A microaerobically induced small heat shock protein contributes to Rhizobium leguminosarum/ Pisum sativum symbiosis and interacts with a wide range of bacteroid proteins

Abstract

During the establishment of the symbiosis with legume plants, rhizobia are exposed to hostile physical and chemical microenvironments to which adaptations are required. Stress response proteins including small heat shock proteins (sHSPs) were previously shown to be differentially regulated in bacteroids induced by Rhizobium leguminosarum bv. viciae UPM791 in different hosts. In this work, we undertook a functional analysis of the host-dependent sHSP RLV_1399. A rlv_1399-deleted mutant strain was impaired in the symbiotic performance with peas but not with lentil plants. Expression of rlv_1399 gene was induced under microaerobic conditions in a FnrN-dependent manner consistent with the presence of an anaerobox in its regulatory region. Overexpression of this sHSP improves the viability of bacterial cultures following exposure to hydrogen peroxide and to cationic nodule-specific cysteine-rich (NCR) antimicrobial peptides. Co-purification experiments have identified proteins related to nitrogenase synthesis, stress response, carbon and nitrogen metabolism, and to other relevant cellular functions as potential substrates for RLV_1399 in pea bacteroids. These results, along with the presence of unusually high number of copies of shsp genes in rhizobial genomes, indicate that sHSPs might play a relevant role in the adaptation of the bacteria against stress conditions inside their host.IMPORTANCEThe identification and analysis of the mechanisms involved in host-dependent bacterial stress response is important to develop optimal Rhizobium/legume combinations to maximize nitrogen fixation for inoculant development and might have also applications to extend nitrogen fixation to other crops. The data presented in this work indicate that sHSP RLV_1399 is part of the bacterial stress response to face specific stress conditions offered by each legume host. The identification of a wide diversity of sHSP potential targets reveals the potential of this protein to protect essential bacteroid functions. The finding that nitrogenase is the most abundant RLV_1399 substrate suggests that this protein is required to obtain an optimal nitrogen-fixing symbiosis.

Keywords: NCR; Rhizobium; nitrogen fixation; oxidative stress; sHSP.

Fig 1

Phylogenetic tree of rhizobial sHSPs. The phylogenetic tree of sequences from rhizobial sHSPs was performed using IQ-Tree from Clustal Omega protein alignment. The alpha-crystallin A2 chain isoform 1 from the human protein CRYAA (NP_001300979_1) was used as a root of the tree. Rlv UPM791 sHSPs are highlighted in gray. Two main groups containing Class A (light green) and Class B (light gray) sHSPs are indicated. Numbers in red indicate the bootstrapping support value and numbers in gray indicate the phylogenetic distance. Sequence names shown in the tree contain the abbreviated name of the rhizobial species and strains followed by the name of the sHSP protein, indicating whether the localization of the coding sequence is chromosomic (chr) or plasmidic (p), and their accession numbers from GenBank. Bd: Bradyrhizobium diazoefficiens; Em: Ensifer meliloti; Re: Rhizobium etli; Rj: Rhizobium johnstonii; Rlp: R. leguminosarum bv. phaseoli; Rlt: R. leguminosarum bv. trifolii; Rlv: R. leguminosarum bv. viciae.

Fig 2

Expression levels of symbiotic sHSP-encoding genes. The histogram shows the relative expression levels of the shsp genes in bacteroids induced by Rlv UPM791 in pea and lentil nodules using the rpoD gene as standard for normalization. Values are the means of three replicates ±standard error. Values significantly different at P < 0.01 (*) are indicated.

Fig 3

Sequence alignment of anaeroboxes from Rlv UPM791 FnrN-type promoters. Shaded box denotes potential FnrN-binding site (consensus anaerobox TTGA-N₆-TCAA) from DNA upstream of the rlv_1399 gene. Underlined DNA sequences show anaeroboxes from the promoter regions of fnrN1, fnrN2, fixN, and hypB genes. Numbers on the anaeroboxes indicate their centered position to the translation start sites. Promoter sequences were obtained from the corresponding genes in the GenBank database.

Fig 4

Analysis of the expression of rlv_1399 in Rlv under free-living conditions. (A) β-Glucuronidase activity of aerobic (21% O₂) and microaerobic (1% O₂) cells. Data are the mean of three replicates ± standard error. Values significantly different at P < 0.01 (*) are indicated. (B) Immunodetection of RLV_1399_ST in cell extracts from bacterial cultures grown under aerobic and microaerobic conditions, as indicated. Proteins were resolved by SDS-PAGE in 15% polyacrylamide gels. Each line was loaded with 90 µg of protein. Numbers on the left margin indicate the position of molecular weight standards (kDa). The arrow on the right indicates the position of RLV_1399_ST. Left and right panels are portions of the same western blot membrane.

Fig 5

Effect of RLV_1399 expression on the tolerance of Rlv to oxidative stress. The graph shows the growth of UPM1421 derivative strains carrying empty vector pLMB51 (blue lines) or pLMB1399_ST plasmid (orange lines) grown under microaerobic (1% O₂) conditions in the absence (−) or presence (+) of 2 mM H₂O₂ in UMS medium. Each OD determination represents the mean of three replicates ± standard error.

Fig 6

Effect of RLV_1399 expression on cell viability of bacterial cultures exposed to NCR peptides. Microaerobic cells of UPM1421 derivatives strains carrying pLMB51 or pLMB1399_ST were exposed to chemically synthesized cationic (G35, L36) and anionic (G39, L40) NCR peptides from pea (G) and lentil (L) nodules at the indicated concentrations during 2 h at 28°C. Cells were fivefold serially diluted and 5 µL from each dilution were spotted onto TY plates. The plates were incubated at 28°C for 72 h.

Fig 7

Pull-down assays with sHSP RLV_1399_ST protein. Soluble extracts obtained from pea bacteroids induced by UPM1421/pBBR1399_ST (+) or UPM1421/pBBR1MCS-5 (−) strains were applied to a Strep-Tactin column, and eluted fractions were pooled and concentrated. Proteins were resolved in an 8%–16% polyacrylamide gradient SDS-PAGE gel. (A) Coomassie blue stained gel. The analysis was performed with 10 µg from soluble and pooled eluted fractions. The arrows indicate the bands corresponding to RLV_1399_ST and to nitrogenase subunit NifDK and NifH. (B) Immunoblot membranes containing 5 µg of protein from soluble fractions or 2 µg from pooled eluted fractions were revealed using antisera against NifDK, NifH, or with StrepTactin-alkaline phosphatase conjugate to detect RLV_1399_ST, as indicated. (C) Plots from Coomassie blue stained gels of soluble and eluate fractions from UPM1421(pBBR1399_ST). The arrow indicates the plot peak corresponding to RLV_1399_ST. The value shows the mean of the percentage of the RLV_1399_ST peak area to the total plot area of three independent stained gels ± standard error. (D) Plots from reactive bands in immunoblots of soluble and eluate fractions from UPM1421(pBBR1399_ST). Values of increased levels of NifDK, NifH, and RLV_1399_ST proteins in the eluate fraction compared to the soluble fraction are indicated. Values are the mean of three independent immunoblot analyses ± standard error. Numbers on the left margins of panels A and B indicate the position of the molecular weight standards (kDa). Plots in panels C and D were generated using Image J software. SF, soluble fraction; W, wash; E, eluate fraction.