Article Text

Nitric oxide as an antimicrobial agent: does NO always mean NO?
  1. Centre for Clinical Pharmacology and Molecular Biology Laboratory,
  2. Wolfson Institute for Biomedical Research,
  3. University College London,
  4. 140 Tottenham Court Road,
  5. London W1P 9LN, UK

    Statistics from

    See article on page 334

    Nitric oxide (NO) synthesised from l-arginine subserves multiple physiological functions in the cardiovascular, respiratory, gastrointestinal, genitourinary, and central and peripheral nervous systems.1 But synthesis of NO also contributes to host defence and seems to have cytostatic and cytotoxic effects against certain pathogens, and even against host cells themselves.1 How is this double act achieved? What determines the switch from physiological mediator to lethal gas and how is bacterial killing achieved?

    The simple and standard answer to the dual action of NO is that its effects depend on the amounts generated and the local concentrations achieved. In the nanomolar concentrations generated by constitutive NO synthase (NOS) isoforms, NO acts as a cell signalling molecule and interacts preferentially with its physiological target enzymes—the most significant of which seem to be soluble guanylyl cyclase and possibly cytochrome C oxidase. At the higher concentrations generated when the other enzymes become targets for NO action the cytokine induced isoform of NOS (iNOS) is expressed in cells. Haem containing enzymes, enzymes with Fe-S clusters including aconitase, NADH dehydrogenase and succinate dehydrogenase, the non-haem metalloenzymes, ribonucleotide reductase and DNA itself are all susceptible to inhibition or damage when NO output is high. Consistent with this high versus low output explanation, soluble guanylyl cyclase and cytochrome C oxidase are reversibly affected by NO (stimulation and inhibition respectively) with near maximal effects occurring in the nanomolar range of NO concentrations whereas effects on other enzymes occur only in the micromolar range and are often irreversible.

    Despite the simplicity and attractiveness of the low versus high output theory, it is almost certainly only partially correct. Many reports indicate that organisms which seem to be killed following induction of the l-arginine/NO pathway in immune cells are resistant to the effects of NO itself (at least under aerobic conditions). For example, although the growth of Staphylococcus aureus is potently inhibited by NO, Salmonella typhimurium, Escherichia coli and Listeria monocytogenes seem to be resistant.2 Perhaps it should come as no surprise that certain bacteria are resistant to the actions of NO. NOS has been detected in bacteria3 and presumably the NO serves some useful physiological function within the microbe. Furthermore, many bacteria are able to denitrify nitrate to nitrite and then to gaseous products including NO.4 Pseudomonas aeruginosa is a good example of a clinically important denitrifying bacterium. Such bacteria possess a NO reductase pathway which converts NO to nitrous oxide (N2O) and ultimately to nitrogen. Mutation of NO reductase is lethal for certain bacteria,4 suggesting that even for NO producers it is important to ensure that the intracellular concentration of NO does not rise too far. Certain bacteria that are not considered as denitrifying also seem to have the ability to reduce nitrate to NO. The enteric bacteria E coli, Klebsiella pneumonia,S typhimurium, Shigella sonnei, andProteus mirabilis all possess nitrate reductase and when grown under anaerobic conditions can metabolise nitrate to NO via nitrite.5 Given that nitrate concentrations in many biological fluids are in the order of 50 μM, significant amounts of NO might be generated through the NOS independent pathway.

    If NO itself is often not the toxic species, or if bacteria are able to metabolise and deactivate NO, how does activation of thel-arginine/NO pathway cause bacterial stasis and death? One possibility is that NO becomes cytotoxic only when generated together with superoxide (O2 -). NO and O2 - interact to form peroxynitrite (ONOO-) a powerful oxidant species that is toxic to many prokaryotic and eukaryotic cells. Inducible NOS and the phagocyte NADPH oxidase that generates O2 - are differentially regulated but both systems are induced by inflammatory stimuli and so co-production of NO and O2 - is common in activated phagocytes. Alternatively, peroxynitrous acid (ONOOH) may form as the result of interaction between nitrite and hydrogen peroxide under mildly acidic conditions, and this can kill E coli.6

    A third possible route for NO mediated cytotoxicity relies upon the formation of S-nitrosothiols (RSNO).7 Paradoxically, many RSNOs seem to have profound bacteriostatic effects on bacteria including those such as E coli and S typhimurium that are resistant to authentic NO. RSNOs react with sulphydryl groups on proteins and this can alter protein function to lead to cell stasis or death. The potential importance of RSNOs as bacteriostatic agents is exemplified by the finding that salmonella mutants that are deficient in active peptide transport and cannot transport glutathione are resistant to the toxic effects of the RSNO S-nitrosoglutathione.7

    It is clear that it is often not NO itself, but rather a number of NO related species that are important for antibacterial effects. Recently, the concept of nitrosative stress has been proposed.8 9Cellular damage may result from oxidative stress, largely mediated by oxidation of thiol groups, or by nitrosative stress induced by nitrosation of thiol groups and NO related inhibition of enzyme activity. Generation of nitrogen oxides may be involved in both processes with formation of peroxynitrite producing oxidative and nitrosative stress (through generation of OH. and NO2 .) and NO and nitrosothiols producing predominantly nitrosative stress.

    Why are some bacteria sensitive to one NO related compound and others sensitive to a different NO congener? It seems as though bacteria relatively deficient in low molecular weight thiols (for example, glutathione) are unable to protect themselves against S-nitrosothiols, peroxynitrite and even NO itself; Staphylococci spp fall into this group. Bacteria with high thiol concentrations are resistant to NO itself but are still susceptible to peroxynitrite and S-nitrosothiols; E coli and S typhimuriumfall into this group. It is also clear that bacteria may increase their resistance to nitrosative stress and these may be the most virulent of all. Certain transcriptional regulators that determine the expression of antioxidant defence enzymes (for example, superoxide dismutase, glucose-6-phosphate dehydrogenase, DNA repair enzymes, catalase, glutathione reductase) are activated in E coli exposed to NO related species. In addition to activation by oxidative stimuli, the transcriptional activator oxyR is activated by S-nitrosothiols,8 and the redox sensitive transcriptional regulator soxRS is activated by authentic NO.9 The net effect is to lead to the expression of antioxidant defence mechanisms that render the E coli more resistant to subsequent oxidative and/or nitrosative stress. Are these resistant “superbugs”? Certainly transcriptional activation ofsoxRS can also lead to antibiotic resistance.9

    Enzymatic generation of NO by induction of NOS in phagocytic cells is only one form of host defence reliant upon NO. It has been reported in this journal that free NO is generated when the nitrite present in saliva is acidified in the stomach. It is possible that peroxynitrous acid, and nitrosothiols are also formed within the stomach and that together these agents form a first line of defence against ingested microorganisms. In this issue a paper from Benjamin’s group (see page 334) demonstrates that Helicobacter pylori is killed by acidified nitrite. The authors speculate that an adequate dietary intake of nitrate may be important to ensure sufficient chemical generation of NO in the stomach to protect against colonisation of the gastric mucosa by H pylori. The experiments described were done under conditions of relatively low oxygen tension and even then effects were seen only at high concentrations of acidified nitrite (IC50 in the order of 200 μM). From these experiments it is not possible to tell whether the organisms are most sensitive to NO itself or to another NO species. Either way, the organisms seem relatively resistant and it would be important to understand which of the possible mechanisms contribute to resistance: high concentrations of thiols, activation of oxyR or soxRS, low glutathione transport, expression of NO reductase, etc. It would also be of interest to know whether the organisms are more sensitive to RSNOs or peroxynitrite rather than to authentic NO.

    Generation of NO and related compounds within the stomach may be a significant antimicrobial mechanism. Some organisms will not be affected, either because they are innately resistant to nitrosative stress or because they have been primed and have acquired resistance. Some will acquire resistance as a result of exposure to nitrosative stress in the stomach. On the basis of the limited data availableH pylori seems to be a relatively resistant strain although this would need to be tested directly in studies in vivo. It would be important to understand the mechanisms of resistance since this might provide a suitable target for drug therapy (for example, an inhibitor of NO reductase). It would also be interesting to determine whether a glass of nitrite or nitrosoglutathine would eradicate the organism. In the enthusiasm to explore these effects it should not be forgotten that NO and related compounds may damage and be mutagenic to host cells. N-nitroso compounds are carcinogens. Furthermore, H pyloriinduce expression of NOS in the stomach and it has been suggested that this might underlie the association between chronic infection withH pylori and gastric cancer.10

    See article on page 334


    View Abstract

    Request permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Linked Articles