Accumulation of and Response to Auxins in Roots and Nodules of the Actinorhizal Plant Datisca glomerata Compared to the Model Legume Medicago truncatula

Abstract

Actinorhizal nodules are structurally different from legume nodules and show a greater similarity to lateral roots. Because of the important role of auxins in lateral root and nodule formation, auxin profiles were examined in roots and nodules of the actinorhizal species Datisca glomerata and the model legume Medicago truncatula. The auxin response in roots and nodules of both species was analyzed in transgenic root systems expressing a beta-glucuronidase gene under control of the synthetic auxin-responsive promoter DR5. The effects of two different auxin on root development were compared for both species. The auxin present in nodules at the highest levels was phenylacetic acid (PAA). No differences were found between the concentrations of active auxins of roots vs. nodules, while levels of the auxin conjugate indole-3-acetic acid-alanine were increased in nodules compared to roots of both species. Because auxins typically act in concert with cytokinins, cytokinins were also quantified. Concentrations of cis-zeatin and some glycosylated cytokinins were dramatically increased in nodules compared to roots of D. glomerata, but not of M. truncatula. The ratio of active auxins to cytokinins remained similar in nodules compared to roots in both species. The auxin response, as shown by the activation of the DR5 promoter, seemed significantly reduced in nodules compared to roots of both species, suggesting the accumulation of auxins in cell types that do not express the signal transduction pathway leading to DR5 activation. Effects on root development were analyzed for the synthetic auxin naphthaleneacetic acid (NAA) and PAA, the dominant auxin in nodules. Both auxins had similar effects, except that the sensitivity of roots to PAA was lower than to NAA. However, while the effects of both auxins on primary root growth were similar for both species, effects on root branching were different: both auxins had the classical positive effect on root branching in M. truncatula, but a negative effect in D. glomerata. Such a negative effect of exogenous auxin on root branching has previously been found for a cucurbit that forms lateral root primordia in the meristem of the parental root; however, root branching in D. glomerata does not follow that pattern.

Keywords: DR5:GUS; IAA; PAA; cytokinins; hairy roots; nodules; root branching.

Figure 1

Auxin profiles of roots and nodules of the legumes Medicago truncatula and Cicer arietinum and the actinorhizal species Datisca glomerata.Endogenous concentrations of free and conjugated auxins were determined for roots (black) and nodules (white) of M. truncatula, C. arietinum, and D. glomerata. (A) indole-3-acetic acid; (B) phenylacetic acid; (C) IAA-alanine; (D) IAA-aspartate; (E) IAA-phenylalanine; (F) IAA-valine; (G) IAA-tryptophan; (H) IAA-leucine/IAA-isoleucine. Values represent means ± standard deviation, SD (n = 3). For each auxin/plant species, statistically significant differences between roots and nodules are labeled with an asterisk (Mann–Whitney U test, p ≤ 0.05). The numerical results are shown in Supplementary Table S1 .

Figure 2

Cytokinins in roots and nodules of Medicago truncatula and Datisca glomerata. Endogenous concentrations of free and conjugated auxins were determined for roots (black) and nodules (white) of M. truncatula and D. glomerata. (A) trans-zeatin, (B) cis-zeatin; (C) trans-isopentenyladenine; (D) trans-dehydrozeatin; (E) trans-zeatin-7-glucoside; (F) trans-zeatin-9-glucoside; (G) trans-zeatin-O-glucoside; (H) trans-zeatin riboside O-glucoside; (I) tDZR—trans-dehydrozeatin-riboside; (J) trans-zeatin riboside. Values represent means ± SD (n = 3). Statistically significant differences between roots and nodules are labeled with an asterisk (Mann–Whitney U test, p ≤ 0.05). No significant different between roots and nodules were detected for M. truncatula (when no error bar is shown, the cytokinin was only quantifiable in one of the biological samples). The numerical results are shown in Supplementary Table S1 .

Figure 3

Transgenic hairy root systems expressing a DR5:GUS fusion. Light micrographs of (A) transgenic roots of composite Medicago truncatula plants, (B–C) root sections, and (D–F) nodule sections and (G–I) transgenic roots of composite Datisca glomerata plants and (J) a nodule and (K) a nodule section are shown. (A) In M. truncatula roots, GUS activity was detected in the root tips, lateral root primordia, vascular bundles, and at lower levels in the root cortex. (B) During lateral root formation, GUS activity was observed in the lateral root primordia, and (C) lower activity was found in the lateral root vascular system during lateral root emergence (arrow). (D) During nodule development, GUS activity was localized in the nodule primordium (black arrowhead) and in the incipient nodule meristem (white arrowhead) as well as at the nodule base (arrows) at 4 dpi. (E) In differentiated nodules at 9 dpi, GUS activity was present in the nodule meristem and in the nodule vascular system (black arrowheads). (F) At 3 wpi, no GUS activity could be detected in the nodule anymore (black arrowheads point at the nodule vascular system).In transgenic roots of composite D. glomerata plants, (G) GUS staining was restricted to the root tip and to the peripheral part of the xylem (arrowhead). (H) GUS staining was found in lateral root primordia; during lateral root development, it became restricted to the root tip and the vascular system (arrows). (I) Nodule primordia—distinguishable from lateral root primordia since they are flatter—also showed GUS staining (arrowhead points at a nodule primordium that is positioned at the base of a lateral root). (J) Mature nodules only showed GUS staining at the tips of actively growing lobes (arrowhead), the weak staining on parts of the periderm represents epiphytic bacteria with endogenous GUS activity. (K) Analysis of longitudinal sections showed that GUS activity was restricted to a cell layer distal of the meristem of the nodule lobe (arrowhead). No DR5:GUS-dependent GUS staining was ever detected associated with a D. glomerata nodule vascular bundle (an arrow points at the nodule vascular bundle in K). Size bars: (A), 1 mm; (B) and (C), 100 µm; (D), 50 µm; (E), 100 µm; (F), 200 µm; (G), 100 µm; (H), 200 µm; (I), 100 µm; (J), 1 mm; (K), 200 µm.

Figure 4

Effect of NAA (A) and PAA (B) on average main root length (cm) in Medicago truncatula. Values represent means ± SEM. One-way ANOVA with Tukey’s post hoc test was used to assess significant differences between treatment groups. Values labeled with different letters are significantly different (p ≤ 0.05). n = 9–12.

Figure 5

Effect of NAA (A) and PAA (B) on root branching in Medicago truncatula. The figure shows the number of emerged and unemerged lateral roots (primordia) per main root. Values represent means ± SEM. One-way ANOVA with Tukey’s post hoc test was used to assess significant differences between treatment groups. Values labeled with different letters are significantly different (p ≤ 0.05). n = 9–12.

Figure 6

Effect of NAA (A) and PAA (B) on average main root length (cm) in Datisca glomerata. Values represent means ± SEM. One-way ANOVA with Tukey’s post hoc test was used to assess significant differences between treatment groups. Values labeled with different letters are significantly different (p ≤ 0.05). n = 9–12.

Figure 7

Effect of NAA (A) and PAA (B) on branching of the seedling root of Datisca glomerata. The figure shows the number of emerged and unemerged lateral roots (primordia) per main root. Values represent means ± SEM. One-way ANOVA with Tukey’s post hoc test was used to assess significant differences between treatment groups. Values labeled with different letters are significantly different (p ≤ 0.05). n = 9–12.

Figure 8

Effect of NAA (A) and PAA (B) on adventitious root formation in Datisca glomerata. Values represent means ± SEM. One-way ANOVA with Tukey’s post hoc test was used to assess significant differences between treatment groups. Values labeled with different letters are significantly different (p ≤ 0.05). n = 9–12.