Nisin ZP, a Bacteriocin and Food Preservative, Inhibits Head and Neck Cancer Tumorigenesis and Prolongs Survival

Abstract

The use of small antimicrobial peptides or bacteriocins, like nisin, to treat cancer is a new approach that holds great promise. Nisin exemplifies this new approach because it has been used safely in humans for many years as a food preservative, and recent laboratory studies support its anti-tumor potential in head and neck cancer. Previously, we showed that nisin (2.5%, low content) has antitumor potential in head and neck squamous cell carcinoma (HNSCC) in vitro and in vivo. The current studies explored a naturally occurring variant of nisin (nisin ZP; 95%, high content) for its antitumor effects in vitro and in vivo. Nisin ZP induced the greatest level of apoptosis in HNSCC cells compared to low content nisin. HNSCC cells treated with increasing concentrations of nisin ZP exhibited increasing levels of apoptosis and decreasing levels of cell proliferation, clonogenic capacity, and sphere formation. Nisin ZP induced apoptosis through a calpain-dependent pathway in HNSCC cells but not in human oral keratinocytes. Nisin ZP also induced apoptosis dose-dependently in human umbilical vein endothelial cells (HUVEC) with concomitant decreases in vascular sprout formation in vitro and reduced intratumoral microvessel density in vivo. Nisin ZP reduced tumorigenesis in vivo and long-term treatment with nisin ZP extended survival. In addition, nisin treated mice exhibited normal organ histology with no evidence of inflammation, fibrosis or necrosis. In summary, nisin ZP exhibits greater antitumor effects than low content nisin, and thus has the potential to serve as a novel therapeutic for HNSCC.

Fig 1. Nisin ZP and nisin AP induce significant HNSCC cell apoptosis dose-dependently and beyond that in low content nisin.

A and B, Graphs showing changes in percentage of apoptotic HNSCC cells (HSC-3 and UM-SCC-17B) treated with control media or media containing 2.5% nisin, 95% nisin AP, or 95% nisin ZP for 24 h. Cells were then stained using an ethidium bromide and acridine orange (EB/AO) stain and counted. Comparisons between groups relative to controls were analyzed by ANOVA with the level of significance set at *p<0.05. C, Microscopic images showing morphologhy of HNSCC cells treated with control media or media containing nisin ZP (100 to 800 μg/mL) for 24 h then stained with EB/AO (Scale bars, 100 μm). D-F, Graphs showing changes in percentage of vital, apoptotic, and necrotic cells (UM-SCC-17B, HSC-3 and normal oral keratinocytes) treated with control media or media containing nisin ZP (400 or 800 μg/mL) for 24 h. Comparisons between groups relative to controls were analyzed by ANOVA with the level of significance set at *p<0.05 vital cells; # p<0.05 apoptotic cells.

Fig 2. Nisin ZP induces calpain, caspase-8 and PARP activation in HNSCC cells dose-dependently.

A, Graphs showing changes in percentage of apoptotic cells (UM-SCC-17B, HSC-3, UM-SCC-14A and OSCC-3) treated with control media or media containing nisin ZP (400 μg/mL) for 24 h. Cells were then stained with Annexin V and apoptosis was assessed by flow cytometry. Comparisons between groups relative to controls were analyzed by ANOVA with the level of significance set at *p<0.05. B, Immunoblots showing calpain 1, caspase-8, caspase-3, and PARP protein levels in HNSCC (UM-SCC-17B) and C, normal human oral keratinocytes cell lysates after treatment with nisin ZP for 18 h. D, Immunoblots showing PARP and calpain 1 activation in HNSCC cell lysates after treatment for 18 h with control media, media containing nisin ZP (400 μg/mL), media containing a calpain inhibitor (ALLN, 500 nM), or media containing both a calpain inhibitor (ALLN, 500 nM) and nisin (400 μg/mL). D, An immunoblot for ß-actin shows the loading controls.

Fig 3. Nisin ZP and nisin AP significantly reduce HNSCC cell proliferation time- and dose-dependently and clonogenic capacity.

A. Graphs showing changes in cell proliferation in UM-SCC-17B cells treated with control media or media containing nisin AP or ZP (400 or 800 μg/mL) for 48 h. Comparisons between groups relative to controls were analyzed by ANOVA with the level of significance set at *p<0.05. # Comparison between nisin AP and nisin ZP *p≤0.05. B. Graphs showing changes in cell proliferation in HNSCC cells (UM-SCC-17B, HSC-3 and UM-SCC-14A) treated with control media or media containing increasing doses of nisin ZP (100 to 800 μg/mL) for 48 h. Comparisons between groups relative to controls were analyzed by ANOVA with the level of significance set at *p<0.05. ¶Comparison between the 100 μg/ml group relative to the control for UM-SCC-17B cells *pμ0.05. C. Graphs showing changes in cell proliferation in UM-SCC-17B cells treated with control media or media containing nisin ZP (400 μg/mL) for 6, 12, 24 and 48 h. Comparisons between groups relative to controls were analyzed by ANOVA with the level of significance set at *p<0.05. D, The images are of UM-SCC-17B cells treated for 48h with control media or media containing nisin ZP (400 or 800 μg/mL) then cultured for 10 days, stained with crystal violet and imaged to evaluate clonogenic capacity. E, The graph shows the percentage of colonies present relative to controls for assays measuring clonogenic capacity. Comparisons between groups relative to controls were analyzed by ANOVA with the level of significance set at *p<0.05. F. Graphs showing changes in cell proliferation in normal human oral keratinocytes treated with control media or media containing nisin ZP (100, 200, 400 or 800 μg/mL) for 48 h. Comparisons between groups relative to controls were analyzed by ANOVA with the level of significance set at *p<0.05.

Fig 4. Nisin ZP inhibits orasphere formation in HNSCC cells.

A, Phase contrast images and B, graph showing percentage of orasphere inhibition (total area) in HNSCC cells (UM-SCC-17B) cultured under suspension conditions and treated with control media or media containing nisin ZP (100 to 800 μg/ml) for 36 h. Comparisons between groups relative to controls were analyzed by ANOVA with the level of significance set at *p<0.05.

Fig 5. Nisin ZP induces endothelial cell apoptosis and inhibits angiogenic sprouting dose-dependently.

A, Morphological observation of HUVEC cells treated with control media or media containing nisin ZP (100 to 800 μg/mL) for 24h. Cells were then stained with an ethidium bromide and acridine orange (EB/AO) stain, imaged, and counted (Scale bars, 100 μm). B, Graph showing percentage changes in vital, apoptotic, and necrotic HUVEC cells treated with control media or media containing nisin ZP (100 to 800 μg/mL) for 24 h. Comparisons between groups relative to controls were analyzed by ANOVA with the level of significance set at *p<0.05 vital cells; # p<0.05 apoptotic cells. C, Microscopic images of sprouting assays for cells (HUVEC) treated with control media or media containing nisin ZP (100 to 800 μg/mL) for 24h then imaged and counted (Scale bars, 100 μm). D, Graphs showing the total sprout length achieved by cells treated as indicated in C. Comparisons between groups relative to controls were analyzed by ANOVA with the level of significance set at *p<0.05.

Fig 6. Nisin ZP reduces HNSCC tumor burden in mice.

Images show the dissected tumors obtained from mice injected with UM-SCC-17B cells then treated with either water (control) or nisin AP or nisin ZP (400 or 800 mg/kg body weight/day) for 3 weeks.

Fig 7. Nisin ZP reduces tumor microvessel density.

A, Representative images of histological sections stained with H&E (top), and histological sections immunostained for CD31 (bottom) to identify blood vessels. B, Graph showing the results of microvessel quantification from five high power fields per tumor (*p<0.05).

Fig 8. Nisin ZP does not elicit histological signs of toxicity in the liver, lung and kidney of mice.

Images of H&E-stained histological sections of liver, lung and kidney from mice injected with UM-SCC-17B cells then treated with either water (control) or nisin ZP (800 mg/kg body weight/day) for 3 weeks.