Mycothiol Import by Mycobacterium smegmatis and Function as a Resource for Metabolic Precursors and Energy Production

Abstract

Mycothiol ([MSH] AcCys-GlcN-Ins, where Ac is acetyl) is the major thiol produced by Mycobacterium smegmatis and other actinomycetes. Mutants deficient in MshA (strain 49) or MshC (transposon mutant Tn1) of MSH biosynthesis produce no MSH. However, when stationary phase cultures of these mutants were incubated in medium containing MSH, they actively transported it to generate cellular levels of MSH comparable to or greater than the normal content of the wild-type strain. When these MSH-loaded mutants were transferred to MSH-free preconditioned medium, the cellular MSH was catabolized to generate GlcN-Ins and AcCys. The latter was rapidly converted to Cys by a high deacetylase activity assayed in extracts. The Cys could be converted to pyruvate by a cysteine desulfhydrase or used to regenerate MSH in cells with active MshC. Using MSH labeled with [U-¹⁴C]cysteine or with [6-³H]GlcN, it was shown that these residues are catabolized to generate radiolabeled products that are ultimately lost from the cell, indicating extensive catabolism via the glycolytic and Krebs cycle pathways. These findings, coupled with the fact the myo-inositol can serve as a sole carbon source for growth of M. smegmatis, indicate that MSH functions not only as a protective cofactor but also as a reservoir of readily available biosynthetic precursors and energy-generating metabolites potentially important under stress conditions. The half-life of MSH was determined in stationary phase cells to be ∼50 h in strains with active MshC and 16 ± 3 h in the MshC-deficient mutant, suggesting that MSH biosynthesis may be a suitable target for drugs to treat dormant tuberculosis.

Mycothiol ([MSH] AcCys-GlcN-Ins, where Ac is acetyl) (Fig. 1) is the dominant thiol found in most actinomycetes and appears to play a role similar to that of the glutathione found in some other prokaryotes and eukaryotes (17, 27). MSH is maintained in a reduced state by MSH disulfide reductase utilizing reducing equivalents from NADPH (25, 26). MSH forms S-conjugates (MSR) with a variety of alkylating agents, and the S-conjugates are efficiently cleaved by MSH S-conjugate amidase (Mca) to produce the mercapturic acid, AcCySR, that can be excreted from the cell, and 1-O-(2-amino-2-deoxy-α-d-glucopyranosyl)-d-myo-inositol (GlcN-Ins), that is utilized for resynthesis of MSH (15, 30, 33). An S-nitrosomycothiol reductase has been identified in mycobacteria (37), and MSH is a cofactor for maleylpyruvate isomerase in Corynebacterium glutamicum (6).

FIG. 1.

Structure of MSH and pathway for its biosynthesis from 1-l-myo-inositol-1-phosphate (Ins-P).

The pathway of MSH biosynthesis from 1-l-myo-inositol-1-phosphate (Ins-P) has now been fully elucidated, as shown in Fig. 1 (17, 19). The glycosyltransferase, MshA, transfers GlcNAc from UDP-GlcNAc to the three position of Ins-P to produce GlcNAc-(α1-1)-1l-myo-inositol-3-P, which is alternatively designated GlcNAc-α(1-3)-1d-myo-inositol-1-P. The latter is dephosphorylated by MshA2 to generate GlcNAc-Ins that is deacetylated by MshB to produce GlcN-Ins. Cysteine is linked to GlcN-Ins in a reaction utilizing ATP and catalyzed by MshC, and the product, Cys-GlcN-Ins, is acetylated by acetyl coenzyme A (acetyl-CoA) under catalysis by MshD to produce MSH.

Mycobacterium smegmatis and other soil actinomycetes appear not to require MSH for growth. Mutants deficient in MshB, MshC, and MshD have been generated for M. smegmatis (10, 28, 29), C. glutamicum (6), and Streptomyces coelicolor (24) and in MshA for M. smegmatis (18, 22) and S. coelicolor (24). The MshB-deficient mutants still produce some MSH, and the M. smegmatis MshD-deficient mutant produces a low level of MSH along with larger amounts of a homolog of MSH (20).

The synthesis of MSH or a close homolog is essential for growth of M. tuberculosis. Disruption of the mshB gene in M. tuberculosis reduces, but does not prevent, synthesis of MSH because an alternative deacetylase is able to act on GlcNAc-Ins (5). Inactivation of mshD is also not lethal, and the M. tuberculosis mutant produces a low level of MSH along with larger amounts of the MSH homolog N-formyl-Cys-GlcN-Ins that serves as a modest substitute for MSH in some biochemical reactions (4). In the soil bacteria MshA and MshC enzymes are essential for MSH production, and their loss sensitizes the bacteria to various toxins but does not prevent their growth. However, disruption of either the mshA or the mshC gene is lethal for M. tuberculosis (3, 31). It therefore appears that the soil bacteria have protective enzymes absent from the M. tuberculosis genome that allow them to survive without MSH.

The available data suggest that MshA and MshC may represent viable targets for tuberculosis drugs. However, to be effective such drugs would have to rapidly eliminate MSH in the dormant state, and the turnover rate of MSH in nonreplicating cells has not been determined. The objective of the present study was to obtain an estimate of the half-life of MSH in nonreplicating M. smegmatis. This was facilitated when it was found that M. smegmatis mutants lacking MSH have an active transport system that will import MSH from the medium in stationary phase. When the mutants were transferred to MSH-free medium, the decay of MSH could be measured. Results obtained with MshA- and MshC-deficient mutants provide important insights on the turnover rate of MSH and show how MSH serves as a resource capable of generating important biosynthetic precursors and of fueling energy-producing pathways.

MATERIALS AND METHODS

Reagents.

Middlebrook 7H9 broth and Middlebrook OADC (oleic acid, albumin, dextrose, and catalase) enrichment were obtained from Becton Dickinson, and monobromobimane (mBBr) was from Invitrogen. Trifluoroacetic acid (TFA) was from Halocarbon, sodium dodecyl sulfate was from Bio-Rad Laboratories, and methanesulfonic acid was purchased from Fluka. MSH was isolated from M. smegmatis mc²155 (36), and GlcN-Ins was prepared as previously described (21). Other chemicals were obtained from Sigma or Fisher and were American Chemical Society reagent grade or the highest grade available unless otherwise specified.

Bacterial strains and culture conditions.

M. smegmatis strain mc²155 (provided by W. R. Jacobs, Jr., Albert Einstein College of Medicine, The Bronx, NY), strain 49 (22), the Tn1 mutant (29), and the mshA::Tn5 strain (18) were started in Middlebrook 7H9 broth containing 0.05% Tween 80 and 10% OADC in the presence of 20 μg of kanamycin per ml for the Tn1 and mshA::Tn5 mutants. Starter cultures were expanded into Middlebrook 7H9 medium containing 0.05% Tween 80 and 0.4% glucose, supplemented with antibiotics as needed. All cultures were grown at 37°C with shaking at 225 rpm.

MSH uptake.

Early-stationary-phase (24- to 36-h cultures; A₆₀₀ of ∼2) cells (12.5 ml) were transferred to 25-ml Erlenmeyer flasks. MSH was added to the desired concentration, and dithiothreitol (DTT) was added from a freshly prepared stock solution to a final concentration of 50 μM. Incubation at 37°C with shaking at 225 rpm was continued, and 2.5-ml samples to determine cellular and medium MSH content were taken at time zero (immediately after addition of MSH and DTT) and at 60, 120, and 180 min and transferred to ice-cold 15-ml Falcon tubes. The cells were pelleted at 4°C for 10 min at 5,000 × g, and a 0.5-ml portion of the supernatant medium was mixed with 10 μl each of 1 M HEPES (pH 8) and 100 mM mBBr; the mixture was allowed to react in the dark for 15 min at 23°C. The samples were quenched with 2 μl of 5 M methanesulfonic acid and analyzed for thiol content by high-performance liquid chromatography (HPLC) (10) to determine the MSH content of the incubation medium. For determination of total medium MSH content (MSH plus disulfide forms of MSH), a portion of the supernatant medium was treated with 1 mM DTT for 30 min at 23°C and then derivatized with 3 to 5 mM mBBr, the amount chosen to provide a 1 mM excess over the total thiol concentration. To determine cellular thiol content, the pellet was washed once with ice-cold 7H9 medium with 0.05% Tween 80 and resuspended in 0.5 ml of 50% aqueous acetonitrile containing 20 mM HEPES (pH 8.0) and 2 mM mBBr. After 10 min at 60°C in the dark, the labeling reaction was quenched with 2 μl of 5 M methanesulfonic acid, the mixture was centrifuged to pellet the cellular debris, and the supernatant was removed for thiol analysis by HPLC. The pellet was vacuum dried and weighed to obtain the residual dry weight (RDW).

An analogous experiment to determine the uptake of MSH disulfide (MSSM) was carried out with M. smegmatis strain 49. The final concentration of MSSM in the medium was 6.5 μM, and samples were taken at time zero and at 180 min. The pellet and medium were derivatized with mBBr after reduction in the presence of 1 mM DTT, as described above.

MSH utilization.

All depletion experiments were performed in preconditioned medium prepared from Middlebrook 7H9 broth containing 0.4% glucose and 0.05% Tween 80. Early-stationary-phase cells (36- to 40-h cultures; A₆₀₀ of 1.4 to 2.4) were equally divided into separate 50-ml Erlenmeyer flasks. Bacteria in one flask were incubated with 50 μM DTT for 3 h at 37°C with shaking at 225 rpm. The medium from this flask was used to resuspend MSH-loaded cells and is referred to as preconditioned medium. Bacteria in the second flask were incubated with MSH at the desired concentration and 50 μM DTT. After 3 h at 37°C with shaking at 225 rpm, the cells from each flask were transferred to two ice-cold Falcon tubes and pelleted at 4°C at 5,000 × g. The MSH-loaded cells were washed once with ice-cold Middlebrook 7H9 medium containing 0.05% Tween 80, collected by centrifugation at 4°C and 5,000 × g, and resuspended at the original density in the preconditioned medium (containing no MSH) preheated to 37°C. Samples were taken immediately, and sampling continued for up to 45 h. Each sample (2 to 3 ml) was collected in an ice-cold Falcon tube and pelleted. A portion (7.5 μl) of the supernatant medium was analyzed for GlcN-Ins as described elsewhere (2). The remaining medium was reduced with 1 mM DTT in 20 mM HEPES (pH 8.0) for 30 min at 23°C; half was directly derivatized with 3 mM mBBr (for 10 min at 60°C) for thiol analysis, and half was reacted for 10 min at 60°C with 5 mM N-ethylmaleimide prior to derivatization with 3 mM mBBr to serve as a control used to identify nonthiol fluorescent components. The reactions were quenched by mixing with 2 μl of 5 M methanesulfonic acid.

Synthesis of [GlcN-6-³H]MSH.

A crude cell extract of M. smegmatis mc²155 was utilized as a source of MshA, MshA2, and MshB (Fig. 1) for synthesis of [6-³H]GlcN-Ins (19). Stationary phase cells were extracted with four passes through a French pressure cell (10,000 lb/in²) in 10 mM MgSO₄ and 25 mM HEPES, pH 7.5. The extract was clarified by centrifugation at 30,000 × g, and the supernatant was dialyzed against 100 volumes of the extraction buffer to generate extract supernatant with 12 mg of protein per ml used for synthesis of [6-³H]GlcN-Ins. UDP-N-acetylglucosamine containing 5 μC_i of [glucosamine-6-³H]UDP N-acetyl-d-glucosamine (NET 434; Perkin Elmer Life Sciences) and 1-l-inositol-1-phosphate (19) were added at 1 mM each to 5 ml of the extract supernatant, and the mixture was incubated at 37°C for 20 h. An assay for GlcN-Ins showed that 38% (1.9 μmol) of the [³H]UDP-GlcNAc had been converted to [6-³H]GlcN-Ins. Acetonitrile (5 ml) was added, the mixture was heated at 60°C for 15 min, and the protein was pelleted by centrifugation at 13,000 × g and 4°C. The supernatant was reduced to dryness on a SpeedVac, and the residue was dissolved in 1.7 ml of water. Conversion to Cys-[6-³H]GlcN-Ins was accomplished by the addition of DTT (2 mM), l-cysteine (1.5 mM), ATP (2 mM), and 21 μg of purified M. tuberculosis MshC (21) and incubation at 37°C. Production of Cys-[6-³H]GlcN-Ins was monitored by mBBr labeling and HPLC analysis and showed the reaction to be complete in 3 h. Conversion of Cys-[6-³H]GlcN-Ins to [GlcN-6-³H]MSH was accomplished by the addition of 1.5 mM acetyl-CoA and 30 μg of Ni affinity-purified M. tuberculosis MshD (10) and incubation at 37°C, and MSH production was monitored as described previously (10). Conversion was complete in 20 min, the mixture was acidified to a pH of <3 with TFA, and precipitate was removed by centrifugation. The [GlcN-6-³H]MSH was purified by HPLC as previously described (36) to yield 1.4 μmol of product with a specific activity of 0.59 mC_i per mmol.

Synthesis of [Cys-U-¹⁴C]MSH.

[Cys-U-¹⁴C]MSH was generated from GlcN-Ins, [U-¹⁴C]cysteine, ATP, and acetyl-CoA using purified M. tuberculosis MshC and MshD. A solution (1.2 ml) containing 50 mM HEPES, pH 8.0; a 2 mM concentration of MgCl₂, DTT, and ATP; 1 mM GlcN-Ins (15); 1 mM l-cysteine; and 2 μC_i of l-[U-¹⁴C]cysteine (NEC 465; Perkin Elmer) was prepared, and the reaction was initiated by addition of 30 μg of purified M. tuberculosis MshC. Analysis of Cys-GlcN-Ins showed the reaction to be complete after 2 h at 37°C. The addition of 1.2 mM acetyl-CoA and 30 μg of purified MshD followed by incubation for 30 min at 37°C produced complete conversion to [Cys-U-¹⁴C]MSH. The reaction mixture was acidified with TFA, and MSH was purified by preparative HPLC (36) to generate 0.9 μmol of [Cys-U-¹⁴C]MSH with a specific activity of 1.7 mC_i per mmol.

Uptake and utilization of [GlcN-6-³H]MSH and [Cys-U-¹⁴C]MSH in the Tn1 mutant.

The experiments were conducted as described above with modifications for the use of radioactive compounds. At each time point during the incubation of loaded cells in preconditioned medium, a 2.5-ml sample of cell suspension was removed and pelleted at 4°C. A 1-ml aliquot of the medium was removed for counting. The washed pellet was extracted with 100 μl of 20 mM HEPES, pH 8.0, in 50% acetonitrile at 60°C. After centrifugation 2 μl of the supernatant was added to 1 ml of water for counting, and the mixture was counted; the remaining supernatant was analyzed for the presence of GlcN-Ins and MSH as described above. During the HPLC separations 1-min fractions were collected and counted to determine radioactive metabolites.

Metabolism of [Cys-U-¹⁴C]MSH in cell extract of the Tn1 mutant.

The M. smegmatis mshC Tn1 mutant, grown to an optical density at 600 nm of ∼2.0 (∼40-h culture), was harvested by centrifugation (5,000 × g at 4°C) and resuspended in twice the volume of ice-cold 25 mM HEPES, pH 7.5, and 10 mM MgSO₄. The cells were passed three times through a French pressure cell (Aminco) at 24,000 lb/in². The extract was centrifuged at 30,000 × g at 4°C for 30 min; the supernatant was collected and dialyzed against 100 volumes of 25 mM HEPES (pH 7.5) and 10 mM MgSO₄, with one change after 12 h. The extract was centrifuged after dialysis to remove any precipitate. The dialyzed cell extract was reacted at 37°C with 1 mM [Cys-U-¹⁴C]MSH and 1 mM NADPH in a final volume of 100 μl. At 0, 1, 2, and 4.5 h, 20 μl was withdrawn from the reaction mixture, mixed with an equal volume of acetonitrile, and heated for 10 min at 60°C to precipitate proteins. After centrifugation, 7.5 μl of the supernatant was assayed for the presence of GlcN-Ins as referenced above. A 20-μl aliquot of the remaining supernatant was reduced with DTT at a final concentration of 2.5 mM (10 min at 60°C), followed by derivatization with 6.8 mM mBBr. The reaction was quenched with 1 μl of 5 M methanesulfonic acid and subjected to HPLC analysis as described above with counting of 1-min fractions.

Assay of AcCys deacetylase and cysteine desulfhydrase activities.

The deacetylation of AcCys was examined in dialyzed extracts of strain mc²155 prepared as described above and diluted ∼10-fold into the extraction buffer (10 mM MgSO₄, 25 mM HEPES, pH 7.5) to a protein concentration of 1 mg per ml. Duplicate 0.5-ml extract samples were preheated to 37°C, and N-acetylcysteine was added to a final concentration of 1 mM. Samples (100 μl) were taken at 0, 5, 15, and 30 min, mixed with 100 μl of acetonitrile, and incubated for 15 min at 60°C to precipitate protein. The samples were iced, and the supernatant was clarified by centrifugation at 13,000 × g for 5 min. A portion of the sample (25 μl) was mixed with 25 μl of 25 mM HEPES, pH 8.0, containing 1 mM DTT, incubated for 15 min at 23°C, and derivatized by the addition of 4 mM mBBr. This sample was assayed for Cys and AcCys by HPLC as previously described (10).

Cysteine desulfhydrase activity was assayed as described for extracts of M. tuberculosis by Wheeler et al. (39). The activity was determined in dialyzed M. smegmatis mc²155 extracts by the analysis of pyruvate as the 1,2-diamino-4,5-dimethoxybenzene (DDB; Invitrogen) derivative. The DDB derivative of pyruvate was detected by fluorescence with picomolar sensitivity by a minor modification of the method of Ohmori et al. (23). Cysteine (1 mM) was added to 0.5 ml of prewarmed (37°C) dialyzed extract prepared as described above. Samples (100 μl) were removed from the incubation at 0, 5, and 15 min and added to an equal volume of acetonitrile. This sample was incubated at 60°C for 15 min to inactivate the cysteine desulfhydrase activity. The samples were iced, and the protein was pelleted by centrifugation for 3 min at 13,000 × g. A 50-μl aliquot of the cleared extract was mixed with 4.5 μl of TFA and 5.5 μl of 3.6 mM DDB in aqueous 1.18 M TFA and incubated at 40°C for 2 h. The sample was diluted with 180 μl of water, and 100 μl was used for HPLC. The pyruvate-DDB derivative eluted at 26 min (55% methanol) in 0.1% TFA-water with a linear gradient from 0 to 100% methanol over 40 min using a reversed-phase C₁₈ column (Beckman 235335).

Generation and fate of [Cys-U-¹⁴C]MSH from l-[U-¹⁴C]cysteine in M. smegmatis mc²155.

A 1-liter culture of strain mc²155 was grown to an A₆₀₀ of ∼3. A 150-ml portion of the cell suspension was removed, and 0.54 μC_i of l-[U-¹⁴C]cysteine (NEC 465; Perkin Elmer) was added at a final concentration of 12 nM. The remainder of the culture was centrifuged, and the supernatant medium was collected (preconditioned medium). The cell suspension containing l-[U-¹⁴C]cysteine was incubated at 37°C for 2 h and centrifuged, and the pellet was washed at 4°C with preconditioned medium, with preconditioned medium containing 5 μM cysteine, or with preconditioned medium containing 1 mM DTT. The cells were resuspended in preconditioned medium and incubated at 37°C. Samples were taken at intervals for up to 48 h and centrifuged. The medium was counted, and the cell pellet was extracted and derivatized with mBBr for thiol analysis as described above. The labeled extract was analyzed for thiol content by HPLC with counting of fractions as described above. Inclusion of cysteine or DTT in the wash medium had no significant effect upon the results obtained.

Radioactivity measurements.

Radioactive MSH samples were counted in 1 ml of aqueous sample added to 8 ml of Fisher Scintiverse BD liquid scintillation cocktail. The samples were counted in glass vials using a Beckman LS1701 liquid scintillation counter. The counting efficiency of ³H was estimated to be 32% using [glucosamine-6-³H]UDP N-acetyl-d-glucosamine. The counting efficiency of [Cys-U-¹⁴C]MSH was estimated using the sample channels ratio method (13). A quench curve was developed using [U-¹⁴C]cysteine and various amounts of mBBr as the quenching agent. The counting efficiency for most supernatant and HPLC samples was 41%. A significantly decreased sample channel ratio with cell pellet samples required correction of the counting efficiency for those samples.

RESULTS

Uptake of MSH by MSH-deficient mutants.

Preliminary studies with the MSH biosynthesis mutant mshA::Tn5, that does not produce MSH or its precursors (18), indicated that incubation in medium containing low amounts of MSH results in accumulation of MSH in cells to levels comparable to that of the parental strain mc²155. Similar results were obtained with strain 49, shown to be defective in MSH production due to a mutation in mshA (18, 22), and this strain was used initially for a more detailed study. Stationary phase cells of strain 49 were incubated and sampled over 3 h following addition of MSH (12 μM) and DTT (50 μM), the latter to prevent oxidation of MSH. Analysis of the MSH content of the medium and the cells revealed that the MSH content of the medium declined to zero whereas the cell level increased by an amount equivalent to half that lost from the medium (Fig. 2A). As seen in Fig. 2B, at 3 h the cellular content of MSH was 11 μmol per g of RDW, comparable to that present in M. smegmatis mc²155 during exponential or stationary phase growth, ∼12 μmol per g of RDW (20).

FIG. 2.

(A) Total MSH content of medium (○) and of cells (▵) during incubation of stationary phase M. smegmatis MshA-deficient strain 49 (A₆₀₀ of ∼2) in preconditioned medium (12.5 ml) containing 50 μM DTT and 12 μM MSH. (B) Cellular MSH content for incubation as in panel A for medium containing MSH at 12 (○) and 94 (□) μM MSH. Error bars show range for duplicate determinations (where greater than the symbol).

Increasing the initial concentration of MSH in the medium to 94 μM MSH permitted uptake to occur to a level more than double the typical MSH content of the parental strain mc²155 (Fig. 2B). In this case, the medium content of MSH was 46 μM after 3 h, and about two-thirds of the loss was accounted for by the cellular uptake. Similar experiments were conducted with the mshA::Tn5 mutant and produced nearly identical results (Fig. 3). Loss of MSH in the medium could occur as the result of oxidation to disulfide forms. Reduction of medium samples with additional DTT prior to analysis for MSH showed that such loss is not significant in the first hour (<5%) but does become significant in the second and third hours, such that all of the medium content is in disulfide forms at 3 h for the lowest initial MSH concentrations.

FIG. 3.

Cellular level of MSH after a 3-h incubation in medium containing MSH at the given initial medium MSH content. □, strain 49; ▪, mshA::Tn5 mutant; ○, Tn1 mutant.

The cellular level of MSH after a 2-h uptake by strain 49 in 12 μM MSH was determined to be 11 μmol per g of RDW (Fig. 2B), versus <0.1 for cells harvested at time zero, and this corresponds to an intracellular concentration of ∼4 mM. The mean medium MSH concentration during this 2-h interval was measured at 7.5 μM. Thus, MSH uptake can occur against a ∼500-fold concentration gradient and must be an active transport process.

Uptake of MSSM was also examined with stationary phase M. smegmatis strain 49. After a 3-h incubation in 6.5 μM MSSM, the cells were determined to have an MSH content of 1.0 μmol per g of RDW, compared with <0.1 μmol per g of RDW immediately after addition of MSSM, and a content of 11 μmol per g of RDW for 3-h uptake in 12 μM MSH. Thus, assuming uptake is proportional to concentration for both substrates under these conditions, uptake of the disulfide is at least 12-fold slower than uptake of MSH.

MSH uptake was also determined with the M. smegmatis Tn1 mutant (29). This mutant produces substantial amounts of GlcN-Ins during exponential growth (∼3 μmol per g of RDW) but is unable to ligate it with Cys to produce Cys-GlcN-Ins owing to the lack of MshC activity. Accumulation of MSH from the medium in stationary cells of the Tn1 mutant was similar to but slower than that found for the MshA-deficient mutants (Fig. 3). Results from multiple experiments are summarized in Fig. 3 and show that uptake is three- to fivefold faster in the MshA-deficient mutants.

MSH utilization by MshA-deficient strain 49 in stationary phase.

The fate of MSH and GlcN-Ins was examined in strain 49 cultured to stationary phase, incubated with MSH added to the medium, resuspended in MSH-free stationary phase medium, and analyzed during continued incubation at 37°C. In separate experiments cells loaded to a high (∼24 μmol of MSH per g of RDW, about twice the normal level in M. smegmatis mc²155) and to an approximately normal level of MSH (∼14 μmol per g of RDW) were analyzed for up to 44 h (Fig. 4A). The half-life for MSH loss was estimated at 67 ± 13 h at the high initial concentration and at 37 ± 16 h at the normal initial MSH level. The cellular GlcN-Ins content remained nearly constant at 0.14 ± 0.03 μmol per g of RDW during incubation at the high initial MSH content (Fig. 4B). Least squares fit of the data exhibited a tendency toward lower values corresponding to a half-life on the order of 90 h. At the normal level of initial MSH, the GlcN-Ins content was below the detectible level (<0.02 μmol per g of RDW for the sample size available in these experiments) except at 20 h, where a value of 0.06 μmol per g of RDW was measured.

FIG. 4.

Utilization of MSH and GlcN-Ins in stationary phase M. smegmatis MshA-deficient strain 49. (A) Loss of cellular MSH with time. (B) Loss of cellular GlcN-Ins with time. Cells were loaded at high (▪) and normal (•) levels of initial MSH content, as shown by the time zero values. At normal initial MSH content, GlcN-Ins was below the limit of detection (<∼0.02 μmol per g of RDW) except at 44 h (∼0.06 μmol per g of RDW). Data points represent single determinations, and the high MSH content data are representative of two independent experiments.

Since strain 49 does not produce detectable GlcN-Ins (<0.004 μmol per g of RDW) (22), the present results demonstrate that GlcN-Ins can be derived from imported MSH. Mca has been shown to efficiently cleave a wide range of MSR to produce GlcN-Ins plus the mercapturic acid AcCySR, but Mca will also cleave MSH, and this provides a source of GlcN-Ins in the absence of MSR formation (15, 33). Since GlcN-Ins does not accumulate with time in strain 49, it must be utilized by other enzymatic processes. One such process is the resynthesis of MSH catalyzed by MshC and MshD (Fig. 1).

MSH utilization by the MshC-deficient Tn1 mutant in stationary phase.

Analogous studies to those with strain 49 were conducted with the Tn1 mutant at normal (∼10 μmol per g of RDW) and subnormal (∼6 μmol per g of RDW) levels of initial MSH loading (Fig. 5A). Loss of MSH from this mutant followed approximate first-order kinetics with half-lives of 16 and 19 h for normal and subnormal initial loadings, respectively. These MSH half-lives indicate that MSH loss is two- to fourfold faster in the Tn1 mutant than for strain 49. This difference presumably derives in part from the ability of strain 49 to resynthesize MSH from GlcN-Ins, an ability lacking in the Tn1 mutant owing to the absence of MshC activity.

FIG. 5.

MSH and GlcN-Ins in stationary phase cultures of the MshC-deficient Tn1 mutant. (A) Loss of cellular MSH with time. (B) Loss of cellular GlcN-Ins with time. Cells were loaded at normal (•) and low (○) levels of initial MSH content, as shown for the time zero values. Data points represent single determinations, and the two experiments presented are representative of five independent studies.

A similar pattern was seen for the change in the cellular GlcN-Ins content to that found for MSH (Fig. 5B). Prior to loading of Tn1 mutant cells, the GlcN-Ins content was typically 1.6 to 2.1 μmol per g of RDW. After loading of MSH to a normal initial MSH content, the GlcN-Ins content was 1.7 μmol per g of RDW, but it fell steadily over a 20-h interval after transfer to MSH-free medium (Fig. 5B). After loading to a lower MSH content, the GlcN-Ins level was 0.8 μmol per g of RDW and remained relatively constant, declining at a rate of only ∼0.004 μmol per g of RDW per h (Fig. 5B). This value for GlcN-Ins, 0.8 μmol per g of RDW, is nearly two orders of magnitude above the value, ∼0.01 μmol per g of RDW, previously reported for M. smegmatis mc²155 in early stationary phase (2) and at least one order of magnitude above the upper limit value (0.06 μmol per g of RDW) for MshA-deficient mutant 49 loaded to a normal level of MSH. These results indicate that MshC is a key enzyme in maintaining the GlcN-Ins at or below ∼0.1 μmol per g of RDW but that an additional enzyme that metabolizes GlcN-Ins is present and prevents the level from increasing much above ∼2 μmol per g of RDW.

The fate of [GlcN-6-³H]MSH in the MshC-deficient Tn1 mutant.

In order to further investigate the disposition of the GlcN residue during the metabolism of MSH, [GlcN-6-³H]MSH was enzymatically synthesized from UDP-[6-³H]GlcN. Early-stationary-phase Tn1 mutant cells were loaded with [GlcN-6-³H]MSH to a level of MSH content (9.8 μmol per g of RDW) typical of the parental strain and then incubated in MSH-free preconditioned medium at 37°C. Samples were taken at 0, 1, 4.8, and 25 h, and the distribution of radiolabel was determined in the medium, 50% acetonitrile extract, and in the cell extract pellet (Fig. 6A and B). Samples of the extract were analyzed for thiol after reduction with DTT and labeling with mBBr to determine MSH content and for amines after labeling with AccQFluor to determine GlcN and GlcN-Ins content.

FIG. 6.

Analysis of stationary phase MshC-deficient Tn1 mutant cells loaded with [GlcN-6-³H]MSH as a function of time during incubation in preconditioned medium at 37°C. Percentage of disintegrations per minute at time (t) zero for the total of all components (▪), MSH (•), and medium (○) (A) and for the cell extract pellet (▴), void fraction from thiol analysis (▾), and GlcN-Ins in amine analysis (⧫) (B). (C) Cell content of MSH (•) and of GlcN-Ins (⧫). (D) Cell content of void volume components in the thiol analysis (▾) and of the pellet fraction (▴) as calculated from [GlcN-6-³H]MSH specific activity. Data points represent single determinations.

All of the radioactivity present at time zero was accounted for at later times (99% ± 9%). At time zero the bulk of the radiolabel from [GlcN-6-³H]MSH was found in the cell extract (83%), and this declined to 28% during the 25-h incubation. As expected, little activity (3%) was found in the medium at time zero, but this increased markedly over the course of the incubation, corresponding to the loss of activity from the cell extract. A small amount (∼10%) was found in the insoluble residue from extraction of the cells (pellet), and this increased to ∼13% during the incubation. About 5% of the total activity was present in the wash of the cells prior to extraction and in the wash of the extracted pellet prior to determination of the RDW and subsequent counting.

HPLC analysis of the cell extract showed that the radioactivity in MSH (Fig. 6A) and the MSH content (Fig. 6C) fell in parallel during the incubation such that the specific activity remained constant at 0.93 ± 0.03 dpm/pmol over the course of the experiment, as expected in the absence of MSH synthesis. The half-life for MSH loss was 15 h. One other peak of radioactivity was seen in the thiol HPLC analysis after reduction of disulfides. It eluted in the void volume, indicating that it is not a thiol and therefore not labeled by mBBr (Fig. 6B). The molar content for the void component was calculated using the MSH specific activity and is shown in Fig. 6D. The GlcN-Ins content (Fig. 6C) determined by HPLC was ∼1 μmol per g of RDW and decreased only slightly with time. The level of radioactivity in GlcN-Ins was 0.4 dpm/pmol at time zero and uniformly increased with time to 0.9 dpm/pmol at the final time point, the latter value agreeing with that determined for MSH (0.93 dpm/pmol). This demonstrates that the rate of production of GlcN-Ins from MSH is substantially greater than its production via MshA, MshA2, and MshB (Fig. 1) under these conditions.

The increase in medium radioactivity might be explained by export of GlcN-Ins derived from MSH. However, in the experiment shown in Fig. 6, this would produce ∼6 μM GlcN-Ins in the medium at 25 h, and the measured level was <0.6 μM. Thus, export of GlcN-Ins is at most a minor process. The results are consistent with the metabolism of the GlcN residue of GlcN-Ins to intermediate substances not retained on a reversed-phase C₁₈ HPLC column after treatment with mBBr (Fig. 6, void component) and with these intermediates ultimately generating products excreted into the medium.

The fate of [Cys-U-¹⁴C]MSH in the MshC-deficient Tn1 mutant.

To examine the utilization of Cys derived from MSH, purified MshC and MshD were utilized to synthesize [Cys-U-¹⁴C]MSH from [U-¹⁴C]cysteine, GlcN-Ins, ATP, and acetyl-CoA (Fig. 1). The M. smegmatis Tn1 mutant was grown to stationary phase, loaded with [Cys-U-¹⁴C]MSH to a level of 9 μmol per g of RDW, and incubated in MSH-free preconditioned medium at 37°C. Samples were taken at 0, 1, 4.9, and 25 h and treated as described in the previous section. Two marked differences from the [GlcN-6-³H]MSH experiment are apparent in the data for [Cys-U-¹⁴C]MSH (Fig. 7A). First, over half of the radiolabel present at time zero is lost over 25 h, and, second, only a very small fraction of the counts end up in the medium. The only plausible explanation for the loss of ¹⁴C, distributed uniformly among the carbons of the Cys residue, is that it was metabolized to CO₂, but the small amounts of [Cys-U-¹⁴C]MSH available precluded a specific determination.

FIG. 7.

Analysis of stationary phase MshC-deficient Tn1 mutant cells loaded with [Cys-U-¹⁴C]MSH as a function of time during incubation in preconditioned medium at 37°C. (A) Percentage of disintegrations per minute at time (t) zero for the total of all components (▪), MSH (•), medium (○), cell extract pellet (▴), and void fraction from thiol analysis (▾). (B) MSH content from thiol analysis (•) and GlcN-Ins from amine analysis (⧫). (C) Cell extract pellet content (▴) and thiol analysis void volume content (▾) as calculated from [Cys-U-¹⁴C]MSH specific activity. Data points represent single determinations.

The analyses for MSH and GlcN-Ins produced results (Fig. 7B) very nearly the same as obtained in the experiment with [GlcN-6-³H]MSH (Fig. 6C), except that no radioactivity was found associated with GlcN-Ins. The half-life for MSH loss was 13 h. The specific activity calculated for MSH remained constant at 2.30 ± 0.05 dpm per pmol, and this value was used to quantitate the content in the pellet fraction and in the void components of the thiol analysis (Fig. 7C). The void components vary little over the course of the experiment and presumably consist largely of intermediates in the metabolism to generate CO₂.

Roughly 10% of the radioactivity is associated with the pellet after the 3-h loading with [GlcN-6-³H]MSH (Fig. 6B) and with [Cys-U-¹⁴C]MSH (Fig. 7A). This might suggest that MSH itself can become efficiently associated with protein in early stationary phase. A small increase in the pellet fraction occurs over the ensuing 25 h.

Utilization of [Cys-U-¹⁴C]MSH in dialyzed extracts of the MshC-deficient Tn1 mutant.

[Cys-U-¹⁴C]MSH was employed to trace the fate of Cys derived from MSH in studies using dialyzed cell extract. In a preliminary study it was found that 1 mM MSH produced GlcN-Ins at a rate of 1.3 ± 0.1 μmol per g of protein per h in the presence of dialyzed supernatant extract of the M. smegmatis Tn1 mutant and 1 mM NADPH at 37°C. The experiment was repeated using 1 mM [Cys-U-¹⁴C]MSH. Samples were taken at intervals and labeled with mBBr for thiol analysis and with AccQFluor for GlcN-Ins determination. Fractions were collected and counted during the HPLC analysis for thiols. The initial MSH content was ∼110 μmol per g of protein and decreased ∼10% during the 4.5-h incubation. Radioactivity in the peak for MSH, combined with the determination of MSH content by fluorescence, corresponded to a specific activity of 0.45 ± 0.01 dpm per pmol. A small peak for Cys was observed at 1 h and later intervals, and the mean specific activity was estimated at 0.50 ± 0.15 dpm per pmol. The remaining activity eluted in the void volume and was quantitated based upon the specific activity for MSH. The increase in the quantities with time is shown in Fig. 8. The increase in Cys plus the increase in void components amounted to 72% ± 2% of the increase in GlcN-Ins. The Cys increased most in the first hour and then changed little whereas the void components increased slowly at first and then more rapidly. This indicates that Cys is an intermediate in the production of the void components. Surprisingly, AcCys was not detected in these studies.

FIG. 8.

Increase in radiolabeled products from incubation of 1 mM [Cys-U-¹⁴C]MSH (∼100 μmol per g of protein) with dialyzed cell extract of the MshC-deficient Tn1 mutant at 37°C in the presence of 1 mM NADPH. •, GlcN-Ins from amine analysis; ▴, Cys from thiol analysis; ○, components in void volume from thiol analysis as calculated from [Cys-U-¹⁴C]MSH specific activity. Data points represent single determinations.

Conversion of AcCys to pyruvate.

Since cleavage of MSH by Mca generates GlcN-Ins plus AcCys but very little or no AcCys was detected in the thiol analyses, it was expected that the extract must also contain a high AcCys deacetylase activity. The dialyzed extract of the Tn1 mutant was assayed for AcCys deacetylase activity. A specific activity of 34 ± 4 (n = 2) nmol per min per mg of protein with 1 mM AcCys was determined with a 1:10 dilution of dialyzed extract. This corresponds to ∼50% per min for undiluted extract (10 mg per ml of protein) and indicates that the deacetylation must be rapid under cellular conditions. This is important since AcCys rapidly diffuses through the mycobacterial membrane and could not be retained in the cell. Immersion of cells of strain mc²155 in 10 mM AcCys was found to rapidly generate a cellular level of 30 μmol AcCys per g of RDW, the expected equilibrium value; this led to a 50-fold elevation of the Cys content and a marked suppression of the GlcN-Ins content (2), consistent with facile cellular deacetylation of AcCys.

Wheeler et al. (39) have measured a cysteine desulfhydrase activity in extracts of Mycobacterium bovis and M. tuberculosis that generates pyruvate as the product. Since conversion of Cys to pyruvate was a plausible step in conversion of Cys to CO₂, we assayed the dialyzed extract of the Tn1 mutant for cysteine desulfhydrase activity. At 1 mM Cys the rate of production of pyruvate was 1.5 ± 0.3 nmol per min per mg of protein (n = 2). This compares favorably with the rates 3.6 ± 2.7 and 2.9 ± 1.0 nmol per min per mg of protein reported for Mycobacterium bovis BCG and M. tuberculosis H37Rv, respectively, assayed at the same Cys concentration (39).

The half-life for MSH turnover is the same in M. smegmatis mc²155 as in strain 49.

The turnover of MSH in the parental strain mc²155 was examined using [U-¹⁴C]cysteine to generate [Cys-U-¹⁴C]MSH in situ in stationary phase. After unincorporated [¹⁴C]cysteine was washed away, the cells were resuspended in preconditioned medium lacking cysteine and sampled for analysis over 42 h at 37°C. Although the MSH content remained constant with time, the specific radioactivity of the MSH increased during the first 2 h, while that of Cys decreased sharply. The specific radioactivity of MSH decreased with further incubation. The loss of radioactivity from MSH paralleled the loss in total radioactivity (Fig. 9), consistent with the generation of CO₂ from the [¹⁴C]Cys derived from [Cys-U-¹⁴C]MSH.

FIG. 9.

Distribution of radioactivity and analysis of thiol content for stationary phase M. smegmatis mc²155 incubated with [U-¹⁴C]cysteine, resuspended in preconditioned medium without label, and analyzed as a function of time of incubation at 37°C with shaking. (A) Percentage of total counts at the 2-h time point (t = 2 h) for MSH (•), the cell pellet (▴), the void volume from thiol analysis (▾), the medium (○), and the total of all components (▪). (B) Cellular content of MSH (•) and Cys (□). Data points represent mean and range of values in three independent experiments.

Since MSH can be regenerated from the GlcN-Ins and Cys in mutant strain 49 and in strain mc²155 and since Mca activity on [Cys-U-¹⁴C]MSH generates [¹⁴C]Cys at a specific activity equal to that of the MSH, the loss of radioactivity from MSH is a measure of the MSH catabolism not including the portion recycled via MshC and MshD to regenerate MSH. Thus, the half-life for radioactivity loss from MSH in mc²155 should be equivalent to the half-life for MSH loss in mshA mutant 49 if the parental and mutant strains behave in a comparable fashion, and this proved to be the case. The half-life for loss of radioactivity from MSH in mc²155 (Fig. 9) was 49 ± 17 h. This value is comparable to the values (67 ± 13 and 37 ± 16 h) for loss of MSH from strain 49 (Fig. 4A). These findings demonstrate that the results found with the MSH biosynthesis mutants are consistent with the behavior of the parental strain mc²155.

DISCUSSION

We consider first the active transport of MSH into M. smegmatis. The observed uptake against a strong concentration gradient rules out uptake mediated by a porin system such as MspA (34). A possible candidate for the transporter is MSMEG1642, the homolog of Rv1747 that has been shown to be an ATP-dependent ABC transporter functional with glucosamine and maltose (R. Whalan and R. S. Buxton, personal communication). The faster rate of accumulation of MSH by MshA-defective strain 49 than by the MshC-deficient Tn1 mutant (Fig. 3) is too great to be explained by the difference in utilization rates (Fig. 5 versus Fig. 4) and may result from inhibition of the transport system in the MshC-deficient Tn1 mutant by its high intracellular GlcN-Ins level.

The most important findings of the present studies involve the turnover of MSH in stationary phase M. smegmatis. The key enzyme initiating catabolism of MSH is Mca (Fig. 10). It had previously been shown that purified Mca is able to cleave MSH to generate GlcN-Ins and AcCys (15, 33). From the MSH specific activity of Mca, it was estimated for M. smegmatis that the cellular rate for cleavage of 3 mM MSH would be ∼20% per h (15), or ∼2 μmol per g of RDW per h at 10 μmol MSH per g of RDW. The mean half-life for loss of MSH in Tn1 mutant cells in the various experiments reported here at near normal MSH content is 16 ± 3 h (n = 4), corresponding to a rate of ∼0.3 μmol per g of RDW per h. Thus, the expected activity of Mca is more than adequate to account for the observed rate. The generation of GlcN-Ins from MSH loaded into MshA-deficient strain 49 (Fig. 5), of labeled GlcN-Ins from [GlcN-6-³H]MSH in Tn1 mutant cells (Fig. 6), and of labeled Cys from [Cys-U-¹⁴C]MSH in dialyzed extracts of Tn1 (Fig. 8), together with the demonstration of high AcCys deacetylase activity in Tn1 mutant cell extracts, document the course of the degradation of MSH to produce GlcN-Ins and Cys as shown in Fig. 10.

FIG. 10.

Biosynthesis and degradative metabolism of MSH. MSH biosynthetic enzyme functions are defined in Fig. 1. Pdh, pyruvate dehydrogenase.

The results for utilization of [GlcN-6-³H]MSH in the Tn1 mutant show that all of the initial ³H label is accounted for and that most of the label lost from MSH appears in the medium (Fig. 6A). Two components present in the extract, GlcN-Ins and the void component (Fig. 6B), remain at relatively constant levels over the course of the experiment. They were generated during the 3-h loading of cells with [GlcN-6-³H]MSH, and the near constant values indicate that they represent steady-state intermediates in the metabolism of [GlcN-6-³H]MSH. The specific steps involved in the catabolism of GlcN-Ins have not been identified but presumably involve ultimate conversion of GlcN to pyruvate via the glycolytic pathway (38). myo-Inositol can serve as the sole carbon source for growth of M. smegmatis (K. Bzymek, G. Newton, and R. Fahey, unpublished results) so it also can be catabolized to generate energy and biosynthetic precursors. The 6-³H label of GlcN is predicted to end up in the methyl group of pyruvate and is transferred to acetyl-CoA by pyruvate dehydrogenase (Fig. 10, Pdh). Further metabolism via the citric acid cycle distributes the ³H to water, NAD(P)H, and reduced flavin adenine dinucleotide (38, 40). Transfer of electrons from NADH and reduced flavin adenine dinucleotide into the electron transport pathway leads to production and exchange of protons with water. Reduction of disulfides by NADPH generates thiols that can exchange protons with water. These processes explain why most of the label from [GlcN-6-³H]MSH becomes transferred to the medium in the case of the Tn1 mutant (Fig. 6A).

The results with [Cys-U-¹⁴C]MSH were different in several respects from those with [GlcN-6-³H]MSH but have, in part, a common metabolic explanation. With [Cys-U-¹⁴C]MSH there was a similar loss in labeled MSH to that found with [GlcN-6-³H]MSH, but little of the radioactivity appeared in the medium, and the loss from MSH was associated with a net loss of radioactivity from the cell culture (Fig. 7A). GlcN-Ins was produced at a level comparable to that in other experiments but was unlabeled (Fig. 7B). A peak of radioactivity associated with the HPLC void volume remained at a relatively constant level (Fig. 7A and C) and plausibly represents metabolites of the uniformly labeled [¹⁴C]Cys. Wheeler et al. (39) have shown that a cysteine desulfhydrase activity converting Cys to pyruvate (Fig. 10) is present in M. tuberculosis and M. bovis BCG, and this activity was shown here to be present in extracts of M. smegmatis. Once converted to pyruvate, the label from [¹⁴C]Cys is then lost as CO₂, first by the action of pyruvate dehydrogenase to produce acetyl-CoA plus CO₂ and subsequently via the Krebs cycle that, in multiple rounds, converts both carbons of the acetyl residue to CO₂ (Fig. 10). The failure to find significant activity from [Cys-U-¹⁴C]MSH in the culture medium rules out substantial conversion of acetyl-CoA to acetate and loss of the latter to the medium, a process not excluded by the results with [GlcN-6-³H]MSH.

A common feature of the [GlcN-6-³H]MSH and [Cys-U-¹⁴C]MSH studies was the finding of ∼10% of the total radioactivity associated with the pellet at time zero. This must have been generated during the 3-h loading period but increased only slightly during the subsequent 25-h incubation (Fig. 6B and 7A). The most plausible explanation of these results is that MSH itself becomes bound to protein in the stationary phase cells during the 3-h loading interval as the MSH content increases from zero to normal levels. Considerable recent interest has focused upon the glutathionylation of proteins to produce protein-glutathione disulfides thought to be involved in regulating enzyme activity (7, 8, 12). It is possible that analogous mycothiolation of proteins occurs in mycobacteria. Further studies are needed to substantiate this hypothesis.

It is clear from the present results that M. smegmatis has the catabolic activity to generate Cys plus GlcN-Ins from MSH as well as the ability to resynthesize MSH from these precursors (Fig. 10). This potentially creates a futile cycle wasteful of cellular resources and must be subject to regulatory controls that optimize these processes. In the MshC-deficient Tn1 mutant, the GlcN-Ins level accumulated to 1.5 to 2 μmol per g of RDW, a level apparently limited by the activity of the enzyme that degrades GlcN-Ins (Fig. 10). Although the pathway leading to GlcN-Ins is blocked in MshA-deficient strain 49, MshC can utilize Cys and GlcN-Ins to regenerate MSH with the aid of MshD. As a result the GlcN-Ins level falls to 0.15 μmol per g of RDW (Fig. 4) at a high MSH level and is markedly lower at normal MSH content. Thus, the GlcN-Ins level is influenced by at least four separate processes, two of which produce it and two others that utilize it. Degradation of GlcN-Ins likely generates GlcN and Ins, important components of the cell envelope (11, 14).

The present results show that MSH can serve as a reservoir for generation of Cys. Accumulation of intracellular Cys is undesirable since it undergoes rapid autoxidation producing peroxide as a toxic by-product. The amino, sulfhydryl, and carboxyl groups of Cys facilitate the binding of heavy metals that catalyze this autoxidation. Since the amino and carboxyl groups are blocked in glutathione, it undergoes autoxidation much more slowly than Cys (9, 35), and autoxidation of MSH is slower still (16). The incorporation of Cys into MSH represents a pathway for preserving Cys and limiting its concentration within the cell. The conversion of Cys to pyruvate by the desulfhydrase (Fig. 10) constitutes another pathway for Cys utilization that has been proposed by Wheeler et al. as a key process for keeping the Cys level low in M. tuberculosis (39). The sulfide produced by this reaction may be one of the resources utilized to reconstitute the sulfur centers of WhiB-type regulators (1, 32). The biosynthesis of Cys from sulfate and methionine and the utilization of Cys in protein and CoA biosynthesis are additional processes that influence the pool of Cys. It appears that a complex regulatory system is required to coordinate the multiple Cys-generating processes with the many demands for Cys utilization while keeping the level of Cys low. The ability to incorporate Cys into MSH and to rapidly regenerate it therefrom is an important component of this system.

The original goal of these studies was to obtain an estimate of the turnover rate of MSH in nonreplicating mycobacteria. The results show that the half-life for turnover of MSH in early-stationary-phase M. smegmatis cells is ∼50 h in cells with active MshC where resynthesis of MSH from GlcN-Ins and Cys is possible. However, in the MshC-deficient Tn1 mutant, where such resynthesis is blocked, the half-life of MSH is reduced to 16 h. To the extent that these values are reliable indicators of the lifetime of MSH in dormant M. tuberculosis, MSH biosynthesis could be a useful target for drugs to treat dormant tuberculosis and reduce treatment times.

In conclusion, a function for MSH as a readily accessible storage form for cysteine and GlcN-Ins has been demonstrated. These components of MSH can be catabolized under stress conditions to generate a host of valuable intermediates to meet biosynthetic requirements or be fully metabolized for energy production. Further studies are needed to identify the genes encoding the N-acetylcysteine deacetylase and the cysteine desulfhydrase and to establish the initial biochemical steps involved in GlcN-Ins catabolism.