- Top of page
- Materials and Methods
Arbuscular mycorrhizal (AM) fungi are obligate symbiotic microorganisms that form associations with plant roots in a host nonspecific manner. Most terrestrial plant species live in symbiosis with AM fungi, and AM fungi take up inorganic phosphate (Pi) from soils through extraradical hyphae and transfer the Pi to the host plants. In return, AM fungi receive carbon compounds derived from photosynthesis in host plants (Smith & Gianinazzi-Pearson, 1988; Smith & Read, 1997; Harrison, 1999). This function of AM symbiosis has garnered attention from the viewpoint of sustainable agriculture (Bethlenfalvay & Linderman, 1992). It has been thought that the nutrient exchange between AM fungi and host plants occurs at the arbuscules, which are arbuscular fine hyphae and symbiosis-specific organs.
AM symbiosis is characterized by a specific organ, the arbuscule, which is formed in root cortex cells by the penetration of the finely branched hyphae of AM fungi (Bonfante-Fasolo et al., 1986). This organ is speculated to be a site of nutrient exchange between the host plant and AM fungi (Cox et al., 1980). Phosphate efflux from the fungi to the host plant at arbuscules is supported by the recent discovery of novel plant Pi transporters that are localized around arbuscules and acquire Pi from the fungi (Rausch et al., 2001; Harrison et al., 2002; Paszkowski et al., 2002).
The alkaline phosphatase (ALP) of AM symbiosis has been investigated since a mycorrhizal-specific ALP was identified electrophoretically in a crude extract of mycorrhizal roots (Gianinazzi-Pearson & Gianinazzi, 1976, 1978). Histochemical evidence of strong ALP activities at arbuscules (Gianinazzi et al., 1979; Tisserant et al., 1993; Ezawa et al., 1995) has raised the idea that the ALP is involved in phosphate efflux from arbuscules. ALP activity was not observed in host plants and may originate in AM fungi in mycorrhizal roots (Gianinazzi-Pearson & Gianinazzi, 1978). This hypothesis was later confirmed through a combined electrophoretic–enzymatic hyphal separation technique (Kojima et al., 1998). Therefore, ALP activity has been used as a marker for the metabolic activity of AM fungi (Tisserant et al., 1993; Guillemin et al., 1995).
Recently it was found that the efflux of phosphate from intraradical hyphae separated from roots by enzymatic digestion was partly suppressed by inhibitors of ALP (T. Kojima, unpublished data), suggesting that ALP in arbuscules may have an important role in the transfer of phosphate from AM fungi to host plants. Because purification of the enzyme has so far been unsuccessful (Kojima et al., 2001), little is known about the enzymatic characteristics of the ALP in AM fungi (Gianinazzi-Pearson & Gianinazzi, 1976). A specific inhibitor, Be2+, was used to characterize the ALP in intraradical hyphae, and these efforts revealed that this ALP has increased affinity for sugar phosphate and does not hydrolyse polyP (Ezawa et al., 1999). However, the function of the arbuscular ALP in symbiosis is still little known, and cloning of the enzyme may shed light on its unknown function.
Molecular genetic information on AM fungi has been accumulated. The development of high-throughput DNA sequencing techniques and high-quality RT–PCR kits have enabled the construction of expressed sequence-tagged (EST) libraries from relatively small samples of AM fungi, and EST libraries from the mycorrhizal roots, germinating spores, and extraradical hyphae of these organisms have been constructed (Franken & Requena, 2001; Sawaki & Saito, 2001). Such an extensive approach may facilitate or accelerate the identification of as-yet unknown genes specific to mycorrhizal symbiosis. In this regard, a cDNA clone showing similarity to a yeast ALP gene (PHO8) (Kaneko et al., 1987) was found in an EST library constructed from the extraradical hyphae of Glomus intraradices. Using this clone, we succeeded in cloning the ALP genes from the AM fungi G. intraradices and Gigaspora margarita for the first time. Here we present the characteristics of these ALP genes and their expression.
- Top of page
- Materials and Methods
We successfully cloned full-length cDNAs for the ALPs of two AM fungi by using an EST library of G. intraradices and a cDNA similar to the yeast ALP PHO8. PHO8-like ALPs occur in a wide range of organisms (with the exception of plants), constituting the ALP family (accession number in the BLOCKS database, IPB001952). The amino acid sequences of the known PHO8-like ALPs included in this family share six conserved domains. These six domains are also well conserved in the AM fungal ALPs that we isolated, GiALP and GmALP (Fig. 1).
The secondary structures of GiALP, GmALP, and yeast ALPs are very similar. In particular, the hydrophobic domain, which is thought to be the transmembrane domain, is present in the N-terminal regions of GiALP, GmALP, and yeast ALPs (Fig. 1). The yeast PHO8 enzyme is localized in vacuoles and attached to tonoplasts (Clark et al., 1982). In this regard, the enzymes encoded by GiALP and GmALP may also be membrane proteins.
The overall homology between the amino acid sequences of GiALP, GmALP, and yeast ALPs is not high (Table 2), because the sequences beyond the six conserved domains are not well conserved (Fig. 1). However, the phylogenetic tree reveals that GiALP, GmALP, and yeast ALPs are in the same cluster (Fig. 2). In light of the similarity in the positions of the conserved domains, similar secondary structures, and phylogenetic relationship, we are convinced that GiALP and GmALP are ALP genes of AM fungi.
It is interesting that there is only one copy of the GmALP gene (Fig. 3). Because spores of AM fungi contain multiple nuclei, it is surprising that we found only one copy of GmALP even though we used the full-length cDNA as a probe.
We tried to complement ALP activity in a PHO8 deletion mutant of S. cerevisiae (Saccharomyces deletion project, http://www-deletion.stanford.edu/) by introducing GiALP or GmALP. Although the gene was expressed in the transformed yeast, ALP activity was not complemented (data not shown). Yeast ALP is a N-glycosylated protein composed of two identical subunits with an Mr of 66 000, of which the carbohydrate component accounts for about 8% (Onishi et al., 1979). Yeast ALP is synthesized as an inactive form. The ALP protein becomes an active form by carbohydrate modification and removal of C-terminal peptide through delivery of ALP protein from ER to vacuole (Klionsky & Emr, 1989). Probably the GiALP and GmALP proteins in yeast cells may not be processed appropriately into an active form.
Are the ALPs encoded by GiALP and GmALP identical with the ALP revealed by the enzymatic–cytohistochemical technique? ALP activity was cytochemically detected in the vacuoles in arbuscules of AM fungi (Gianinazzi et al., 1979), and the ALP activity in the insoluble fraction was higher than that in the soluble fraction when proteins were extracted from AM fungal tissues (Ezawa et al., 1999). These results suggest that like the yeast PHO8 ALP, the cytohistochemically detected ALP in AM fungi may be a membrane protein presumably attached to tonoplast. Biochemical analysis revealed that the ALP crudely extracted from AM fungi was nonspecific as to substrate, although the enzyme had a greatly increased affinity for sugar phosphate (Ezawa et al., 1999). Yeast PHO8 ALP is known to be nonspecific regarding substrate (Kaneko et al., 1985).
The transcript levels of GiALP and GmALP in mycorrhizal roots were much higher than those in extraradical hyphae and germinating spores (Figs 5 and 6). Microscopic observation showed that the intraradical hyphae had a higher proportion of ALP-active hyphae than did the extraradical hyphae (Zhao et al., 1997; Kjøller & Rosendahl, 2000; van Aarle et al., 2002). Enzyme activities in crude extracts showed higher ALP activity in intraradical hyphae than in germinating spores (Saito, 1995). Furthermore, the GiALP and GmALP genes were constitutively expressed irrespective of growth stage and colonization rate (Fig. 5). Tisserant et al. (1993) and Zhao et al. (1997) showed that the proportions of intraradical hyphae of Glomus sp. and Gigaspora sp. having ALP activity approximately correlated with the colonization rates. In fact, the levels of ALP activity in the intraradical hyphae of Glomus sp. and Gigaspora sp. were relatively stable regardless of the growth stage (Boddington & Dodd, 1999). These tendencies in the localization and activity of the ALPs in AM fungi agreed with the observed expression patterns of GiALP and GmALP, suggesting that these genes may encode the ALP observed with the enzymatic–histochemical technique.
The yeast ALP encoded by PHO8 is a repressive type, and its transcription is enhanced by Pi deficiency (Kaneko et al., 1985; Oshima, 1997), indicating that the PHO8 ALP has a role in the hydrolysis of stored phosphorus compounds under conditions of phosphorus deficiency. Because the ALPs in AM fungi might have a similar function, we evaluated the effect of external Pi conditions on the expression of GiALP and GmALP (Fig. 6). However, the expression of both genes was constitutive irrespective of Pi addition.
Olsson et al. (2002) examined the effect of external Pi on the metabolism of G. intraradices in carrot hairy roots and found that the proportion of ALP-active extraradical hyphae was increased with high-Pi medium. Boddington & Dodd (1998, 1999) used a conventional soil pot culture system and reported that the ALP activity in the intraradical hyphae of Gl. manihotis grown under Pi-sufficient conditions was lower than that under Pi-deficient conditions, although the ALP activity in the intraradical hyphae of Gi. rosea was not influenced by the Pi levels. van Aarle et al. (2002) found that Pi addition to established AM-inoculated onion did not influence the proportion of ALP in either the intraradical or extraradical hyphae of G. margarita. The apparent inconsistency among these experiments may be due to differences between the experimental system or the plant–fungus combination. Because we followed the same experimental system as van Aarle et al. (2002), it is plausible that our results (Fig. 6) agree with theirs.
In the yeast S. cerevisiae, the expression of several genes associated with the PHO regulatory system, which responds to changes in the Pi concentration, are controlled at the transcriptional level. The yeast PHO regulatory system includes the expression of genes encoding structural proteins such as ALP (PHO8), acid phosphatase, and a high-affinity Pi transporter, and positive and negative regulatory factors of various structural proteins (Lemire et al., 1985; Toh-e et al., 1988; Bun-ya et al., 1991; Lenburg & Oshea, 1996; Oshima et al., 1996). Maldonado-Mendoza et al. (2001) showed that, as in yeast, the expression of a Pi transporter gene in the extraradical hyphae of G. intraradices was regulated in response to the environmental Pi concentration. This finding suggests that AM fungi may have a regulatory system similar to the yeast PHO system, implying that the expression of the ALP genes of AM fungi may be regulated in response to the environmental Pi concentration. However, the expression of GiALP and GmALP was not affected by the environmental Pi concentration. Unlike yeast cells, AM fungal cells are differentiated, and the environments of the intraradical and extraradical hyphae are very different. The ALP genes we identified were mainly expressed in mycorrhizal roots (i.e. intraradical hyphae). However, the Pi transporter gene was expressed in extraradical hyphae (Maldonado-Mendoza et al., 2001). These results suggest the regulatory system in AM fungi may be more complex than the yeast PHO regulatory system.
The present study revealed that the ALP genes in AM fungi (especially in the intraradical hyphae) are constitutively expressed under symbiotic conditions, irrespective of colonization rates and environmental Pi concentration. These results suggest that ALP may have a significant function specific to the intraradical hyphae.