In vivo metal selectivity of metal-dependent biosynthesis of cobalt-type nitrile hydratase in Rhodococcus bacteria: a new look at the nitrile hydratase maturation mechanism?^†

Abstract

This study highlights the effect of heavy metal ions on the expression of cobalt-containing nitrile hydratase (NHase) in Rhodococcus strains, which over-produce this enzyme. Both metal-dependent derepression of transcription and maturation of NHase were considered. We demonstrated that nickel ions can derepress the NHase promoter in several Rhodococcus strains. The cblA gene of a cobalt-dependent transcriptional repressor was shown to be indispensable for nickel-mediated derepression. As for maturation, we showed that nickel ions could not replace cobalt ions during the synthesis of active NHase. We also revealed that the amount of β-subunit decreased during NHase expression without added cobalt. We showed this using three variants of NHase in vivo synthesis: by using nickel- or urea-induced synthesis in cblA⁺ strains, and by using metal-independent constitutive synthesis in cblA⁻ strains. In all cases, we found that the amount of β-subunit was significantly lower than the amount of α-subunit. In contrast, equimolar amounts of both subunits were synthesized after growth in the presence of added cobalt. Nickel did not affect NHase synthesis in mixtures with cobalt. This suggests that the metal selectivity in cblA-dependent regulation of NHase transcription was too low to discriminate between cobalt and nickel, but the selectivity of the NHase maturation mechanism was high enough to do so. Moreover, we can assume that the β-subunit is more subject to proteolytic degradation without the addition of cobalt, than the α-subunit. This indicates that cobalt ions presumably play an unknown role in the stability of the β-subunit in vivo.

Graphical Abstract

Metal-dependent cblA-mediated mechanism of transcription regulation of NHase could not discriminate Ni and Co, but mechanism of NHase enzyme maturation could do this.

Significance to metallomics

The work provides deeper insight into the mechanisms of regulation of the highly expressing and biotechnologically important enzyme, nitrile hydratase. Probing the in vivo metal selectivity of cobalt-dependent transcriptional regulation and NHase maturation enabled us to reveal different “resistance” mechanisms to competing metal ions. The unknown role of cobalt ions in the stability of NHase subunits (particularly β) was uncovered. The findings reported here will help to further investigate the metallation of NHase, and contribute to understanding the mechanisms of correct metal insertion into metalloenzymes.

1. Introduction

Nitrile hydratases (NHases) are heteromeric metal-containing enzymes, which are able to transform organic nitriles, compounds containing the –CN group, into amides, compounds containing the –CONH₂ group.¹ NHases are widespread in nature,² but have been experimentally studied mostly in eubacteria.³ The metal ion (Fe or Co) in NHases is required for catalysis.⁴

The roles of bacterial NHases in nature have not been satisfactorily elucidated yet. It is presumed that they exist to utilize naturally occurring nitriles, which are widespread as secondary metabolites of plants.^5–7 Due to their high and inducible synthesis in response to the presence of nitriles in growth media, NHases are expected to play a role in plant-microbial interactions. Additionally, due to the fact that a lot of herbicides are nitriles, NHases might play a significant role in their degradation in soil. In industry, NHases are used as biocatalysts for the biotechnological production of acrylamide from acrylonitrile, and nicotinamide from 3-cyanopyridine.⁸ A growing number of studies today are aimed at expanding the currently available set of NHases in order to meet the different needs of the chemical and pharmaceutical industries.^9,10

Co-NHases can be synthesized in wild type bacterial strains up to approx. 50% of soluble cell proteins.¹¹ The mechanisms of regulation of this high-level expression are a rather interesting but poorly studied topic. The expression of Co-NHase genes nhmBAG (Fig. 1) in R. rhodochrous M8 is subjected to several independent regulatory mechanisms, including catabolite repression, induction by substrates, and cobalt-dependent activation.^12–14 Cobalt derepresses transcription, and is required for the biosynthesis of the NHase holoenzyme (post-translationally modified enzyme with the cobalt ion incorporated within the active center¹⁵). The transcription of NHase in R. rhodochrous M8 and M33 strains is derepressed by cobalt via the action of ArsR-type cobalt-dependent transcriptional regulator cblA (Fig. 1).¹⁴

Fig. 1

Schematic representation of the structure of NHase gene clusters in R. rhodochrous strains used in this work. M8: R. rhodochrous strain M8; AC-2063: R. aetherivorans strain VKPM AC-2063; M33: R. rhodochrous strain M33; M33 aam: R. rhodochrous strain M33 aam; J1: R. rhodochrous strain J1. nhmCD/nhhCD: substrate dependent transcriptional regulator genes; aam: acylamidase gene from R. erythropolis TA37; nhmBA/nhhBA: NHase subunit genes; nhmG/nhhG: NHase accessory protein (cobalt metal-chaperone) genes; cblA/ncoR: cobalt-dependent transcriptional regulator genes.

An important stage of the biosynthesis of holo-NHase is the post-translational maturation, which includes the metallation of apo-NHase. A novel self-subunit swapping mechanism was proposed for this process by Zhou and coworkers.¹⁵ According to them, NHase accessory protein (coded by nhmG/nhhG in M8/J1, Fig. 1) forms a complex with the α-subunit and catalyses coordination of cobalt in the α-subunit in the first stage. Secondly, this holo-α-subunit is transferred from this complex to the complex with the β-subunit, resulting in the formation of functional holo-NHase.

We suppose that both the mechanism of metal-dependent transcription regulation and the mechanism of NHase metallation can be selective for cobalt ions. Such selectivity can be physiologically significant for NHase-producing bacteria, due to known facts of inactivating metalloproteins by non-cognate metals.¹⁶ To the best of our knowledge, the in vivo metal selectivity of both mechanisms has not been studied till date. It is only known that the enzyme is fully inactive when NHase-producing bacteria grow in the presence of other ions instead of cobalt ions in nutrient media.^13,17

Thus, the main objective of this work was to estimate the in vivo metal selectivity of the two above-mentioned NHase regulation mechanisms in R. rhodochrous. Our approach was to reveal the effect of heavy metal ions on the cblA-dependent activity of the NHase promoter, clarify the separate roles of two downstream genes (nhmG and cblA) in this regulation, and study the effect of heavy metal ions on the formation of the Co-NHase enzyme.

2. Methods

2.1. Bacterial strains, plasmids, and growth conditions

The strains and plasmids, used in this work, are listed in Table 1.

Table 1

Strains and plasmids used in this work

Strain or plasmid	Characteristic	Source
Strains
R. rhodochrous M8	nhmCDBAG cblA, natural soil isolate, contains the cluster of Co-type NHase structural and regulator genes	13
R. rhodochrous M33	ΔnhmCD nhmBAG cblA, derivative of R. rhodochrous M8	14
R. rhodochrous M33 aam	ΔnhmCD ΔnhmBA::aam nhmG cblA, derivative of R. rhodochrous M33, in which NHase genes nhmBA substituted with acylamidase gene aam from R. erythropolis TA37	14
R. rhodochrous M33 delta	ΔnhmCDBAG ΔcblA, derivative of R. rhodochrous M33	This work
R. rhodochrous M33 delta-1	ΔnhmCD aam nhmG cblA, derivative of R. rhodochrous M33 delta	This work
R. rhodochrous M33 delta-6	ΔnhmCD aam nhmG, derivative of R. rhodochrous M33 delta	This work
R. rhodochrous M33 delta-8	ΔnhmCD aam cblA, derivative of R. rhodochrous M33 delta	This work
R. rhodochrous M33 delta-const	ΔnhmCD nhmBAG+cblA−, derivative of R. rhodochrous M33 delta	This work
R. rhodochrous J1	nhhCDBAG ncoR, natural soil isolate, contains the cluster of Co-type NHase structural and regulator genes	Received from Dr M. Kobayashi18
R. aetherivorans VKPM AC-2063	nhmCDBAG cblA, natural soil isolate, contains the cluster of Co-type NHase structural and regulator genes	Russian National Collection of Industrial Microorganisms (http://eng.genetika.ru/service-offer/vkpm/), number VKPM-AC-2063 (the same as VKM AC2610D19)
E. coli XL1 Blue	recA1 endA1 gyrA96 thi-1 hsdR17 supE44 relA1 lac [F′ proAB lacIqZΔM15 Tn10 (Tetr)] was used as a host for gene cloning procedures	Obtained from Stratagen
E. coli S17-1	TpR SmR recA, thi, pro, hsdR-M + RP4:2-TcMu:KmTn7 λpir, was used for conjugative transfer of plasmids to R. rhodochrous strains	Obtained from Stratagen
Plasmids
pRY1	pTZ19R-based vector, used for integration into the R. rhodochrous chromosome (contains only pMB1 replicon for replication in E. coli), Ampr Tsrr	20
pRY1-1	Derivative of pRY1 with the Pnh-aam-nhmG-cblA cassette for integration into the R. rhodochrous chromosome, Ampr Tsrr	This work
pRY1-6	Derivative of pRY1 with the Pnh-aam-nhmG cassette for integration into the R. rhodochrous chromosome, Ampr Tsrr	This work
pRY1-8	Derivative of pRY1 with the Pnh-aam-cblA cassette for integration into the R. rhodochrous chromosome, Ampr Tsrr	This work

Rhodococcus strains were cultured in a liquid mineral synthetic (MS) medium of the following composition: glucose, 5 g l⁻¹; Na₂HPO₄·12H₂O, 2.8 g l⁻¹; KH₂PO₄, 0.8 g l⁻¹; MgSO₄·7H₂O, 0.5 g l⁻¹; FeSO₄·7H₂O, 0.005 g l⁻¹. NH₄NO₃ (2 g l⁻¹) or NH₄Cl (2 g l⁻¹) or urea (6 g l⁻¹) was used as the source of nitrogen. In some cases, the metal ions were added to the cultures in the form of the following salts: CoCl₂·6H₂O, NiCl₂·6H₂O, ZnSO₄·7H₂O, CdCl₂·2,5H₂O, Pb(CH₃COO)₂·3H₂O, NaAsO₂, CuSO₄·5H₂O. The additions were calculated to obtain certain molar concentration of particular metal ion in the growth media. Selection of recombinant R. rhodochrous variants after conjugation was conducted on LB medium of the following composition: yeast extract, 5 g l⁻¹; tryptone, 10 g l⁻¹; NaCl, 5 g l⁻¹, with antibiotics in the following concentrations: thiostrepton (25 μg ml⁻¹) and nalidixic acid (10 μg ml⁻¹). E. coli strains with plasmids were grown on LB medium with ampicillin (100 μg ml⁻¹). Growth of the R. rhodochrous cultures was conducted at 30 °C, and E. coli was grown at 37 °C, with constant mixing (300 rpm) in the case of liquid cultures.

The purified water, used for growth experiments, contained no more than 0.1 ppm of heavy metals in total. Additives to growth media were of analytical grade or higher, and contained no more than 0.007% of heavy metals in total. Considering these background concentrations of heavy metals, the medium could contain no more than 0.2 μM of heavy metals in total (calculations are not shown).

2.2. Sequences of genes used in this study

All sequences are available in NCBI GenBank under the following accession numbers: AY654301.1 — NHase cluster nhmCDBAG from R. rhodochrous M8, MLYX02000005.1 – the contig of the R. rhodochrous M8 whole genome shotgun sequence, containing this NHase cluster and also cblA (nucleotide positions 59473–64431), D67027.1 – NHase cluster nhhCDBAG from R. rhodochrous J1, LC385649.1 – NHase cluster nhmCDBAGcblA from R. aetherivorans VKPM AC-2063, LC379280.1 – ncoR gene for arsR-type transcription regulator (homologue of cblA) from R. rhodochrous J1, JX894244.1 – acylamidase gene aam from R. erythropolis TA37, OLL19885.1 – DNA gyrase subunit B from R. rhodochrous M8, LC384630 – DNA gyrase subunit B from R. rhodochrous J1.

2.3. Molecular genetic methods

The conventional techniques of molecular biology (restriction, ligation and PCR amplification) were performed according to the recommendations of the enzymes’ manufacturer (ThermoFisher Scientific, Waltham, MA, USA). Chromosomal DNA isolation was conducted as described in ref. 21 and plasmid DNA was isolated using the GeneJet Plasmid Miniprep Kit (ThermoFisher Scientific). Plasmids were introduced into E. coli cells by electroporation using a Gene Pulser Xcell electroporator (BioRad). All stages of plasmid construction were confirmed by sequencing. DNA sequencing was done using an automatic ABI PRISM 3500 sequencer (ThermoFisher Scientific) from the Joint Use Centre in the Research Institute for Genetics and Selection of Industrial Microorganisms.

2.4. SDS-PAGE and MALDI-TOF analysis of proteins

The cells were washed twice with 0.1 M phosphate buffer (pH 7.5), resuspended in the same buffer with the addition of 5% acetamide (for stabilization of NHase), and sonicated on Soniprep 150 (Sanyo, Japan). Cell debris was removed by centrifugation, and the supernatant was used as the cell-free extract for SDS-PAGE analysis. Approximately 10 μg of total soluble protein was loaded in one well, and relative amounts of NHase were estimated visually by Coomassie Blue stained SDS-PAGE electrophoresis, performed according to Laemmli.²² Target protein bands were excised from the gel for mass spectrometric measurements. The latter were performed using a MALDI-TOF mass-spectrometer (Ultraflex II, Bruker Daltonics, USA), in “Human Proteome” Core Facility (http://proteocenter.ibmc.msk.ru/).

2.5. Determination of acylamidase activity of cells

The cells were washed twice with 0.01 M Tris-HCl buffer (pH 8.0) and resuspended in the same buffer. To measure the activity, 150 μl of the cell suspension was mixed with 150 μl of 1 × 10⁻³ M 4′-nitroacetanilide solution (p-nitroanilide of acetic acid) and incubated for 20 min at 37 °C. Thereafter, the reaction was stopped by centrifugation at 0 °C. The supernatants were separated from the cells, and the concentrations of the produced p-nitroaniline in the supernatants were determined spectrophotometrically at 410 nm (the molar extinction coefficient of p-nitroaniline was 8900 mol⁻¹ cm⁻¹). The specific acylamidase activities were expressed as the amount (μM) of p-nitroaniline synthesized by 1 mg of dry cells per minute (μM per min per mg cdw).

2.6. Determination of NHase activity of cells

The cells were washed twice with 10 mM phosphate buffer (pH 7.6) and resuspended in the same buffer. NHase activity was assayed in the reaction mixture (500 μl) consisting of 10 mM phosphate buffer (pH 7.6), 188 mM acrylonitrile, and an appropriate concentration of cells. The reaction was carried out at 20 °C for 20 min and stopped by the addition of 10 μl of HCl. The amount of acrylamide formed was determined by GC. The specific NHase activities were expressed as the amount (μM) of acrylamide synthesized by 1 mg of dry cells per minute (μM per min per mg cdw).

2.7. Quantitative real-time PCR (qRT-PCR) analysis

Total RNAs were extracted from Rhodococcus cells by using the RNeasy Mini Kit (QIAGEN, Hilden, Germany) with a step of cell lysis using liquid nitrogen. The extracted RNA was treated with DNase I (ThermoFisher Scientific) to thoroughly remove residual genomic DNA. The synthesis of cDNA was performed by using the MMLV RT kit (Eurogen, Russia) according to the manufacturer's protocol. Quantitative analyses of the transcripts were performed on an Applied Biosystems 7500 Fast Real-Time PCR System (ThermoFisher Scientific) by using qPCRmix-HS SYBR + LowRox (Eurogen, Russia). A typical 20 μl PCR mixture contained 15 μl of SQ, 2 μl of qPCR mix-HS SYBR + LowRox, 1 μl of serially diluted template cDNA (<200 ng), and 1 μl of each primer. Typical cycling conditions were 95 °C for 5 min, followed by 40 cycles at 95 °C for 30 s, 60 °C for 20 s, and 72 °C for 30 s. The amounts of a specific mRNA target were normalized to the amounts of a housekeeping gene gyrB in different Rhodococcus strains, using the gyrB-M8-F, gyrB-M8-R, gyrB-J1-F, and gyrB-J1-R primers. Relative activities of the NHase promoter in all strains were determined by amplification of fragments of nhmG/nhhG genes cDNA, using the nhmG-F and nhmG-R primers. Relative amounts of mRNA were determined using the 2^−ΔΔCT method²³ with the standard calculation algorithms provided by 7500 Fast Software version 2.3 (Applied Biosystems). The quantification of target genes was performed after four or more independent growth experiments and the results were represented as an averaged value ± standard deviation (SD). All sequences of the primers are listed in Table S1, ESI.†

2.8. Construction of recombinant R. rhodochrous strains

R. rhodochrous strain M33 delta, the basic strain for construction of deletion variants, was obtained from the R. rhodochrous M33 aam strain. The ΔnhmD aam nhmG cblA genes of this strain were deleted from the chromosome using double homologous recombination. DNA fragments Sh1 (0.5 kbp flanking ΔnhmD left, primers Sh1-F, Sh1-R) and Sh2 (0.5 kbp flanking cblA right, primers Sh2-F, Sh2-R) were used to perform two recombination crosses and delete the cluster. The PCR joined fragment Sh1–Sh2 was introduced into BamHI and EcoRI restriction sites of the pRY1 vector, which could not autonomously replicate in Rhodococcus cells. Plasmid pRY1-Sh1-Sh2 was introduced into R. rhodochrous M33 via conjugation from E. coli S17-1, as described in a previous study.¹⁴ Clones with integrated plasmids after the first cross were selected with thiostrepton resistance. Clones which lost the plasmids after the second cross were selected with thiostrepton sensitivity. Finally, clones with ΔnhmBAG ΔcblA were checked with PCR and sequencing.

The R. rhodochrous strains expressing aam with nhmG and cblA deletions were constructed by integrating the variants of the expression cassette (delta-1, delta-6, delta-8, Fig. 5) into the Sh3 locus of the R. rhodochrous M33 delta chromosome. Every variant was obtained by PCR-joining of the Sh3 fragment (primers Sh3-F, Sh3-R), 2Tfd fragment (containing the doubled transcription terminator from phage fd, primers 2Tfd-F, 2Tfd-R), and the variant of the expression cassette. The delta-1 variant (Pnh-aam-nmhG-cblA) was amplified with primers F1, R1, and delta-6 (Pnh-aam-nmhG) with primers F1, R2. The delta-8 variant (Pnh-aam-cblA) was PCR-joined from Pnh-aam (primers F1, R3) and cblA (primers cblA-F, R1).

The R. rhodochrous strain expressing nhmBAG without cblA (M33 delta-const) was constructed by integrating the Sh3-Pnh-nhmBAG cassette into the Sh3 locus of the R. rhodochrous M33 delta chromosome (the Pnh-nhmBAG fragment was obtained with the primers F6-del, R6-del).

All sequences of the primers are listed in Table S1 in ESI.†

3. Results and discussion

3.1. Features of the Rhodococcus strains used in this work

To correctly study the effect of metals on the cobalt-dependent activity of the NHase promoter (Pnh) and the cobalt-dependent biosynthesis of NHase, we used a set of Rhodococcus strains with different combinations of the promoter, reporter enzyme, and metal-dependent regulator genes (R. rhodochrous M33 aam strain and its derivatives, Fig. 1 and 5).

The M33 aam strain is a derivative of the NHase over-producer R. rhodochrous M33 strain. Cobalt-dependent NHase genes in M33 aam are substituted with the aam gene of metal-independent acylamidase from R. erythropolis TA37 in such a way that aam is placed under the control of the NHase promoter. Earlier we showed that the specific acylamidase activities of whole cells of the R. rhodochrous M33 aam strain correlate perfectly with the relative levels of specific mRNA in the cells, transcribed from the NHase promoter.¹⁴ According to this, detailed studies of NHase promoter activity in this work were based on the enzyme activity of reporter acylamidase, fused with this promoter. Simultaneously, important findings were verified by measuring the relative levels of specific mRNA (by RT-qPCR). Additionally, wild type NHase producers, R. rhodochrous strains M8 and J1, and R. aetherivorans strain VKPM AC-2063, were used to verify the effects of metals on the NHase promoter, located within its natural expression cassette.

The effects of metals on the biosynthesis of the NHase enzyme were studied using the R. rhodochrous strain M33 and its cblA⁻ derivative.

3.2. Optimization of conditions for studying the effect of metals on the expression of NHase

3.2.1. Concentrations of metal ions in nutrient media, suitable for growth experiments

Elevated concentrations of heavy metal ions can inhibit the growth of bacterial culture. In order to study the expression of NHase in the growing culture, we tested the inhibitory effects of Ni²⁺, Zn²⁺, Cu²⁺, Cd²⁺, Pb²⁺, As³⁺ and Co²⁺ on the growth rate of the R. rhodochrous M33 aam strain. The inhibitory effects of these ions were investigated during the growth in liquid culture (MS/urea medium), with the addition of 21–504 μM of each ion, with 2-fold increments. All metals, except copper, were not inhibitory, up to 252 μM. The maximum concentration of copper which allowed the culture to grow was 42 μM. Thus, this concentration was chosen to test the effect of all the metals on promoter activity.

3.2.2. Optimal composition of nutrient media for studying the effect of metals on the activity of the NHase promoter

Recently, we showed that the NHase promoter had substantial basal activity without added cobalt, and 5-fold derepression in the presence of cobalt was observed.¹⁴ The basal activity was observed during the growth of the R. rhodochrous M33 aam strain on the MS medium with urea as the nitrogen source. This basal activity can interfere with the possible effects of other heavy metal ions on the promoter activity. Considering that urea is an amide, which can participate in the induction of the NHase promoter, we studied the effect of non-amide sources of nitrogen, to reveal the possibility of lowering the basal activity.

The growth and specific acylamidase activity curves were obtained during cultivation of the R. rhodochrous M33 aam strain on liquid MS medium with urea or NH₄NO₃ or NH₄Cl as the nitrogen source and without added heavy metal ions (Fig. 2).

Fig. 2

Growth and acylamidase activities of the R. rhodochrous M33 aam strain on MS medium with urea, NH₄NO₃ or NH₄Cl as the source of nitrogen and without added heavy metal ions. Dotted line with triangles: OD of cultures grown on urea/NH₄NO₃/NH₄Cl (the values did not differ in these three variants, so separate data are not shown). Solid line with triangles: acylamidase activity on the medium with urea. Solid line with circles: acylamidase activity on the medium with NH₄NO₃/NH₄Cl (the values did not differ in these two variants, so separate data are not shown). Error bars represent the SD calculated from four or more independent experiments.

Unlike urea, NH₄NO₃ and NH₄Cl did not induce NHase promoter activity, compared with the level at the start of the experiment. The absence of induction of activity was detected at both the exponential and stationary phases of culture growth. The relative levels of specific mRNA, determined at 11 and 29 hours of cultivation, correlated well with the acylamidase activities (data not shown).

These results reveal the possibility of tightly regulating the activity of the NHase promoter which has low basal activity without amides and added cobalt, and can be induced by both. Interestingly, amides induced maximum promoter activity at the stationary growth phase. Hence, the MS/NH₄NO₃ medium was chosen for further study of heavy metal ion effects on the activity of the NHase promoter, due to low level NHase promoter activity in this medium.

3.3. Effect of metal ions on the NHase promoter activity

3.3.1. Effect of metal ions on NHase promoter activity in the cassette with the reporter acylamidase in R. rhodochrous M33 aam

The effects of heavy metal ions on the NHase promoter activity were compared during the growth of the R. rhodochrous M33 aam strain on MS/NH₄NO₃ medium. The cultures were grown with and without the initial addition of corresponding metal salts (42 μM of metal ion), and specific acylamidase activities of cells were measured at regular intervals during 2-day experiments (Fig. 3).

Fig. 3

Effect of metals on the activity of the NHase promoter. (A) Growth and acylamidase activity of the R. rhodochrous M33 aam strain in the presence of cobalt and nickel ions. The cultures were grown on MS medium, supplemented with 42 μM of each ion. The activities of cultures with Zn²⁺, Cu²⁺, Cd²⁺, Pb²⁺ and As³⁺ ions were similar to activity without added metal (data not shown). (B) Relative levels of specific mRNA in the presence of cobalt, nickel, and without added metals, determined after 10 hours of cultivation. The mRNA level without added metals was set to 1. Error bars represent the SD, calculated from a minimum of four independent experiments.

Data on the activity of reporter acylamidase showed that nickel and cobalt ions derepressed the promoter, while other heavy metal ions had no effect. The specific mRNA levels, determined during the first day of cultivation, correlated well with the levels of acylamidase activity.

The effects of nickel and cobalt on the promoter activity were compared in more detail, specifically upon addition of 0.42–168 μM of these metal ions to the growth medium. The acylamidase activities of the R. rhodochrous M33 aam strain grown under these conditions are presented in Fig. 4.

Fig. 4

Acylamidase activities of R. rhodochrous M33 aam grown upon supplementation of non-inhibitory concentrations of nickel and cobalt ions (in the range of 0.42–168 μM). Error bars represent the SD, calculated from a minimum of four independent experiments.

We observed that 4.2 μM of added cobalt ions was sufficient to obtain maximum activity, and even 0.42 μM had visible effects. The effect of nickel ions was revealed only at 42 μM, and 126 μM was sufficient to obtain maximum activity similar to that of cobalt.

3.3.2. Effect of cobalt and nickel on the activity of the NHase promoter in wild type NHase-producing Rhodococcus strains

The effect of nickel on the activity of the NHase promoter was further verified using wild type NHase-producing R. rhodochrous strains M8 and J1, and R. aetherivorans strain VKPM AC-2063. Additionally, the effect of cobalt, reported for the R. rhodochrous M8/M33 strains,¹⁴ was studied for the first time in R. rhodochrous J1 and R. aetherivorans VKPM AC-2063 strains.

NHase-producing Rhodococcus strains contain the NHase promoter within natural expression cassettes, unlike the recombinant M33 aam strain (Fig. 1). The annotated NHase loci in R. rhodochrous M8 and J1 strains contain similar sets of genes, except the absence of the cobalt-dependent regulator in J1 (sequence truncated just after the nhhG, see GenBank D67027.1). We sequenced the 1 kbp region downstream the nhhG, and found the ORF, which had 99% nucleotide similarity to cblA (ncoR in Fig. 1). The Co-NHase cluster had not been sequenced in the NHase-producing R. aetherivorans strain VKPM AC-2063. We sequenced it and found out that the nhmCDBAGcblA cassette in it is similar to that of the R. rhodochrous M8 strain (Fig. 1).

NHases cannot be used as a reporter for monitoring the effect of metals on promoter activity, because NHase activity requires cobalt ions. Thus, the RT-qPCR detection of relative levels of specific mRNA was used to reveal the effects of nickel and cobalt. We grew R. rhodochrous M8 and J1, and R. aetherivorans VKPM AC-2063 strains on MS/urea medium (and with 2 g l⁻¹ of yeast extract for R. rhodochrous J1), with the addition of cobalt or nickel, and found that both ions induced NHase transcription in the strains (data not shown). As expected, there were no detectable NHase activities of the strains grown with the addition of nickel.

3.4. Effect of nhmG and cblA deletions on the metal-dependent regulation of promoter activity in R. rhodochrous

The requirement of the cblA gene for cobalt-dependent derepression of NHase transcription in the M33 aam strain was shown in our previous work.¹⁴ However, it was not clear whether the presence of only cblA, without nhmG, was sufficient for cobalt derepression. It was also important to investigate if cblA was responsible for the newly identified nickel-dependent derepression.

To clarify these, we constructed three isogenic R. rhodochrous strains with Pnh-aam expression cassettes, which differ in the presence of nhmG and cblA. First, the NHase-negative R. rhodochrous M33 delta strain was constructed on the basis of the R. rhodochrous M33 strain, as a result of double-crossover excision of nhmDBAG and cblA genes. Then, three Pnh-aam expression cassettes were integrated into its chromosome, resulting in R. rhodochrous strains M33 delta-1 (nhmG⁺cblA⁺), M33 delta-6 (nhmG⁺ ΔcblA), and M33 delta-8 (ΔnhmG cblA⁺) (Fig. 5).

Fig. 5

Schematic representation of the structure of the variants of Pnh-aam expression cassettes, integrated into the Sh3 locus of the R. rhodochrous M33 delta chromosome. Ω Ω: doubled transcription terminator from phage fd; aam: acylamidase gene from R. erythropolis TA37; nhmG: NHase accessory protein gene (cobalt metal-chaperone); cblA: cobalt-dependent transcriptional regulator gene.

The strains were grown with the addition of cobalt or nickel ions (42 μM), and specific acylamidase activities were determined (Fig. 6).

Fig. 6

Acylamidase activities of R. rhodochrous strains M33 delta-1 (Phn-aam-nhmG-cblA), M33 delta-6 (Phn-aam-nhmG-ΔcblA), and M33 delta-8 (Phn-aam-ΔnhmG-cblA) grown with the addition of nickel and cobalt ions (42 μM). Error bars represent the SD, calculated from a minimum of four independent experiments.

Both cobalt and nickel induced the activity in both cblA⁺ strains, irrespective of the presence of nhmG. However, both ions had no effect on activity in nhmG⁺ strains with deleted cblA. The data support the suggestion that nhmG does not cooperate with cblA in its action, and only cblA is responsible for both cobalt- and nickel-dependent derepression of the NHase promoter.

Summarizing the results described in Sections 3.3 and 3.4, we can state that (1) metal-dependent derepression of the NHase promoter is related only with the presence of cblA, and (2) in vivo nickel ions could replace the cobalt ions in this regulation. However, approx. 10-fold higher concentration of nickel was required to obtain the derepression level similar to that obtained with cobalt.

The CblA seems to be not selective enough to discriminate between Co and Ni ions. This regulator belongs to the SmtB/ArsR family of transcriptional repressors, known to dissociate from the cognate promoters in the presence of elevated concentrations of heavy metals. Dissociation of the regulator allows more effective transcription initiation.^24–29 The nickel/cobalt response was reported for several ArsR-type regulators of metal efflux pumps: DmeR from Agrobacterium tumefaciens (in vivo Ni was a more potent inducer than Co³⁰), NmtR from Streptomyces coelicolor (Ni-dependent regulation was shown in vitro and in vivo, but Co-dependent regulation was shown only in vitro³¹), KmtR from Mycobacterium tuberculosis (no difference between Ni and Co in vivo response³²), NmtR from M. tuberculosis (in vivo Ni was a slightly more potent derepressor than Co³³). It is interesting to look into some details of the works mentioned here. Some studies^31–34 considered only the maximum permissive concentrations of metals. The absence of the difference between in vivo response to cobalt and nickel in these cases was similar to our results in the presence of maximum tested concentrations of Ni and Co. The results obtained by other authors at maximum permissive concentrations of metals do not seem to reflect the normal physiology of the cell, because normal Co and Ni cell requirements can differ substantially. The authors tested the response of the DmeR-regulated reporter in Agrobacterium tumefaciens to a wide range of Ni and Co concentrations (10–500 μM).³⁰ Similar to our results, they revealed different responses at lower concentrations, and similar responses at higher metal concentrations.

The requirement for high concentrations of Ni for cblA-dependent derepression of NHase transcription can be explained by the putative intrinsic ability of CblA to bind Ni weaker than Co. This binding preference can be sequence dependent (see Fig. S1 and additional short discussion in ESI†).

3.5. Effect of metal supplementation on the biosynthesis of NHase

Metal ions often play both catalytic and structural roles within the metalloenzymes. Thus, the absence of the required metal or its displacement with another ion can destabilize the three-dimensional structure of an enzyme. The latter can increase the rate of proteolytic degradation of the enzyme, which, in turn, can decrease the detectable amount of the enzyme. We attempted to investigate the effect of Ni²⁺, Zn²⁺, Cu²⁺, Cd²⁺, Pb²⁺, and As³⁺ ions on in vivo formation of NHase both with and without added cobalt. Additionally, we tested the formation of NHase without added metals, using the nhmBAG⁺cblA⁻ strain (M33 delta-const), in which the NHase promoter is constitutively active without cobalt.

NHase activities and SDS-PAGE analysis of soluble proteins of R. rhodochrous M33 and M33 delta-const grown with cobalt and nickel mixtures, and without added metals, are presented in Fig. 7.

Fig. 7

Effect of metal supplementation on the biosynthesis of NHase in R. rhodochrous M33 nhmBAG⁺cblA⁺ and nhmBAG⁺cblA⁻ strains. The characteristic SDS-PAGE of soluble intracellular proteins is presented. The cells for this analysis were grown independently four times, and their average specific NHase activities are shown below the picture (the relative SD was no more than 15%). The identity of NHase subunit proteins was additionally proven using MALDI-TOF (the proteins are indicated by arrows, see ESI† for more details).

The amount of β-subunit was substantially lower than the amount of α-subunit in the case of both nhmBAG⁺cblA⁺ strain grown with nickel supplementation and nhmBAG⁺cblA⁻ strain grown without added metals. This shifted proportion of NHase subunits was contrary to nearly equal amounts of subunits in the case of cobalt supplementation. We also tested the variant of metal-free NHase expression, based on our finding that urea induces the activity of the NHase promoter (see Section 2.2). Despite the lower level of NHase synthesis in this case, we still observed a shifted proportion of NHase subunits (data not shown).

Neither the NHase activity nor subunit proportion was affected by nickel in the nhmBAG⁺cblA⁺ strain after growth with the addition of an equimolar mixture of cobalt and nickel (42 μM of each ion). When 4.2 μM of cobalt was added, the 20-fold excess of nickel (84 μM) resulted in the α-subunit excess. The effects of other metals on the NHase formation in nhmBAG⁺cblA⁺ were tested using the same mixing scheme. We observed that addition of Zn²⁺, Cu²⁺, Cd²⁺, Pb²⁺, and As³⁺ ions did not induce the synthesis of NHase, and had no effect on subunit formation (data not shown).

Hence, NHase synthesis without the addition of cobalt led to low stability or low synthesis of the NHase β-subunit, but did not affect the stability or synthesis of the α-subunit. Probably the increased rate of proteolytic degradation of the β-subunit, due to destabilization of the protein without cobalt, is the main reason for the observed effects. This indicates that cobalt ions presumably play an unknown role in the stability of the β-subunit in vivo.

Nickel ions could not replace the cobalt ions in this case. Thus, the NHase maturation mechanism is more specific to metal ions, than the mechanism of CblA-dependent activation of NHase transcription. The nhmG-(nhhG)-dependent mechanism of Co-NHase maturation is rather complicated,^15,35 and our data point out that further investigations are required for deeper understanding of the particular roles of the participants of NHase maturation.

4. Conclusion

The goal of this work was the complex study of the effect of metals on Co-NHase expression in R. rhodochrous. It was revealed that metal-dependent regulation of NHase promoter activity seems to be single-gene dependent (depends on the presence of only cblA, but not nhmG). The use of several heavy metal ions allowed us to probe and compare the in vivo metal selectivity of cblA-dependent regulation of both NHase transcription and NHase maturation processes. We showed that selectivity of the former was not high enough to discriminate between cobalt and nickel, but the selectivity of the latter was higher and allowed this discrimination. It is likely associated with the higher complexity of the NHase maturation process, which includes the joint action of NHase accessory protein, the α-subunit, and, probably, the β-subunit.

Conflicts of interest

There are no conflicts of interest to declare.

Abbreviations

Abbreviations
 
• Pnh

  Promoter of the nitrile hydratase genes

 
• NHase

  Nitrile hydratase enzyme

 
• SD

  Standard deviation

 
• cdw

  Cells dry weight

Acknowledgements

This work was funded by Russian Science Foundation, project 16-14-00216. The authors are grateful to Dr Alena Grechishnikova from the University of Utah (Salt Lake City) for careful revision of the manuscript.

Footnotes

†

Electronic supplementary information (ESI) available. See DOI: 10.1039/c8mt00129d