The meristem-associated endosymbiont Methylorubrum extorquens DSM13060 reprograms development and stress responses of pine seedlings

Abstract

Microbes living in plant tissues-endophytes-are mainly studied in crop plants where they typically colonize the root apoplast. Trees-a large carbon source with a high capacity for photosynthesis-provide a variety of niches for endophytic colonization. We have earlier identified a new type of plant-endophyte interaction in buds of adult Scots pine, where Methylorubrum species live inside the meristematic cells. The endosymbiont Methylorubrum extorquens DSM13060 significantly increases needle and root growth of pine seedlings without producing plant hormones, but by aggregating around host nuclei. Here, we studied gene expression and metabolites of the pine host induced by M. extorquens DSM13060 infection. Malic acid was produced by pine to potentially boost M. extorquens colonization and interaction. Based on gene expression, the endosymbiont activated the auxin- and ethylene (ET)-associated hormonal pathways through induction of CUL1 and HYL1, and suppressed salicylic and abscisic acid signaling of pine. Infection by the endosymbiont had an effect on pine meristem and leaf development through activation of GLP1-7 and ALE2, and suppressed flowering, root hair and lateral root formation by downregulation of AGL8, plantacyanin, GASA7, COW1 and RALFL34. Despite of systemic infection of pine seedlings by the endosymbiont, the pine genes CUL1, ETR2, ERF3, HYL, GLP1-7 and CYP71 were highly expressed in the shoot apical meristem, rarely in needles and not in stem or root tissues. Low expression of MERI5, CLH2, EULS3 and high quantities of ononitol suggest that endosymbiont promotes viability and protects pine seedlings against abiotic stress. Our results indicate that the endosymbiont positively affects host development and stress tolerance through mechanisms previously unknown for endophytic bacteria, manipulation of plant hormone signaling pathways, downregulation of senescence and cell death-associated genes and induction of ononitol biosynthesis.

Keywords: intracellular; metabolism; plant–microbe interactions; symbiosis; transcription network.

Figure 1.

Colonization of Scots pine (Pinus sylvestris L.) roots and stem by Methylorubrum extorquens DSM13060 at 90 DPI. (A) Bacterial colonies reside in the cylindrical sheath and epidermis (arrowheads), accumulating in the non-vascular parenchyma and in the xylem vessels (circled) of root. (B) Biofilm-like bacterial growth covering the root surface (arrowheads) and individual bacterial cells in the non-vascular parenchyma. (C) Bacteria inside chlorenchymal cells (arrows) of stem. Note: The bacterial cells carrying a fluorescent GFP are visualized in bright green. Microscopic sections: Cross (A, B); longitudinal (C). Ch = chlorenchyma, Co = cortex, Cs = cylindrical sheath, E = epiderm, En = endoderm, NVp = non-vascular parenchyma, Xy = xylem. Scale bar—20 μm.

Figure 2.

Heatmap of differentially expressed genes (adj. P < 0.01) between six replicates of Methylorubrum extorquens DSM13060-inoculated (‘M ext’) and control (‘untreated’) Scots pine seedlings. The heatmap is showing differences in expression compared with gene averages. Normalized intesites were scaled with log1p, and the gene averages were subtracted from individual expressions.

Figure 3.

Gene set enrichment analysis for gene ontology terms of up- or downregulated genes in Scots pine due to Methylorubrum extorquens DSM13060 infection. Genes were sorted by fold change and analyzed for enrichment using 100 permutations without a P value cutoff.

Figure 4.

Multivariate analysis of the complete data set using PCA (A) and pairwise comparisons of 90 days old Methylorubrum extorquens-inoculated (black symbols) and untreated (white symbols) pine seedlings for needles (B, circles), roots (C, rectangles) and stems (D, triangles) using partial least square-discriminant analysis (PLS-DA).

Figure 5.

A bar graph showing normalized relative gene expression of Methylorubrum extorquens DSM13060-inoculated pine seedlings 7 and 90 DPI analyzed by real-time quantitative PCR. Fold changes are relative to GAPDH, *, P < 0.05, **, P < 0.01 indicate statistical significance between comparisons of treatment and control. CUL1 = Cullin1, CYP71 = cytochrome P450, ERF3 = ethylene-responsive transcription factor 3, HYL1 = Hyponactic leaves 1, PR-5 = pathogenesis-related protein 5.

Figure 6.

In situ hybridization of pine seedlings inoculated with Methylorubrum extorquens and controls 90 DPI. (A) Shoot tip meristem, (B) needle and (C) root tissue of an inoculated seedling hybridized with a probe specific for CUL1. (D) Shoot tip meristem of control seedling hybridized with a probe specific for CUL1. Am = apical meristem, Sp = scale primordium, Me = mesophyll, V = vascular tissue, Co = cortex and Cs = cylindrical sheath. Scale bar—20 μm.

Figure 7.

In situ hybridization of pine seedlings inoculated with Methylorubrum extorquens and controls 90 DPI. (A) Shoot tip meristem of an inoculated and (B) a control seedling, hybridized with a probe specific for the ETR2. (C) Shoot tip meristem, (D) needle and (E) root tissue of an inoculated seedling, hybridized with a probe specific for the ERF3. (F) Shoot tip meristem of a control seedling hybridized with a probe specific for the ERF3. (G) Shoot tip meristem of an inoculated and (H) a control seedling, hybridized with a probe specific for HYL1. Am = apical meristem, Sp = scale primordium, Me = mesophyll, V = vascular tissue, Co = cortex and Cs = cylindrical sheath. Scale bar—20 μm.

Figure 8.

In situ hybridization of pine seedlings inoculated with Methylorubrum extorquens and controls 90 DPI. (A) Shoot tip meristem and (B) needle tissue of an inoculated seedling, hybridized with a probe specific for GLP1-7. (C) Shoot tip meristem and (D) needle tissue of a control seedling, hybridized with a probe specific for GLP1-7. (E) Needle tissue of an inoculated seedling, hybridized with a sense probe specific for GLP1-7. (F) Shoot tip meristem and (G) needle tissue of an inoculated seedling, hybridized with a probe specific for CYP71. (H) Shoot tip meristem of a control seedling, hybridized with a probe specific for CYP71. Am = apical meristem, Sp = scale primordium, Me = mesophyll, V = vascular tissue, Co = cortex, Cs = cylindrical sheath. Scale bar—20 μm.

Figure 9.

Proposed key mechanisms of Methylorubrum extorquens DSM13060 in affecting hormonal signaling of Scots pine and subsequent gene networks. (A) the bacterium induces host CUL1, potentially through phospholipase A2 enzymes or neddylation, which activates auxin signaling by submitting the AUX/IAA repressor to the ubiquitin degradation pathway. (B) Bacterial induction of HYL1 can affect auxin, abscisic acid (ABA) and cytokinin signaling via miRNA and siRNA processing. (C) Developmental genes behind flowering, root hair and lateral root formation are repressed, whereas genes for shoot and root growth are induced in pine seedlings. Biosynthesis genes of flavones are induced, and anthocyanin synthesis genes are repressed. Decreased expression of genes involved in senescence and cell death occurs in colonized seedlings potentially due to production of methyl-esterified 3-hydroxybutyrate oligomers (Koskimäki et al. 2016) and ononitol. ALE2 = abnormal leaf shape 2, ARF = auxin-responsive factor, ASK = apoptosis signal-regulating Kinase-1, AUX/IAA = auxin/indole acetic acid, COW1 = phosphatidylinositol transporter, CRSP = CO₂ response secreted protease, CUL1 = Cullin-1, DCL = dicer-like 1, Pri-miRNA = primary transcript micro-RNA, RBX = RING H2-type domain, RUB = related to ubiquitin, SE = serrate protein, Ub = ubiquitin.