Arbuscular mycorrhizal fungi differentially affect expression of genes coding for sucrose synthases in maize roots

Authors

  • Sabine Ravnskov,

    Corresponding author
    1. Department of Crop Protection, Danish Institute of Agricultural Sciences, Research Centre Flakkebjerg DK-4200 Slagelse, Denmark;
      Author for correspondence: Sabine Ravnskov Tel: +45 581 13469 Fax: +45 581 13301 Email: sabine.ravnskov@agrsci.dk
    Search for more papers by this author
  • Yong Wu,

    1. Citrus Research and Education Center, University of Florida, Institute of Food and Agricultural Sciences, Lake Alfred, Florida 33850, USA
    Search for more papers by this author
  • James H. Graham

    1. Citrus Research and Education Center, University of Florida, Institute of Food and Agricultural Sciences, Lake Alfred, Florida 33850, USA
    Search for more papers by this author

Author for correspondence: Sabine Ravnskov Tel: +45 581 13469 Fax: +45 581 13301 Email: sabine.ravnskov@agrsci.dk

Summary

  • •   The effects of three arbuscular mycorrhizal (AM) fungal isolates on the expression of genes coding for sucrose synthase and on the nonstructural carbohydrate (CHO) status of maize roots were evaluated. Gene expression and CHOs were compared with their status in nonmycorrhizal plants grown at three soil phosphorus (P) concentrations.
  • •   The AM fungi and soil P supply influenced expression of genes coding for sucrose synthase in maize roots. In general, up to 18 d after plant emergence, AM colonization increased the expression of genes coding for sucrose synthase in maize roots, whereas increasing soil P decreased this gene expression. The responses in gene expression were detected earlier than other effects of AM fungal colonization, such as increased leaf P status and plant growth response under limiting P supply.
  • •   Higher sucrose-synthase gene expression was not related to the concentration of sucrose, reducing sugars or starch in the root tissue.
  • •   Higher gene expression in AM roots confirms that there is greater allocation of sucrose from nonstructural CHO pools in roots for the AM fungus during the earliest phase of colonization than in nonmycorrhizal roots.

Introduction

The arbuscular mycorrhizal (AM) symbiosis is functionally characterized by the reciprocal exchange of nutrients between the symbionts: Carbon (C) in form of nonstructural carbohydrates (CHO) from plants to the AM fungus and phosphorus (P) from AM fungi to plants. The outcome of the symbiosis measured as plant growth or reproduction depends on the balance in the nutrient flow between the symbionts, which can be influenced by plant species/variety (Ravnskov & Jakobsen, 1995; Syvertsen & Graham, 1999), AM fungus species/isolate (Pearson & Jakobsen, 1993; Graham et al., 1996; Boucher et al., 1999) and external factors such as soil P supply (Eissenstat et al., 1993). Thus, the functioning of AM symbioses may occur along a continuum from mutualism to parasitism (Johnson et al., 1997).

Previous studies have quantified the C cost of AM symbiosis as between 4% and 20% of the plant's C photoassimilate (Pang & Paul, 1980; Paul & Kucey, 1981; Kucey & Paul, 1982; Snellgrove et al., 1982; Koch & Johnson, 1984; Harris et al., 1985; Douds et al., 1988; Wang et al., 1989; Jakobsen & Rosendahl, 1990; Pearson & Jakobsen, 1993; Peng et al., 1993). The CHO allocation from host to fungus varies with plant age and developmental stage of the mycorrhizal symbiosis (Wright et al., 1998). The course of gene expression for key enzymes involved in carbon utilization in root cells during root colonization may provide a greater understanding of the dynamics of CHO allocation from the plant to the AM fungus at a given developmental stage of the symbiosis.

Alteration of CHO allocation is known to be influenced by changing P status of AM plants compared with nonmycorrhizal plants (Graham et al., 1997; Wright et al., 1998; Black et al., 2000; Valentine et al., 2001). This emphasizes the importance of comparing AM and nonmycorrhizal plants with varying P status to study the dynamics of CHO allocation to the symbiosis in relation of P nutrient supply to plant tissues (Eissenstat et al., 1993).

In most plants, CHO is transported from the photosynthetic to nonphotosynthetic tissue as sucrose and before being utilized in cells, sucrose must be cleaved into hexoses for further metabolism (Sturm, 1999). The only two known enzymes that catalyse this reaction are invertase and sucrose synthase (Sturm, 1999). Sucrose synthase is a cytoplasmic enzyme and the important isoforms Sus1 and Sh1 have been identified in maize (Echt & Chourey, 1985). Studies of expression of genes coding for Sus1 and Sh1 have shown that expression of Sus1 was low or nondetectable under sugar-depleted conditions, whereas Sh1 was maximally expressed under conditions of low carbohydrate supply (Koch et al., 1992). Similarly, Zeng et al. (1998) have shown differential expression of the two genes by hypoxia and anoxia: Sus1 was upregulated by hypoxia, while Sh1 was upregulated by anoxia.

Transport of CHO from plants to AM fungi occurs in the roots after degradation of sucrose to hexoses (Shachar-Hill et al., 1995). Comparison of the expression of the genes that code for invertase or sucrose synthase in nonmycorrhizal and mycorrhizal plants may be used to evaluate more precisely temporal aspects of CHO metabolism in roots related to developmental stages of AM symbiosis. Blee & Anderson (2002) studied gene expression of sucrose-degrading enzymes on a cellular level and found that expression of genes coding for sucrose synthase and vacuolar invertase was increased in arbusculated cells compared with nonmycorrhizal cells. However, measurements of invertase and sucrose synthase in extracts of root tissue of mycorrhizal bean, mung bean and maize showed no difference compared with nonmycorrhizal roots. They concluded that the presence of the AM fungal hyphae stimulates discrete alterations in expression of sucrose-metabolizing enzymes to increase the sink potential of the cell.

The objective was to study CHO status of maize roots during the early colonization of three AM fungal isolates by measuring the expression of genes coding for sucrose synthase and, furthermore, to compare these gene expressions with the expression of the same genes in nonmycorrhizal plants grown at three different soil P supplies.

Materials and Methods

Biological material and experimental setup

Maize plants (Zea mays, ssp. McNare 508) were grown in symbiosis with three AM fungal isolates; two isolates of Glomus intraradices, Schenck & Smith, FL208 (Florida) and BEG87 (Denmark), and Glomus mosseae (Nicol. & Gerd.) Gerd. & Trappe isolate BEG83 (Denmark).

The experiment was designed with six treatments: three control treatments with nonAM plants watered with three different P supplies (0P, 0.25 mm P and 0.5 mm P) and three AM fungal treatments (G. intraradices isolates FL208 and BEG87, and G. mosseae BEG83). Plants with AM fungal treatments were watered as 0P supply treatment.

All treatments had five replicates and the experiment of the same design was repeated in order to confirm the results.

Growth conditions

Plants were grown in 280 g Candler fine sand soil with Mehlich I extractable P 4–6 µg g−1 (Mehlich, 1953) in tubes (diameter 5 cm, length 18 cm, volume 262 ml (D16 Deepot Cells; Stuewe & Sons, Inc., Corvallis, OR, USA). The soil in the tubes consisted of three layers: bottom (150 g sand soil), middle (100 g sand soil mixed with AM inoculum), and top (30 g of sand soil). Inoculum of G. intraradices FL208 consisted of colonized root fragments. One gram of root fragments was mixed into 100 g sand soil for each pot before use. Inoculum of G. intraradices (BEG87) and G. mosseae was dry pot cultures consisting of sand soil, colonized roots and spores. For these AM fungi, 50 g of this inoculum was mixed with 50 g sand soil before use. In control treatments, AM inoculum was substituted with the same amount of sand soil. Inoculum filtrate of each AM fungus was prepared to add the same microorganisms, except for AM fungi to each pot. The filtrate was a suspension of 50 g of each inoculum in 1.5 l water that had been mixed slowly for 24 h at 23–27°C and sieved twice through a 44-µm mesh screen. Ten milliliters of the inoculum filtrate from all three AM fungi was added to each pot just before they were watered the first time. Two maize seeds (four in pots used for preharvest) were planted and thinned to one plant per pot after emergence in pots used for harvests 2, 3 and 4. Plants used in ‘preharvests’ and in harvest 1 were not thinned in order to have sufficient plant material for analyses.

Plants were placed in a completely randomized design in a glasshouse in central Florida with temperatures varying from 20 to 32°C and relative humidity from 60 to 100% for 25 d from 16 March to 11 April 2001. Pots were watered as previously described (Graham et al., 1996) with a modified Hoagland's nutrient solution (Hoagland & Arnon, 1939) with or without P, depending on the treatment. The position of the plants on the bench was changed every second day to minimize effects of variation in glasshouse conditions.

Five replicates of each treatment were harvested 3, 6 (preharvests), 9, 13, 18 and 25 d (harvest 1–4), respectively, after emergence. For preharvests (3 d and 6 d) replicate plants were pooled before analyses of one sample per treatment.

Plant harvest, nutrient analyses and AM fungal colonization

Plants were analysed for shoot dry weights after oven drying at 70°C for 48 h. Dried shoots were ground and analysed for P concentration by an ammonium molybdovanadate method (Stuffin, 1967). Fresh roots were washed, cut into 1-cm pieces and randomly divided into three weighed subsamples: one for Northern blot analysis, one for determination of AM fungal colonization after staining in Trypan blue, and one for dry weight and carbohydrate analyses. Subsamples for Northern blot analysis were placed in liquid nitrogen as soon as possible after the shoots were cut off and stored in a −80°C freezer until use. Subsamples of roots for determination of AM fungal colonization were stored in 2.5% acetic acid, and the remainder of the roots were dried and weighed for nonstructural CHO analyses.

The AM fungal colonization was evaluated under the dissecting microscope at ×25 magnification by scoring for the presence or absence of root internal AM fungal structures at 200 gridline intersections with root pieces after staining the roots as described by Kormanik & McGraw (1982) with Trypan blue instead of acid fuchsin.

Analysis of gene expression

Total RNA was extracted from 0.5 g subsamples of roots by grinding the tissue with mortar and pestal in liquid nitrogen with TRIzol reagent added according to the manufacturer's instructions (Cat. No. 15596; Life Technology, Carslbad, CA, USA). The RNA concentration and quality was determined at 260 nm and 280 nm. Five micrograms of total RNA was separated by electrophoresis in a denaturing 1.2% (w : v) agarose gel containing 0.2 m formaldehyde. Equal loading per lane was assessed by ethidium bromide staining. The RNA was transferred to a nylon membrane by capillary blotting in 10× standard saline citrate (SSC) and crosslinked by UV irradiation.

Two probes that code for sucrose synthase activity (Sus1 and Sh1 (see Koch et al., 1992)) were used to study gene expression with nonradioactive Northern blot. The gene-specific pattern of Sus1 and Sh1 is described in Koch et al. (1992) and in Zeng et al. (1998). Probe cDNA fragments were labeled with digoxigenin (dig)-dUTP via polymerase chain reaction (PCR) incorporation. Membranes were hybridized with dig-labeled probes (approximately 10 mg probe ml−1 hybridization buffer) in Perfecthyp™ plus hybridization buffer (Sigma, St. Louis, MO, USA) at 68°C overnight and were washed twice with high stringency solution (0.5× SSC, 0.1% (w : v) sodium dodecyl sulfate (SDS)) at 68°C for 20 min For chemiluminescent detection of dig-labeled probes, diluted (75 mU ml−1) antidig antibody conjugated with alkaline phosphatase (Roche, Indianapolis, IN, USA) was incubated with membranes for 30 min at room temperature. Substrate for alkaline phosphatase, Disodium 3-(4-methoxyspiro{1,2-dioxetane-3,2′-(5′-chloro)tricyclo[3.3.1.13,7]decan}-4-y1) phenyl phosphate (CSPD), was used to generate luminescent signals, which were detected by an X-ray film. Each Northern blot procedure was performed two or three times.

Analyses of nonstructural CHO in roots

After weighing, dried roots were finely ground, resuspended in water, boiled in 2 min for extraction of nonstructural CHOs, and centrifuged in 2 min at 800 g (Eissenstat & Duncan, 1992). Supernatant and pellet were analysed for soluble and insoluble starch, respectively, by using amyloglucosidase digestion test after correcting for free glucose (Haissig & Dickson, 1979). The supernatant were also used for determination of reducing sugars by using arsenomolybdate (Nelson, 1944) and sucrose by using anthrone sulfuric acid after alkali treatment (see van Handel, 1968). Glucose was used as a standard for analysis of starch and sucrose, whereas fructose was used as a standard in Nelson's test for reducing sugars.

Statistical analysis

Plant dry weights, AM fungal colonization of plant roots, P concentrations in shoots and CHO concentrations in roots were analysed with one-way analysis of variance followed by a multiple range test (LSD0.05). In order to test homogeneity of variance before analyses, data were tested by Bartletts test (P < 0.05).

Results

Preharvest

Plants harvested 3 d and 6 d after emergence were treated as ‘preharvests’ as plants from each treatment were pooled before analyses and analysed for shoot dry weights and gene expression, only. There were no differences in plant growth or expression of genes coding for sucrose synthase in plants among treatments in these two harvests (data not shown).

Plant growth and AM fungal colonization of plant roots

Plant growth was similar in all treatments at the first harvest. After 18 d, dry weights of 0.5 mm P plants were significantly higher than those of the other treatments, whereas plants inoculated with G. intraradices FL208 grew significantly less than plants in the other treatments. After 25 d, G. intraradices BEG87 and G. mosseae BEG83 increased plant growth to the level of control plants fertilized with 0.25 mm P, whereas plants inoculated with G. intraradices FL208 were comparable to the 0P nonmycorrhizal treatment (Fig. 1). Among treatments, there was no difference in root–shoot ratio of plants during the course of the experiment (data not shown).

Figure 1.

Dry weight of maize (Zea mays) plants grown without mycorrhiza at different phosphorus levels or in symbioses with three different isolates of arbuscular mycorrhizal fungi, respectively, harvested 9, 13, 18 and 25 d after emergence. Nonmycorrhizal and 0P addition, small open circles; nonmycorrhizal and 0.25 mm P addition, medium-size open circles; nonmycorrhizal and 0.5 mm P addition, large open circles; Glomus intraradices (FL 208), closed squares; G. intraradices (BEG87), closed triangles; Glomus mosseae, closed circles. Values are means ± SE, n = 5.

Colonization of plant roots by G. intraradices BEG87 was significantly greater than with either of the other AM fungi during the time course (Fig. 2). The rate of colonization of G. intraradices FL208 and G. mosseae BEG83 was similar until 18 d and 25 d after emergence, when colonization of plant roots with G. mosseae BEG83 was significantly greater than that of G. intraradices FL208 (Fig. 2).

Figure 2.

Arbuscular mycorrhizal fungal colonization of maize (Zea mays) roots grown in symbiosis with three different isolates of arbuscular mycorrhizal fungi, respectively, harvested 9, 13, 18 and 25 d after emergence. Glomus intraradices (FL 208), closed squares; G. intraradices (BEG87), closed triangles; Glomus mosseae, closed circles. Values are means ± SE, n = 5.

Sucrose synthase gene expression

In 9-d-old plant roots, increasing soil P supply decreased the expression of both Sus1 and Sh1 (Fig. 3), whereas all three AM fungi, especially G. mosseae BEG83 increased the expression of these genes (Fig. 3). Expression of Sus1 in plants colonized with the two G. intraradices isolates was comparable to Sus1 gene expression of the 0P nonmycorrhizal treatment, whereas G. mosseae BEG83 appeared to further upregulate gene expression after 18 d compared with all nonmycorrhizal P treatments. After 18 d, Sh1 was downregulated by increased P supply and upregulated by AM colonization. In the last harvest, the pattern of gene expression changed markedly; the downregulation of Sus1 expression by increasing P supply faded, and P began to upregulate Sh1. At 25 d after emergence, AM still upregulated Sus1, but downregulated Sh1 (Fig. 3).

Figure 3.

Gene expression of Sus1 and Sh1 coding for sucrose synthase in roots of maize (Zea mays) grown without mycorrhiza at different phosphorus levels or in symbioses with three different isolates of arbuscular mycorrhizal fungi (Glomus intraradices FL 208; G. intraradices BEG87; Glomus mosseae BEG83) harvested 9, 18 and 25 d after emergence.

Phosphorus concentration in shoots

Phosphorus concentrations in shoots were generally higher in P-fertilized plants compared with unfertilized plants during the experiment (Fig. 4a). Mycorrhizal treatments did not affect P concentrations in the shoots compared with the 0P nonmycorrhizal treatment. However, total P content in maize shoots increased in the following order in harvests after 18 d: nonmycorrhizal 0P and G. intraradices FL208; G. mosseae BEG83; G. intraradices BEG87; 0.25 mm P; 0.5 mm P (Fig. 4b).

Figure 4.

Concentrations (a) and contents (b) in shoots of maize (Zea mays) grown without mycorrhiza at different phosphorus levels or in symbioses with three different isolates of arbuscular mycorrhizal fungi, respectively, harvested 9, 13, 18 and 25 d after emergence. Nonmycorrhizal and 0P addition, small open circles; nonmycorrhizal and 0.25 mm P addition, medium-sized open circles; nonmycorrhizal and 0.5 mm P addition, large open circles; Glomus intraradices FL 208, closed squares; G. intraradices BEG87, closed triangles; Glomus mosseae BEG83, closed circles. Values are means ± SE, n = 5.

Concentration of CHO in roots

The sucrose fraction in roots was unaffected by treatment until harvest 4. At 25 d after emergence, sucrose concentration in roots colonized with G. intraradices FL208, G. mossseae BEG83 and 0P treatment was significantly lower than in roots with G. intraradices BEG87, 0.25 mm P and 0.5 mm P treatments (Fig. 5). These data were correlated with data of dry weight of shoots (r = 0.59, P < 0.0001). The concentration of reducing sugars in plant roots showed the same pattern as sucrose, but with no significant differences among treatments (Fig. 6). Starch concentration in the roots was measured in the last two harvests only; no significant differences occurred among treatments, but in the last harvest there was a tendency (P = 0.068) for starch concentration in roots of the 0.5 mm P and G. intraradices BEG87 treatments to be higher than in the other treatments (data not shown).

Figure 5.

Concentration of sucrose in roots of maize (Zea mays) grown without mycorrhiza at different phosphorus levels or in symbioses with three different isolates of arbuscular mycorrhizal fungi, respectively, harvested 9, 13, 18 and 25 d after emergence. Nonmycorrhizal and 0P addition, small open circles; nonmycorrhizal and 0.25 mm P addition, medium-size open circles; nonmycorrhizal and 0.5 mm P addition, large open circles; Glomus intraradices FL 208, closed squares; G. intraradices BEG87, closed triangles; Glomus mosseae, closed circles. Values are means ± SE, n = 5.

Figure 6.

Concentrations of reduced sugars in roots of maize (Zea mays) grown without mycorrhiza at different phosphorus levels or in symbioses with three different isolates of arbuscular mycorrhizal fungi, respectively, harvested 9, 13, 18 and 25 d after emergence. Nonmycorrhizal and 0P addition, small open circles; nonmycorrhizal and 0.25 mm P addition, medium-size open circles; nonmycorrhizal and 0.5 mm P addition, large open circles; Glomus intraradices (FL 208), closed squares; G. intraradices (BEG87), closed triangles; Glomus mosseae, closed circles. Values are means ± SE, n = 5.

Discussion

The present study demonstrates that early colonization of AM fungi and soil P supply influenced expressions of genes coding for sucrose-degrading enzymes, and thereby, may alter the time-course of sucrose utilization in maize roots. The AM fungi increased expression of genes for sucrose synthase in 9- to 18-d-old maize roots, whereas additional P supply decreased sucrose gene expression. The effects of AM fungi and P supply on the gene expression in young maize roots occurred before other effects of AM (except colonization) were detectable.

Stimulation of expression of genes coding for sucrose synthase by AM colonization in the maize roots is consistent with the results of Wright et al. (1998), who reported higher sucrose synthase activity in mycorrhizal clover plants compared with nonmycorrhizal plants. In both plant species, this sucrose synthase activity by AM fungal colonization decreased as shoot growth of mycorrhizal plants exceeded that of nonmycorrhizal plants. These findings suggest that C allocation by the plant to the AM fungus is reduced as root colonization becomes well established.

Differential expression pattern of Sus1 and Sh1 in response to external factors (as in this case P supply and inoculation with different AM fungi especially in harvest after 25 days) has been seen in response to sugar availability (Koch et al., 1992) and oxygen levels (Zeng et al., 1998). Koch et al. (1992) found that expression of Sus1 was stimulated by high carbohydrate availability, whereas Sh1 was maximally expressed under conditions of low carbohydrate supply in root tips. These findings were not confirmed in present experiment as there was no correlation between gene expression and concentrations of sugars in the roots. Koch et al. (1992) also found that the effect of sugars on Sus1 and Sh1 was less pronounced on the protein level and total enzyme activity did not change in response to the sugar availability. Similarly, Blee & Anderson (2002) found increased expression of genes coding for sucrose synthase and invertase in arbusculated cells compared with nonmycorrhizal cells in 28-d-old plants, but could not confirm these findings by biochemical measurements of sucrose synthase and invertase in extracts from the root tissue. They concluded that this could be because they extracted from whole roots where the proportion of arbusculated cells was small.

In present experiment, the P status of the plants was measured in the shoots, but not in roots, where it would be impossible to distinguish between P in the plant vs the fungus. Plants with AM colonization upregulated sucrose metabolism genes, reflecting increased catabolism and utilization of sucrose in the maize roots. However, for two of the fungal isolates (BEG 83 and 86) concentration of P in leaves was lower than in nonmycorrhizal plants of the same size. By contrast, Black et al. (2000) found that both AM and soil P supply increased photosynthetic rate in AM cucumber compared with nonmycorrhizal cucumber owing to an increased P status rather than as a consequence of a mycorrhizal sink for assimilates. However, Jifon et al. (2002) found that the photosynthetic rate of AM plants compared with nonAM plants with same P status was higher in some cases and not in others, suggesting that the effect of AM on photosynthetic rate could be attributed to nonnutritional factors such as fungal sink strength. They concluded that the CHO sink strength of the AM fungus depends on the mycorrhizal dependency of the plant. Similarly, Wright et al. (1998) found that mycorrhizal clover with same nutritional status as nonmycorrhizal clover had a higher photosynthetic rate and increased sucrose synthase and invertase activity in the roots. These differences in C source–sink relationships between host and AM fungus could be due to the varying functional compatibility between the symbionts. Earlier experiments have shown that the flow of nutrients between the organisms in the mycorrhizal symbiosis depends on the species of plant (Ravnskov & Jakobsen, 1995; Jifon et al., 2002), and the isolate of the AM fungus (Pearson & Jakobsen, 1993; Graham et al., 1996). The present experiment confirms that C utilization as measured here by gene expression in maize roots probably depends on species/isolate of the AM fungus involved in the symbiosis.

Comparative studies of whole-plant C budgets of nonmycorrhizal and mycorrhizal plants with same nutrient status have shown an increased allocation of C to mycorrhizal roots (Peng et al., 1993; Wright et al., 1998). No significant differences among treatments in pools of sucrose, starch and reducing sugars in roots were found before 25 d after planting in the present experiment. Similarly, Wright et al. (1998) found no difference in sucrose and starch contents in roots of clover before 38 d after planting. By contrast, increased glucose and fructose in mycorrhizal clover plants compared with nonmycorrhizal plants occurred by 18 d after planting. Graham et al. (1997) studied C allocation patterns in various citrus genotypes with different mycorrhizal dependency under high and low P conditions; in general, they found that mycorrhizal plants allocated more C to pools of nonstructural CHOs in AM shoots and roots than did nonmycorrhizal plants, even under high P supply. This response was related to mycorrhizal dependency of the genotype of the plant (e.g. less AM-dependent genotypes had consistently lower starch concentrations – less C allocation to pools of nonstructural CHOs – than more AM-dependent species under high P conditions).

In conclusion, AM fungi caused changes in the regulation of sucrose synthase genes in maize roots. The response varied with different developmental stages of the root colonization process, and was apparently not related to P metabolic effects. The finding that gene regulation responses are not straightforwardly related to sugar status may not be surprising according to the literature (Koch et al., 1992). Our study dictates that future investigations should be coupled with more precise real-time measurements of metabolic activity (e.g. NMR; see Shachar-Hill et al., 1995) and that it is important to include arbuscular mycorrhizal symbioses of different functional compatibility to draw general conclusions (Ravnskov & Jakobsen, 1995).

Acknowledgements

We wish to thank Diana Drouillard, Diane Bright, Mauricio Rubio and Jessica Cook for skilful technical assistance. This research was supported by Danish Agricultural and Veterinary Research Council, Grant 9901594.

Ancillary