Laser Capture Microdissection Transcriptome Reveals Spatiotemporal Tissue Gene Expression Patterns of Medicago truncatula Roots Responding to Rhizobia

Abstract

We report a public resource for examining the spatiotemporal RNA expression of 54,893 Medicago truncatula genes during the first 72 h of response to rhizobial inoculation. Using a methodology that allows synchronous inoculation and growth of more than 100 plants in a single media container, we harvested the same segment of each root responding to rhizobia in the initial inoculation over a time course, collected individual tissues from these segments with laser capture microdissection, and created and sequenced RNA libraries generated from these tissues. We demonstrate the utility of the resource by examining the expression patterns of a set of genes induced very early in nodule signaling, as well as two gene families (CLE peptides and nodule specific PLAT-domain proteins) and show that despite similar whole-root expression patterns, there are tissue differences in expression between the genes. Using a rhizobial response dataset generated from transcriptomics on intact root segments, we also examined differential temporal expression patterns and determined that, after nodule tissue, the epidermis and cortical cells contained the most temporally patterned genes. We circumscribed gene lists for each time and tissue examined and developed an expression pattern visualization tool. Finally, we explored transcriptomic differences between the inner cortical cells that become nodules and those that do not, confirming that the expression of 1-aminocyclopropane-1-carboxylate synthases distinguishes inner cortical cells that become nodules and provide and describe potential downstream genes involved in early nodule cell division. [Formula: see text] Copyright © 2023 The Author(s). This is an open access article distributed under the CC BY-NC-ND 4.0 International license.

Keywords: M. truncatula; RNA-Seq; laser capture microdissection; nodulation.

Fig. 1.

Stages of early nodule development and areas of tissue harvest. Images are micrographs of cross sections from root segments grown in an aeroponic system (described in “Materials and Methods”) and harvested at the indicated time points. Roots were fixed and cut as described in Chavan et al. (2018). A, uninoculated root; B, 12 h postinoculation (hpi); C, 24 hpi (arrow indicates early cell divisions); D, 48 hpi; E, 72 hpi. Early (I), Middle (II), and Late (III) refer to categories for rhizobial response gene analysis (see “Results”). F, Schematic of tissues captured for RNA at time points in this experiment. All tissue types were captured at each time point, except for cells forming a nodule (NOD), which were captured only at 48 and 72 hpi. ED, epidermal cells; OC, outer cortical cells; ICA, inner cortical cells across from xylem poles; ICB, inner cortical cells between xylem poles; VS, vasculature cells.

Fig. 2.

Induction of nodulation genes in tissue responding to rhizobia. A, Diagram of the tissues sectioned, from the area of elongation at 0 h postinoculation (hpi), and from the areas corresponding to the 0-hpi area at later time points. B, Corresponding area captured at each time point (also displayed in ePlant). C, Expression patterns of indicated gene (NIN in this figure) at each time point by fragments per kilobase of transcript per million mapped reads. Note the scale is calculated to a local maximum, indicated in red, which varies by the gene. D, Average expression, with error bars of the same gene (NIN) in whole-root segments under the same conditions from data in Schnabel et al. (2023).

Fig. 3.

Induction of nodulation genes in tissue responding to rhizobia. A, Expression patterns of indicated genes at each time point by fragments per kilobase of transcript per million mapped reads. Note the scale is calculated to a local maximum, indicated in red, which varies by the gene. B, Average expression, with error bars of the same genes in part A in whole-root segments under the same conditions from data in Schnabel et al. (2023).

Fig. 4.

Induction of CLE peptide genes in tissue responding to rhizobia. A, Expression patterns of indicated genes at each time point by fragments per kilobase of transcript per million mapped reads. Note the scale is calculated to a local maximum, indicated in red, which varies by the gene. B, Average expression, with error bars of the same genes in part A in whole-root segments under the same conditions from data in Schnabel et al. (2023).

Fig. 5.

Induction of nodule-specific PLAT domain genes in tissue responding to rhizobia. A, Expression patterns of indicated genes at each time point by fragments per kilobase of transcript per million mapped reads. Note the scale is calculated to a local maximum, indicated in red, which varies by the gene. B, Average expression, with error bars of the same genes in part A in whole-root segments under the same conditions from data in Schnabel et al. (2023).

Fig. 6.

Rhizobial response genes show several different temporal expression patterns in root segments. Heat map of the 1,932 rhizobial response genes identified from Schnabel et al. (2023) and sorted by temporal pattern of expression. Shown are genes that increased in A17 (the majority of which also increased in sunn-4 and rdn1–2), increased in sunn-4 but not A17 (including genes that also increased in rdn1–2), or increased in rdn1–2 only (one gene). Yellow (I), blue (II), and green (III) refer to Early, Middle, and Late time points of first observed increase in expression (see text for definition). Letters down the y-axis refer to a pattern of expression defined as: A, transient (up and then decreasing); B, persistent (up and holding steady); and C, increasing (steady increase throughout).

Fig. 7.

Distribution of rhizobial response genes with defined temporal expression patterns (DTEPs) between tissues. See text for description of the DTEP patterns. A, Pie chart indicating percentage of the rhizobial response genes identified with a pattern in each tissue. B, Venn diagram of sharing of gene patterns between tissues. The number of genes shared between tissues is indicated in each overlap, with tissues indicated by color. ED, epidermis (blue); OC, outer cortical cells (red); ICA, inner cortical cells across from xylem poles (green); ICB, inner cortical cells between xylem poles (yellow); and VS, vasculature (tan). C, Pie charts showing distribution of patterns by tissue. Raw data provided in Tables 1 and 2.

Fig. 8.

Tissue expression of Medicago truncatula ACC synthases in early nodulation signaling. Expression patterns of indicated 1-aminocyclopropane-1-carboxylate (ACC) synthase genes at each time point by fragments per kilobase of transcript per million mapped reads. Note the scale is calculated to a local maximum, indicated in red, which varies by the gene.

Fig. 9.

Enrichment of lists of genes differentially expressed between the inner cortical cells across from xylem poles (ICA) and the inner cortical cells between xylem poles (ICB). A, Gene ontology (GO) term functional enrichment of lists from Supplementary Dataset 2D using Classification Super Viewer tool (see “Materials and Methods”). GO categories are displayed for each GO subclass ranked by normed frequency values. Errors bars indicate the standard deviation of the normed frequency. Terms with statistical enrichment determined by hypergeometric enrichment tests (P < 0.05) are indicated in boldface. B, Gene families over-enriched in each category according to GenFam (P < 0.05), which uses a Fisher’s exact test with a Bonferroni corrected P value.