We determined whether SL7838 cell suspensions can generate NO. When supplied nitrate (500 μM) under partial anaerobiosis, the suspensions generated NO for several hours (Figure ). We next determined if SL7838 can generate NO after infecting JC murine cancer cells. L-NAME (20 μM) was used to inhibit the indigenous NO generating activity of JC cells (due to iNOS), and NO formation was detected by DAF-FM diacetate; the latter fluoresces in the presence of NO, as we confirmed experimentally (Figure. ). Addition of nitrate or nitrite (at serum concentrations) to SL7838-infected JC cells generated the fluorescence signal (Figure. and ). Similar results were obtained with infected human MDA-MB-435 breast, or A2780 ovarian cancer cells (not shown).
Figure 1 A. NO production by SL7838 cell suspensions (A660, 1.5; glucose-M9 medium) incubated without shaking in the presence of nitrate. NO concentration was determined by the oxy-haemoglobin method ; B. NO (green fluorescence) production in SL7838-infected (more ...)
We then explored NO generation directly in implanted s.c. 4T1 tumours in living mice after intratumoural administration of SL7838. The tumours were impaled with a microelectrode for NO measurement [18
]; an oxygen microelectrode measured oxygen concentration in parallel. The tumours were examined at three separate locations. In the uninfected control tumours, no NO generation was seen, and the oxygenation levels changed with tumour depth as well as location (Figure. , panels A-C
) revealing, as expected, several regions of anoxia (<~1 μM oxygen). In the infected tumours (Figure. , panels D-F
), NO generation of up to 300 μM was seen (region D); the amount varied in different regions (e.g., lower levels in region E; none in F). Anaerobiosis tended to be more marked in the infected tumours, but there was no obvious correlation between the tumour oxygenation status and NO production.
Figure 2 In vivo measurement of NO, and O2 in dorsal 14-day old 4T1 subcutaneous tumours in BALB/c mice (50-100 mm3-). Symbols: Panels A-C: vertical NO (black circle) and oxygen (white triangle) concentration profiles in 3 different spots of an uninfected tumour; (more ...)
In order to test if NO generation by SL7838 contributes to its lethality to cancer cells, we generated strain SL7842, which was rendered more active in NO generation by deleting the norV and hmp genes that encode the pathways for decomposing this radical; while SL7838 cell suspensions caused complete disappearance of added NO (90 μM) within ca. 200 min (as measured by Greiss assay), those of SL7842 generated no decomposition; the NO donor, DEA NONOate, was used in these experiments. SL7848-infected 4T1 cells exhibited much greater loss of viability than those infected with SL7838 (Figure. ; compare bars 4 and 6 from the left). The killing of the cancer cells was dependent both on nitrate availability and bacterial infection. Invasion/replication assays showed that both the strains were equally competent in invading and replicating in cancer cells (not shown). Similar results were obtained with JC, A2780, or 293T cancer cells (not shown).
Figure 3 Loss of viability at 48 h of 4T1 breast cancer cells in vitro. Viability loss at 48 h of 4T1 breast cancer cells in vitro with no additions ("cells") and as a result of NO3- (150 μM) addition with or without infection by strain SL7838, or SL7842, (more ...)
Strain SL7842 proved more effective also in treating implanted 4T1 tumours in mice. Following the development of subcutaneous (s.c.) tumours (10 - 14 days post inoculation of 106
4T1 cells [5
]), separate groups of mice were injected intratumourally with saline (control), strain SL7838, or SL7842. While treatment with SL7838 did increase survival (Figure. ; Log Rank Test, *p
) as reported before, treatment with SL7842 generated ~100% greater survival (Log Rank Test, **p
). Similar outcome is evident from tumour burden measurements (Figure. ). The bacteria remained localized to the tumour (not shown), as revealed by bioluminescence imaging as we previously reported ([5
]; see Methods); and the mice exhibited no significant change in weight, suggesting non-toxicity of the treatment.
Figure 4 Efficacy of bacterial treatment of implanted cancer in mice. A. Survival (Kaplan-Meier) curves of BALB/c mice bearing s.c. 4T1 tumours (50-100 mm3). Symbols: untreated mice administered saline (control; black circle); SL7838-treated mice (black square); (more ...)