Report

A Nitric Oxide–Inducible Lactate Dehydrogenase Enables Staphylococcus aureus to Resist Innate Immunity

See allHide authors and affiliations

Science  21 Mar 2008:
Vol. 319, Issue 5870, pp. 1672-1676
DOI: 10.1126/science.1155207

Abstract

Staphylococcus aureus is one of the most successful human pathogens, colonizing 2 billion individuals worldwide and causing invasive infections even in immunocompetent hosts. S. aureus can evade multiple components of host innate immunity, including the antimicrobial radical nitric oxide (NO) produced by activated phagocytes. We show that S. aureus is capable of metabolically adapting to nitrosative stress by expressing an NO-inducible l-lactate dehydrogenase (ldh1, SACOL0222) divergently transcribed from the NO-detoxifying flavohemoglobin (hmp). l-Lactate production allows S. aureus to maintain redox homeostasis during nitrosative stress and is essential for virulence. NO-inducible lactate dehydrogenase activity and NO resistance distinguish S. aureus from the closely related commensal species S. epidermidis and S. saprophyticus.

Enzymatic generation of radical nitric oxide (NO) by NO synthase is one of the most broad-spectrum mechanisms of host resistance to pathogenic microorganisms (1). NO synthesis is observed in patients with staphylococcal infections (2), and S. aureus stimulates NO production by human phagocytes (3). Moreover, high NO concentrations are found in the nasopharynx (4), the primary site of S. aureus colonization. NO and its congeners can modify a variety of cellular targets, including heme and nonheme iron, protein thiols, tyrosine residues, lipids, and DNA (59). Despite the abundance of potential NO targets, S. aureus is able to resist the antimicrobial actions of NO through a coordinated nitrosative stress response that includes the Hmp flavohemoglobin and enzymes involved in hypoxic or anaerobic metabolism (10). Mutant strains lacking a two-component regulatory system, SrrAB, which controls the transition to hypoxic/anaerobic metabolism, are extremely sensitive to nitrosative stress (10).

The ability to replicate in the presence of NO distinguishes S. aureus from many other bacterial pathogens (fig. S1), including the commensal species S. epidermidis and S. saprophyticus (Fig. 1A). NO resistance is a general trait of S. aureus observed in methicillin-susceptible (MSSA) and methicillin-resistant (MRSA) clinical isolates from both hospital and community-acquired (CA-MRSA) settings (Fig. 1A).

Fig. 1.

Metabolic adaptation of S. aureus to nitrosative stress. (A) Growth rate μ (hour–1) of S. aureus strains COL (blue circles), Newman (blue diamonds), and MW2 (blue squares); S. epidermidis strains 6293 (cyan circles) and 6903 (cyan diamonds); and S. saprophyticus strains C1L3 (red circles) and C1L4 (red diamonds) in PN medium (10) after NO exposure. (B) Metabolite excretion profile of S. aureus COL and derivatives grown in PN medium aerobically (+O2) or anaerobically (–O2) in the presence or absence of NO. Data represent average rates of acetate (red), l-lactate (dark blue), d-lactate (light blue), ethanol (orange), or formate (cyan) production (hour–1·108 CFU) during 4 hours of NO stress. (C) d- and l-lactate dehydrogenase activity in cell-free extracts from cultures grown in PN medium aerobically (Aer), aerobically with NO, or anaerobically (Anaer). (D) Pyruvate metabolism in S. aureus. Glucose (Glc) is oxidized to pyruvate (Pyr), which can be converted to d- or l-lactate (d-Lac, l-Lac) or oxidatively decarboxylated, generating CO2 and acetyl-CoA (Ace-CoA). Acetyl-CoA is converted to acetate (Ace) to generate ATP. Anaerobically, pyruvate can be converted to 2,3-butanediol (2,3-But) or to formate (For) and acetyl-CoA, the latter of which can be reduced to ethanol (EtOH).

To gain insights into the metabolic state of S. aureus growing under nitrosative stress, we assessed end-product excretion for 4 hours after exogenous NO treatment. Under aerobic conditions, S. aureus primarily excreted acetate, whereas fermenting cells produced a mixture of mainly l-lactate, ethanol, and formate (Fig. 1B). However, upon NO exposure, respiring cells ceased acetate production and fermenting cells no longer excreted measurable amounts of ethanol or formate (Fig. 1B). Moreover, NO exposure resulted in almost exclusive production of l-lactate from both respiring and fermenting S. aureus (Fig. 1B), which suggests that nitrosative stress restricts available glucose catabolic pathways. These results are consistent with increased lactate dehydrogenase gene expression observed in S. aureus exposed to NO or anaerobiosis (Fig. 1C) (10, 11). Interestingly, although both S. epidermidis and S. saprophyticus exhibited increased lactate dehydrogenase activity under anaerobic conditions, neither species displayed lactate dehydrogenase activity after treatment with NO (Fig. 1C). Thus, NO-inducible lactate dehydrogenase activity is specific to S. aureus and correlates with the ability to replicate during nitrosative stress (Fig. 1, A and C).

NO can inhibit aerobic respiration by competitively binding to cytochrome hemes of the terminal oxidases (12). Indeed, exposure of S. aureus to NO completely inhibited oxygen consumption until Hmp-mediated scavenging lowered NO to submicromolar levels (Fig. 2A). Therefore, reducing equivalents generated from glucose oxidation during nitrosative stress must be handled in a respiration-independent manner, as occurs by the reduction of pyruvate to lactate (Fig. 1D). Although NO-dependent inhibition of respiration poses a redox stress leading to increased l-lactate excretion, this cannot account for the absence of acetate production during nitrosative stress. To investigate this phenomenon, we determined enzyme activities in cell-free extracts in the presence or absence of increasing NO concentrations (Fig. 2B). Both l- and d-lactate dehydrogenase activities were found to be resistant to NO, but pyruvate dehydrogenase (PDH) activity was extremely NO-sensitive. Inhibition of PDH limits the generation of acetyl–coenzyme A (CoA), thereby restricting acetate production (Fig. 1D). Thus, in addition to overcoming the redox stress created by impaired respiration, NO-treated S. aureus must adapt to a lower overall adenosine triphosphate (ATP) yield.

Fig. 2.

Physiological consequences of NO exposure in S. aureus. (A) NO inhibition of staphylococcal respiration. Oxygen consumption was measured in cell suspensions in sealed vessels as loss of dissolved oxygen (red) by means of a Clark-type electrode. The presence of NO (blue) was simultaneously monitored via an NO-specific probe and was enzymatically scavenged from the cell suspension in a Hmp-dependent manner; compare top trace (wild type, WT) with the bottom trace from hmp mutant cells. (B) Percent enzyme activity in cell-free extracts as a function of NO concentration. PDH, pyruvate dehydrogenase; LDH, l-lactate dehydrogenase; DDH, d-lactate dehydrogenase; ADH, alcohol dehydrogenase; PFL, pyruvate formate lyase. Enzymatic activity assays are described in (17) and fig. S3. (C) Redox state of cells measured as NAD+/NADH ratio during aerobic and anaerobic growth in the presence or absence of NO. Nucleotide extraction and quantification were performed as described in (17). (D) Transcript levels of ldh1 and ldh2 under varying environmental conditions. We used quantitative real-time reverse transcription polymerase chain reaction to compare ldh1 and ldh2 transcript levels via a ΔΔCt algorithm outlined in (17). Values represent transcript levels relative to those of rpoD, encoding the sigma-70 subunit of RNA polymerase.

In glucose-fermenting S. aureus, pyruvate can serve as an electron acceptor, generating lactate from three different lactate dehydrogenases (Fig. 1D). Acetyl-CoA production proceeds via the pyruvate formate-lyase (PFL) reaction, which converts pyruvate to formate rather than CO2 and thus does not consume oxidized nicotinamide adenine dinucleotide (NAD+) (Fig. 1D). Excess acetyl-CoA is converted to ethanol by an iron-containing alcohol dehydrogenase (ADH), regenerating twice as much oxidizing power as lactate production (Fig. 1D). The tendency of fermenting cells to forgo the additional ATP yield from acetate production in favor of NAD+-generating ethanol synthesis underscores the importance of redox balance in nonrespiring cells (Fig. 1, B and D). PFL is inhibited at physiologically relevant NO concentrations (Fig. 2B), thereby limiting acetyl-CoA production and accounting for the absence of ethanol excretion despite the NO resistance of ADH (Fig. 1B and Fig. 2B). By limiting ethanol production, NO eliminates an efficient mechanism of NAD+ regeneration, forcing fermenting cells to rely on l-lactate dehydrogenase to maintain redox homeostasis (Fig. 1D).

The importance of lactate dehydrogenase activity in maintaining redox homeostasis in both aerobically and anaerobically cultured cells undergoing nitrosative stress is demonstrated by the inability of an ldh1 ldh2 mutant to maintain a physiological balance of NAD+ and its reduced form NADH in the presence of NO (Fig. 2C). Thus, the NO-resistant metabolic state achieved by S. aureus growing under nitrosative stress requires the induction of l-lactate dehydrogenase activity. Transcript levels of ldh2 (SACOL2618) were relatively abundant under aerobic growth and were modestly induced by anaerobiosis, but remained unchanged during nitrosative stress (Fig. 2D). In contrast, ldh1 expression was virtually undetectable in aerobic cultures but predominated upon NO exposure (Fig. 2D and fig. S2). Accordingly, inactivation of ldh1 measurably impaired staphylococcal growth during nitrosative stress in both aerobically and anaerobically cultured cells (Fig. 3A). Both Ldh1 and Ldh2 are resistant to NO (fig. S2); thus, constitutive Ldh2 activity can partially compensate for the loss of Ldh1. However, the complete loss of l-lactate production in an ldh1 ldh2 mutant eliminated the ability of S. aureus to grow in the presence of NO (Fig. 1B and Fig. 3A). Furthermore, providing the ldh1 gene in trans significantly enhanced the growth rate of both ldh1 and ldh1 ldh2 mutants in the presence of NO (Fig. 3B). Collectively, these data show that adaptation to the metabolic constraints imposed by NO under both aerobic and hypoxic conditions is critical for S. aureus resistance to nitrosative stress. NO-inducible ldh1 appears to be primarily responsible for the metabolic adaptation of S. aureus to nitrosative stress in vitro and is present in all S. aureus genomes sequenced to date.

Fig. 3.

Role of l-lactate dehydrogenase activity in S. aureus NO resistance. (A) Growth rate μ (hour–1) of S. aureus COL and derivatives grown aerobically (left) or anaerobically (right) in PN medium in the presence of NO. (B) Complementation of observed NO sensitivity in ldh1 and ldh1 ldh2 S. aureus Newman, with pldh1 harboring the ldh1 allele from S. aureus expressed from its endogenous promoter.

To determine the importance of NO adaptation in staphylococcal pathogenesis, we compared the virulence of wild-type S. aureus and isogenic l-lactate dehydrogenase–deficient mutants in a murine sepsis model. Intravenous (iv) injection of C57BL/6 mice with 5 × 107 colony-forming units (CFU) of wild-type S. aureus induced arthritic symptoms by day 3, followed by 2 weeks of progressive disease (Fig. 4A). Whereas an ldh2 mutation alone had no discernable effect on virulence, strains lacking the NO-inducible ldh1 gene were measurably attenuated, and ldh1 ldh2 mutants were essentially avirulent (Fig. 4A). Furthermore, the ldh1 and ldh1 ldh2 mutant strains exhibited competitive disadvantages (by a factor of 4 and a factor of 67, respectively) when coadministered with wild-type S. aureus (Fig. 4A, inset).

Fig. 4.

Role of NO-induced l-lactate dehydrogenase activity in S. aureus virulence. (A) Survival of 6-week-old female C57BL/6 and isogenic iNOS/ mice inoculated iv with 5 × 107 CFU of S. aureus strain Newman and mutant derivatives. Inset: Competitive indices [C.I. = (mutantOUT:wild-typeOUT)/(mutantIN:wild-typeIN)] of ldh1 or ldh1 ldh2 mutants coinfected iv with wild-type S. aureus strain Newman. Mice were injected with 107 CFU of 1:1 mixtures of mutant and wild-type bacteria; kidneys were harvested at 5 days after infection, homogenized, sonicated to disrupt S. aureus aggregates, and plated to determine viable CFU counts. Colonies were patched onto selective medium to determine the ratio of mutant to wild-type bacteria in renal tissue. (B) Histology of infected murine renal tissue. Female C57BL/6 mice were inoculated iv with 107 CFU of S. aureus strain Newman or indicated derivatives; tissue was harvested 5 days after infection, formalin-fixed, paraffin-embedded, then stained with hematoxylin and eosin. Scale bar, 1 mm.

Histological examination of infected renal tissue 5 days after inoculation revealed massive abscesses surrounding bacterial microcolonies (Fig. 4B). By day 5, abscesses were significantly larger and more numerous in mice inoculated with wild-type S. aureus relative to ldh1 or ldh1 ldh2 mutants (Fig. 4B and fig. S4). The reduced mortality, decreased abscess formation, and competitive disadvantage exhibited by the ldh1 mutant were not apparent in congenic C57BL/6 iNOS–/– mice lacking the ability to produce inflammatory NO (Fig. 4 and fig. S4). Additionally, l-lactate dehydrogenase activity is necessary for S. aureus to resist the cytostatic effects of host NO in cultured murine macrophages (fig. S5). Thus, NO-inducible ldh1 allows S. aureus to evade host-derived NO. Although the attenuation of ldh1 ldh2 mutants was partially abrogated in iNOS–/– mice, virulence was not fully restored (Fig. 4A). Together, these data suggest that an NO-independent requirement for l-lactate dehydrogenase activity during infection can be met by either ldh1 or ldh2, but NO resistance specifically requires high-level l-lactate dehydrogenase activity provided by ldh1.

This work has identified novel enzymatic targets of NO and provided insights into the metabolic state of S. aureus growing in the presence of host-derived NO. NO inhibits respiration as well as acetyl-CoA generation by PDH and PFL. Although the mechanism by which NO inhibits PDH and PFL is uncertain, preferential reactivity of NO with sulfhydryl groups and metals (13) suggests several possible molecular targets. The PDH complex requires the sulfhydryl-containing cofactor lipoamide and the coordinate activity of four redox-active thiols to catalyze the oxidative decarboxylation of pyruvate (14). PFL is an oxygen-labile enzyme whose active site contains both glycyl and thiyl radical moieties during catalysis (15). Moreover, PFL activation requires an additional enzyme, PflA, which contains an iron-sulfur cluster (16).

Metabolic adaptation to circumvent NO-mediated cytostasis is essential for the full virulence of S. aureus in an immunocompetent host. Whether S. aureus is residing within an aerobic or hypoxic niche within the host, NO restricts available options to regenerate oxidizing power. A key adaptive response of S. aureus during nitrosative stress is the induction of ldh1, encoding the dominant l-lactate dehydrogenase (Fig. 2D and fig. S2). Interestingly, NO induction of ldh1 is independent of the SrrAB two-component system, thereby implicating other mechanisms for the NO sensitivity of the srrAB mutant (10). The ldh1 allele is absent from coagulase-negative S. epidermidis and S. saprophyticus (fig. S6), and consequently neither species exhibits NO-inducible l-lactate dehydrogenase activity nor grows under nitrosative stress (Fig. 1, A and C). Thus, S. aureus has evolved NO-inducible l-lactate dehydrogenase activity through the acquisition of ldh1, which is divergently transcribed from the NO-scavenging flavohemoglobin encoded by hmp.

The hmp-ldh1 locus comprises a S. aureus–specific NO-resistance cassette that both detoxifies host-derived NO and circumvents its detrimental metabolic consequences. The pathogenic nature of S. aureus hinges on a previously unappreciated feature of its redox metabolism—the ability to induce homolactic fermentation during nitrosative stress.

Supporting Online Material

www.sciencemag.org/cgi/content/full/319/5870/1672/DC1

Materials and Methods

Figs. S1 to S6

Table S1

References

References and Notes

View Abstract

Stay Connected to Science

Navigate This Article