959 were still significantly larger than the healthy bladders. The collective results show that the engineered L. plantarum WCFS1 strains were able to regress the implanted MB49 bladder tumours, with the combination therapy of Lp-mCX3CL1-LVIS_1959 and Lp-mlL-15-LVIS_1959 delivering the best therapeutic efficacy.

[0318] The engineered OppA sustains cytokine delivery resulting in more significant regression of bladder tumours

[0319] Having observed a reduction in bladder mass and volume for the different treatment groups, an additional experiment was used to investigate if the anchoring of LVIS_1959 on the surface of L. plantarum WCFS1 would translate to an increase in therapeutic efficacy. Two different set of engineered strains were administered to mice with the same treatment schedule shown in Figure 17A. The first set contained L. plantarum WCFS1 engineered to deliver mCX3CL1 and mlL-15 only, and the second set contained L. plantarum WCFS1 engineered to deliver mCX3CL1 and mlL-15 with the surface anchored LVIS_1959 binding protein.

[0320] This experiment revealed that the bladder mass and volume from mice instilled with the combination Lp- mCX3CL1 and Lp-mlL-15 as compared to the Lp-LVIS_1959 control group showed a significant reduction. This experiment also revealed that the bladder mass and volume from mice instilled with the combination of Lp-mCX3CL1-LVIS_1959 and Lp-mlL-15-LVIS_1959 were significantly smaller than the mice instilled with the combination of Lp-mCX3CL1 and Lp-mlL-15, and Lp-LVIS_1959 control group (Figure 19A-B). These results confirm that the delivery of cytokine payloads without an engineered adhesion molecule to the tumour site using L. plantarum successfully regresses bladder tumours. These results also confirm that the greater retention of the L. plantarum WCFS1 strain is due to the surface anchored LVIS_1959 binding protein resulting in higher therapeutic efficacy.

[0321] Sustained cytokine delivery enhances the infiltration of immune cells at the bladder tumour site

[0322] Next, flow cytometry was used to verify if the sustained release of chemokines into the bladder could enhance the infiltration of targeted immune cell subsets responsible for eliciting an anti-tumour response in bladder tumours. The proportion of CD8+T cells, CD4+T cells, NK1.1+cells, and CD103+DCs within the whole resected bladder organ were measured for bladders instilled with the following treatment groups: the Lp-LVIS_1959 control strain and the combination of the Lp-mCXCL9-LVIS_1959 and Lp- mCXCL10-LVIS_1959 strains.

[0323] The analysis showed that all four populations had a significant increase in proportion when administered the Lp-mCXCL9-LVIS_1959 and Lp-mCXCL10-LVIS_1959 strains compared to the Lp-LVIS_1959 control strain (Figure 20). This result suggested that the delivery of the CXCL9 and CXCL10 chemokines was able to elicit their desired effect of greater targeted immune cell infiltration at the bladder tumour site.

[0324] The proportion of infiltrated NK1 ,1+cells and CD8+T cells within the whole resected bladder organ was also measured for bladders instilled with the Lp-mCX3CL1-LVIS_1959 strain (monotherapy and 8880361

[0325] 46 combination therapy), as CX3CL1 has been reported to attract these immune cell populations. The Lp- LVIS_1959 and Lp-mlL-15-LVIS_1959 strains were also analysed to serve as controls.

[0326] The flow cytometry analysis showed that there was a significant increase in proportion of the NK1 .1+CX3CR1+cells for the bladders treated with the Lp-mCX3CL1-LVIS_1959 strain and the combination of the Lp-mCX3CL1-LVIS_1959 and Lp-mlL-15-LVIS_1959 strains, compared to the Lp- LVIS_1959 strain (Figure 21). This was not observed for the Lp-mlL-15-LVIS_1959 strain. This confirmed that the administration of CX3CL1 at the tumour site was essential for enhancing NK cell infiltration. Additionally, the combined administration of the Lp-mCX3CL1-LVIS_1959 and Lp-mlL-15-LVIS_1959 strains resulted in a significantly greater proportion of the CD8+CX3CR1+T cells compared to the Lp- LVIS_1959 strain (Figure 21 B). There was no observable increase for the Lp-mCX3CL1-LVIS_1959 and the Lp-mlL-15-LVIS_1959 monotherapy groups.

[0327] Combination therapy

[0328] While the administration of the engineered bacteria over two doses showed therapeutic efficacy, the bladder masses and bladder volumes of multiple treatment groups were significantly greater than the healthy bladder. Hence, the next step involved extending the treatment to 4 doses (Figure 22A). It was found that the bladder mass and volume of mice instilled with the Lp-mCX3CL1 -LVIS_1959 monotherapy, Lp-mlL-15-LVIS_1959 monotherapy, and the combination therapy of Lp-mCX3CL1-LVIS_1959 and Lp- mlL-15-LVIS_1959 were significantly smaller than the Lp-LVIS_1959 control group (Figure 22B-C).

[0329] Notably, there was also no significant difference in bladder mass and volume between the healthy mice group and the mice group treated with the combination therapy of Lp-mCX3CL1-LVIS_1959 and Lp-mlL- 15-LVIS_1959. The bladder mass and volume for the BCG treated group however was still found to be significantly higher than the healthy mice group. The results reaffirmed that the engineered L. plantarum WCFS1 strains were able to regress the implanted MB49 bladder tumours in the bladder, with the combination therapy of Lp-mCX3CL1-LVIS_1959 and Lp-mlL-15-LVIS_1959 delivered the best therapeutic efficacy.

[0330] EXAMPLE 5 - Identifying further immunomodulators of interest

[0331] In order to identify additional immunomodulators of interest, migration assays were conducted, as described above, in the presence of commercial cytokines and engineered L. plantarum secreted cytokines. The aim of these experiments was to identify downstream cytokines that were produced by J82 carcinoma cells in response to upstream delivered cytokines. A number of downstream cytokines associated with anti-tumour effects were identified as potential targets for bacterial delivery enabling a more precise and tailored anti-tumour response.

[0332] T rends observed in response to the presence of commercial cytokines were generally the same as those observed in response to engineered L. plantarum secreted cytokines. For example, significant upregulation was observed in CXCL11 , TRAIL, GM-CSF and CX3CL1 , in response to commercial and 8880361

[0333] 47 engineered L. plantarum secreted CXCL9 in combination with CXCL10 (Figure 23). IL-15 was the only cytokine significantly upregulated in response to engineered L. plantarum secreted CXCL9 and CXCL10 but not commercial CXCL9 and CXCL10, although some upregulation was still observed (Figure 23I). Significant upregulation was also observed in IL-21 , CXCL11 , IFN-a, TRAIL, CD40L, GM-CSF, TNF-a, CX3CL1 and IL-15, in response to commercial and engineered L. plantarum secreted CCL4 (Figure 23).

[0334] Downstream cytokines produced in response to commercial and engineered L. plantarum secreted CX3CL1 were also investigated. Statistically significant upregulation of downstream cytokines CX3CL1 , CXCL11 , IFN-y, IL-15 and TRAIL was detected in response to commercial CX3CL1 with NK-92 cells (Figure 24A). Statistically significant upregulation of CXCL11 was detected in response to engineered L. plantarum secreted CX3CL1 (Figure 24B).

[0335] Finally, CCL3, CCL5 and CXCL10 were found to be significantly upregulated in response to engineered L. plantarum secreted mCX3CL1 in combination with mlL-15 (Figure 25). These data show the mechanism by which these cytokines exhibit anti-tumour activity and identify further cytokines of interest for anti-tumour delivery.

[0336] Conclusion

[0337] As described herein, the L. plantarum WCFS1 strain was successfully engineered to anchor the LVIS_1959 OppA binding protein on its surface to enhance its binding efficiency to mouse and human bladder carcinoma cells. Further testing in the MB49 murine model showed higher retention of the surface-modified L. plantarum WCFS1 strain than the empty vector control.

[0338] To design the therapeutic strategy for bladder carcinoma, in silico gene expression profiling was performed on a public human bladder carcinoma dataset to look for chemokines and cytokines with diminished expression levels in bladder tumours compared to healthy bladder tissues. The CXCL9, CXCL10, and CX3CL1 chemokines were shortlisted from a larger list of chemokines based on their ability to attract preferred populations of immune cells.

[0339] Finally, the L. plantarum WCSF1 strain was engineered to individually secrete the shortlisted chemokines to attract immune cells to trigger an anti-tumour response. Using migration assays established on a commercially available 3D microfluidic device, the L. plantarum WCFS1 secreted chemokines were confirmed to be bioactive. The engineered L. plantarum WCSF1 strains were then tested fortheir therapeutic efficacy in the MB49 orthotopic syngeneic murine model for bladder carcinoma. The engineered strains were found to significantly reduce bladder mass and volume, indicating therapeutic efficacy and response. Flow cytometry confirmed that the sustained delivery of chemokines into the bladder could significantly enhance the infiltration of targeted immune cell subsets responsible for eliciting an anti-tumour response for eliminating bladder tumours. 8880361

[0340] 48

[0341] EXAMPLE 6 - Engineering L. plantarum to secrete CCL4 and testing bioactivity of CCL4

[0342] A secondary strategy was developed through the engineering of L. plantarum WCSF1 to secrete CCL4 to bind the CCR5 receptor found on CD103+ DCs, T cells and NK cells, triggering these cells to migrate in response to the CCL4 concentration gradient formed at the bladder tumor site.

[0343] Human and mouse CCL4 were each cloned into a gene cassette comprising of the pgm promoter, synthetic RBS, Lp_0373 secretion signal peptide and the terminator, Term 908 (Figure 9A). The gene cassette was cloned into the pTRK892 plasmid backbone that contains the pWV01 replicon and the erythromycin resistance gene (Erm), and transformed into L. plantarum WCFS1 . After successfully cloning the gene constructs, the engineered L. plantarum WCFS1 cultures were grown overnight and the secreted human and mouse CCL4 in the MRS media were extracted using trichloroacetic acid (TCA) precipitation. Western blot was subsequently used to detect protein expression and secretion of CCL4 from engineered L. plantarum WCFS1 strains (Figure 26).

[0344] Next, the following test was used to check if the L. plantarum WCFS1 secreted chemokines were able to attract T cells in a concentration dependent manner. The 3 different concentrations of secreted human CCL4 (hCCL4) were generated by culturing Lp-hCCL4 in AIM-V + 2% human AB serum cultures with the different OD600 = 0.3, 0.6 and 1 .2 (Figure 27A). Lp-empty was also cultured with the different OD600 = 0.3, 0.6 and 1 .2 to serve as a second control to account for native proteins secreted by L. plantarum WCFS1 that could affect T cell migration. The media containing secreted hCCL4 was directly added to the side channel on the 3D microfluidic device. IL-2 (100IU / ml) was supplemented to the AIM-V + 2% AB serum culture media used throughout the 3D microfluidic device to ensure T cell survival. The concentration of hCCL4 for the OD600 = 0.3, 0.6 and 1 .2 cultures were measured using the Luminex assay to check for the amount of hCCL4 used in the 3D microfluidic device (Figure 27B).

[0345] The secreted hCCL4 from all three OD600 = 0.3, 0.6 and 1 .2 cultures were able to significantly increase T cell migration when compared to the blank and empty vector controls, after 48h of incubation. Out of the 3 different OD600 cultures on day 3, OD600 = 0.3 culture had the highest T cell chemotactic response and this response decreased slightly for OD600 = 0.6 and 1 .2 (Figure 28A-C). This indicated that the hCCL4 concentration in the OD600 = 0.3 culture was the most optimal, and that higher concentrations negatively impacted T cell migration.

[0346] Having confirmed that the secreted hCCL4 was functional, the next step involved testing the therapeutic efficacy of the Lp-mCCL4 strain in the MB49 orthotopic bladder tumour mouse model. An additional engineered L. plantarum strain that secreted the CCL5 chemokine was incorporated in the animal model studies. The Lp-mCCL5 strain was added as a comparator to determine whether differences in activity or potency, given that both CCL4 and CCL5 targeted the same CCR5 receptor. The tumour model was established by implanting the MB49 bladder carcinoma cell line in the bladder of each mouse. After allowing the tumour to grow within the bladder for a week, the Lp-mCCL4 and Lp-mCCL5 strains were instilled into the bladder weekly, over the course of four weeks. It was found that the bladder masses and volume of mice instilled with both strains were significantly smaller than the Lp-empty and saline control group (Figure 29). 8880361

[0347] 49

[0348] EXAMPLE 7 - Pairing chemokine therapies with immune-activating cytokines and immune checkpoint inhibitors

[0349] To enhance immune infiltration and cytotoxic activity within the bladder TME, a modular “attract-activate” strategy was built on two classes of immunomodulators: attractors and activators. These effectors serve distinct yet complementary functions in coordinating immune cell trafficking and effector function against carcinoma cells. An 'attractor1is defined as a cytokine that recruits immune cells by binding chemokine receptors and directing their migration along a gradient. Representative attractors used in the examples here include CX3CL1 , CXCL9, CXCL10, CCL4, CCL5, each implicated in promoting lymphocyte migration. An 'activator' here refers to immunostimulatory molecules that enhance the functional potency, persistence, or cytotoxic activity of infiltrating immune cells. This category includes: (i) classical immune- activating cytokines that directly promote immune cell function (representative examples include IL-15 (Figures 15 and 22B-C) and IL-2 (Figures 12, 27 and 29)); and (ii) immune checkpoint inhibitors, which function as activators by blocking inhibitory signalling pathways in immune cells. Despite acting through inhibitory pathways instead of direct stimulatory signals, immune checkpoint inhibition yields a biological outcome of increased T cell cytotoxicity, thereby aligning it with the activator class. Representative examples of immune checkpoint inhibitors include PD-L1 , PD-1 or CTLA-4 inhibitors. The mouse anti-PD- L1 and anti-CTLA-4 nanobodies were cloned into the pTRK892 plasmid backbone in place of where the chemokine gene was supposed to be. This engineered strain was also grown in culture overnight in the buffered MRS media and the secreted proteins were extracted by TCA precipitation. Western blot was used to detect the expression and secretion of both nanobodies (Figure 30).

[0350] The MB49 orthotopic bladder tumour mouse model was used to test if the combination of Lp-mCXCL9 and Lp-mCXCL10 strains could be paired with activators, given that this was previously observed with the Lp-mCX3CL1 + Lp-IL-15 combination. After allowing the tumour to grow within the bladder for a week, the following strain combinations were instilled into the mice bladders weekly, over the course of four weeks: (1) combination of Lp-mCXCL9 and Lp-mCXCL10, (2) combination of Lp-mCXCL9, Lp-mCXCL10 and Lp- mlL-2, and (3) combination of Lp-mCXCL9, Lp-mCXCL10 and Lp-PD-L1 nb, (4) Lp-PD-L1 nb and (5) BCG. It was found that the bladder masses and volume of mice instilled with all strains were significantly smaller than the Lp-empty and saline control group (Figure 29). While not significant, there was an observable lower bladder mass and volume across all treatment groups compared to BCG-treated bladders. The results here also indicated that the Lp-mCXCL9 and Lp-mCXCL10 combination, when coadministered with an additional activator, exhibits activity that is equal to, greater than, or functionally comparable to its activity with or without the activator. 8880361

[0351] 50

[0352] References

[0353] A number of publications are cited above in order to more fully describe and disclose the invention and the state of the art to which the invention pertains. Full citations for these references are provided below. The entirety of each of these references is incorporated herein.

[0354] 1 . Ferlay, J. et al. (2021) Cancer statistics for the year 2020: An overview. Int J Cancer.

[0355] 2. Berdik, C. (2017) Unlocking bladder cancer. Nature 551 , S34.

[0356] 3. Furuse, H. and Ozono, S. (2010) Transurethral resection of the bladder tumour (TURBT) for nonmuscle invasive bladder cancer: basic skills. Int J Urol 17 (8), 698-9.

[0357] 4. Brausi, M. et al. (2011) A Review of Current Guidelines and Best Practice Recommendations for the Management of Nonmuscle Invasive Bladder Cancer by the International Bladder Cancer Group. Journal of Urology 186 (6), 2158-2167.

[0358] 5. Kawai, K. et al. (2013) Bacillus Calmette-Guerin (BCG) immunotherapy for bladder cancer: current understanding and perspectives on engineered BCG vaccine. Cancer Sci 104 (1), 22-7.

[0359] 6. De Boer, E.C. et al. (1992) Induction of urinary interleukin-1 (IL-1), IL-2, IL-6, and tumour necrosis factor during intravesical immunotherapy with bacillus Calmette-Guerin in superficial bladder cancer. Cancer Immunol Immunother 34 (5), 306-12.

[0360] 7. Zuiverloon, T.C. et al. (2012) Markers predicting response to bacillus Calmette-Guerin immunotherapy in high-risk bladder cancer patients: a systematic review. Eur Urol 61 (1), 128-45.

[0361] 8. Hussain, M.H. et al. (2009) Bladder cancer: narrowing the gap between evidence and practice. J Clin Oncol 27 (34), 5680-4.

[0362] 9. Rouanne, M. et al. (2022) BCG therapy downregulates HLA-I on malignant cells to subvert antitumor immune responses in bladder cancer. The Journal of Clinical Investigation 132 (12).

[0363] 10. Decaestecker, K. and Oosterlinck, W. (2015) Managing the adverse events of intravesical bacillus Calmette-Guerin therapy. Res Rep Urol 7, 157-63.

[0364] 11. Zheng, Y.-q. et al. (2015) Applications of bacillus Calmette-Guerin and recombinant bacillus Calmette- Guerin in vaccine development and tumor immunotherapy. Expert review of vaccines 14 (9), 1255-1275.

[0365] 12. Teppema, J.S. et al. (1992) Morphological aspects of the interaction of Bacillus Calmette-Guerin with urothelial bladder cells in vivo and in vitro: relevance for antitumor activity? Urol Res 20 (3), 219-28.

[0366] 13. Lee, Y.C. et al. (2022) The dynamic roles of the bladder tumour microenvironment. Nat Rev Urol 19 (9), 515-533.

[0367] 14. Messing, E.M. (2017) The BCG Shortage. Bladder cancer (Amsterdam, Netherlands) 3 (3), 227-228.

[0368] 15. Liu, Y.T. and Sun, Z.J. (2021) Turning cold tumors into hot tumors by improving T-cell infiltration. Theranostics 11 (11), 5365-5386. 8880361

[0369] 51

[0370] 16. Mariathasan, S. et al. (2018) TGFp attenuates tumour response to PD-L1 blockade by contributing to exclusion of T cells. Nature 554 (7693), 544-548.

[0371] 17. Lebacle, C. et al. (2021) BCG-unresponsive high-grade non-muscle invasive bladder cancer: what does the practicing urologist need to know? World Journal of Urology 39 (11), 4037-4046.

[0372] 18. Oddens, J. et al. (2013) Final results of an EORTC-GU cancers group randomized study of maintenance bacillus Calmette-Guerin in intermediate- and high-risk Ta, T1 papillary carcinoma of the urinary bladder: one-third dose versus full dose and 1 year versus 3 years of maintenance. Eur Urol 63 (3), 462-72.

[0373] 19. Coon, B.G. et al. (2012) Fibronectin attachment protein from bacillus Calmette-Guerin as targeting agent for bladder tumor cells. International journal of cancer 131 (3), 591-600.

[0374] 20. Luo, Y. et al. (2000) Co-expression of interleukin-2 and green fluorescent protein reporter in mycobacteria: in vivo application for monitoring antimycobacterial immunity. Mol Immunol 37 (9), 527-36.

[0375] 21 . Domingos-Pereira, S. et al. (2019) Intravesical Ty21 a Vaccine Promotes Dendritic Cells and T Cell- Mediated Tumor Regression in the MB49 Bladder Cancer Model. Cancer Immunol Res 7 (4), 621-629.

[0376] 22. Domingos-Pereira, S. et al. (2017) Preclinical efficacy and safety of the Ty21a vaccine strain for intravesical immunotherapy of non-muscle-invasive bladder cancer. Oncoimmunology 6 (1), e1265720.

[0377] 23. Glanville, N. et al. (2021) 878 Novel microbial immunotherapy approach for the treatment of bladder cancer. Journal for ImmunoTherapy of Cancer 9 (Suppl 2), A920-A920.

[0378] 24. Green, D.B. et al. (2018) Complications of Intravesical BCG Immunotherapy for Bladder Cancer. RadioGraphics 39 (1), 80-94.

[0379] 25. Silva, A.L. et al., The Impact of Immune Checkpoint-Inhibitors Therapy in Urinary Bladder Cancer, Onco, 2021 , pp. 3-22.

[0380] 26. Narayan, V.M. and Dinney, C.P.N. (2020) Intravesical Gene Therapy. Urol Clin North Am 47 (1), 93- 101.

[0381] 27. Schneewind, O. and Missiakas, D.M. (2012) Protein secretion and surface display in Gram-positive bacteria. Philosophical Transactions of the Royal Society B: Biological Sciences 367 (1592), 1123-1139.

[0382] 28. Martin, R. et al. (2013) Surface glycosaminoglycans mediate adherence between HeLa cells and Lactobacillus salivarius Lv72. BMC microbiology 13, 210-210.

[0383] 29. Munoz-Provencio, D. et al. (2011) Identification of Surface Proteins from Lactobacillus casei BL23 Able to Bind Fibronectin and Collagen. Probiotics and Antimicrobial Proteins 3 (1), 15-20.

[0384] 30. Monnet, V. (2003) Bacterial oligopeptide-binding proteins. Cellular and Molecular Life Sciences CMLS 60 (10), 2100-2114.

[0385] 31. De Klerk, D.P. (1985) The Glycosaminoglycans of Human Bladder Cancers of Varying Grade and Stage. Journal of Urology 134 (5), 978-981. 8880361

[0386] 52

[0387] 32. Juncker, A.S. et al. (2003) Prediction of lipoprotein signal peptides in Gram-negative bacteria. Protein Sci 12 (8), 1652-62.

[0388] 33. Tjalsma, H. and van Dijl, J.M. (2005) Proteomics-based consensus prediction of protein retention in a bacterial membrane. PROTEOMICS 5 (17), 4472-4482.

[0389] 34. Duong, T. et al. (2011) Construction of vectors for inducible and constitutive gene expression in Lactobacillus. Microbial Biotechnology 4 (3), 357-367.

[0390] 35. Callanan, M.J. et al. (2007) Modification of Lactobacillus p-glucuronidase activity by random mutagenesis. Gene 389 (2), 122-127.

[0391] 36. Nguyen, H.-M. et al. (2019) Constitutive expression and cell-surface display of a bacterial p- mannanase in Lactobacillus plantarum. Microbial Cell Factories 18 (1), 76.

[0392] 37. Joseph, M. and Enting, D. (2019) Immune Responses in Bladder Cancer-Role of Immune Cell Populations, Prognostic Factors and Therapeutic Implications. Front Oncol 9, 1270.

[0393] 38. Groom, J.R. and Luster, A.D. (2011) CXCR3 in T cell function. Exp Cell Res 317 (5), 620-31.

[0394] 39. Spranger, S. and Gajewski, T.F. (2016) Tumor-intrinsic oncogene pathways mediating immune avoidance. Oncoimmunology 5 (3), e1086862.

[0395] 40. Spranger, S. et al. (2015) Melanoma-intrinsic p-catenin signalling prevents anti-tumour immunity. Nature 523 (7559), 231-5.

[0396] 41 . Williford, J.M. et al. (2019) Recruitment of CD103(+) dendritic cells via tumor-targeted chemokine delivery enhances efficacy of checkpoint inhibitor immunotherapy. Sci Adv 5 (12), eaayl 357.

[0397] 42. Liu, J.Y. et al. (2015) CTL- vs Treg lymphocyte-attracting chemokines, CCL4 and CCL20, are strong reciprocal predictive markers for survival of patients with oesophageal squamous cell carcinoma. Br J Cancer 113 (5), 747-55.

[0398] 43. Brassard, J. et al. (2018) CD103 expression by DCs: a new player in lung cancer immunity. The Journal of Immunology 200 (1 Supplement), 56.21 .

[0399] 44. Guo, J. et al. (2003) Fractalkine transgene induces T-cell-dependent antitumor immunity through chemoattraction and activation of dendritic cells. Int J Cancer 103 (2), 212-20.

[0400] 45. Imai, T. et al. (1997) Identification and Molecular Characterization of Fractalkine Receptor CX3CR1 , which Mediates Both Leukocyte Migration and Adhesion. Cell 91 (4), 521-530.

[0401] 46. Dichmann, S. et al. (2001) Fractalkine induces chemotaxis and actin polymerization in human dendritic cells. Inflammation Research 50 (11), 529-533.

[0402] 47. Foussat, A. et al. (2000) Fractalkine receptor expression by T lymphocyte subpopulations and in vivo production of fractalkine in human. Eur J Immunol 30 (1), 87-97.

[0403] 48. Grabstein, K.H. et al. (1994) Cloning of a T Cell Growth Factor that Interacts with the β Chain of the lnterleukin-2 Receptor. Science 264 (5161), 965-968. 8880361

[0404] 53

[0405] 49. Burton, J.D. et al. (1994) A lymphokine, provisionally designated interleukin T and produced by a human adult T-cell leukemia line, stimulates T-cell proliferation and the induction of lymphokine-activated killer cells. Proceedings of the National Academy of Sciences 91 (11), 4935-4939.

[0406] 50. Shen, H. et al. (2024) Prodrug-conjugated tumor-seeking commensals for targeted cancer therapy. Nature Communications 15 (4343), 1-15.

[0407] 51 . Michon, C et al. (2016) Display of recombinant proteins at the surface of lactic acid bacteria: strategies and applications. Microbial Cell Factories 15:70.

[0408] 52. Boorjian SA, et al. (2021) Intravesical nadofaragene firadenovec gene therapy for BCG-unresponsive non-muscle-invasive bladder cancer: a single-arm, open-label, repeat-dose clinical trial. Lancet Oncology. 22(1):107-117

[0409] 53. Svatek R. et al. (2024) PIVOT-006: A phase 3, randomized study of cretostimogene grenadenorepvec versus observation for the treatment of intermediate risk non-muscle invasive bladder cancer (IR-NMIBC) following transurethral resection of bladder tumor (TURBT). Journal of Clinical Oncology 42(4).

[0410] 54. Vilaseca A. et al. (2023) Safety and efficacy of the erdafitinib (erda) intravesical delivery system, TAR- 210, in patients (pts) with non-muscle-invasive bladder cancer (NMIBC) or muscle-invasive bladder cancer (MIBC) harboring select FGFR mutations or fusions: Phase 1 first-in-human study. Journal of Clinical Oncology 41 (6).

[0411] 55. Brisuda, A et al. (2021) Bladder cancer therapy using a conformationally fluid tumoricidal peptide complex. Nature Communications 12(3427).

[0412] 56. Karim Chamie et al. (2022) Final clinical results of pivotal trial of IL-15RaFc superagonist N-803 with BCG in BCG-unresponsive CIS and papillary nonmuscle-invasive bladder cancer (NMIBC). Journal of Clinical Oncology 40(16).

[0413] For standard molecular biology techniques, see Sambrook, J., Russel, D.W. Molecular Cloning, A

[0414] Laboratory Manual. 3 ed. 2001 , Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press

Claims

888036154Claims:1 . A composition comprising one or more bacteria, wherein the one or more bacteria are engineered to express a payload, wherein the payload comprises at least two immunomodulators, wherein at least one of the immunomodulators is an attractor, wherein the attractor recruits an immune cell by binding a chemokine receptor on the immune cell.

2. The composition according to claim 1 , wherein the composition comprises one bacterium engineered to express the payload comprising the at least two immunomodulators.

3. The composition according to claim 1 , wherein the composition comprises two or more bacteria each engineered to express at least one immunomodulator, wherein the immunomodulators expressed by the two or more bacteria together constitute the payload.

4. The composition according to any one of claims 1-3, wherein the at least two immunomodulators are selected from the group consisting of a chemokine, an interferon (IFN), an interleukin (IL), a tumour necrosis factor (TNF), a haematopoietin, a transforming growth factor beta (TGF-p), a colony stimulating factor (CSF), VEGF-A, MIF, BAFF, CXCL5, CXCL1 , HGF, CXCL11 , IFN-y, IL-1a, IL-2, IL-3, IL-4, IL-7, IL-15, IL-18, IL-22, M-CSF, CCL2, CCL20, SCF, SDF-1a, CXCL9, CXCL10, CCL4, IL-15, CX3CL1 , IL- 21 , IFN-a, TRAIL, CD40L, GM-CSF, TNF-a, CCL3, CCL5, and an immune checkpoint inhibitor.

5. The composition according to any one of claims 1-4, wherein the at least two immunomodulators are selected from the group consisting of CX3CL1 , CXCL9, CXCL10, CCL4, CCL5, IL-15, IL-2 and an immune checkpoint inhibitor.

6. The composition of any one of claims 1-4, wherein the at least one attractor is / are selected from the group consisting of a chemokine, IL-8, CXCL5, CXCL1 , CXCL11 , CCL2, CCL20, SDF-1a, CXCL9, CXCL10, CCL4, CX3CL1 , CXCL11 , CCL3 and CCL5.

7. The composition according to claim 6, wherein the at least one attractor is / are selected from the group consisting of CX3CL1 , CXCL9, CXCL10, CCL4 and CCL5.

8. The composition according to any one of claims 1-7, wherein the payload comprises CXCL9 and CXCL10.

9. The composition according to any one of claims 1-7, wherein the payload comprises CX3CL1 and IL-15.

10. The composition according to any one of claims 1-7, wherein the payload comprises CXCL9, CXCL10 and CCL4.88803615511. The composition according to any one of claims 1-7, wherein the payload comprises CXCL9, CXCL10 and IL-2.

12. The composition according to any one of claims 1-7, wherein the payload comprises CXCL9, CXCL10, CCL4 and IL-2.

13. The composition according to any one of claims 1-7, wherein the payload comprises CXCL9, CXCL10 and CCL5.

14. The composition according to any one of claims 1-7, wherein the payload comprises CXCL9, CXCL10 and IL-15.

15. The composition according to any one of claims 1-7, wherein the payload comprises CXCL9, CXCL10, IL-15 and IL-2.

16. The composition according to any one of claims 1 and 3-8, wherein the composition comprises a first bacterium engineered to express CXCL9 and a second bacterium engineered to express CXCL10.

17. The composition according to any one of claims 1 , 3-7 and 9, wherein the composition comprises a first bacterium engineered to express CX3CL1 and a second bacterium engineered to express IL-15.

18. The composition according to any one of the preceding claims, wherein at least one of the immunomodulators is an activator, wherein the activator enhances the functional potency, persistence and / or cytotoxic activity of recruited immune cells.

19. The composition according to claim 18, wherein the at least one activator is selected from the group consisting of IL-15, IL-2 and an immune checkpoint inhibitor.

20. The composition according to any one of claims 4-19, wherein the immune checkpoint inhibitor is an antibody, nanobody, peptide, aptamer, RNA-based inhibitor or small molecule.21 . The composition according to any one of claims 4-20, wherein the immune checkpoint inhibitor is a PD-L1 inhibitor, a PD-1 inhibitor or a CTLA-4 inhibitor.

22. The composition according to claims 18 or 19, wherein the activator is IL-2.

23. The composition according to claim 21 or 22, wherein the payload comprises IL-2 and / or a PD-L1 inhibitor and / or a PD-1 inhibitor and / or a CTLA-4 inhibitor and at least one additional activator.

24. The composition according to claim 23, wherein the at least one additional activator is selected from the group consisting of IL-2, IL-15, a PD-L1 inhibitor, a PD-1 inhibitor and a CTLA-4 inhibitor.88803615625. The composition according to any one of claims 18-21 , wherein the activator is a PD-L1 inhibitor.

26. The composition according to any one of claims 18-21 , wherein the activator is a PD-1 inhibitor.

27. The composition according to any one of claims 18-21 , wherein the activator is a CTLA-4 inhibitor.

28. The composition according to any one of the preceding claims, wherein the immune cell is:(a) a lymphocyte, optionally wherein said lymphocyte is a T cell or a natural killer cell (NK), further optionally wherein the T cell is a CD4+ cell, a CD8+ cell, or(b) a natural killer T cell (NKT), or an antigen presenting cell (APC), optionally wherein said APC is a dendritic cell (DC).

29. The composition according to any one of the preceding claims, wherein the one or more bacteria are a commensal bacterium / commensal bacteria.

30. The composition according to any one of the preceding claims, wherein the one or more bacteria are gram-positive bacteria, optionally wherein the gram-positive bacteria is of the genus Lactobacillus.31 . The composition according to claim 30, wherein the one or more bacteria of the genus Lactobacillus is / are of the species:(a) Lactobacillus plantarum, optionally of the strain WCFS1 , or(b) Lactobacillus salivarius, optionally of the strain DSM20555, or(c) Lactobacillus reuteri, optionally of the strain DSM20016.

32. The composition according to any one of the preceding claims, wherein the one or more bacteria are engineered to express an adhesion molecule that is overexpressed from the genome of the bacterium / bacteria as compared to a non-engineered bacterium / bacteria.

33. The composition according to any one of the preceding claims, wherein the one or more bacteria are engineered to express an adhesion molecule that is expressed from a nucleic acid introduced into the bacterium / bacteria.

34. The composition according to any one of the preceding claims, wherein the one or more bacteria are engineered to express an adhesion molecule that is exogenously coated on a surface of the bacterium / bacteria.

35. The composition according to claim 34, wherein said adhesion molecule is an oligopeptide binding protein (OppA).88803615736. The composition according to any one of claims 32-35, wherein the adhesion molecule comprises a membrane anchor with an amino acid sequence having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to any one of SEQ ID NOs: 1-10, optionally wherein the membrane anchor is cleavable.

37. The composition according to any one of claims 32-36, wherein the adhesion molecule comprises a surface adhesin with an amino acid sequence having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to any one of SEQ ID NOs: 11-17.

38. The composition according to any one of claims 33-37, wherein the adhesion molecule comprises a membrane anchor and a surface adhesin that are heterologous to each other.

39. The composition according to any one of claims 33-38, wherein the membrane anchor comprises a secretion signal peptide and a cleavage site, optionally wherein the cleavage site is cleavable by an enzyme, further optionally wherein the enzyme is a signal peptidase II (SPase II).

40. The composition according to any one of the preceding claims, for use as a medicament.41 . The composition according to any one of claims 1 -39, for use in the treatment of cancer in a subject, optionally wherein said cancer is bladder cancer, further optionally wherein said bladder cancer is non-muscle invasive bladder cancer (NMIBC) or muscle invasive bladder cancer (MIBC).

42. The composition according to claim 41 , wherein the cancer exhibits an immune excluded and / or immune desert phenotype.

43. The composition for the use according to claim 41 or 42, wherein the treatment is administered in place of Bacillus Calmette-Guerin (BCG) therapy or alongside BCG therapy.

44. The composition for the use according to claim 41 or 42, wherein the treatment is administered as a follow up therapy suitable for NMIBC patients who have experienced BCG failure.

45. A kit comprising a first container containing a first composition comprising one bacterium engineered to express at least one first immunomodulator, and a second container containing a second composition comprising one bacterium engineered to express at least one second immunomodulator, wherein the first immunomodulator is different from the second immunomodulator, and wherein at least one of the immunomodulators is an attractor, wherein the attractor recruits an immune cell by binding a chemokine receptor on the immune cell.

Images