Spores of Clostridium engineered for clinical efficacy and safety cause regression and cure of tumors in vivo

Abstract

Spores of some species of the strictly anaerobic bacteria Clostridium naturally target and partially lyse the hypoxic cores of tumors, which tend to be refractory to conventional therapies. The anti-tumor effect can be augmented by engineering strains to convert a non-toxic prodrug into a cytotoxic drug specifically at the tumor site by expressing a prodrug-converting enzyme (PCE). Safe doses of the favored prodrug CB1954 lead to peak concentrations of 6.3 µM in patient sera, but at these concentration(s) known nitroreductase (NTR) PCEs for this prodrug show low activity. Furthermore, efficacious and safe Clostridium strains that stably express a PCE have not been reported. Here we identify a novel nitroreductase from Neisseria meningitidis, NmeNTR, which is able to activate CB1954 at clinically-achievable serum concentrations. An NmeNTR expression cassette, which does not contain an antibiotic resistance marker, was stably localized to the chromosome of Clostridium sporogenes using a new integration method, and the strain was disabled for safety and containment by making it a uracil auxotroph. The efficacy of Clostridium-Directed Enzyme Prodrug Therapy (CDEPT) using this system was demonstrated in a mouse xenograft model of human colon carcinoma. Substantial tumor suppression was achieved, and several animals were cured. These encouraging data suggest that the novel enzyme and strain engineering approach represent a promising platform for the clinical development of CDEPT.

Figure 1. The prodrug CB1954 is a low-toxicity monofunctional DNA-alkylating agent, but can be converted to a much more toxic bifunctional DNA-alkylating agent upon 2 x 2-electron reduction of the 4-nitro or 2-nitro group to the corresponding hydroxylamine

The 4-hydroxylamine is more cytotoxic than the 2-hydroxylamine. Some NAD(P)H-dependent nitroreductase (NTR) enzymes can catalyse this reaction, with varying kinetics and nitro group specificity (see Table 1).

Figure 2. Construction and in vitro characterization of C. sporogenes therapeutic strains

(A) Chromosomal pyrE locus of therapeutic strains and control strains. WT, wild type; ΔpyrE, a mutant constructed previously by integrating the ‘empty’ cloning site (lacZ') of pMTL-JH27 [37]; N1, the strain containing a NmeNTR expression cassette; E1, the strain containing an E. coli NfnB expression cassette; angled arrows, fdx promoter. (B) Successful integration shown by PCR using chromosome-specific primer Csp-pyrD-sF2 in combination with insert-specific primer M13F. MW, 2-Log DNA Ladder (NEB) molecular weight marker; plasmid, pMTL-JH27 derivative containing NmeNTR or NfnB cassette accordingly; WT, wild-type C. sporogenes genomic DNA; N1 and E1, genomic DNA from strain N1 or E1 respectively. (C) Functional expression of NmeNTR and NfnB, in strains N1 and E1 respectively, indicated by specific menadione reductase activities in cell lysates. Data shown are the mean of three independent experiments, error bars show standard deviation. N1 and E1 show specific menadione reductase activities elevated above the background activity of endogenous quinone reductases seen in WT and ΔpyrE. (D) NTR-expressing and control strains are equally able to form spores. Strains were grown in complex medium (TYG) for five days, heated to inactivate vegetative cells, and spore titres were determined by plating. A non-sporulating Δspo0A mutant constructed previously [35] is included as a control. Data shown are the mean of three independent experiments, error bars show standard deviation.

Figure 3. The NTR-expressing therapeutic strains are efficacious in CB1954 therapy in vivo

(A) Growth curves of HCT116 tumors treated as indicated in the legend. Spores (5 x 10⁷) were injected on day 0 when tumors reached a volume of ~250 mm³ and CB1954 prodrug (15 mg/kg) or sham treatment was given for 5 consecutive days starting at day 5 post spore injection. Data shown are average ± SEM. (B) Growth delay of individual tumors in the different groups. Growth delay is defined as the time for the tumor to reach three times its volume at the start of the prodrug or sham treatment. Statistical differences between groups are calculated using the non-parametric Mann-Whitney test (*, p<0.05; **, p<0.01; *, p<0.001) (C) Representative gram staining of tumor sections showing colonization of the tumor by C. sporogenes vegetative cells following spore administration, germination in the tumor, and outgrowth. C. sporogenes vegetative cells are visible as gram-positive purple rods and reside in the necrotic area of the tumor. HCT116 nuclei (pink) are counterstained with iodine solution.