WO2021074379A1

WO2021074379A1 - Bacterial delivery of viruses into eukaryotic cells

Info

Publication number: WO2021074379A1
Application number: PCT/EP2020/079214
Authority: WO
Inventors: Christian Schaub; Corinne John; Garyfalia STERGIOU; Marina STUDART; Martyna WROBLEWSKA
Original assignee: Redbiotec Ag
Priority date: 2019-10-18
Filing date: 2020-10-16
Publication date: 2021-04-22

Abstract

The present invention relates to a live bacterium with at least one exogenous polynucleotide integrated into the bacterial genome or maintained episomally. This exogenous polynucleotide encodes a virus. Also included are medical applications using the bacterium of the invention.

Description

BACTERIAL DELIVERY OF VIRUSES INTO EUKARYOTIC CELLS

BACKGROUND

Certain viruses are adapted to infect eukaryotic cells and have evolved strategies to replicate even under harsh conditions. Several years ago, it was observed that certain viral infections are capable of reducing tumor sizes. Subsequent studies have shed light on the oncolytic activity of various families of viruses, as they continue to be developed as anticancer weapons.

Since then, numerous oncolytic viruses have been developed, and many are presently in clinical stage as described in the review by Kaufman et al., 2016. However, systemic application remains challenging for most oncolytic viruses. This is often due to premature clearance of the virus by the human immune system, which results in insufficient titers of the virus reaching the tumors and thereby limiting the efficacy. Further, it was found that increasing the dosage to overcome the immune clearance led to safety issues. Additionally, viruses are often not able to penetrate dense tissues and reach deeper areas of the tumors.

On the other hand, some bacterial strains are known to intrinsically target tumor cells thanks to the nutrient-rich, frequently immunosuppressed and hypoxic tumor environment. Certain naturally invasive bacteria like Listeria monocytogenes can be manipulated, for example, to transfer genetic material from the bacteria into mammalian cells in a process termed bactofection (Pilgrim et al., 2003). Likewise, non-invasive bacterial strains like Escherichia coli (E. coli) can be engineered to achieve bactofection. In one example in vaccination research, E. coli strains were engineered to deliver a murine cytomegalovirus (MCMV) - encoded on a bacterial artificial chromosome - into immunocompromised mice (L. Cicin-Sain et al., 2003).

The concept of combining bacteria with viruses for the purpose of cancer treatment was first contemplated by Krzykawski in 2015. This scientific review proposed that viruses and bacteria could possibly be combined to "fight cancer in concert". However, no details as to how this might be established are provided in this paper, other than the theory of binding or fusing a virus to the bacteria via binding proteins. Thus there remains a need in the art for combining bacteria and viruses in a new delivery system, and methods of delivery of therapeutic agents into eukaryotic cells using such system.

SUMMARY OF THE INVENTION

The present invention provides a live bacterium for the in vivo delivery of viruses into eukaryotic cells.

The invention relates to a live bacterium comprising at least one exogenous polynucleotide encoding a virus integrated into the bacterial genome or maintained episomally. In addition, the bacterium targets a eukaryotic cell; and is capable of releasing the DNA or RNA encoding the virus into the cytoplasm of the targeted eukaryotic cell.

In one embodiment, the bacterium of the present invention comprises a polynucleotide encoding a virus that replicates in and lyses cancer cells.

In another embodiment, the virus encoded by the exogenous polynucleotide is an oncolytic virus.

The virus used in the delivery system can be selected from the group consisting of herpes simplex virus, HSV-1, Adenovirus, Adeno-associated virus (AAV), Adenovirus 3, Adenovirus 5, Parvovirus, Parvovirus H1PV, Mononegavirales, Ebola virus, Measles virus, Nipah virus, Influenza virus, Newcastle Disease Virus, Poxvirus, Alphaviruses, Picornaviruses, Coxsackie virus, Flaviviruses, Zika virus, Vaccinia virus, Vesicular Stomatitis virus, Poliovirus, Reovirus, Senecavirus, Echovirus, Semliki Forest virus and Maraba virus, and a chimeric virus.

The bacterium used in the delivery system of the invention can be selected from the group consisting of Enterobacteriaceae spp., Lactobacillus spp., Yersinia spp., Escherichia spp., Klebsiella spp., Bordetella spp., Neisseria spp., Aeromonas spp., Francisella spp., Corynebacterium spp., Citrobacter spp., Chlamydia spp., Hemophilus spp., Brucella spp., Mycobacterium spp., Legionella spp., Rhodococcus spp., Pseudomonas spp., Helicobacter spp., Salmonella spp., Vibrio spp., Bacillus spp., Leishmania spp. and Erysipelothrix spp., Shigella spp., Listeria spp., Rickettsia spp., Acetoanaerobium spp., Aerococcaceae spp., Carnobacteriaceae spp., Enterococcace spp., Leuconostocaceae spp., Streptococcaceae spp., and bacteria that are generally considered safe (GRAS status) by the US Food and Drug Administration (FDA) preferably wherein the bacterium is Listeria monocytogenes, Salmonella typhimurium, Salmonella typhi, or Escherichia coli. In a preferred aspect, the bacterium is Escherichia coli and the virus is HSV-1, adenovirus or parvovirus.

In addition, the bacterium of the invention can comprise one or more further accessory elements, such as a polynucleotide encoding one or more of:

(i) a polynucleotide encoding an invasion factor to enable the bacterium to invade eukaryotic cell, preferably an invasin (inv) derived from Yersinia spp.;

(ii) a polynucleotide encoding bacteriocin release protein (BRP), and bacteriophage lambda lysozyme, holin and antiholin to enable the said bacterium to self lyse upon entering the eukaryotic phagosomes;

(iii) a polynucleotide encoding a phospholipase, preferably phospholipase C (pic) derived from Clostridium spp. or Listeria spp. and capable of decomposing the eukaryotic phagosomal membranes;

(iv) a polynucleotide encoding a protein capable of lysing the eukaryotic phagosomal membranes, preferably listeriolysin O (hly) derived from Listeria spp. or perfringolysin (pfo) derived from Clostridium spp.; and

(v) a promotor driving the expression of the polynucleotides und (i) and (ii) and which responds to an environmental stimulus, such as a low magnesium, hypoxia, or bacterial quorum sensing.

The bacterium of the invention, may be engineered to be auxotroph, preferably diaminopimelic acid (dap) auxotroph.

Alternately, the bacterium of the invention may be a prototroph.

The bacterium of the present invention may have one or more of the endogenous bacterial endonucleases and/or DNA repair proteins and/or metabolic proteins which are not functional, optionally wherein

(i) the endogenous bacterial endonucleases are EndA and/or HsdR;

(ii) the endogenous bacterial DNA repair protein is RecA, and/or

(iii) the endogenous metabolic proteins are thiamine phosphate synthase (ThiE) and/or thiamine thiazole synthetase (ThiA), and/or a component of the Ton system such as the endogenous TonB receptor. The bacterium may be engineered to avoid or minimize clearance by the host immune system using mechanisms such as the deletion of IpxM gene to express truncated and less virulent versions of surface lipopolysaccharides (LPS), or the coating of the bacterial surface with nanoparticles, such as polyethylene glycol (PEG) nanoparticles.

In a preferred aspect, the bacterium of the invention is an E. coli which comprises invasin (inv), and listeriolysin O (hly) and wherein the bacterium has been engineered so that the endogenous bacterial endonucleases EndA and HsdR and the endogenous bacterial DNA repair protein RecA are not functional and the IpxM gene has been deleted.

In addition, the bacterium may be further engineered to comprise an exogenous polynucleotide encoding a pro-apoptotic element, checkpoint inhibitor, antibody, nanobody, enzyme, antiangiogenic element, immune modulator and/or other anticancer molecule integrated into the bacterial or the viral genome.

The bacterium of the present invention may be used in the diagnosis, prevention or treatment of mammalian disease, preferably human disease.

The bacterium of the invention may be used in therapy, for example in the diagnosis, prevention or treatment of mammalian diseases, preferably human diseases. Specifically, the bacterium of the invention could be used in the prevention or treatment of cancer.

When the bacterium is used to diagnose, prevent or treat cancer, the cancer can be selected from the group consisting of brain tumor, glioblastoma, pituitary tumor, colorectal cancer, ovarian cancer, prostate cancer, pancreatic cancer, lung cancer, gastric cancer, liver cancer, head-and-neck cancer, oropharyngeal cancer, hypopharyngeal cancer, laryngeal cancer, lip cancer, oral cavity cancer, salivary glands cancer), anal cancer, penile cancer, vaginal cancer, vulvar cancer, breast cancer, bladder cancer, kidney cancer, bile duct cancer, skin cancers, melanoma, central nervous system tumors, adrenal cancer, adrenocortical carcinoma, appendix cancer, rectal cancer, endometrial cancer, esophageal cancer, eye cancer, retinoblastoma), gallbladder cancer, heart cancer, small intestine cancer, teratoid tumor, testicular cancer, thymus cancer, thyroid cancer, urethral cancer, uterine cancer, Wilms tumor, mesothelioma, multiple myeloma, sarcomas, Kaposi's sarcoma, uterine sarcoma, osteosarcoma, Lymphomas, Burkitt lymphoma, Hodgkin lymphoma, and non-Hodgkin lymphoma. Administration of the bacterium may be locally or systemically, optionally as oral, inhalation, intratumoral, intravenous, or subcutaneous administration.

In a further aspect, the bacterium may be administered in combination with one or more additional therapies such as an anticancer treatment such as cytotoxic chemotherapies, monoclonal antibodies, cancer vaccines, cytokines, checkpoint inhibitors, immunomodulators, surgery or radiotherapy. Preferably, the bacterium is Escherichia coli, the virus is HSV-1 and the cancer is pancreatic or colon cancer.

Figures

Figure 1 depicts the percentage of mammalian cells invaded by the engineered bacteria after defined periods of time and at different multiplicities of infection (MOIs, which defines the concentration of bacteria to mammalian cells) in Vero and Hela cells, respectively.

Figure 2 shows the cellular entry of invasive E. coli at different time points . The average number of internalized bacteria per mammalian cell is shown. The number of internalized E. coli bacteria increases over time and with increasing MOI in Vero and Hela cells, respectively. Different cell lines vary in their susceptibility to bacterial invasion.

Figure 3 depicts internalized E. coli at 30 min (A) and 6 h (B) post bacterial invasion (white arrow) in Vero cells. After the cell invasion, the bacteria self-lyse within the phagosome. (A) At the 30 min time point, the bacteria are still intact; (B) at the 6h time point, the bacteria are already lysed and disintegrated. Cargo release from the vacuole occurs after bacterial lysis and is trafficked to the nucleus (said E. coli is engineered to trigger this event).

Figure 4 shows the EGFP expression in HeLa cells from E. co//-delivered plasmid DNA.

Figure 5 shows the expression of viral antigen in Vero cells up to 3 days post bactofection. More than 98% of Vero cells show expression of an early HSV-1 viral antigen upon bactofection (MOI 20). Expression occurs at early time points, suggesting it is a result of direct DNA delivery to the nucleus (and not the virus-reinfection process).

Figure 6 depicts viral reconstitution after bactofection using bacteria encoding a virus. Two different bacterial MOIs were used for bactofection. The titers of reconstituted virus increase over time and with increasing bacterial MOI. Virus reconstitution can also be detected with lower MOI and as early as day 1 (data not shown), though normally seen at day 2 or 3 post bactofection.

Figure 7 shows Vero cells infected by E. co//-delivered, in s/fu-reconstituted HSV-1 virus.

Figure 8 shows cell line differences in their permissiveness to bacterial invasion and virus reconstitution. Figure 9 depicts bacterial clearance from the blood. Results from two different blood donors are presented. The titer of the engineered E. coli decreased in the blood by 4 log after 1 hour as shown for two different strains (strain 1 = dap auxotroph, strain 2 = dap prototroph). In both donors, engineered bacteria show the same clearance pattern as wild type, non-pathogenic E. coli, despite their invasiveness and other modifications, suggesting their safety when injected intravenously.

Figure 10 shows bacterial accumulation in tumor and in organs. Days 2, 5 and 7 post application. Engineered E. coli (non-invasive dap prototroph) were recovered from CT26 tumors from syngeneic Balb/C mice following intravenous (IV) injection of 1x10^s bacteria. Bacteria accumulate mainly in the tumor and reach titers of lxlO⁸ at 2 days post administration (p.a.). This experiment was conducted in immunocompetent mice.

Figure 11 shows bacterial accumulation at the tumor site post intravenous (IV) injection, with bacteria carrying the firefly luciferase gene (lux) integrated into their genome. This experiment was conducted in immunocompetent mice.

Figure 12 shows the bacterial accumulation in the tumor at day 2 and 7 post IV injection. The number of E. coli bacteria in the tumor varies at the early time points (2 days p.a.), depending on the initial titer (administered IV). At later time points, the bacteria titer levels off. The titers remain stable for at least 8 days. This experiment was conducted in immunocompetent mice.

Figure 13 shows bacterial accumulation in the tumor at 2 h and 72 h post IV injection. Results from the invasive dap prototroph E. coli strain are presented. Bacteria are distributed in the tumor and organs at 2h post administation (p.a.). At a later time point (72 h p.a.), bacteria numbers increase in tumors and decrease in most organs. This experiment was conducted in immunocompetent mice.

Figure 14 shows bacterial accumulation in the tumor and organs at 8 days post administration. Results from the dap prototroph strain upon IV and intratumoral (IT) injection are presented. Both administration routes led to tumor accumulation. Lower off-target titer values were observed upon IT administration. This experiment was conducted in immunocompetent mice.

Figure 15 shows the delivery of cargo (firefly luciferase) to tumor cells at day 4 post bacteria invasion. Bacteria invade tumor cells very quickly after being systemically applied. Reporter gene expression can be detected upon delivery by engineered E. coli. Cargo reconstitution occurs within the tumor cells as early as day 4 and cargo product persists at least until day 7 post bacteria administration (data not shown). This experiment was conducted in immunocompetent mice.

Figure 16 shows bacterial accumulation in tumor and organs at days 1, 3 and 7 post administration. Result from IT injection of the diaminopimelic acid (dap) prototroph strain are presented. Bacteria numbers decrease with time and are cleared from most organs at day 7 p.a. This experiment was conducted in immunocompetent mice.

Figure 17 shows bacterial accumulation in tumor and organs at days 1, 3 and 4 post administration. The result from IT injection of the dap auxotroph strain is presented. Bacteria numbers decrease with time and are cleared from most organs at day 4 p.a. This experiment was conducted in immunocompetent mice.

Figure 18 A shows tumor volume of all individual animals which received intratumoral HSV-1 treatment. Complete tumor regression was observed in one animal, and a tumor reduction of 60% in another animal. Figure 18 B shows tumor volume of all individual animals which received intratumoral empty bacteria. Treatment with empty E. coli has an effect on the tumor kinetics as well, showing response in one animal and delayed tumor growth in another animal. Figure 18 C shows tumor volume of all individual animals which remained untreated. As opposed to E. coli comprising an FISV-1 BAC and empty E. coli treatment these tumors grew much faster reaching oversize by day 50 in the longest surviving animal.

Figure 19 shows the average tumor volume of all three groups of animals (E. coli comprising an FISV-1 BAC, empty E. coli and untreated) in the day of sacrifice. Treatment with E. coli comprising an FISV-1 BAC shows much broder tumor volume spread, suggesting some animals had much smaller tumors than those in two other groups.

Figure 20 shows improved survival of animals receiving E. coli comprising an FISV-1 BAC as treatment. Treatment with empty E. coli increases median survival of animals as well. In the absence of E. coli cargo the effect seems not to be as long lasting as when FISV is present (effect visible till day 65 post treatment initiation).

Figure 21 shows body weight measurements of all three groups. Intratumoral treatment of PancOl tumors does not result in body weight changes of the animals. The general health status of the animals also points to the safety of the treatment (E. coli comprising an HSV-1 BAC and respective empty E. coli vehicle).

Figure 22 shows CT26 tumors from syngeneic Balb/C mice following triple IT injection of 10⁷ bacteria, up to day 13 post first application. Animals which received E. coli comprising an HSV-1 BAC where further divided to responders and non-responders to therapy and compared to the control group, which received PBS instead of bacteria. Animals were culled on day 13 due to scheduled sacrifice.

Figure 23 depicts CT26 tumors from syngeneic Balb/C mice following triple IT injection of 10⁷ bacteria, up to day 20 post first application. Animals receiving E. coli comprising an HSV-1 BAC were divided into responders to therapy and non-responders, and presented together with the group that received PBS only. In addition, two responding animals were plotted individually (mouse 14 and mouse 30), showing a clear difference in tumor growth in comparison to non-responders and control group. Complete tumor regression was observed in one animal (Mouse 14), and a tumor growth delay for 10 days in another one (Mouse 30). In total, 43% of the animals responded to the HSV-1 treatment.

DETAILED DESCRIPTION

We describe herein an innovative bacterium which combines the use of live bacteria and viruses as encoded into the said bacterial genome or maintained episomally in the bacterium. While in principle any virus could be delivered to any tissue for use in therapy or prophylaxis of a disease, the current focus is on the delivery of an oncolytic virus to cancer cells of solid tumors in humans.

In this scenario, the bacteria specifically deliver the genetic material (DNA or RNA) of a virus to cancer cells through bactofection, followed by the reconstitution of infectious viral particles inside the mammalian cells. Additionally, the bacterium and/or the virus may be further armed with other components as described below. The release of the viral genetic material and reconstitution of infectious viruses in this scenario occurs after the invasion of the targeted cancer cells, which are then destroyed due to the direct oncolytic activity of the viruses in addition to the potential activation of the immune system. This is particularly advantageous in solid tumors, where the deeper inner regions are difficult to reach. One particular focus of the present invention is therefore on the treatment of solid tumors, including but not limited to pancreatic cancer, with the bacterium of the invention.

In example 1 below, it is shown that the engineered bacteria (£. coli) can invade mammalian cells successfully, as seen in Figures 1 and 2. It was found that the number of internalized E. coli bacteria increases over time, and with the increasing bacterial multiplicity of infection (MOI). Figures 2 and 3 illustrate the average number of bacteria counted per cell in Experiment 1 and an example of immunostaining thereto.

Further, experiments described in example 2 below revealed that engineered E. coli cells were able to deliver cargo (in this case EGFP as a plasmid DNA) into cultured mammalian cells (FleLa cells). Specifically, in these experiments, E. coli was engineered to carry EGFP encoded by plasmid DNA. This was demonstrated via immunostaining and the detection of the EGFP signal in the FleLa cells after bactofection. As shown in Figure 3, the bacteria are still intact at 30 minutes after cell invasion (Figure 3 A), whereas at 6h post cell invasion, the bacteria have self-lysed and disintegrated (Figure 3 B). Once inside the phagosome vacuoles, the bacteria undergo self-lysis as mediated by the engineered bacteriocin release protein, bacteriophage lambda lysozyme, holin and antiholin, and further aided by diaminopimelic acid auxotrophy. Next, the listeriolysin O (and optionally, phospholipase C) released from the lysed bacteria cause the vacuoles to lyse, thereby releasing the cargo, which in this case is a plasmid DNA that is then transcribed and expressed by the host cell. The expression of E. co//-delivered plasmid DNA carrying EGFP could be demonstrated and is shown in Figure 4, where the EGFP signal detected in FleLa cells after bactofection. This experiment confirmed that delivery of DNA into tumor cells using the engineered E. coli bacteria was successful.

In a next step, it was tested whether or not engineered E. coli is capable of delivering DNA encoding an oncolytic virus. As can be seen from example 3, E. coli was engineered to carry a BAC encoding FISV-1 and to deliver the DNA into the nucleus of cultured Vero cells. As in Figure 5, more than 98% of the Vero cells show expression of an early FISV-1 viral antigen upon bactofection (MOI 20). This expression occurs at early time points, suggesting that it is a result of direct DNA delivery to the nucleus, and not as a result of the virus-reinfection process. Figure 5 depicts Vero cells stained for DAPI and the viral antigens at 24, 48 and 78 h after bactofection. The viral antigens were detected in the nucleus and the signal intensity for viral antigens increased over time. This indicates that the engineered bacteria are capable of delivering nucleotide cargo that subsequently is expressed by the mammalian cells.

In the next experiment, the same E. coli strain loaded with BAC encoding full virus DNA results in the production of infectious viral particles. As described in example 4, engineered E. coli carrying full FISV-1 virus DNA was used for bactofection, and the viral production was documented after 2, 3 and 7 days. Pictures of Vero cells after bactofection with E. coli which delivered FISV-1 virus were taken and the results for different cell lines were summarized. Specifically, Figure 6 depicts that viral reconstitution increases over time and with increasing bacterial MOI. Figure 7 shows Vero cells after bactofection and in situ reconstitution of FISV-1 virus. The table in Figure 8 summarizes the permissiveness of all tested cell lines with regard to bactofection and virus reconstitution tested so far in the current invention.

In order to assess the behavior of the bacteria in real patient situation, the stability of the engineered E. coli of this invention in whole blood was tested as shown in example 5. Specifically, human blood was mixed with the engineered E. coli and bacterial survival was measured at different time points by counting colonies. Three different bacterial strains were tested here, strain 1 is a dap auxotroph and an invasive strain (the invasiveness being mediated by the Yersinia invasin) with the following capabilities: lysis of the phagosome through insertion of listeriolysin O (hly), specific modifications to allow stable cargo maintenance via the deletion of genes encoding EndA, RecA and FHsdR, reduced immunogenicity via the deletion of IpxM gene, and the dap auxotrophy (dap stands for diaminopimelic acid, an amino acid component of the bacterial cell wall) which renders its replication-incompetent (or competent for a few cycles only) and is aiding in the self-lysis of bacteria (due to the impaired cell wall synthesis). Strain 2 is dap prototroph and also invasive (the invasiveness being mediated by the Yersinia invasin) with the following capabilities: lysis of the phagosome through insertion of listeriolysin O (hly), specific modifications to allow stable cargo maintenance via the deletion of EndA, RecA and HsdR genes, reduced immunogenicity via the deletion of IpxM, but being a prototroph, this strain is able to synthesize dap that is required for its growth. It was found that engineered E. coli are cleared from the blood of both donors, marked by a roughly 4 log decrease over the span of 1 hour, as shown for two different strains (strain 1 the dap auxotroph, and strain 2 the dap prototroph). Despite the fast clearance profile, sufficient accumulation of bacteria at target site was observed as shown in further examples. As shown in Figure 9, E. coli concentrations are reduced in human blood of donor 1 and donor 2 within 60 minutes from 6.5 loglO CFU/mL to 3.4-3.6 loglO CFU/mL for wild type E. coli and to roughly 1.9-2.3 loglO CFU/mL for the engineered bacterial strains 1 and 2.

Finally, the engineered E. coli strains were tested in immunocompetent mice showing that engineered bacteria efficiently target tumors and reach high intra-tumoral titers as described in Example 6. Flere, E. coli bacterial accumulation in tumor and different organs of mice after the administration were measured by tissue homogenization and bacteria recovery on agar plates as well as by detecting a luciferase (Lux) signal coming from the bacteria.

Tumors were induced in immunocompetent mice using colon carcinoma cells (CT26) following standard laboratory procedures and the engineered bacteria were injected as described in Example 6. Figure 11 shows the mouse injected with the bacteria. As shown in Figure 10, bacterial accumulation specifically in the tumor was observed. Engineered E. coli were recovered from CT26 tumors. The bacteria were found to accumulate in the tumor and reach titers of 10^s at 2 days p.a. with low off-target values. Furthermore, the number of E. coli bacteria in the tumor exhibited a dose response in accordance with the initial MOI (administered IV) as observed at 2 days p.a. albeit near saturation at higher MOI (Figure 12). The titers remained stable for at least 8 days. Besides, Figure 13 displays bacterial accumulation in tumors and organs at 2 hours and 72 hours post IV injection. Results from dap prototroph strain are presented. Bacteria are evenly distributed in the tumor and organs at 2h p.a. At a later time point (72h p.a.), the bacteria titers increase in the tumor but decrease in most organs.

Figure 14 compares bacterial accumulation in the tumor versus the organs at 8 days post administration in immunocompetent mice. Results from dap prototroph strain upon IV (intravenous) vs IT (intratumoral) injection are presented. Both administration routes, IT and IV, led to tumor accumulation. Lower off-target effect as shown in the lower titers were observed upon intratumoral administration. In further experiments as shown in Figure 15, it was found that cargo delivery (a CMV-driven firefly luciferase plasmid in this case) to tumor cells can be achieved at day 4 post administration of the bacteria The bacteria invade tumor cells very quickly after being systemically applied. A reporter gene expression can be detected upon delivery by engineered E. coli.

Figure 16 illustrates bacterial accumulation in tumor and organs at days 1, 3 and 7 post administration in immunocompetent mice. Results from IT injection of dap prototroph strain are presented. Bacterial numbers decrease with time and are cleared from most organs at day 7 p.a., in contrast to the stable titer observed in the tumor.

Figure 17 displays bacterial accumulation in tumor and organs at days 1, 3 and 4 post administration in immunocompetent mice. Results from IT injection of dap auxotroph strain are presented. Bacterial numbers decrease with time and are cleared from most organs by day 4 p.a., in contrast to the stable titer observed in the tumor.

Figures 18-21 and example 7 show cancer treatment in mice induced with pancreatic cancer cells and treated with empty E. coli bacteria and with an E. coli comprising the oncolytic virus FISV-1 as defined herein. In detail, Figure 18 A shows complete tumor regression in one animal, and a tumor reduction of 60% in another animal when treated with E. coli comprising an FISV-1 BAC. In addition, Figure 18 B shows that treatment with an empty E. coli has an effect on the tumor kinetics. Furthermore, Figure 18 C shows tumor volume of all untreated individual animals. As opposed to E. coli comprising an FISV-1 BAC and empty E. coli treatment, the tumors in untreated animals grew much faster reaching oversize by day 50 in the longest surviving animal. In Figure 19, the average tumor volume of all three groups of animals (E. coli comprising an FISV-1 BAC, empty E. coli and untreated) on the day of sacrifice can be seen. Animals treated with E. coli comprising a FISV-1 BAC show much smaller tumor volumes. Figure 20 indicates that the treatment with empty E. coli increases the median survival of animals as well, but not as effectively as treatment with E. coli comprising an FISV-1 BAC. In the absence of E. coli cargo, the effect seems not to be as long lasting as when FISV-1 is present (effect visible till day 65 post treatment initiation). The body weight measurements of all three groups can be seen in Figure 21. Intratumoral treatment of pancreatic tumors in mice does not result in body weight changes of the animals. The general health status of the animals also points to the safety of the treatment (E. coli comprising an FISV-1 BAC and respective empty E. coli vehicle). To further investigate the effect of the treatment with E.coli carrying HSV-1, mice were induced with colon cancer cells as described in Example 8, Specifically, the inventors compared responding and non-responding animals (Figure 22 and 23 and example 8). Figure 22 depicts CT26 tumors from syngeneic Balb/C mice following triple IT injection of 10⁷ bacteria, up to day 13 post first application. Animals which received E. coli comprising an FISV-1 BAC were further divided to responders and non responders to therapy and compared to the control group, which received PBS instead of bacteria. Animals were sacrificed on day 13. Finally, Figure 23 shows that animals receiving E. coli comprising an FISV-1 BAC were divided into responders to therapy and non-responders, and presented together with the group that received PBS only. Two high responders were plotted individually (mouse 14 and mouse 30), showing a clear difference in tumor growth in comparison to non-responders and the control group. Complete tumor regression was observed in one animal (Mouse 14), and a tumor growth delay for 10 days in another one (Mouse 30). In total, 43% of the animals responded to the E. coli comprising an FISV-1 BAC treatment.

From these experiments, it is clear that the engineered bacteria of the invention successfully invade mammalian cells, accumulate in the tumor tissues at much higher titer compared with the healthy organs, and the bacterial titer remains high in the tumors but decreases in other organs over time. In addition, the bacteria successfully deliver viral DNA or RNA to the target mammalian cells where the virus could then be reconstituted. In two mouse models of cancer, it was shown that the bacteria could deliver an oncolytic virus to tumor cells which led to reduction of the tumor size.

As discussed above, the present invention, therefore, relates to a bacterium which can be used as a bacterial delivery system. In the context of the present invention, the "bacterial delivery system" is defined as a live bacterium that has been genetically modified by integration of at least one exogenous polynucleotide into the bacterial genome or maintained episomally. The term bacterium and bacteria are used interchangeably and in the context of the invention encompass bacteria specifically engineered to release the genetic material (DNA or RNA) encoding a virus into targeted eukaryotic cells. In this regard, the term "a" bacterium of course also covers a bacterial cell line. Different types of bacteria can serve as vehicle in this delivery system as discussed below. The design of the bacteria may include refined cell targeting mechanisms and genetic modifications to increase invasiveness or to increase cargo stability. Further, the bacteria may be genetically engineered to increase the safety of administration and reduce immunogenicity.

The term "live" bacteria refers to bacteria that have an active metabolism and are usually able to replicate depending on whether the bacterium is prototroph or auxotroph. In this regard, a live bacterium of the invention may be either a wild type or engineered prototroph or an auxotroph. A prototroph is an organism able to synthesize all the compounds needed for its growth. Prototrophic bacteria do not have specific requirements in nutrients (e.g. amino acids and nucleotides) for their growth and division. In contrast, an auxotroph is an organism unable to synthesize a particular organic compound (for example, diaminopimelic acid (dap)) required for its growth. For proper growth (and division) of an auxotrophic bacterium, said compound needs to be provided.

The live bacterium used in the delivery system of the invention can be selected from the group consisting of Enterobacteriaceae spp., Lactobacillus spp., Yersinia spp., Escherichia spp., Klebsiella spp., Bordetella spp., Neisseria spp., Aeromonas spp., Francisella spp., Corynebacterium spp., Citrobacter spp., Chlamydia spp., Hemophilus spp., Brucella spp., Mycobacterium spp., Legionella spp., Rhodococcus spp., Pseudomonas spp., Helicobacter spp., Salmonella spp., Vibrio spp., Bacillus spp., Leishmania spp. and Erysipelothrix spp., Shigella spp., Listeria spp., Rickettsia spp., Acetoanaerobium spp., Aerococcaceae spp., Carnobacteriaceae spp., Enterococcace spp., Leuconostocaceae spp., Streptococcaceae spp., and bacteria that are generally regarded as safe (GRAS status) by the US Food and Drug administration (FDA). The list of bacteria with GRAS status is constantly updated. In a preferred aspect, the bacterium is Listeria monocytogenes, Salmonella typhimurium, Salmonella typhi, or Escherichia coli.

The term "live bacterium" also refers to genetically engineered bacteria of the species discussed above, with distinct features which can vary from the bacterial strain they originated from.

In order to allow for the transfer of the viral DNA or RNA into the eukaryotic cell, the bacterium must be capable of transferring the viral genetic material into the cytoplasm or the nucleus of a eukaryotic cell. Methods for DNA or RNA transfer are well-known. Invasive bacteria strains, such as Listeria, Shigella, and Salmonella can invade the host cells naturally.

This property can be transferred to other bacteria such as E. coli through the transfer of genes encoding one or more invasion factor(s), preferably an invasin (inv) derived from Yersinia spp.. Thus, in one embodiment of the invention, the bacterium of the invention comprises an invasion factor. Other accessory elements can be included in the bacterium to facilitate transfer of the cargo into the eukaryotic cell. For example, the bacterium can include a bacteriocin release protein (BRP) and bacteriophage lambda lysozyme, holin and antiholin to enable the said bacterium to self lyse upon entering the eukaryotic host cell. Specifically the BRP protein included can be mutated as known in the art and disclosed in van der Wal et al., (1998). The bacterium can further comprise a polypeptide encoding a protein capable of a decomposing the eukaryotic phagosomal membranes such as phospholipase, preferably phospholipase C (pic), preferably derived from Clostridium spp. or Listeria spp. Another option is a polypeptide encoding a protein capable of lysing the eukaryotic phagosomal membranes, preferably listeriolysin O (hly) derived from Listeria spp. or perfringolysin (pfo) derived from Clostridium spp. Furthermore, the bacterium can comprise a promoter driving the expression of these genes and controlled by an environmental stimulus, such as low magnesium, hypoxia, or bacterial quorum sensing.

As used in the present invention the term "invasin" describes a class of proteins associated with the penetration of pathogens into host cells. Invasins play a role in promoting entry during the initial stage of infection. The inv originated from Yesinia pseudotuberculosis binds to integrins on cell surface such as betal-integrin, thus triggering the inv-mediated bacteria uptake. In one scenario the invasion gene may be combined with a constitutive promoter or an inducible promoter, for example through a hypoxia-driven promoter as defined below.

The term "bacteriocin release protein" used in the context of the present invention refers to a small lipoprotein of 28 amino acids required for the release of the bacteriocin colicin DF13 into the extracellular medium of Escherichia coli cultures. As a result of BRP expression, periplasmic proteins are released. Specifically the BRP protein included can be mutated as known in the art and disclosed in van der Wal et al., (1998).

In the present invention, the term "holin" describes a diverse group of small proteins produced by dsDNA bacteriophages in order to trigger and control the degradation of the host's cell wall at the end of the lytic cycle. Holins form pores in the host's cell membrane, allowing lysins to reach and degrade peptidoglycan, a component of bacterial cell walls.

The term "antiholin" as used in the present invention describes a protein that interacts with and inhibits holin, thereby delaying the host cell lysis timing. Lysis inhibition is imposed when a cell infected by a T4-like virus is superinfected by new incoming viruses.

In the context of the present invention the term "phospholipase" and "phospholipase C" refers to enzymes that hydrolyze phospholipids to fatty acids and other lipophilic substances. Phospholipase C cleaves before the phosphate, releasing diacylglycerol and a phosphate-containing head group. In one scenario of the present invention, phospholipase C from Clostridium perfrigens is used to enhance lysis of the phagosome membrane.

As used herein the term "listeriolysin O (hly)" and "perfringolysin (pfo)" describes proteins capable of destroying cell membranes. In the context of the present invention these hemolysins are preferably isolated from Listeria spp. and/or Clostridium spp., and capable of lysing the eukaryotic phagosome thus releasing the cargo of the bacterium into the targeted cell.

As used within the present invention, the term "promoter" means a nucleotide sequence, which initiates and regulates transcription of a polynucleotide. An inducible promoter is a nucleotide sequence, wherein expression of a genetic sequence operably linked to the promoter is controlled by an analyte, co-factor, regulatory protein, etc. It is intended that the term "promoter" or "control element" includes full-length promoter regions and functional (e.g., controls transcription or translation) segments of these regions. In order to allow for transcription of the exogenous polynucleotide and production of the exogenous RNA in the bacterium, the exogenous polynucleotide is operably linked to a promoter that is functional in the bacterium. Promoters for gene expression in bacteria are well-known.

The bacterium of the present invention may be a prototroph or an auxotroph. In one aspect, the bacterium is an auxotroph, preferably the bacterium is a diaminopimelic acid (dap) auxotroph.

When used in the present invention the term "prototroph" or "auxotroph" describes the abilities of bacteria to live in a self-sustaining or dependent manner. It refers to a condition in which the bacteria are capable of synthesizing all organic nutrition factors needed on their own (prototroph) or are dependent on the uptake of certain nutrition factors from outside (auxotroph) as defined above.

The bacteria of the present invention may additionally have one or more of the endogenous bacterial endonucleases, DNA repair proteins, and/or metabolic proteins which are not functional. Specifically, the endogenous bacterial endonucleases can be EndA and/or HsdR. In this context, the endogenous bacterial DNA repair protein can be RecA. Furthermore, the endogenous metabolic proteins can be thiamine phosphate synthase (ThiE) and/or thiamine thiazole synthase (ThiA), and/or a component of the Ton system such as TonB receptor.

The term "endonucleases" used in the context of the present invention means enzymes that cleave the phosphodiester bond within a polynucleotide chain. Some, such as deoxyribonuclease I, cut DNA relatively nonspecifically (without regard to sequence), while many, typically called restriction endonucleases or restriction enzymes cleave only at very specific nucleotide sequences. The endonuclease EndA cleaves within dsDNA while HsdR is a restriction endonuclease and part of the EcoKI restriction-modification system. HsdR cleaves foreign, non-methylated DNA modifies hemimethylated DNA.

In the context of the present invention the term "RecA" refers to the endogenous DNA repair protein of bacteria used in the delivery system of the present invention. In E. coli, homologous recombination events mediated by RecA can occur during the period after DNA replication when sister loci remain close. RecA can also mediate homology pairing, homologous recombination and DNA break repair between distant sister-loci that had segregated to opposite halves of the E. coli cell. To stabilize the transported material bacterial endonucleases and DNA repair proteins may be downregulated, functionally disabled or deleted. Methods to achieve such nonfunctional protein products are known. Non-limiting examples of bacterial endonucleases are EndA and HsdR while RecA is an example of DNA repair proteins. Precautions might stem from engineered bacteria with nonfunctional thiamine synthase leading to an external dependency of thiamine (vitamin Bl) which is essential for bacterial growth. Thus, bacteria lacking this enzyme are growth arrested. A similar approach is leading to a growth disadvantage through the deletion of tonB that encodes a crucial subunit of the iron-siderophore and vitamin B12 acquisition system in E. coli (Ton). Without the Ton- system, the modified bacteria are restricted further without the administration of the above- mentioned micronutrients.

In another embodiment of the present invention the bacterium is engineered to avoid or minimize clearance by the host immune system using mechanisms such as the deletion of IpxM gene to express truncated and less virulent versions of surface lipopolysaccharides (LPS), or the coating of the bacterial surface with nanoparticles, such as polyethylene glycol (PEG) nanoparticles.

In the present invention, the term "surface lipopolysaccharides (LPS)," means peptides bound to the outer membrane of bacteria. Membrane-associated proteins play a key role in bacterial physiology and pathogenesis and are the major targets for vaccine development.

The term "polyethylene glycol (PEG) nanoparticles" as used in the present invention describes the PEGylation of the bacterial surface membrane. This means a process of covalent and non-covalent attachment or amalgamation of polyethylene glycol (PEG) polymer chains to molecules and macrostructures. In a specific example of the invention, the bacterium is an E. coli which is either a dap auxotroph or a prototroph, with the following further features. The bacterium comprises invasin (inv) derived from Yersinia spp., and listeriolysin O (hly) derived from Listeria spp. In addition, the bacterium has been engineered so that the endogenous bacterial endonucleases EndA and HsdR and the endogenous bacterial DNA repair protein RecA are not functional. Furthermore, the IpxM gene has been deleted to avoid or minimize the bacterium being prematurely cleared by the host immune response before reaching the target site.

E. coli strain may be a common cloning strain as for example E. cloni(r) 10G (Lucigen) described as F- (not carrying a plasmid), mcrA (Mutation eliminating restriction of DNA methylated at the sequence CmCGG (possibly mCG), mrr (Mutation eliminating restriction of DNA methylated at the sequence CmAG or GmAC) A(mrr-hsdRMS-mcrBC), (D80dlacZAM15, AlacX74, recAl, deoR (regulatory gene that allows constitutive expression of deoxyribose synthesis genes; permits uptake of large plasmids.), A(ara,leu)7697, araD139, galU, galK, nupG, rpsL, and l-. This E. coli strain preferably comprises DlpxM (a deleted IpxM gene), inv::endAl-FRT (endonuclease A gene has been deleted), hly::tonB (listeriolysin O and tonB genes have been deleted), plc::dapA-FRT (dap auxothroph), and Lysis2.1::dapB-FRT). This E. coli may carry an oncolytic virus, preferably a modified FISV-1 BAC of strain F (AICP34.5, AICP47). Another modified strain of FISV-1 was approved by the FDA in 2015 for treatment of melanoma. In phase I, I/ll and III melanoma, breast, head and neck, gastrointestinal cancer: induction of local and systemic tumour specific T cell responses, decreased regulatory T cells; prolonged progression-free survival, tumour-specific responses and complete remission.

In addition, the bacterium comprises at least one exogenous polynucleotide integrated into its genome or maintained episomally that encodes a virus or part thereof. The term "exogenous polynucleotide" as referred to in the present invention means a DNA molecule which is naturally not present in the living bacteria used in the bacterium of the present invention. The polynucleotide may be viral DNA. It may encode subunits of viral particles or fully functional viruses. These exogenous polynucleotides may be integrated into the bacterial genome or maintained episomally.

As used in the present invention the term "integrated into the bacterial genome" describes that the exogenous polynucleotide is inserted as a DNA sequence into the bacterial chromosome described in the present invention. Integrated means the circular chromosome is opened and the exogenous genetic material is fitted into the DNA. This is a process well-known in the art and can be established for example by using site-specific recombinases or recognition motifs for integrases in combination with enzymes capable of cutting DNA strands.

The term "maintained episomally" as used within the present invention refers to the viral DNA transported as a cosmid, plasmid, bacmid, close-ended double-stranded DNA (ceDNA), or as a double-stranded DNA within the cytoplasm of the bacteria used in the delivery system of the present invention.

In a preferred embodiment of the present invention, the bacterium of the invention targets cancer cells.

As used in the context of the present invention the term "virus" means a small infectious particle that replicates inside living cells of an organism. The term "oncolytic virus" as used in the present invention describes a type of virus capable of infecting and lysing cancer cells, but not normal cells. The viral replication destroys the cancer cell through oncolysis and induces the release of new viral particles which then preferably infect other cancer cells until the tumor shrinks or is destroyed. Various oncolytic viruses occur naturally or can be genetically engineered to become oncolytic. The oncolytic virus used in the present invention can be a DNA or an RNA virus. Once directed to the tumor the oncolytic virus may be able to replicate within the cancer tissue.

The viruses described in the present invention can be (1) a double-stranded DNA virus, such as human herpes virus, poxvirus, and adenovirus; (2) a single-stranded DNA virus, such as parvovirus; (3) a single stranded negative-sense RNA virus, such as Mononegavirales, influenza viruses and Newcastle Disease Virus; or (4) a single-stranded positive-sense RNA virus, such as Flaviviruses, alphaviruses and picornaviruses. Preferably, the viral DNA is integrated within the DNA of the bacterium but can however, also be maintained episomally.

The viruses comprised by the present invention are selected from the group consisting of herpes simplex virus, HSV-1, Adenovirus, Adeno-associated virus (AAV), Adenovirus 3, Adenovirus 5, Parvovirus, Parvovirus H1PV, Mononegavirales, Ebola virus, Measles virus, Nipah virus, Influenza virus, Newcastle Disease Virus, Poxvirus, Alphaviruses, Picornaviruses, Coxsackie virus, Flaviviruses, Zika virus, Vaccinia virus, Vesicular Stomatitis virus, Poliovirus, Reovirus, Senecavirus, Echovirus, Semliki Forest virus and Maraba virus , and a chimeric virus, but are not limited to these examples. In the context of the present invention, the virus may or may not be engineered to specifically target cancer cell. To achieve selectiveness for cancer cells, the virus may be modified. Different viral modifications are known to target, or refine targeting of cancer cells. Possible examples include, but are not limited to modifications of viral coat proteins, use of tumor-specific promoters, and recognition of tumor surface-proteins or metabolic products of cancer tissue. However, the system of the present invention could grant additional cell specificity by targeting the vehicle bacteria to the tumor.

In a specific example of the present invention, the bacterium may be Escherichia coli and the virus may be HSV-1, adenovirus or parvovirus.

Thus, the combination of the abovementioned E. coli (F- mcrA A(mrr-hsdRMS-mcrBC) endAl recAl (D80dlacZAM15 AlacX74 araD139 A(ara,leu)7697 galU galK rpsL nupG l- tonA) with DlpxM, inv::endAl-FRT, hly::tonB, plc::dapA-FRT, and Lysis2.1::dapB-FRT may further comprises a HSV-1 BAC of strain 7 (AICP34.5, AICP47) and may be used to treat any cancer as defined herein. Preferably the cancer is colon or pancreatic cancer.

In another aspect, the bacterium may be engineered to deliver further elements. As defined above further elements can be the bacterium is further engineered to comprise an exogenous polynucleotide encoding a tumor antigen, pro-apoptotic factor, immune checkpoint inhibitor, antibody, nanobody, enzyme, antiangiogenic factor, other immune modulators and/or other anticancer molecules integrated into the bacterial and/or the viral genome.

The above mentioned exogenous polynucleotide preferably encodes an antibody or nanobody against one or more of the following molecules: PD-1, PD-L1, PD-L2, CTLA-4, LAG3, Siglec-7, Siglec-9, Siglec-15, IDO, TIM-3, VISTA, A2AR, B7-H3, B7-H4, BTLA, KIR, NOX2, or molecules implicated in the CD47/SIRPa signaling axis.

In another embodiment, the bacterium of the invention may be used in therapy, for example in the prevention or treatment of mammalian disease, preferably human disease.

Specifically, in another embodiment, the bacterium of the invention could be used in the prevention or treatment of cancer. The bacterium can be used in a method for diagnosis, the prevention and/or treatment of the medical conditions described herein. As such, the term "treating" or "treatment" includes administration of the bacterium to a patient to modify the course of a disease.

Furthermore, the term "prevention" or "prophylaxis" as used interchangeably herein, refers to any medical or public health procedure whose purpose is to prevent a medical condition described herein. As used herein, the terms "prevent", "prevention" and "preventing" refer to the reduction in the risk of acquiring or developing a given condition. Also meant by "prevention" is the reduction or inhibition of the recurrence of a medical condition.

The bacterium of the present invention can be administered preferably to a human subject, or other species for veterinary treatment. The bacterium, methods and uses described herein are applicable to both human therapy and veterinary applications.

When the bacterium is used to prevent or treat cancer, the cancer can be selected from the group consisting of brain tumor, glioblastoma, pituitary tumor, colorectal cancer, ovarian cancer, prostate cancer, pancreatic cancer, lung cancer, gastric cancer, liver cancer, head-and-neck cancer, oropharyngeal cancer, hypopharyngeal cancer, laryngeal cancer, lip cancer, oral cavity cancer, salivary glands cancer), anal cancer, penile cancer, vaginal cancer, vulvar cancer, breast cancer, bladder cancer, kidney cancer, bile duct cancer, skin cancers, melanoma, central nervous system tumors, adrenal cancer, adrenocortical carcinoma, appendix cancer, rectal cancer, endometrial cancer, esophageal cancer, eye cancer, retinoblastoma), gallbladder cancer, heart cancer, small intestine cancer, teratoid tumor, testicular cancer, thymus cancer, thyroid cancer, urethral cancer, uterine cancer, Wilms tumor, mesothelioma, multiple myeloma, sarcomas, Kaposi's sarcoma, uterine sarcoma, osteosarcoma, Lymphomas, Burkitt lymphoma, Hodgkin lymphoma, and non-Hodgkin lymphoma.

Administration of the bacterium may be locally or systemically, optionally as oral, inhalation, intratumoral, intravenous, or subcutaneous administration.

In a preferred embodiment, the cancer is pancreatic cancer, and the administration is intravenous.

Suitable oral formulations may be in the form of a solid dosage form, suspension or granules for suspension, syrup, elixir, and the like. Pharmaceutically acceptable excipients, lubricants, and sweetening or flavoring agents may be included in the oral pharmaceutical compositions. If desired, conventional agents for modifying tastes and colors may also be included. For injectable formulations, a pharmaceutical composition may be in admixture with suitable excipients in a suitable vial or tube to form a composition suitable for injection.

In another embodiment, the bacterium is administered in combination with one or more additional therapies such as an anticancer treatment such as cytotoxic chemotherapies, monoclonal antibodies, cancer vaccines, cytokines, checkpoint inhibitors, immunomodulators, surgery or radiotherapy.

It must be noted that as used herein, the singular forms "a", "an", and "the", include plural references unless the context clearly indicates otherwise. Thus, for example, a reference to "a gene modification" includes one or more of such different modifications and reference to "the method" includes reference to equivalent steps and methods known to those of ordinary skill in the art that could be modified or substituted for the methods described herein.

Unless otherwise indicated, the term "at least" preceding a series of elements is to be understood to refer to every element in the series. Those skilled in the art will recognize or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the invention described herein. Such equivalents are intended to be encompassed by the present invention.

Throughout this specification and the claims which follow, unless the context requires otherwise, the word "comprise", and variations such as "comprises" and "comprising", 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 integer or step. When used herein the term "comprising" can be substituted with the term "containing" or sometimes when used herein with the term "having".

When used herein, "consisting of" excludes any element, step, or ingredient not specified in the claim element. EXAMPLES

Example 1: Engineered E. coli efficiently invade mammalian cells. Cultured HeLa cells were incubated with engineered E. coli and the percentage of invaded cells was quantified from immunostainings at different time points.

Material and Methods

In vitro bactofection assay protocol

On the day before the experiment, mammalian cells were seeded in 12-well plates using (1ml in each well, number of cells depend on the cell line, but usually 0.4xl0^s/ml works very well) and bacteria were grown in BHI medium (or other, such as 2YT). The following day, the experiment continues with these overnight bacterial cultures. The OD₆₀o values of bacteria cultures were measured and the number of bacteria/mL of culture was calculated. A bacteria-to-mammalian-cell-ratio (MOI) of 5, 20 and 50 bacteria/cell was calculated and if necessary the bacteria were diluted in DMEM medium. A appropriate number of bacteria was added to each experimental well, mixed gently and incubated for 2h (or different time, e.g. 0.5h, lh, 6h) in the C0₂ incubator. After 2h (or other) of incubation, the plates were washed 4 times with PBS (room temperature). After washing, 1 ml of mammalian culture media (such as DMEM) containing proper antibiotic (e.g. gentamicin) was added to each well and the plates were placed back in the incubator. Then 3 samples of each bacteria strain were collected on the desired days. The samples were centrifuged at 4000 rpm for 5 min to separate cell supernatant from the cell debris.

Immunostaining Protocol

The cells were washed with lx PBS and fixed 20 minutes at RT with Fixative Solution (44.1 mL DMEM w/o serum, 5.4 mL HCHO (37 %), 0.5 mL 1M Hepes pH 6.8) and washed with PBS four times, then the cells were permeabilized 3 minutes at RT with Permeabilization Buffer (PBS lx, 0.1 % Triton X-100, 0.5 % BSA) and washed again with lx PBS four times. Then the cells were blocked 10 minutes at RT with Blocking Buffer (BSA 0.5 % in lx PBS) and the first antibody was diluted in blocking buffer and samples were incubated for 1.5 h. For E. coli staining the Abeam antibody (ab25823) was used in 1:250 dilution, while for VP16 staining the Abeam antibody (abl37967) was used in 1:100 dilution. The cells were washed again with lx PBS four times before the second antibody was diluted in 1:500 in blocking buffer and incubated for 45 minutes at RT. The cells were washed with lx PBS, stained with DAPI, washed again with lx PBS four times and mounted on a slide using mounting solution.

Cells were visualized under the microscope. Pictures were taken, merged and analyzed using ImageJ software. Results

As can be seen from Figure 1, E. coli invades mammalian cells as early as a few minutes post administration. Coincubation of bacteria with mammalian cells for 2h results in 80-100 % invasion efficiency, depending on the bacterial MOI used. Bactofection efficiency seems to be MOI-dependent (the higher the MOI the more efficient the bactofection), but seems to be advantageous up to MOI 50. MOI higher than 50 seems to have deleterious effect on cells in in vitro culture. Figure 1 shows the percentage of FleLa cells invaded by E. coli for 30 minutes, 1 and 2 hours after bactofection. The number of internalized E. coli bacteria increases over time and with increasing MOI. Different cell lines vary in their susceptibility to bacterial invasion. Figures 2 and 3 illustrate the average number of bacteria counted per cell and an example of immunostaining thereto.

Example 2: Engineered E. coli deliver cargo in cultured mammalian cells

E. coli was engineered to carry EGFP as plasmid DNA. The cellular uptake of bacteria was demonstrated by immunostaining and the EGFP signal could be detected in the FleLa cells and Vero cells after bactofection.

Materials and Method Immunostaining Protocol

The cells were washed with lx PBS and fixed 20 minutes at RT with Fixative Solution (For 50 mL stock use: 44.1 mL DMEM w/o serum, 5.4 mL HCHO (37 %), 0.5 mL 1M Hepes pH 6.8). The cells were washed with PBS four times, then the cells were permeabilized 3 minutes at RT with Permeabilization Buffer (PBS lx, 0.1 % Triton X-100, 0.5 % BSA) and washed again with lx PBS four times. Then the cells were blocked 10 minutes at RT with Blocking Buffer (BSA 0.5 % in lx PBS (same as the permeabilization buffer w/o Triton) and the first antibody was diluted in blocking buffer for 1.5 h. For E. coli staining an Abeam antibody (abl37967) was used in 1:250 dilution, while for VP16 staining the Abeam antibody (abll0226) was used in 1:100 dilution. The cells were washed again with lx PBS four times before the second antibody was diluted in 1:500 in blocking buffer for 45 minutes at RT (do it on parafilm with 40-50 pL). The cells were washed with lx PBS, stained with DAPI, washed again with lx PBS four times and mounted on a slide, a drop of mounting solution was used.

Results

As can be seen from Figure 3, at 30 minutes after cell invasion bacteria are still intact (Figure 3 A), while at 6h after cell invasion bacteria are lysed and disintegrated (Figure 3 B). Figure 3 shows the uptake of E. coli into the cultured Vero cells. Cargo release from the vacuole occurs after bacterial lysis and is trafficked to the nucleus (E. coli is engineered to trigger this event). The expression of E. coli delivered plasmid DNA carrying EGFP could be demonstrated and is shown in cultured HeLa cells in figure 4, where the EGFP signal detected in FleLa cells after bactofection is shown.

Example 3: Engineered E. coli deliver viral antigens

E. coli was engineered to carry FISV-1 BAC and to deliver it into the nucleus of cultured Vero cells. An early viral antigen was stained.

Material and Method Immunostaining Protocol

Results

>98 % of Vero cells show expression of an early HSV-1 viral antigen upon bactofection (MOI 20). The expression occurs at early time points suggesting it is a result of direct DNA delivery to the nucleus (and not the virus-reinfection process). Figure 5 depicts Vero cells stained for DAPI and viral antigens 24, 48 and 78 h after bactofection. The viral antigens were detected in the nucleus, and an increasing signal for viral antigens 78 h after bactofection could be observed.

Example 4: E. coli loaded with full virus DNA cargo results in the production of infectious viral particles

Engineered E. coli carrying full HSV-1 virus DNA was used for bactofection, and the viral production was documented after 2, 3 and 7 days. Pictures of infected Vero cells after bactofection with E. coli which delivered HSV-1 virus were taken and the results of bactofection for different cell lines are summarized. Material and Methods

Plaque assay protocol (to determine viral titer as plaque-forming units per ml (pfu/ml))

On the day before the experiment cells were seeded in 6-well plates, 0.4x10^s cells/mL and 2mL per well. On the following day, the cell media was discarded and replaced with 900 pL of fresh, warm medium just before the experiment. Serial dilutions of the bactofection supernatant were prepared as follows: 900 pL PBS + 100 pL virus = 1/10 dilution, up to 10 ^s. Monolayers were infected with 100 pL of each dilution (pipetting dropwise) and the plate was mixed gently. The plate was incubated 2h at 37 °C and 5 % C0₂ and mixed from time to time. At the end of the incubation time, 4 % agarose (Gibco) was heated in the microwave. The agarose was mixed with warm plaquing media (DMEM + 2% FBS, no glucose) of 1:10 dilution to a final concentration of 0.4 %. After 2 h incubation time, the cell supernatant was discarded from the plate and gently overlaid with agar (37 °C). The plates were incubated at RT for 15-30 minutes. The plates were further incubated at 37 °C and 5 % C0₂ for 6 days.

After the plaquing period (6 days), 2 mL of fixing/staining solution (0.5 g Crystal Violet (0,05 % w/v), 27 ml 37 % Formaldehyde (1 %), 100 ml lOx PBS, 10 ml Methanol (1 %), 863 ml dFI20 to 1L) was pipetted into each well. The mixture was incubated under the hood for lh. The agar and the solution were aspirated using the long plastic pipette tip and the plates dried in the hood for a few minutes. The plaques were counted and pfu/ml was calculated.

Results

Figure 6 depicts how the titers (pfu/mlxlO⁶) increased after 3 and 7 days from an original MOI 20 or 50 used, reflecting an increase in virus production over time. Figure 7 shows Vero cells after bactofection and in situ reconstitution of FISV-1 virus. The table in Figure 8 summarizes the permissiveness of all tested cell lines to bactofection and virus reconstitution.

Example 5: Stability of E. coli in whole blood

Fluman blood was contacted with the engineered E. coli and the bacteria survival rate was measured at different time points by counting colonies.

Material and Method Whole blood killing assay

The bacteria were grown overnight in LB medium and diluted to lxlO⁷ in a way that this amount did not exceed 5pL. 200 pL of blood was added into 96-well round-bottom plates and the bacteria were added to the blood and incubated under constant shaking for the desired amount of time (e.g. 5, 10, 20, 30 and 60 min) at 37 °C. At time point "0" samples were taken to determine whether a proper number of bacteria were added and the samples were seeded with antibiotic selection on agar plates in 3 serial dilutions (10-fold). The following day all colonies were counted.

Blood collection was drawn by a qualified professional in the medical clinic and samples were transported with lithium heparin as an anticoagulatory agent. After delivery, blood was incubated on a weaving platform for 15 min to mix well and adjust to room temperature. Donors were tested for infectious diseases (HIV, FIBV, FICV) prior to blood donation.

Strain 1 is dap auxotroph, invasive (insertion of inv gene), escape out of the phagosome through insertion of listeriolysin O (hly), has specific modifications to allow stable cargo maintenance (EndA, RacA, HsdR), has deletion of IpxM (in order to avoid clearance by the immune system), and possesses a dap auxotrophy (dap = diaminopimelic acid, amino acid, part of the cell wall of bacteria) which renders its replication-incompetent (or competent for a few cycles only) and is aiding in self-lysis of bacteria in the phagosome (due to the impaired cell wall synthesis).

Strain 2 is dap prototroph, invasive (insertion of inv gene), escape out of the phagosome through insertion of listeriolysin O (hly), has specific modifications to allow stable cargo maintenance (EndA, RecA, HsdR), has deletion of IpxM (in order to avoid clearance by the immune system), does NOT have dap auxotrophy (dap = diaminopimelic acid, amino acid, part of the cell wall of bacteria).

Readout

Colony count

Results

The titer of the engineered E. coli decreased in the blood by 4 log at 1 hour post incubation=, as shown for two different strains (strain 1 = dap auxotroph, strain 2 = dap prototroph) in the blood from both donors. Engineered bacteria showed rapid clearance from blood, suggesting their safety when injected intravenously.

As shown in Fig. 9 E. coli concentrations reduce in human blood of donor 1 and donor 2 within 60 minutes from 6 loglO CFU/mL to 4 loglO CFU/mL for wildtype E. coli, and to 2 loglO CFU/mL for the modified bacteria strains 1 and 2. Example 6: Engineered bacteria efficiently target tumors and reach high intra-tumoral numbers

E. coli bacteria accumulation in tumor and different organs after the administration were measured by bacteria recovery and detection of the signal from luciferase (Lux) expressed by the bacteria.

Material and Method Lux integration

Preparation of all bacterial growth media was carried out using deionized water and according to the manufacturer's specific instructions. All media was autoclaved at 121 °C for 15 minutes prior to use. All broths were stored at room temperature & agar in the fridge.

Initial overnight cultures (OVNC) of E. coli TOPIO glycerol stocks containing the pl6sLux plasmid non- integrated, were incubated at 30 °C in 30 ml of LB broth with Erythromycin at a concentration of 500 pg/ml. This OVNC was mini-prepped for the pl6sLux plasmid using New England Biolabs (NEB) Monarch Plasmid Miniprep Kit. The DNA concentration of the pl6sLux plasmid was measured using a Nanodrop spectrophotometer. Preparation of bacterial strains to be made electrocompetent was carried out as per protocol. Where strains were auxotrophic for diaminopimelic acid (dap), dap was added to the media at a concentration of 0.5 mM to all media. 40 pi of freshly prepared electrocompetent cells was combined with 150 ng of pl6sLux plasmid and left on ice for 15 minutes. Electroporation was then carried out using the following parameters 2.5 kV, 25 pF at 200 W. Transformed cells were allowed to recover in 950 pi of Super Optimal Broth (SOB) at 30 °C for 3 hours. This cell suspension was serially diluted 5 times at a 1:10 ratio with IOOmI plated on selective media plates with erythromycin at a concentration of 500 pg/ml for each dilution at 30 °C. After 24 hours of growth, plates were checked for colony growth and luminescence in the IVIS^* 100 in vivo imaging system from Perkin Elmer. Automatic settings for luminescence were selected with a maximum exposure time of 3 minutes. Positive luminescing colonies were marked on the outside of the plate using a laboratory pen and then cultured overnight at the integrating temperature of 42 °C with an erythromycin concentration of 500 pg/ml. This OVNC was serially diluted 5 times at a 1:10 ratio with 100 mI plated on selective media plates with erythromycin at a concentration of 500 pg/ml and kept at 42 °C. After 24 hours of growth plates were checked for colony growth and luminescence in the IVIS^* 100 in vivo imaging system from Perkin Elmer. Automatic settings for luminescence were selected with a maximum exposure time of 3 minutes. This process was continued until 100 % of the colonies growing on the plates showed a luminescent signal with the colony with strongest signal being chosen for OVNC. Integration of the pl6sLUX plasmid was confirmed by growing the transformed E. coli on plates without selective media whereby no colonies showed any loss of light signal. A growth curve for each strain transformed was carried out by measuring optical density (OD 600) every 20 minutes and plating serially diluted samples on selective plates with erythromycin at a concentration of 500 pg/ml. Plates were subsequently counted and growth curve extrapolated.

In vivo studies

4 week old Female Balb/c mice were ordered from Envigo U.K. with a 2 week lead time for delivery. The mice should be afforded an adaptation period of at least 7 days in the animal unit prior to tumor induction. Standard laboratory food and water provided ad libitum. Animals are caged in groups of five maximum with a minimum of 2 mice per cage. Mice in good condition, without fungal or other infections, weighing 16-22 g and of 6-8 weeks of age were included in experiments.

Tumor induction

The day before induction, the animals were divided into appropriate groups, and their flanks were shaved with a small beard trimmer. Stocks of CT26 tumor cells were taken from liquid nitrogen and washed in an appropriate media containing 10 % FCS. These cells were inoculated into a tissue culture flask and grown till 80 % confluency under normal cell culture conditions (37 °C, 5 % C0₂). For tumor induction, the cells were washed and spun down twice in serum-free DMEM (1,000 g for 5 minutes) and the minimum tumorigenic dose of 1 x 10⁵ cells was re-suspended in final volume of 200 pi per mouse of serum-free culture media. The viability of cells used for inoculation was greater than 95 % as determined by count using a Nucleocounter. The test tube containing the cells was kept in a glove in a pants pocket to keep at 37 °C. Pasteur pipettes were used to mix cells immediately before injection to ensure equal distribution of cells. For routine tumor induction, the minimum tumorigenic dose of cells suspended in 200 mI of serum-free culture medium was injected subcutaneously (s.c.) into the flank, of infection-free 6-8 week old mice using a 26-gauge syringe needle. More than 20 animals should not be induced at one time per induction session as cell clumping and variation in tumor growth can occur. Following tumor establishment, tumors were allowed to grow and develop and generally monitored twice weekly until they reached a size of 0.5cm x 0.5cm. Tumor volume calculated according to the formula V=(ab²) P/6, where a is the longest diameter of the tumor and b is the longest diameter perpendicular to the diameter a. Prior to bacterial inoculation all tumors were measured and mice weighed.

Escherichia coli growth conditions and injection into mice

E. coli strains transformed with the pl6sLUX integrated or transiently with a CMV-driven Firefly expression plasmid (used as cargo in Fig. 15) were taken from -80 °C stocks and streaked onto LB agar and left to grow overnight at 37 °C. Ampicillin at 100 pg ml ¹ was used to maintain the strains transformed with the Firefly plasmid and erythromycin at 500 pg ml ¹ for the integrated pl6sLUX plasmid. A single colony was then picked and inoculated into LB broth and placed at 37 °C where it was left shaking at 200 rpm overnight. Antibiotics were added where necessary. A 1 % dilution was inoculated into fresh LB (~OD₆₀o 0.05) and left grow till it reached early log phase (OD_¥o 0.6). Once it reached an OD₆₀o according to its growth curve corresponding to 10⁹ cfu mi l, 1 ml of the culture was centrifuged at 13,500 g for 1 minute and the supernatant was removed. The pellet was then re suspended in 1 ml PBS and spun once again at 13,500 g for 1 minute. This was repeated twice more. This was then diluted down to 10⁷ cfu ml ¹ or 10^s cfu ml ¹. 100 pi of this was then injected into the mouse tail vein giving a final concentration injected into the mice at 10^s cfu/100 mI or 10⁷ cfu/100 mI respectively. Mice were then monitored on a daily basis for weight and tumor dimensions.

In vivo imaging

In vivo bioluminescent imaging (BLI) of bacteria containing the pl6sLUX plasmid in tumors Animals will be anesthetized using a 2.5 % mixture of isoflurane with oxygen before being placed in the light-tight unit of the I VIS imaging system. Following image acquisition, mice will be returned to their cages with adequate food and water and allowed to recover. In vivo BLI imaging was performed using the I VIS 100 (Perkin Elmer). At defined time points post bacterial administration, animals were anesthetized using a 2.5 % mixture of isoflurane with oxygen, and whole-body image analysis was performed in the I VIS 100 system for up to 3 minutes at high sensitivity. The IVIS 100 allows for continuous delivery of isoflurane with oxygen so that the mice can remain knocked out for the duration of the imaging procedure. Within this unit the mice will be positioned so that their noses and mouths were fully covered by the tubes delivering gaseous anesthesia. Mice will then be subjected to non-invasive optical imaging with acquisition times ranging from 1 second to 3 minutes depending on the strength of the optical signal produced. In the case of mice which had been inoculated with E. coli transformed with pl6sLUX plasmid, no substrate was necessary for a luminescent signal to be produced. Animals were allowed to recover after in their box and will be awake within a few minutes. Regions of interest were identified and quantified using Living Image software (Perkin Elmer).

In vivo bioluminescent imaging (BLI) of bacteria containing the CMV-driven Firefly plasmid in tumors In vivo BLI imaging was performed using the IVIS 100 (Perkin Elmer). At defined time points post bacterial administration, animals were anesthetized using isoflurane, and whole-body image analysis was performed in the IVIS 100 system for up to 3 minutes at high sensitivity. The IVIS 100 allows for continuous delivery of isoflurane with oxygen so that the mice can remain knocked out for the duration of the imaging procedure. In the case of mice which had been inoculated with E. coli transformed with CMV driven Firefly, the substrate D-Luciferin is required for a light signal to be produced. Prior to imaging taking place, a fresh stock solution of Luciferin should be prepared at 15 mg/mL in Dulbecco's Phosphate-Buffered Saline (DPBS). Filter sterilized through a 0.2 pm filter. Inject 10 pL/g of body weight. Each mouse should receive 150 mg Luciferin/kg body weight (e.g. For a 10 g mouse, inject 100 pL to deliver 1.5 mg of Luciferin). Inject the Luciferin intra-peritoneal (i.p.) 10-15 minutes before imaging. Animals were allowed to recover after in their box and were awake within a few minutes. Regions of interest were identified and quantified using Living Image software (Perkin Elmer).

Bacterial recovery from tumors and organs

At defined time points, post-bacterial administration the mice were imaged for their final BLI and then the animals anesthetized using Ketamine (75 mg/kg) and Medetomidine (1 mg/kg). Complete anesthesia will be confirmed by pitching the mouse feet for lack of response (no pedal withdrawal reflex). The mouse was then held vertically and a cardiac puncture performed. The 26G needle was advanced in the notch just to the left of the animal's xiphoid. The needle was positioned parallel to the spine and placed just under the ribs. The heart is located approximately at the level of the elbow. The needle was placed, bevel up, into the chest, and the heart punctured. A slight back pressure was applied with the syringe. When the needle was in the heart, blood will flow into the syringe. More back pressure was applied as the blood fills the syringe. Once the required blood volume was taken, the needle was removed and the blood stored in the appropriate tube with 100 pi of Ethylenediaminetetraacetic acid (EDTA). This was a terminal procedure and the mouse was humanely euthanized by cervical dislocation. The tumor and various internal organs including liver, spleen, kidneys, lungs, heart, and brain were aseptically dissected in a class II laminar flow hood with subsequent recovery of viable bacteria (cfu). All tissue samples were weighed and then divided in half with one of half of all of the samples flash-frozen in liquid nitrogen and stored in the -80 °C freezer for later analysis. To enumerate the total number of bacteria the dissected tumors and organs were immediately placed in 0.5 ml of PBS. Tumors and organs were then homogenized by fine mincing with a scalpel followed by pushing through a 20 mm pore nylon filter in sterile PBS (supplemented with 0.5mM dap for dap- auxotrophic strains). Serial dilutions were plated in triplicate on selective agar; E. coli with integrated pl6sLUX had plates with an erythromycin concentration of 500 pg/ml. E. coli transiently transformed with Firefly had plates with 100 pg/ml of ampicillin. Resulting colonies were used to calculate the number of bacterial cells per tissue sample. Again dap auxotrophic strains had all growth media supplemented with dap at 0.5 mM. Readout

In vivo bioluminescent imaging (BLI) of bacteria containing the pl6sLUX plasmid in tumors In vivo bioluminescent imaging (BLI) of bacteria containing the CMV driven Firefly plasmid in tumors

Bacterial recovery from tumors and organs

In Figures 10, 12, 13, 14, 16 and 17 bacteria were quantified using tissue homogenization and plating.

Fig 15 shows the delivery of luminescent cargo.

Results

Figure 10 and 11 shows Bacteria accumulation in tumor and organs. Days 2, 5 and 7 post application. Engineered E. coli was recovered from CT26 tumors from syngeneic Balb/C mice following IV injection of 106 bacteria. The bacteria accumulate in the tumor and reach titers of lxlO⁸ at 2 days p.a. with lower off-target values.

Figure 12 depicts bacteria accumulation in the tumor at day 2 and 7 post IV injection. The number of E. coli bacteria in the tumor varied according to the initial titer (administered IV) as observed at 2 days p.a., with a tendency to level off at later time points. The titers remain stable for at least 8 days.

In addition, Figure 13 displays bacteria accumulation in tumors and organs at 2 hours and 72h post IV injection. Results from dap prototroph strain are presented. Bacteria are evenly distributed in the tumor and organs at 2h p.a. At a later time point (72h p.a.), bacteria numbers increase in tumors and decrease in most organs.

Figure 14 compares bacteria accumulation in tumor and organs at 8 days post administration. Results from the dap prototroph strain upon IV vs IT injection are presented. Both administration routes, IT and IV, led to tumor accumulation. Lower off-target titer values were observed upon intratumoral administration.

Figure 15 shows cargo delivery to tumor cells at day 4 post bacteria invasion. The bacteria invade tumor cells very quickly after being systemically applied. A reporter gene expression can be detected upon delivery by engineered E. coli. Cargo reconstitution occurs within the tumor cells as early as day 4 and cargo product persists until day 7 after bacterial invasion. Figure 16 illustrates bacteria accumulation in tumor and organs at days 1, 3 and 7 post administration. Results from IT injection of dap prototroph strain are presented. Bacteria numbers decrease with time and are cleared from most organs at day 7 p.a.

Figure 17 displays bacteria accumulation in tumor and organs at days 1, 3 and 4 post administration. Results from IT injection of the dap auxotroph strain are presented. Bacteria numbers decrease with time and are cleared from most organs at day 4 p.a.

Example 7: Bacterial delivery of oncolytic virus in pancreatic cancer model

Material and Methods In vivo studies

Six-eight-week old female Balb/c- Nude mice were afforded an adaptation period of at least 7 days in the animal unit prior to tumor induction. Animals were caged in groups of five maximum, and standard laboratory food and water provided ad libitum. Mice in good condition, without fungal or other infections, weighing >15 g and of 7-9 weeks of age were included in experiments.

Tumor induction

The day before induction, animals were divided into appropriate groups, and their flanks were shaved. Stocks of PANC01 cells were taken from liquid nitrogen and washed in appropriate media containing 10 % FBS + 2mL L-Glutamine. These cells were inoculated into a tissue culture flask and grown till 80 % confluency under normal cell culture conditions (37 °C, 5 % C02). For tumor induction, the cells were washed and spun down twice in serum-free DMEM (1,000 g for 5 minutes) and the minimum tumorigenic dose of 5 x 10^s cells was re-suspended in final volume of 200 pi per mouse of serum-free culture media. The viability of cells used for inoculation was greater than 95 % as determined by count using a Nucleocounter. The test tube containing the cells was kept in a glove in a pants pocket to keep at 37 °C. Pasteur pipettes were used to mix cells immediately before injection to ensure equal distribution of cells. For routine tumor induction, the minimum tumorigenic dose of cells suspended in 200 mI of serum-free culture medium was injected subcutaneously (s.c.) into the flank, of infection-free 7-9-week old mice using a 26-gauge syringe needle. Following tumor establishment, tumors were allowed to grow and develop, and generally monitored twice weekly until they reached a size of 80-250 mm³. Tumor volume were calculated according to the formula V=(ab2) P/6, where a is the longest diameter of the tumor and b is the longest diameter perpendicular to the diameter a. Prior to bacterial inoculation all tumors were measured and mice weighed. All animals will be monitored latest till day 97. Escherichia coli growth conditions and injection into mice

FISV-1 E. coli strain (F-, mcrA, A(mrr-hsdRMS-mcrBC), (D80dlacZAM15, AlacX74, recAl, deoR, A(ara,leu)7697, araD139, galU, galK, nupG, rpsL, l-, DlpxM, inv::endAl-FRT, hly::tonB, plc::dapA-FRT, Lysis2.1::dapB-FRT) carrying a modified FISV-1 BAC of strain F (AICP34.5, AICP47) (used in Fig. 18, Fig. 19 and Fig. 20) and empty bacteria (F-, mcrA, A(mrr-hsdRMS-mcrBC), O80dlacZAM15, AlacX74, recAl, deoR, A(ara,leu)7697, araD139, galU, galK, nupG, rpsL, l-, DlpxM, inv::endAl-FRT, hly::tonB, plc::dapA-FRT, Lysis2.1::dapB-FRT) were cultured in in 50mL vented conical tubes in 37°C with agitation using a plate mixer and set up as overnight culture of in 5mL medium. Chloramphenicol at 50 pg mL-1 , diaminopimelic acid at 0.5mg mL-1 and MgS0₄ at 20mM were used to maintain the strains. A single colony was then picked and inoculated into LB broth and placed at 37 °C where it was left shaking at 200 rpm overnight. Antibiotics (chloramphenicol) and supplements (MgS04) were added where necessary. Cells were harvested by centrifugation and resuspended in fresh medium and used to seed 5mL fresh medium at ~OD600=0.2. Cells were further cultured in 50mL vented conical tubes at 37^QC with agitation using a plate mixer to ~OD600= 1.0. Harvested by centrifugation (lOOOx g, 4°C, 5mins) washed 2x with PBS and resuspend in PBS+ 20mM MgS0₄ + 0.5mM 2,6- diaminopimelic acid and resuspended to desired dose concentration assuming OD600= 1.0 = 6x108 cells/mL. 100 pi of bacteria suspention was then injected into the mouse tumors giving a final concentration injected into the mice at 107 cfu/100 pL, on days 1, 4 and 7. Mice were then monitored on a daily basis for weight and tumor dimensions. Control group of animals received 100 mI of 107 cfu empty bacteria on days 1, 4 and 7 (Fig. 18, Fig. 19, Fig. 20 and Fig. 21) directly into the tumors. A group of animals was left untreated (Fig. 18, Fig. 19, Fig. 20 and Fig. 21).

Mice from all groups were monitored on a daily basis for weight and tumor dimensions.

Readout

Tumor volume measurements (daily) in Fig. 18 - 21

Animal survival in Fig. 20 (Kaplan-Meier Curves generated using GraphPad Prism 8 Software)

Animal body weight in Fig. 21

Result

Figure 18 A shows a complete tumor regression in one animal, and a tumor reduction of 60% in another animal. Figure 18 B shows that treatment with an empty E. coli has an effect on the tumor kinetics. Figure 18 C shows tumor volume of all untreated individual animals. As opposed to E. coli comprising an FISV-1 BAC and empty E. coli treatment, these tumors grew much faster reaching oversize by day 50 in the longest surviving animal. E. coli [HSV-1] refers to E. coli comprising an HSV-1 BAC.

Figure 19 depicts average tumor volume of all three groups of animals (£. coli comprising an HSV-1 BAC, empty E. coli and untreated) on the day of sacrifice. Animal treatment with E. coli comprising an HSV-1 BAC leads to much smaller tumor volumes. E. coli [HSV-1] refers to E. coli comprising an HSV-1 BAC.

Figure 20 shows the treatment with empty E. coli increases the median survival of animals as well. In the absence of E. coli cargo the effect seems not to be as long lasting as when oHSV is present (effect visible till day 65 post treatment initiation). E. coli [HSV-1] refers to E. coli comprising an HSV-1 BAC.

Figure 21 shows body weight measurements of all three groups. Intratumoral treatment of PancOl tumors does not result in body weight changes of the animals. The general health status of the animals also points to the safety of the treatment (HSV-1 and respective empty E. coli vehicle). E. coli [HSV-1] refers to E. coli comprising an HSV-1 BAC.

Example 8: Bacterial delivery of oncolytic virus in a colorectal cancer model

Material and Methods In vivo studies

Four-week old female Balb/c mice were afforded an adaptation period of at least 7 days in the animal unit prior to tumor induction. Animals were caged in groups of five maximum, and standard laboratory food and water provided ad libitum. Mice in good condition, without fungal or other infections, weighing 16-22 g and of 6-8 weeks of age were included in experiments.

Tumor induction

The day before induction, animals were divided into appropriate groups, and their flanks were shaved with a small beard trimmer. Stocks of CT26 tumor cells were taken from liquid nitrogen and washed in appropriate media containing 10 % FCS. These cells were inoculated into a tissue culture flask and grown till 80 % confluency under normal cell culture conditions (37 °C, 5 % C0₂). For tumor induction, cells were washed and spun down twice in serum-free DMEM (1,000 g for 5 minutes) and the minimum tumorigenic dose of 1 x 10⁵ cells was re-suspended in final volume of 200 pL per mouse of serum-free culture media. The viability of cells used for inoculation was greater than 95 % as determined by count using a Nucleocounter. The test tube containing the cells was kept in a glove in a pants pocket to keep at 37 °C. Pasteur pipettes were used to mix cells immediately before injection to ensure equal distribution of cells. For routine tumor induction, the minimum tumorigenic dose of cells suspended in 200 mI of serum-free culture medium was injected subcutaneously (s.c.) into the flank, of infection-free 6-8-week old mice using a 26-gauge syringe needle. Following tumor establishment, tumors were allowed to grow and develop and generally monitored twice weekly until they reached a size of 0.5cm x 0.5cm. Tumor volumes were calculated according to the formula V=(ab²) P/6, where a is the longest diameter of the tumor and b is the longest diameter perpendicular to the diameter a. Prior to bacterial inoculation all tumors were measured and mice weighed.

Escherichia coli growth conditions and injection into mice

FISV-1 E. coli strain (F-, mcrA, A(mrr-hsdRMS-mcrBC), (D80dlacZAM15, AlacX74, recAl, deoR, A(ara,leu)7697, araD139, galU, galK, nupG, rpsL, l-, DlpxM, inv::endAl-FRT, hly::tonB, plc::dapA-FRT, Lysis2.1::dapB-FRT) carrying a modified FISV-1 BAC of strain F (AICP34.5, AICP47) (used in Fig. 22 and Fig. 23) were taken from -80 °C stocks and streaked onto LB agar and left to grow overnight at 37 °C. Chloramphenicol at 50 pg mL ¹ , diaminopimelic acid at 0.5mg mL ¹ and MgS0₄ at 20mM were used to maintain the FISV-1 strain and diaminopimelic acid at 0.5mg mL ¹ and MgS0₄ at 20mM were used to maintain the empty bacteria. A single colony was then picked and inoculated into LB broth and placed at 37 °C where it was left shaking at 200 rpm overnight. Antibiotics (chloramphenicol) and supplements (MgS0₄) were added where necessary. A 1 % dilution was inoculated into fresh LB (~OD₆₀O 0.05) and left grow till it reached early log phase (OD₆₀o 0.6). Once it reached an OD₆₀o according to its growth curve corresponding to 10⁹ cfu mL ¹, 1 mL of the culture was centrifuged at 13,500 g for 1 minute and the supernatant was removed. The pellet was then re-suspended in 1 mL PBS and spun once again at 13,500 g for 1 minute. This process was repeated twice more and the final pellet was diluted down to 10^s cfu mL ¹. 100 pL of this volume was injected into the mouse tumors (intratumorally, IT) giving a final concentration of 10⁷ cfu/100 pL, injected into the mice on days 1, 4 and 7. Control group of animals received 100 pL of lx PBS on days 1, 4 and 7 directly into the tumors.

Readout

Tumor volume measurements (daily) in Fig. 22 and Fig. 23

Result

Figure 22 shows CT26 tumors from syngeneic Balb/C mice following triple IT injection of 10⁷ bacteria, up to day 13 post first application. Animals which received E. coli comprising an FISV-1 BAC were further divided to responders and non-responders to therapy and compared to the control group, which received PBS instead of bacteria. Animals were culled on day 13 due to scheduled sacrifice. Figure 23 shows that animals receiving E. coli comprising an HSV-1 BAC were divided into responders to therapy and non-responders, and presented together with the group that received PBS only. Two responding animals were plotted individually (mouse 14 and mouse 30), showing a clear difference in tumor growth in comparison to non-responders and control group. Complete tumor regression was observed in one animal (Mouse 14), and a tumor growth delay for 10 days in another one (Mouse 30). In total, 43% of the animals responded to the HSV-1 treatment.

References

Cicin-Sain L, Brune W., Bubic I., Jonjic S and Koszinovski U. H. (2003) : Vaccination of Mice with Bacteria Carrying a Cloned Herpesvirus Genome Reconstituted In Vivo. J. Virol., Vol. 203, p. 8249-8255

Kaufman H.L., Kohlhapp F.J., Zloza A. (2016): Oncolytic viruses: a new class of immunotherapy drugs. Nat. Rev. Drug Discov., Vol.14

Krzykawski M. (2015): Combined bacterial and viral treatment: a novel anticancer strategy. Cent EurJ Immunol., Vol. 40(3)

Pilgrim S., Stritzker J., Schoen J., Kolb Maurer A., Geiginat G., Loessner M.J., Gentschev I., Goebel W. (2003) : Bactofection of mammalian cells by Listeria monocytogenes: improvement and mechanism of DNA delivery. Gene Therapy, Vol. 10, p. 2036-2045 - van der Wal, F. J., Koningstein, G., ten Hagen, C. M., Oudega, B., and Luirink, J., "Optimization of Bacteriocin Release Protein (BRP)-Mediated Protein Release by Escherichia coli: Random Mutagenesis of the pCloDF13-Derived BRP Gene To Uncouple Lethality and Quasi-Lysis from Protein Release," Appl Environ Microbiol, vol. 64, no. 2, pp. 392-398, Feb. 1998.

Claims

1. A live bacterium comprising at least one exogenous polynucleotide integrated into the bacterial genome and/or maintained episomally, wherein

(i) the exogenous polynucleotide encodes a virus or part thereof;

(ii) the bacterium targets a eukaryotic cell; and

(iii) the bacterium is capable of releasing the DNA or RNA encoding the virus into the targeted eukaryotic cell.

2. The bacterium of claim 1, wherein the bacterium targets cancer cells.

3. The bacterium of claim 1 or 2, wherein the virus encoded by the exogenous polynucleotide is an oncolytic virus.

4. The bacterium of any one of claims 1 to 3, wherein the virus is selected from the group consisting of herpes simplex virus, HSV-1, Adenovirus, Adeno-associated virus (AAV), Adenovirus 3, Adenovirus 5, Parvovirus, Parvovirus H1PV, Mononegavirales, Ebola virus, Measles virus, Nipah virus, Influenza virus, Newcastle Disease Virus, Poxvirus, Alphaviruses, Picornaviruses, Coxsackie virus, Flaviviruses, Zika virus, Vaccinia virus, Vesicular Stomatitis virus, Poliovirus, Reovirus, Senecavirus, Echovirus, Semliki Forest virus and Maraba virus , and a chimeric virus.

5. The bacterium of any one of the preceding claims, wherein the bacterium is selected from the group consisting of Enterobacteriaceae, Lactobacillus spp., Yersinia spp., Escherichia spp., Klebsiella spp., Bordetella spp., Neisseria spp., Aeromonas spp., Francisella spp., Corynebacterium spp., Citrobacter spp., Chlamydia spp., Hemophilus spp., Brucella spp., Mycobacterium spp., Legionella spp., Rhodococcus spp., Pseudomonas spp., Helicobacter spp., Salmonella spp., Vibrio spp., Bacillus spp., Leishmania spp. and Erysipelothrix spp., Shigella spp., Listeria spp., Rickettsia spp., Acetoanaerobium spp., Aerococcaceae spp., Carnobacteriaceae spp., Enterococcace spp., Leuconostocacease spp., Streptococcaceae spp., and bacteria that are generally considered safe (GRAS status) by the US Food and Drug administration (FDA), preferably wherein the bacterium is Listeria monocytogenes, Salmonella typhimurium, Salmonella typhi, or Escherichia coli.

6. The bacterium of any one of the preceding claims, wherein the bacterium is Escherichia coli and the virus is HSV-1, adenovirus or parvovirus.

7. The bacterium of any one of the preceding claims, wherein the bacterium further comprises one or more polynucleotides selected from the group consisting of:

(i) a polynucleotide encoding invasion factors to enable the bacterium to invade eukaryotic cell, preferably an invasin (inv) derived from Yersinia spp.

(ii) a polynucleotide encoding bacteriocin release protein (BRP), bacteriophage lambda lysozyme, holin and antiholin to enable the said bacterium to self-lyse upon entering the eukaryotic phagosomes;

(v) a promoter driving the expression of the polynucleotides under (i) and (ii) and which responds to an environmental stimulus, such as a low magnesium, hypoxia, or bacterial quorum sensing.

8. The bacterium of any one of claims 1 to 7, wherein the bacterium is an auxotroph, preferably wherein the bacterium is a diaminopimelic acid (dap) auxotroph.

9. The bacterium of any one of claims 1 to 7, wherein the bacterium is a prototroph.

10. The bacterium of any of the claims 1 to 9, wherein one or more of the endogenous bacterial endonucleases and/or DNA repair proteins and/or metabolic proteins are not functional, optionally wherein:

(i) the endogenous bacterial endonucleases are EndA and/or HsdR;

(ii) the endogenous bacterial DNA repair protein is RecA, and/or;

(iii) the endogenous metabolic proteins are thiamine phosphate synthase (ThiE) and/or thiamine thiazole synthase (ThiA), and/or a component of the Ton system such as TonB receptor.

11. The bacterium of any of the claims 1 to 10, wherein the bacterium is engineered to avoid or minimize clearance by the host immune system using mechanisms such as the deletion of IpxM gene to express truncated and less virulent versions of surface lipopolysaccharides (LPS), or the coating of the bacterial surface with nanoparticles, such as polyethylene glycol (PEG) nanoparticles.

12. The bacterium of any one of the claims 1 to 11, wherein the bacterium is an E. coli which comprises invasin (inv), and listeriolysin O (hly) and wherein the bacterium has been engineered so that the endogenous bacterial endonucleases EndA and HsdR and the endogenous bacterial DNA repair protein RecA are not functional and the IpxM gene has been deleted.

13 The bacterium of any one of claims 1 to 12, wherein the bacterium is further engineered to comprise an exogenous polynucleotide encoding a tumor antigen, pro-apoptotic factor, immune checkpoint inhibitor, antibody, nanobody, enzyme, antiangiogenic factor, other immune modulators and/or other anticancer molecules integrated into the bacterial and/or the viral genome.

14. The bacterium of any one of claims 1 to 13, for use in the diagnosis, prevention or treatment of mammalian disease, preferably human disease.

15. The bacterium of any one of claims 1 to 14 for use in the diagnosis, prevention or treatment of cancer.

16. The bacterium for the use of claim 15, wherein the cancer is preferably a solid tumor, and is selected from the group consisting of brain tumor, glioblastoma, pituitary tumor, colorectal cancer, ovarian cancer, prostate cancer, pancreatic cancer, lung cancer, gastric cancer, liver cancer, head-and-neck cancer, oropharyngeal cancer, hypopharyngeal cancer, laryngeal cancer, lip cancer, oral cavity cancer, salivary glands cancer), anal cancer, penile cancer, vaginal cancer, vulvar cancer, breast cancer, bladder cancer, kidney cancer, bile duct cancer, skin cancers, melanoma, central nervous system tumors, adrenal cancer, adrenocortical carcinoma, appendix cancer, rectal cancer, endometrial cancer, esophageal cancer, eye cancer, retinoblastoma), gallbladder cancer, heart cancer, small intestine cancer, teratoid tumor, testicular cancer, thymus cancer, thyroid cancer, urethral cancer, uterine cancer, Wilms tumor, mesothelioma, multiple myeloma, sarcomas, Kaposi's sarcoma, uterine sarcoma, osteosarcoma, Lymphomas, Burkitt lymphoma, Hodgkin lymphoma, and non- Hodgkin lymphoma.

17. The bacterium for the use of any one of claims 14 to 16, wherein the bacterium is administered locally or systemically, optionally as oral, inhalation, intratumoral, intravenous, or subcutaneous administration.

18. The bacterium for the use of any one of claims 14 to 17, wherein the bacterium is administered in combination with one or more additional therapies such as an anticancer treatment such as cytotoxic chemotherapies, monoclonal antibodies, cancer vaccines, cytokines, checkpoint inhibitors, immunomodulators, surgery or radiotherapy.

19. The bacterium for the use of any one of claims 15 to 18, wherein the bacterium is Escherichia coli, the virus is HSV-1 and the cancer is pancreatic or colon cancer.