Factors governing attachment of Rhizobium leguminosarum to legume roots at acid, neutral, and alkaline pHs

Abstract

Rhizobial attachment to host legume roots is the first physical interaction of bacteria and plants in symbiotic nitrogen fixation. The pH-dependent primary attachment of Rhizobium leguminosarum biovar viciae 3841 to Pisum sativum (pea) roots was investigated by genome-wide insertion sequencing, luminescence-based attachment assays, and proteomic analysis. Under acid, neutral, or alkaline pH, a total of 115 genes are needed for primary attachment under one or more environmental pH, with 22 genes required for all. These include components of cell surfaces and membranes, together with enzymes that construct and modify them. Mechanisms of dealing with stress also play a part; however, exact requirements vary depending on environmental pH. RNASeq showed that knocking out the two transcriptional regulators required for attachment causes massive changes in the bacterial cell surface. Approximately half of the 54 proteins required for attachment at pH 7.0 have a role in the later stages of nodule formation. We found no evidence for a single rhicadhesin responsible for alkaline attachment, although sonicated cell surface fractions inhibited root attachment at alkaline pH. Our results demonstrate the complexity of primary root attachment and illustrate the diversity of mechanisms involved.

Importance: The first step by which bacteria interact with plant roots is by attachment. In this study, we use a combination of insertion sequencing and biochemical analysis to determine how bacteria attach to pea roots and how this is influenced by pH. We identify several key adhesins, which are molecules that enable bacteria to stick to roots. This includes a novel filamentous hemagglutinin which is needed at all pHs for attachment. Overall, 115 proteins are required for attachment at one or more pHs.

Keywords: adhesin; attachment; legume; pea; rhizobium; root.

Fig 1

Primary attachment to pea roots of wild-type and mutant strains of R. leguminosarum at pH 7.0. Attachment of Rlv3841 (WT) and strains mutated in pssA (RL3752) and praR (RL0390) was measured using the Lux-based whole root attachment assay where each strain is labeled by the introduction of a plasmid constitutively expressing Lux. Luminescence (RLU/g of root) shows bacterial attachment after 1 h. n ≥ 10. All data points are shown with the box indicating the interquartile range and the median shown. Maximum and minimum values are indicted by the whiskers. ****P < 0.0001, **P < 0.01 in comparison to WT using Student’s t-test.

Fig 2

Effect of pH on primary attachment to pea roots. Attachment of Rlv3841 (WT) and a strain mutated in gmsA at acidic, neutral, and alkaline pH using the Lux-based whole root attachment assay. Luminescence (RLU/g of root) shows bacterial attachment after 1 h. pH is indicated. n ≥ 10. All data points are shown with the box indicating the interquartile range and the median shown. Maximum and minimum values are indicted by the whiskers. ***P < 0.001 using Student’s t-test. There is no statistically significant difference between primary attachment to roots of WT at the three different pHs.

Fig 3

Rlv3841 genes required for primary attachment to pea roots under acidic, neutral, and alkaline pH. A total of 115 genes were classified as required for attachment (ES/DE) in INSeq experiments performed at pH 6.5, pH 7.0, and pH 7.5 (Table S3), with 22 genes being required for attachment at all pHs (Table 1). The color of the circle indicates pH; on the left-hand side (red) = pH 6.5, at the top (green) = pH 7.0, and on the right (blue) = pH 7.5. Genes are listed in Tables S4 to S9.

Fig 4

Summary of Rlv3841 genes required for primary attachment to pea roots at acidic, neutral, and alkaline pH grouped by function. From INSeq, the 115 genes required at pH 6.5 to pH 7.5 (Fig. 3; Table S3) are shown grouped by their involvement in different cellular structures or processes, and color-coded. Genes required at all pHs (the core attachome) are shown in black and those required under specific conditions are as follows: pH 6.5 (red), pH 7.0 (green), pH 7.5 (blue), at pH 6.5 and pH 7.0 (orange), pH 7.0 and pH 7.5 (gray), and pH 6.5 and pH 7.5 (purple). In addition, there are 24 genes that have an, as yet, unidentified function (Table S3). Abbreviations: TA, toxin:antitoxin; PTS, phosphotransferase system; RNAP, RNA polymerase; PHB, poly-hydroxy butyrate.

Fig 5

Involvement of genes required for root attachment at pH 7.0 at other stages of nodule development (in the rhizosphere, for root colonization, in nodule bacteria and bacteroids). Fifty-four genes are required for primary root attachment (Fig. 3) and their involvement at other stages of infection and symbiosis was determined from the INSeq study of nodule development (26). Twenty-five genes are root-attachment specific, with the remainder required at one or more stages of symbiosis (attachment to roots (1 h) is represented by a dark blue box, while those of other developmental stages are pale blue). Two genes (RL2400 and RL2642) are not represented; they are INSeq-classified as AD in nodule bacteria (26) (Table 3). *Rhizosphere-progressive genes, in total 146 (26), with six of them also required for primary attachment.

Fig 6

Effect of co-inoculation with wild-type bacteria on primary attachment to pea roots of a mutant at different pHs. Comparison of Rlv3841 (WT) and a strain mutated in RL2969, as single inoculum (Mutant), or in 1:1 or 1:100 co-inoculation with unlabeled Rlv3841 (Mutant:WT) using the Lux-based whole root attachment assay at (A) pH 6.5, (B) pH 7.0, and (C) pH 7.5. Luminescence (RLU/g of root) shows bacterial attachment after 1 h. n ≥ 9. All data points are shown with the box indicating the interquartile range and the median shown. Maximum and minimum values are indicated by the whiskers. ***P < 0.001, **P < 0.005 using Student’s t-test.

Fig 7

Purification of a crude adhesin protein fraction and its effect on primary attachment of Rlv3841 (WT) to pea roots. (A) SDS-gel of Rlv3841 cell fractions (soluble, membrane, and crude adhesin) stained with SYPRO Ruby. The putative rhicadhesin fraction, a protein band at approx. 14 kDa (circled in blue), is visible in both soluble and crude adhesin fractions, the latter of which was excised from the gel and used for proteomic analysis and pre-incubation of roots. Ladder protein sizes (kDa) are indicated. (B) Root attachment of Rlv3841 (WT) using the Lux-based whole root attachment assay at pH 7.5 either with no addition (black) or pre-incubation of roots with crude adhesin for 1 h (+ Crude adhesin, gray). Luminescence (RLU/g of root) shows bacterial attachment after 1 h. Data are displayed as mean ± SEM, n = 5. ***P < 0.001 using Student’s t-test.