Transcript profiling coupled with spatial expression analyses reveals genes involved in distinct developmental stages of an arbuscular mycorrhizal symbiosis

Abstract

The formation of symbiotic associations with arbuscular mycorrhizal (AM) fungi is a phenomenon common to the majority of vascular flowering plants. Here, we used cDNA arrays to examine transcript profiles in Medicago truncatula roots during the development of an AM symbiosis with Glomus versiforme and during growth under differing phosphorus nutrient regimes. Three percent of the genes examined showed significant changes in transcript levels during the development of the symbiosis. Most genes showing increased transcript levels in mycorrhizal roots showed no changes in response to high phosphorus, suggesting that alterations in transcript levels during symbiosis were a consequence of the AM fungus rather than a secondary effect of improved phosphorus nutrition. Among the mycorrhiza-induced genes, two distinct temporal expression patterns were evident. Members of one group showed an increase in transcripts during the initial period of contact between the symbionts and a subsequent decrease as the symbiosis developed. Defense- and stress-response genes were a significant component of this group. Genes in the second group showed a sustained increase in transcript levels that correlated with the colonization of the root system. The latter group contained a significant proportion of new genes similar to components of signal transduction pathways, suggesting that novel signaling pathways are activated during the development of the symbiosis. Analysis of the spatial expression patterns of two mycorrhiza-induced genes revealed distinct expression patterns consistent with the hypothesis that gene expression in mycorrhizal roots is signaled by both cell-autonomous and cell-nonautonomous signals.

Figure 1.

Array Quality and Variation within and between Experiments. (A) TIFF image of a cDNA array hybridized with a ³³P-labeled cDNA probe from mock-inoculated M. truncatula roots harvested at 31 days after inoculation. The image shown is from filter C. (B) Scatterplot comparing the LRs of duplicate spots on the arrays. Ninety-nine percent of the spots fall within ±0.2 LR of the mean. The data shown are from filter C. Data from filters A and B fall within the same limits. (C) Scatterplot comparing the average LRs from two experiments involving independent biological samples. Ninety-four percent of the spots fall within ±0.34 LR of the mean. The data shown are from filter C. Values from filters A and B fall within ±0.33 and ±0.37 LR of the mean, respectively.

Figure 2.

Cluster Analysis of 92 Transcripts during the Development of the AM Symbiosis and during Growth with 2 mM Versus 0.02 mM Phosphate. Clustered expression ratios visualized by Treeview (Eisen et al., 1998). The rows represent the individual clones, and the GenBank numbers of the ESTs are indicated. The rows labeled 53, 41, and 89 represent the AW587100, AW587100 3′ UTR, and Mt4 controls, respectively. The ratios of expression in mycorrhizal/nonmycorrhizal roots at 8, 15, 22, 31, and 36 days postinoculation (dpi) and in high/low phosphate (Pi) are represented. Biological replicates are adjacent to each other. The color reflects the LR. Red represents a positive LR, green represents a negative LR, and black represents a LR of zero. In each case, the saturation reflects the magnitude of the change. Seven groups of transcripts with distinct expression patterns are indicated.

Figure 3.

RNA Gel Blot Analysis of a Selection of the Genes Identified on the Arrays. (A) and (B) RNA gel blot analysis with total RNA (12 μg) from noncolonized M. truncatula roots (Mt) and M. truncatula roots colonized with G. versiforme (Mt/Gv) at 8, 15, 22, 31, and 36 days postinoculation (dpi). The blots were probed, stripped, and reprobed sequentially with the probes indicated at right. All of the genes except AW586356 are members of cluster 1. AW586356 is a member of cluster 5. (C) RNA gel blot analysis with root RNA (12 μg) from M. truncatula plants grown for 31 days (experiment 1) and 42 days (experiment 2) with differing levels of phosphate fertilization. The blots were probed, stripped, and reprobed sequentially with the probes indicated at right. The Mt4 gene was included as a control, and transcript levels reflect the degree of phosphate starvation (Burleigh and Harrison, 1998).

Figure 4.

Expression of a Selection of the Mycorrhiza-Regulated Genes in Nodulated Roots. (A) RNA gel blot analysis with total RNA (12 μg) from M. truncatula roots and M. truncatula roots colonized with G. versiforme (31 days after inoculation) (lanes 1 and 2) and M. truncatula roots and M. truncatula nodulated roots harvested at 11 (lanes 3 and 4) and 18 (lanes 5 and 6) days after inoculation. The blots were probed, stripped, and reprobed sequentially with the probes indicated at right. The leghemoglobin gene is induced in nodules and was included as a control (Barker et al., 1988). (B) RT-PCR analysis of the same RNA samples shown in (A) with primer pairs for AW585594, AW587040, and EF-1α.

Figure 5.

M. truncatula Roots Expressing the uidA and GFP Genes under the Control of the MtCel1 Promoter. (A) and (B) Light and corresponding epifluorescence micrographs of a M. truncatula root expressing GFP under the control of the MtCel1 promoter. Roots were colonized with G. versiforme. Arrows indicate cells containing arbuscules, and GFP is present in cells containing arbuscules. Bar = 150 μm. (C) Histochemical staining for GUS activity in a M. truncatula root expressing the uidA gene under the control of the MtCel1 promoter. Arrows indicate positive staining in cortical cells containing arbuscules. Bar = 200 μm. (D) and (E) Epifluorescence and corresponding light micrographs of cortical cells from a M. truncatula root expressing the uidA gene under the control of the MtCel1 promoter. The root was stained histochemically to detect GUS activity and then counterstained with acid fuschin to reveal the fungus. Cortical cells were released by squashing. Acid fuschin fluoresces yellow when viewed via epifluorescence microscopy. Arrows indicate positive GUS staining in cells containing arbuscules. Bar = 40 μm.

Figure 6.

M. truncatula Roots Expressing GFP under the Control of the MtSCP1 Promoter. Roots were colonized with G. versiforme. (A) and (B) Light and corresponding epifluorescence micrographs. Arrows indicate cells containing arbuscules, and a strong green fluorescence signal is present in these cells. Arrowheads indicate cells within the same cell file that do not contain arbuscules but show strong green fluorescence. Bar = 40 μm. (C) and (D) Light and corresponding epifluorescence micrographs. Arrowheads indicate cells in the inner cortex that do not contain arbuscules but display strong green fluorescence. Cells containing arbuscules are present but are not in the field of focus. Arrows indicate cells in the outer cortex that show weak green fluorescence. Hyphae, but not arbuscules, are present in these cells. Bar = 40 μm. (E) and (F) A laser scanning confocal microscopy image and the corresponding bright-field image of a hypha traversing the epidermis (e) and outer cortical cells. Arrows indicate outer cortical cells penetrated by the hypha, and green fluorescence is visible in these cells. Green fluorescence is not visible in the epidermal cell. (G) Merged image showing both green fluorescence and bright-field views. Bar = 75 μm.