Distribution of hyphal biomass and P in extraradical and intraradical parts of an AM fungus
This is the first report on the distribution of fungal P in the extraradical and the intraradical parts of AM fungi, although a large number of reports on P transfer and P translocation in AM have been published (Smith & Read, 1997). The hyphal biomass of AM fungi has been investigated by means of microscopic, chitin and phospholipid analysis (Tinker, 1975; Hepper, 1977; Bethlenfalvay et al., 1982; Toth et al., 1991; Olsson et al., 1999). Assuming that the moisture contents of fresh hyphae and roots were almost the same, the proportion of the hyphal mass in roots in the present study was lower than that reported by Hepper (1977) and that reported by Toth et al. (1991) but was within the range reported by Olsson et al. (1999). We obtained the specific root length of the onion, about 3 m g−1 f. wt basis, and specific hyphal length of 5.2 km g−1 hyphae from published data (Solaiman et al., 1999). Use of these conversion factors gave 23 m extraradical hyphae m−1 colonised root at 6 wk and 12 m m−1 at 9 wk. These data are within the range of 71 m m−1 for effective Glomus spp. and 7.1 m m−1 for ineffective Scutellospora calospora in onion (Sanders et al., 1977).
In the present study, the P contents in the hyphae were obtained on a fresh weight basis. To compare our data with data on a d. wt basis in the literature, we assumed that the hyphal moisture content was 80%. The estimated hyphal P concentration determined on a dry basis ranged from 1.9 to 3.2% for the intraradical part and from 0.5 to 1.3% for the extraradical part. These hyphal P concentrations were higher than those reported for other fungi, i.e., 0.5% to 2% P in ectomycorrhizal fungi in culture (Cairney & Smith, 1992; Ashford et al., 1994). To our knowledge, there is only one report on the P content in AM fungal hyphae. In this study by Capaccio & Callow (1982), it was suggested that the intraradical hyphae of G. mosseae is about 0.6% P; however, they did not provide any experimental values. In the present study, the total P concentrations in the intraradical hyphae were much higher than 0.6% and were almost equivalent to those in poly-P accumulating bacteria in activated sludge (Nakamura et al., 1991). Even if the P concentration in the intraradical hyphae was slightly overestimated because a certain amount of host contamination during the hyphal isolation process is inevitable, we can be certain that a large proportion of P in mycorrhizal roots was retained in the intraradical hyphae.
Another important finding is that the amount of P retained in the fungal cells did not change between 6 and 9 wk although the fungal biomass doubled during the same period. It has been shown previously that only a relatively small portion of AM fungal cells were active when examined with viable staining methods (Kough et al., 1987; Schubert et al., 1987; Sylvia, 1988; Hamel et al., 1990; Saito et al., 1993). Arbuscules tend to become senescent quickly and degenerate within 1–2 wk (Cox & Tinker, 1976; Toth & Miller, 1984). Assuming that only active fungal cells retain high P concentrations, the present results agree with these observations and imply that only a small fraction of the intraradical hyphae are involved in P translocation.
Phosphate efflux from the intraradical hyphae separated from roots
This study has demonstrated phosphate efflux from intraradical hyphae in vitro for the first time. A large amount of phosphate was effluxed during 4 h of incubation even without the addition of glucose. The intraradical hyphae collected by the method used in the present study were physiologically active but were already fragmented, so some cytoplasmic phosphorus may have leaked out and been hydrolysed by phosphatases (Joner & Johansen, 2000) during the incubation. However, it is possible that the separated hyphae had the potential to efflux phosphate themselves without any C sources. P efflux was partly inhibited by cyanide, which is an inhibitor of respiratory activities (unpublished results), suggesting that part of the P efflux was not leakage from damaged tissue.
It is significant that both glucose and its analogue, 2-deoxyglucose, enhanced phosphate efflux. At the same time, hyphal poly-P content decreased with the addition of the sugars. Furthermore, the enhancement of phosphate efflux caused by the sugars, 3.5 µg P g−1 fresh hyphae h−1, was comparable to the decrease of hyphal poly-P content, 3.7 µg P g−1 fresh hyphae h−1, suggesting that the addition of sugars increased poly-P hydrolysis and that the phosphate released was effluxed. An analogue of glucose, 2-deoxyglucose, can be absorbed and phosphorylated to glucose-6-phosphate by fungal cells but cannot be further metabolised. Therefore, at least a part of the phosphate efflux from the intraradical hyphae was closely related to glucose uptake and its subsequent phosphorylation (Woolhouse, 1975). In fact, phosphotransferase systems for sugar transport are common among fungi, but there is a diversity of transport systems (Griffin, 1994).
The role of alkaline phosphatase (ALPase) expressed in arbuscules in relation to the C-P exchange has been disputed (Gianinazzi et al., 1979; Tisserant et al., 1992; Larsen et al., 1996). Larsen et al. (1996) found that benomyl inhibited P translocation through hyphae but did not inhibit ALPase; however, the inhibition of ALPase with benomyl has been reported elsewhere (Thingstrup & Rosendahl, 1994). However, inhibition of overall P translocation by means of benomyl may not necessarily indicate that ALPase is not involved in phosphate efflux from the intraradical hyphae. Ezawa et al. (1999) investigated the characteristics of ALPase extracted from intraradical hyphae of G. etunicatum. They hypothesised that some of the phosphate released from intraradical hyphae may be provided by ALPase via hydrolysis of intermediary sugar phosphates because of its low Km to sugar phosphate compounds and its inability to hydrolyse pyrophosphate compounds. Assuming that ALPase is indeed involved in phosphate efflux, the present findings suggest that a part of glucose-6-phosphate derived from the absorbed glucose is hydrolysed with the ALPase and that the phosphate is then effluxed. For further studies, stoichiometric approaches to absorbed C and effluxed P and specific inhibitors for transport processes and other analogue compounds of glucose should be used.
Based upon the localisation of H+-ATPase, it has been suggested that phosphate efflux may occur in arbuscular hyphae while carbon compounds may be absorbed in the intercellular hyphae (Gianinazzi-Pearson et al., 1991; Smith & Read, 1997). The present results show that these processes may be closely related to each other and may be recognised as a coupled exchange process. However, we have no experimental observations so far that indicate whether phosphate efflux and C absorption occur at the same site.
Estimation of P transfer based upon Experiments 1 and 2
We have now estimated the hyphal P distribution in the AM system and the phosphate efflux from intraradical hyphae in vitro. Based upon these data and the following assumptions, we estimated the translocation and transfer of P from soil to host plant during the period from 6 to 9 wk after transplanting.
P in the fungus was composed of three pools: total P excluding poly-P, poly-P and spore P.
Translocation of P from the extraradical to the intraradical part and transfer of P from the fungus to host passed through the poly-P pool in each part. Although poly-P was composed of various chain lengths, poly-P pools were homogenous.
The P transfer from the intraradical hyphae to the host was equivalent to the increase in phosphate efflux caused by the addition of glucose in experiment 2. The phosphate efflux was constant from 6 to 9 wk.
Poly-P contents in the hyphae were presented in our previous report (Solaiman et al., 1999). The amount of P given to the host by the fungus was estimated by the following formula. The original formula was developed for estimation of nutrient inflow through growing roots (Nye & Tinker, 1977). Instead of length, we used hyphal fresh weight since the hyphal specific length was almost constant during the growing period (Solaiman et al., 1999).
- (Eqn 1 )
(W1, W2, fresh weight of intraradical hyphae at harvesting time T1 and T2, respectively; ΔP, P transfer from fungus to host during the 3 wk (P mg pot−1).) Incorporating W1, 99 mg pot−1, W2, 231 mg pot−1, T2−T1, 3 wk, and P efflux, 3.5 µg P g−1 fresh hyphae h−1 from experiment 2, ΔP, 275 mg P pot−1 was obtained.
Figure 5 schematically represents the estimated P translocation and transfer in AM symbiosis. Vigorous protoplasmic streaming has been observed in the extraradical hyphae of AM fungi, and its movement is bi-directional (Cine-document, 1998). Therefore, the flow of P translocation in Figure 5 indicates the net flow. This figure demonstrates several interesting points. First, about 12% of P absorbed by the host plant was received from the fungus. This is markedly lower than the values obtained by the hyphal compartment experiments (Li et al., 1991a,b; Pearson & Jakobsen, 1993). However, in those compartment experiments, available P in the root compartment tended to be depleted so that the contribution of hyphae to total P uptake may have been higher than that under non-compartment growing conditions as in the present study.
Figure 5. Estimated phosphorus flow in arbuscular mycorrhizal onion roots between 6 and 9 wk after transplanting (µg P per pot). Numbers in italics are the estimated values. Numbers in roman are values obtained in the experiments or from the literature. The experimental values are at 6 (top) and 9 (bottom) wk, respectively.
Download figure to PowerPoint
Second, the turnover of the poly-P pool was estimated as follows.
- (Eqn 2)
By using the mean value of the poly-P pool size at 6 and 9 wk, the turnover was estimated to be about 0.74 day−1 for extraradical hyphae and 0.32 day−1 for intraradical hyphae. It is still uncertain if poly-P is the only form of P translocation. Glomus manihotis was found to contain no poly-P granules in its hyphae based upon a DAPI stain (Boddington & Dodd, 1999). However, the hyphae of G. margarita contained substantial amounts of poly-P (Solaiman et al., 1999), and the poly-P contents in the hyphae of G. coronatum were found to depend on P availability in the culture medium (Ezawa et al., 2001). Therefore, the fast turnover of poly-P is not surprising.
Third, we estimated P inflow in the extraradical hyphae according to the following formula (Nye & Tinker, 1977):
- (Eqn 3)
(L1, L2, length of extraradical hyphae at harvesting time T1 and T2, respectively; ΔP, estimated phosphorus absorption by the extraradical hyphae (187 µg pot−1 3 wk−1).) Incorporating L1, 270 m pot−1, L2, 388 m pot−1, T2−T1, 3 wk, a P inflow of 10.9 f mol P m−1 s−1 was obtained. This value was within the range of previous estimates (1–430 f mol P m−1 s−1) (Tinker, 1975; Li et al., 199a; Jakobsen et al., 1992) and was most similar to that estimated for indigenous AM fungi of field-grown wheat (38 f mol P m−1 s−1, Schweiger & Jakobsen, 1999). We used the fragmented hyphae separated from roots, so the actual P transfer in situ might be higher than the experimental values obtained in vitro. This suggests that the actual P transfer may not be very different from the values obtained in the present in vitro experiment. In the present study we did not evaluate arbuscular surface area, so we could not estimate P flux across the arbuscular surface (Smith et al., 1994).
The present study is the first report on the distribution of fungal P in extraradical and intraradical parts of the G. margarita colonising onion, and on the phosphate efflux from the intraradical hyphae separated from host roots and its enhancement with glucose. Based upon these data and a number of assumptions, we estimated the translocation and transfer of P in AM symbiosis. Our estimates more or less agree with the estimates previously arrived at by other methodologies. In conclusion, by applying the technique of separating intraradical hyphae from host roots, we were able to clarify the translocation and transfer of P in AM symbiosis in a quantitative manner. Further investigation of phosphate efflux using the present experimental system will shed light on the mechanisms of nutrient exchange.