Reduction of a heme cofactor initiates N-nitroglycine degradation by NnlA

Abstract

The NnlA enzyme from Variovorax sp. strain JS1663 degrades the linear nitramine N-nitroglycine (NNG)—a natural product produced by some bacteria—to glyoxylate and nitrite. Ammonium (NH4+) was predicted as the third product of this reaction. A source of non-heme Fe(II) was shown to be required for initiation of NnlA activity. However, it was unclear if this Fe(II) was being used as a metallocofactor or a reductant. This study reveals that NnlA contains a b-type heme cofactor. Reduction of this heme is required to initiate NnlA activity. Reduction can occur either by addition of a non-heme Fe(II) source or by reduction with dithionite. Therefore, Fe(II) is not an essential substrate for holoenzyme activity. Data are presented showing that reduced NnlA [Fe(II)-NnlA] can catalyze at least 100 turnovers. In addition, this catalysis occurred in the absence of O2. Finally, NH4+ was verified as the third product, accounting for the complete nitrogen mass balance. Size exclusion chromatography showed that NnlA is a dimer in solution. Additionally, FeII-NnlA is oxidized by O2 and nitrite and binds carbon monoxide (CO) and nitric oxide (NO). These are characteristics shared with PAS domains; NnlA was previously shown to exhibit homology with such domains. Providing further evidence, a structural homology model of NnlA was generated based on the structure of the PAS domain from Pseudomonas aeruginosa Aer2. The structural homology model suggested His73 is the axial ligand of the NnlA heme. Site-directed mutagenesis of His73 to alanine decreased the heme occupancy of NnlA and eliminated NNG activity, providing evidence that the homology model is valid. We conclude that NnlA forms a homodimeric heme-binding PAS domain protein that requires reduction for initiation of the activity.

Activity assays of NnlA homologs suggest the natural product N-nitroglycine is degraded by diverse bacteria

“…To characterize these oligomer states of native protein, analytical size exclusion chromatography data were collected (Figure 2B). As previously reported, Vs NnlA exhibited two major peaks, a lower molecular weight peak consistent with a dimer and a second peak consistent with a large oligomer [21]. By contrast, the chromatograms of the purified Ps, Mr, Ms, and Pd NnlA were dominated by a single peak ranging in molecular mass from 35.9 to 49.0 kDa, masses consistent with dimers (Table S2 in Supporting Information File 1).…”

“…By this method, the His73 is located near the heme, likely acting as its proximal ligand. As described above, the H73A Vs NnlA variant lacked NNG degradation activity and had reduced iron content [21]. The AlphaFold model also predicts a distal pocket composed of Tyr99, Ala101, Ile109, Val111, and Gln127.…”

“…A preliminary screen of these homologs for NNG degradation activity was performed to identify homologs for further characterization. The H73A Vs NnlA variant, previously shown to lack NNG degradation activity, was used as a negative control [21]. E. coli transformants containing the expression vectors were incubated at 37 °C overnight in diluted lysogeny broth (LB) containing NNG and IPTG, the latter component was used to induce NnlA expression.…”

“…The discovery of the nnlA gene resulted from screening a JS1663 genomic library by monitoring for E. coli transformants that produced nitrite (NO 2 − ) in the presence of NNG. Analysis of the mass balance via in vitro experiments showed that NnlA degraded NNG into NO 2 − , ammonium (NH 4 + ), and glyoxylate (Scheme 1) [20,21]. Vs NnlA contains a Per-Arnt-Sim (PAS) domain -protein domains that often bind heme and function as gas or redox sensors [20].…”

“…Vs NnlA contains a Per-Arnt-Sim (PAS) domain -protein domains that often bind heme and function as gas or redox sensors [20]. Indeed, Vs NnlA was shown to contain a heme cofactor [21]. Mutagenesis of a predicted histidine ligand to this heme resulted in loss of the heme and the variant could not degrade NNG.…”