The Effect of Nitric Oxide J. L. SHANK, J. H. SILLIKER,

AND

on

Bacteria

R. H. HARPER

Research Laboratories, Swift and Company, Chicago, Illinois

Received for publication October 23, 1961

ABSTRACT SHANK, J. I,. (Swift and Company, Chicago), J. H. SILLIKER, ANDRI). H. HARPER. The effect of nitric oxide on bacteria. Appl. Nicrobiol. 10:185-189. 1962. Nitric oxide, as well as several other oxides of nitrogen, were assayed for their antibacterial action. It is shown that nitric oxide has virtually no effect on bacteria, whereas both NaNO3 and NaNO2 appear to have either neutral or stimulatory effects. It is suggested that the formation of nitrous acid is mainly responsible for the quantitative as well as the qualitative changes that occur in the bacterial flora of cured meat. A pH-dependent "nitrite cycle" is presented to accouint for the production of nitrous acid in cured meat systems. The addition of nitrate or nitrite to fresh meat ultimately results not only in pigment changes, but also, along with sodium chloride, brings about quantitative changes as well as qualitative changes in the bacterial population. MIembers of the genera Pseudomonas and Achromobacter are mainly responsible for the spoilage of fresh meat (Stewart, 1932; Haines, 1933; Ayres, Ogilvy, and Stewart, 1954). In cured meats, on the other hand, this predominately gram-negative flora of aerobic saprophytes is replaced by gram-positive facultative bacteria of the lactic acid group (unpublished data). Since the chemical and bacteriological aspects of curing occur as a function of nitrate or nitrite addition to meat, it is not unreasonable to postulate that the same general mechanism may be involved in both cases. In meats, nitrate (NO3-) is converted to nitrite (NO2-) through bacterial reduction. Nitrite, in the slightly acid environment of meat (pH 5.5 to 6.5), exists in equilibrium with nitrous acid (HN02). Nitrous acid, under the prevailing reducing conditions of the meat, is reduced to nitric oxide (NO) which reacts with myoglobin to yield the precursor of myochromogen, the stable red pigment of cured meat (Haldane, 1901; Hoaglund, 1914; Urbain and Jensen, 1940; Jensen, 1954). That nitric oxide may also play an important role in the bacteriology of curing thus becomes an inviting hypothesis. Since nitric oxide reacts with the heme pigments of meat, it could react with the heme-containing enzymes of the gram-negative bacteria, thereby poisoning them. This hypothesis could account, a priori, for the overgrowth of gram-positive, catalase-negative bacteria on cured meat. 185

Although Hatton (1881) states that bacteria connected with the spoilage of meat extracts are able to develop in an atmosphere of nitric oxide, Warburg (1927) concluded that nitric oxide is capable of two distinct reactions with the respiratory enzyme of yeast cells. One reaction is "reversible" and does not permanently affect the cell, whereas the other is "irreversible" and kills the cell. Ingram (1939) reported that the oxygen uptake by Bacillus cereus was inhibited by nitrite (at pH 6), thus implying that the cytochrome system of this organism was involved. This hypothesis was denied by Tarr (1941a) whose data show that it is unlikely that nitrite inhibits growth by affecting the aerobic respiratory enzymes of bacteria. Tarr's conclusion is substantiated by the work of Castellani and Niven (1955) who were able to inhibit streptococci with nitrite, even though these bacteria have no heme pigments. Since nitrite is most bacteriostatic under acid conditions (Tarr, 1941a, b) and since nitric oxide would certainly be involved in the pathway of nitrite reduction at reduced pH ranges (Corbet, 1934), a series of experiments was conducted to study the effect of the various oxides of nitrogen, including pure nitric oxide, on bacteria. MATERIALS AND METHODS Analytical. Nitric oxide' was washed through 40 % sodium hydroxide before use to remove any residual

nitrogen dioxide. High purity nitrogen,2 passed through hot copper threads, was used for flushing the system. Where reported, nitrate was determined quantitatively using the ferrous chloride method (AOAC, 1960). Nitrite was determined colorimetrically with modified Griess reagent (AOAC, 1960). Nitric oxide was qualitatively estimated by entrapment of the gas in agar. That this was nitric oxide could be inferred by the immediate formation of a brown gas upon exposure to oxygen (NO2). Bacteriological. Staphylococcus aureus, Streptococcus durans, and Proteus vulgaris were obtained from the American Type Culture Collection, Washington, D. C. Clostridium 3679, species PA-29, was originally obtained from C. F. Schmidt, Continental Can Company, Chicago, Ill. Lactobacillus K7B was obtained from R. H. Deibel of the American Meat Institute Foundation, Chicago, Ill. Pseudomonas fluorescens was isolated from meat in our laboratory and identified according to Bergey's Manual of 1 The Matheson Company, Inc., East Rutherford, N. J. 2 National Cylinder Gas Company, Chicago, Ill.

186

J. L. SHANK, J. H. SILLIKER, AND R. H. HARPER

[VOL. 10

Effect of Nitrous and Nitric Acids on Bacteria That pH alone was not responsible for the results shown in Table 1 is indicated by experiments in which 1 % aqueous solutions of sodium nitrite and sodium nitrate were reacted with an equivalent amount of HCl to form nitric and nitrous acids plus sodium chloride. The acid solutions were prepared individually and used immediately. For this test, 1.0 ml from 18-hr broth cultures was added to 10 ml of the acid solutions. Three trials were run for each test organism. Contact time between the cells and the acid during these trials averaged 86 sec, after which the cells were absorbed onto sterile Millipore discs and washed with 100 ml of 0.001 M phosphate buffer. The Millipore filter discs were subsequently broken up and counts made on TGY agar. It will be noted from Table 2 that the pH of the nitric acid suspensions was lower than the nitrous acid suspensions, but that the greatest antibacterial RESULTS was associated with the nitrous acid. effect Effect of Nitric Oxide and Nitrogen Dioxide of bacteria to > 200 ppm NO. Since nitric oxide Exposure on Various Bacteria has a limited solubility in water (at 25 C, 760 mm, only This preliminary experiment was designed to ascertain 56.3 ppm nitric oxide will be dissolved), it was decided to the gross effects of NO and NO2 on bacterial cells. Eighteen- determine the effect of materially increasing the amount hour broth cultures of the test bacteria were centrifuged, of nitric oxide available to react with the bacteria. Accordwashed in distilled water, and resuspended in 0.001 M phos- ingly, two stainless steel bombs were set up, each containphate buffer, pH 7.2. The organisms were individually absorbed onto separate 12.7-mm penicillin assay filter discs and the discs placed in a vacuum desiccator. The desiccator TABLE 1. Effect of NO and NO2 on various bacteria was evacuated to 29 in. Hg and backfilled with oxygenTotal count per ml free nitrogen gas. This operation was repeated three times. Organism After the fourth vacuumization, nitric oxide was introduced NO + NO + NO Vacuum only to atmospheric pressure. The bacteria on the discs were 0.18% NO2 0.36% NO2 kept in this atmosphere for 30 min. The desiccator was < 100 < 100 100 800,000 then evacuated and refilled with nitrogen. This operation Pseudomonas fluorescens was repeated three times. The discs were then removed Staphylococcus 100 500 160,000,000 9,000,000 from the desiccator under flowing nitrogen and transferred aureus to individual tubes of sterile Trypticase soy broth. After Streptococcus 1,000,000 1,000,000 100,000 1,000 durans breaking up the filter discs with sterile broken glass, an 41,000,000 29,000,000 1,000,000 195,000 aliquot was removed from each tube for bacteriological Lactobacillus K7B