Metabolism of the Aliphatic Nitramine 4-Nitro-2,4-Diazabutanal by Methylobacterium sp. Strain JS178

Organisms and growth conditions.The aerobic RDX degrader Rhodococcus sp. strain DN22 and Methylobacterium strain JS178 were cultivated in M-succinate (with 20 mM of succinate) as previously described (9). Bacteria were incubated at 30°C in a rotary shaker with agitation at 175 rpm.

Chemicals.RDX (>99% purity), [U-¹⁴C]RDX (98% purity), HMX (99% purity), NDAB, and methylenedinitramine (MEDINA) were obtained as described previously (9).

Enrichment and isolation of aerobic bacteria able to degrade NDAB.A sample of sludge taken from an aerobic treatment plant at Tyndall Air Force Base (AFB), Fla., a garden soil from Washington, D.C. (pH 6.9), a garden soil from Lynn Haven, Fla. (pH 6.8), a forest soil from Tyndall AFB (pH 4.9), and an agricultural soil from a farm near Memphis, Tenn. (pH 7.8), were used as inocula (20% [wt/vol] soil slurries in water) for shake flask suspensions tested for the ability to degrade NDAB (100 μM). Slurries were incubated at room temperature under aerobic conditions, and NDAB concentrations were measured at intervals by high-performance liquid chromatography, as described previously (9). Suspensions in which NDAB disappeared were subjected to selective enrichment consisting of sequential transfer to M medium amended with 20 mM pyruvate and increasing concentrations of NDAB (100 to 400 μM) as the nitrogen source. Dilutions of the final suspension were plated on the above medium solidified with agarose (1.5%, wt/vol). Plates were incubated at 30°C, and a representative of the predominant small pink colonies was purified and designated strain JS178.

Mineralization assay.[¹⁴C]NDAB was prepared from [¹⁴C]RDX in M-succinate medium inoculated with RDX-induced strain DN22 (9). When the mineralization reached a plateau (30% of the radiolabel released as ¹⁴CO₂ and 64% remaining as [¹⁴C]NDAB), DN22 cells were removed by filtration through a 0.22-μm-pore-size membrane (Millex-GP; Millipore, Bedford, Mass.) and the culture supernatant was inoculated with NDAB-grown Methylobacterium sp. strain JS178 to an initial A₆₀₀ of 0.05. Parallel sets of microcosms were prepared with [¹⁴C]RDX in M-succinate medium inoculated either with NDAB-induced strain JS178 alone or with both NDAB-induced strain JS178 and RDX-induced strain DN22. At each sampling, microcosms were aerated with 60 ml of 0.20-μm-filtered air (polytetrafluoroethylene membrane; Cole-Parmer, Vernon Hills, IL).

Growth of strain JS178 with NDAB as the sole N source.Cultures were performed in 10 ml of M-succinate medium with NDAB as the sole nitrogen source (150 μM or as indicated) in 125-ml serum bottles. The inoculum consisted of the addition of a washed stationary-phase JS178 culture grown in nutrient broth to an initial A₆₀₀ of 0.04. The bottles were sealed with Teflon-coated septa and crimped with aluminum caps to allow the sampling of the headspace for N₂O analyses. The cultures were incubated as described above and were monitored for growth (A₆₀₀), residual NDAB, and possible metabolites such as nitrite (NO₂⁻), nitrate (NO₃⁻), formaldehyde (HCHO), methanol (CH₃OH), formate (HCOO⁻), ammonium (NH₄⁺), formamide (NH₂CHO), and nitramide (NH₂NO₂). Analyses were performed as described previously (9).

Resting-cell assays.Resting-cell assays were performed in sealed 6-ml vials containing M medium and NDAB (150 μM), without the addition of a carbon source. The biotransformation reaction (1 ml) was started with the addition (to a final A₆₀₀ of 1.0) of strain JS718 (pregrown in M-succinate and various N sources, harvested at early to mid-log phase, and washed twice in M medium).

Molecular biology techniques.DNA was extracted from a JS178 colony grown on an NB plate, and the 16S rRNA gene was amplified and sequenced (2). Sequence comparisons were done using BLASTN (1).

Nucleotide sequence accession number.The 16S rRNA gene sequence of isolate JS178 was deposited in GenBank under accession number AY795971 .

Isolation of strain JS178.Selective enrichment was performed to isolate an aerobic bacterium that could use the RDX ring cleavage product NDAB as a nitrogen source. Among the several soil and wastewater sludge samples tested, the rate of NDAB disappearance in Tyndall AFB forest soil was twice that in the other soils (approximately 100 μM within 10 days). The Tyndall soil was chosen for the enrichment of an aerobic bacterium (strain JS178) that is able to grow using NDAB as the sole source of nitrogen. Strain JS178 is not the only microorganism that is able to transform NDAB (9), but it is the first that is able to grow at the expense of the nitramine.

Characterization of strain JS178.Strain JS178 grew on solidified M medium with succinate and NDAB, as pink-pigmented (nondiffusible) small colonies (diameter, 1 to 3 mm). The strain is gram negative, nonsporulating, rod-shaped, aerobic, and catalase and oxidase positive. In liquid culture, a ring of biomass formed around the wall of the incubation flask. Strain JS178 was able to grow on methanol as the sole C source. JS178 satisfied the morphological and biochemical criteria consistent with the genus Methylobacterium.

The 16S rRNA sequence comparisons revealed a close affiliation (99% similarity) of strain JS178 with Methylobacterium populi (19). A number of compounds, including substrates commonly used in the differentiation of Methylobacterium species (11), were tested as potential carbon sources. The results with strain JS178 were identical to those with M. populi. Both strains are able to use methanol, formate, acetate, betaine, ethanol, fructose, succinate, pyruvate, and methylamine (19). On the other hand, JS178 and M. populi did not grow on d-glucose, d-fucose, d-xylose, l-arabinose, or citrate.

Growth of strain JS178 on NDAB.In M-succinate medium with NDAB as the sole N source, the growth of JS178 was concomitant with the degradation of the nitramine and the production of N₂O (Fig. 2). No nitrite, nitrate, ammonium, nitramide, formaldehyde, formate, or formamide was detected during the degradation of NDAB by JS178. No methanol was detected, but since the initial concentration of NDAB was 150 μM, the maximum theoretical yield of methanol is 300 μM, which corresponds to the detection limit of the method used. The doubling time was approximately 1.8 days, and after 4 days of incubation, it significantly increased, probably due to oxygen limitation in the serum bottles. The degradation of NDAB and the production of N₂O were reflected in the growth kinetics, and the ratio of N₂O produced to NDAB consumed was 0.9 at both 4 and 7 days of incubation, clearly showing that approximately 1 mol of N₂O, or 2 N equivalents, was released per mol of NDAB degraded.

• Open in new tab
• Download powerpoint

FIG. 2.

Growth of Methylobacterium sp. strain JS178 in M-succinate and NDAB (1,300 nmol) as a sole N source. Shown are amounts of NDAB (▵), N₂O (□), and NO₂⁻ (×). Growth is expressed as A₆₀₀ (○). Error bars, standard deviations from duplicate experiments.

Number of nitrogen atoms assimilated from NDAB.The specific growth yields obtained with NDAB, NaNO₂, and (NH₄)₂SO₄ (Fig. 3) were 0.00173 (R = 0.999), 0.00165 (R = 0.995), and 0.00158 (R = 0.999) A₆₀₀ unit/μmol of nitrogen substrate, respectively. The results clearly indicate that 1 N atom of a total of 3 in NDAB (Fig. 1) was assimilated during the growth of JS178.

• Open in new tab
• Download powerpoint

FIG. 3.

Growth yield of Methylobacterium sp. strain JS178 in M-succinate, with nitrite (▵), ammonium (□), and NDAB (○) as N sources. The linear regressions have the following gradients: 0.00165 (R = 0.995) for nitrite, 0.00158 (R = 0.999) for ammonium, and 0.00173 (R = 0.999) for NDAB. Error bars, standard deviations from triplicate experiments.

Number of carbon atoms released from NDAB.The number of carbon atoms released during the degradation of NDAB was investigated as described previously (9). Briefly, [¹⁴C]RDX was incubated in M-succinate with Rhodococcus sp. strain DN22 until approximately 30% of the radiolabel was liberated as CO₂. At this stage, RDX was totally depleted from the culture medium, and 92 to 96% of the [¹⁴C]RDX added initially was transformed into [¹⁴C]NDAB (9). A longer incubation period with strain DN22 did not result in more CO₂ production, even when more succinate (10 mM) was added to the medium (Fig. 4). The results confirm the inability of strain DN22 to degrade the NDAB produced. The incubation of [¹⁴C]RDX with Methylobacterium sp. strain JS178 did not lead to production of CO₂, which indicated that it is unable to degrade RDX (Fig. 4). Compared to the incubation of [¹⁴C]RDX with DN22, the simultaneous inoculation of M-succinate-[¹⁴C]RDX medium with both strains DN22 and JS178 resulted in only a slight increase in CO₂ production. After the reinoculation of the microcosms with JS178 (performed at day 42), the CO₂ augmentation was more perceptible but remained low (Fig. 4). This could be explained by the occurrence of competition between the two bacterial species, possibly for carbon or for oxygen. However, when NDAB-grown Methylobacterium sp. strain JS178 was added to a 36-day-old filtered DN22 [¹⁴C]RDX culture supernatant, the liberation of CO₂ increased from 31.1% ± 0.1% to 65.5% ± 2.4% after 26 additional days of incubation, indicating mineralization of NDAB. In these sequential cultures, the following carbon mass balance was calculated after 108 days of incubation: 66.9% ± 3.2% of the carbon evolved as CO₂, 1.2% ± 0.1% was incorporated into the biomass, and 26.3% ± 1.4% remained in the liquid phase. As mentioned above, the carbon remaining in the liquid phase is possibly methanol or unavailable formaldehyde; the latter is known to form covalent bonds or polymerize with a wide range of chemical species.

• Open in new tab
• Download powerpoint

FIG. 4.

Incubation of [¹⁴C]RDX with Rhodococcus sp. strain DN22 (medium was supplemented with 10 mM succinate at day 25) (○), with Methylobacterium sp. strain JS178 (×), with both strains DN22 and JS178 (JS178 was reinoculated to the microcosms at day 42) (▵), and with the sequential addition of strain DN22 (used to transform [¹⁴C]RDX to [¹⁴C]NDAB) and NDAB-grown JS178 (added when 30% of [¹⁴C]CO₂ was liberated from [¹⁴C]RDX, after the removal of DN22 cells by filtration) (□). Values are averages and standard deviations of duplicate experiments.

Regulation of NDAB degradation in strain JS178.To identify which nitrogen sources are utilized by isolate JS178, we incubated it in M-succinate with various nitrogen compounds. Nitrogen sources that supported growth were then tested for their ability to up-regulate the degradation of NDAB in resting-cell assays. The maximal degradation rate was defined as that obtained with resting JS178 pregrown in M-succinate and NDAB (2.52 μmol/liter/h ± 15%).

The growth of strain JS178 was strongly positive in M-succinate with NaNO₂, (NH₄)₂SO₄, or urea. However, none of the compounds induced NDAB degradation.

RDX and HMX did not support growth of JS178. As reported for M. populi (20), strain JS178 was able to transform RDX only in a rich medium, via a cometabolic process. For both strains, the detection of an RDX mono-nitroso derivative (MNX) and MEDINA indicates that an RDX reductive attack, rather than the known aerobic mechanism based on denitration, is occurring (data not shown) (20).

To test whether an amidohydrolase catalyzes the degradation of NDAB (which contains an amide linkage, R-NH-CHO [Fig. 1]), cells were grown on allantoin, an intermediate in the purine catabolic pathway known to require an amidohydrolase for its decomposition by Escherichia coli (7). The compound supported good growth of JS178. The NDAB degradation rate produced by allantoin was as high as that produced by NDAB.

The possibility that a purine degradative enzyme(s) might be involved in the degradation of NDAB is particularly interesting, because it may lead to a better understanding of the commensalism reported to occur between plants and methylobacteria (16). Allantoin is known to be a key intermediate in the assimilation, metabolism, transport, and storage of nitrogen in plants (Fig. 1) (17). The fact that allantoin stimulates the synthesis of an enzyme(s) that degrades NDAB suggests that the same enzymes are involved in both pathways. We are currently investigating the relationship between the enzymes involved in allantoin and NDAB degradation.

In conclusion, the biotic and abiotic cyclic nitramine ring cleavage product NDAB is degradable by the soil isolate Methylobacterium sp. strain JS178. NDAB served as a source of nitrogen for growth of the isolate and was converted to N₂O and CO₂. Our preliminary results indicate that one or several amidohydrolases, naturally involved in the purine catabolic pathway, might be involved in the degradation of the xenobiotic compound. The results reveal the potential for the sequential participation of rhodococci and methylobacteria to effect the complete degradation of RDX. Methylobacterium species might play a major role in phytoremediation of nitramine compounds, especially in view of the recent report that NDAB formed in leaves of hydroponic cultures of reed canary grass contaminated with RDX and exposed to simulated sunlight (15). The role of methylobacteria as endophytes would be particularly relevant.