Ectomycorrhizas increase apoplastic water transport and root hydraulic conductivity in Ulmus americana seedlings


  • T. M. Muhsin,

    1. Department of Renewable Resources, 4–42 Earth Science Building, University of Alberta, Edmonton, Alberta T6G 2E3, Canada
    Search for more papers by this author
  • J. J. Zwiazek

    Corresponding author
    1. Department of Renewable Resources, 4–42 Earth Science Building, University of Alberta, Edmonton, Alberta T6G 2E3, Canada
      Author for correspondence: J. J. Zwiazek Tel: +1 780 492 2358 Fax: +1 780 492 1767 Email:
    Search for more papers by this author

Author for correspondence: J. J. Zwiazek Tel: +1 780 492 2358 Fax: +1 780 492 1767 Email:


  • • The extent to which water channel transport is responsible for the observed increases in root water flow of ectomycorrhizal plants is reported here.
  • • To examine the contribution of water channel transport to root hydraulic conductance, temperatures in the range 4–20°C and mercuric chloride (HgCl2) were used to study the kinetics of water transport in ectomycorrhizal and nonmycorrhizal roots of American elm (Ulmus americana) seedlings.
  • • Hydraulic conductance declined with decreasing temperatures in both mycorrhizal and nonmycorrhizal seedlings. However, hydraulic conductance and conductivity were higher in the mycorrhizal than the nonmycorrhizal roots at all temperatures studied. Mercuric chloride had a relatively greater impact on root hydraulic conductance in nonmycorrhizal than mycorrhizal roots and activation energy for root hydraulic conductance was significantly higher in mycorrhizal than nonmycorrhizal plants.
  • • The results suggest that ectomycorrhizal hyphae increase hydraulic conductance of roots by decreasing water flow resistance of the apoplast rather than by water channel-mediated transport. The high rates of hydraulic conductance at low root temperatures might be important to plants growing in cold soils and under other challenging environmental conditions that inhibit metabolism and limit water transport.


Ectomycorrhizal associations can have positive, negative or neutral effects on plant water relations (Sands & Theodorou, 1978; Dixon et al., 1980; Coleman et al., 1990). Sands & Theodorou (1978) found no difference in transpiration rates between inoculated and uninoculated plants, but leaf water potential was more negative and hydraulic resistance was higher in the inoculated plants. Water uptake by ectomycorrhizal mycelium strands has been demonstrated (Duddrige et al., 1980), but the mechanism of water transport in ectomycorrhizal roots remains to be examined. Coleman et al. (1990) reported that root hydraulic conductivity of ectomycorrhizal plants increased due to an increase of phosphorus level. Altered plant morphological characteristics, specifically root : shoot ratio, were also proposed to be responsible for the mycorrhizal effect on plant water flux (Anderson et al., 1988) and root hydraulic conductivity was related to a reduced root : shoot ratio in mycorrhizal plants (Dixon et al., 1980). Other studies with arbuscular mycorrhizal fungi suggested that the water extraction and transport by plant roots could be enhanced by mycorrhizae due to high transpirational demand (Kothari et al., 1990; Koide, 1993).

Radial water transport in roots occurs via apoplastic and cell-to-cell pathways (Steudle & Peterson, 1998). A recent discovery of water channels and subsequent reports that root water channels may play an important role in regulating plant water flow (Chrispeels & Maurel, 1994; Daniels et al., 1994; Wan & Zwiazek, 1999; Kamaluddin & Zwiazek, 2001) have challenged traditional views of water transport. In our previous studies (Wan & Zwiazek, 1999; Wan et al., 1999; Kamaluddin & Zwiazek, 2001; Wan et al., 2001), we showed a link between root metabolism and root water flow, largely through the effects on water channel proteins. Similarly, N and P deprivation has been shown to affect water channel activity and reduce root hydraulic conductivity (Carvajal et al., 1996; Clarkson et al., 2000). Therefore, it is possible that the increases in root water flow observed in mycorrhizal plants could be due to nutritional or metabolic effects of these associations on the activity of root water channels.

In the present study, we investigated to what extent water channel transport is responsible for the observed increase in root water flow of ectomycorrhizal plants. We studied the hypothesis that the increase in hydraulic conductivity is due to changes in root water channel transport that is altered by fungal associations with roots. Therefore, we used different temperatures and HgCl2 to study the kinetics of water transport in ectomycorrhizal and nonmycorrhizal roots of American elm (Ulmus americana) to examine the contribution of water channel transport to root hydraulic conductance.

Materials and Methods

Plant growth

American elm (Ulmus americana L.) seeds were germinated and seedlings grown in a growth chamber for 12 wk using (55 cm3) Spencer-Lemaire containers (Spencer-Lemaire Industries Ltd, Edmonton, Alta, Canada). The containers were uniformly filled with a mixture of autoclaved sphagnum peat moss and sand (3 : 1, by volume) with or without added ectomycorrhizal inoculum. The seeds were collected from the trees at the University of Alberta campus (Edmonton, Alta). Seeds were surface-sterilized with 30% H2O2 for 10 min, washed with sterilized distilled water and one seed was placed per container cavity. Following germination, seedlings were placed in a growth chamber under 20/15°C (day/night) temperatures and 16-h photoperiod with approx. 250 µmol m−2 s−1 of photosynthetically active radiation. Seedlings received half-strength Hoagland’s solution twice a week for 4  wk. Since elm is frequently found in nutrient-poor sites, we reduced nutrient level by growing both nonmycorrhizal (control) seedlings and those inoculated with Hebeloma crustuliniforme for a further 8 wk without added fertilizer before measurements. The seedlings were irrigated every second day using spray nozzles to avoid moisture saturation and possible nutrient leaching.

Fungal inoculum

A pure culture of ectomycorrhiza Hebeloma crustuliniforme (Bull. Ex St. Amans) Quel. (Isolate no. 5247) was obtained from Dr D. Khasa, University of Alberta. Fungal culture was maintained on Potato Dextrose Agar medium. Fungus inoculum from the pure culture was prepared in Melin Norkrans Nutrient liquid medium and used for seedling inoculation by mixing the autoclaved soil with the fungal inoculum. At the end of the experiment, root samples were washed gently to estimate the ectomycorrhizal colonization percentages per root segment length using a direct microscope examination of the new roots colonized by hyphal sheath as described by Schenck (1982). Control experimental group consisted of seedlings that were not inoculated with fungus.

Hydraulic conductance (Kr) measurements

We measured Kr using the High Pressure Flow Meter (HPFM) (Dynamax Inc., Houston, TX, USA) as described by Tyree (1997). This method makes the measurements of hydraulic conductance possible in undisturbed roots. Root plugs were gently removed from the Spencer-Lemaire containers to avoid root disturbance and the plugs were immersed in half-strength modified Hoagland’s solution (Epstein, 1972). Whole intact roots were connected to the HPFM system through the shoot excised 2 cm above the root collar. Root systems were pressurized to 0.4 MPa to obtain a pressure-flow relationship. Kr measurements were conducted in eight mycorrhizal and eight nonmycorrhizal root systems (n = 8). For Arrhenius plot determinations, root temperature was controlled by a water bath (Frigomix B. Braun, Germany) and changed from 20°C to 4°C in 4°C steps (Wan & Zwiazek, 1999). The activation energy (Ea) was calculated in descending temperature curves for mycorrhizal and nonmycorrhizal roots using the Arrhenius equation. The Arrhenius plots were obtained by plotting the logarithm of hydraulic conductance against the inverse of absolute temperatures and Ea was calculated from the slope of the whole curve. To calculate root hydraulic conductivity (Lp), the root surface areas were obtained following scanning of roots spread on a ScanMaker III computer scanner (Microtek Laboratory, Inc., Redondo Beach, CA, USA) equipped with a transparent media adapter, and using Sigma Scan 3.0 scanning software (Jandel Scientific, San Rafael, CA, USA) as earlier described (Wan & Zwiazek, 1999).

Inhibition of Kr by HgCl2 and its reversibility by 2-mercaptoethanol (ME)

Hydraulic conductance was measured in mycorrhizal and nonmycorrhizal root systems (n = 8) and the HPFM was depressurized. HgCl2 was injected into the root medium containing half-strength Hoagland’s solution, to reach a concentration of 0.1 mM. Kr was measured every 15 min for 60 min and then mercaptoethanol (ME) was injected into the root medium to provide a concentration of 50 mM (Wan & Zwiazek, 1999). The measurements of Kr continued until stable readings were reached (approx. 2 h). Distilled water was injected in place of HgCl2 and ME for control roots. The roots were maintained at 20°C throughout the measurements.

Hydraulic resistance and water viscosity

The hydraulic resistance for mycorrhizal and nonmycorrhizal roots was calculated for temperatures ranging from 4°C to 20°C as a reciprocal of Kr and the values that were calculated for 20°C were set to 100%. Similarly, water viscosity was calculated for the same range of temperatures (Reynolds & Richards, 1996) and compared with the value calculated for 20°C that was set to 100%.

Data analysis

ANOVA and linear regression analyses of the data were performed using SAS version 6.12 (SAS Institute Inc., Cary, NC, USA). Means were compared for significant differences at 0.05 level using the Tukey’s test.


Mean colonization of roots by the inoculated fungus in mycorrhizal seedlings was approx. 55% and control roots showed no signs of fungal colonization at the end of the experiment. Both mycorrhizal and nonmycorrhizal seedlings were approx. 9–10-cm tall at the time of measurements. Nonmycorhizal seedlings had larger root system with an average surface area of 7.3 ± 1.1 cm2 compared with 4.1 ± 0.9 cm2 in mycorrhizal plants, resulting in a larger root : shoot ratio. Visual examination of roots did not reveal apparent differences in the proportion of woody to nonwoody roots. Although the measurements were conducted in a few seedlings only, the analysed samples showed reductions of up to 50% in leaf chlorophyll, nitrogen and phosphorus levels in nonmycorrhizal seedlings compared with the mycorrhizal plants.

In both mycorrhizal and nonmycorrhizal roots, Kr increased with the increasing root area (Fig. 1). Mycorrhizal roots exhibited several-fold higher Kr at all studied temperatures compared with nonmycorrhizal roots (P = 0.05) (Fig. 2). In both root types, Kr declined with declining temperature from 20°C to 8°C. There was no significant difference between the conductance values within each group when measured at 4°C and 8°C (Fig. 2). The lowest Kr value measured in mycorrhizal roots at 4°C was still higher than that of the highest Kr measured in nonmycorrhizal roots at 20°C. Arrhenius plots for Kr hydraulic conductance were constructed by plotting ln Kr against 1/T (K) (Fig. 3). Ea values calculated from the regression lines of Arrhenius plots were 4.637 and 2.442 kcal mol−1 for mycorrhizal and nonmycorrhizal root systems, respectively, and the values were significantly different (P = 0.05)

Figure 1.

Relationship between root hydraulic conductance (Kr) of mycorrhizal (circles) and nonmycorrhizal (squares) seedlings and root area. Each point represents one root system.

Figure 2.

Root hydraulic conductance (Kr) of mycorrhizal (circles) and nonmycorrhizal (squares) seedlings exposed to decreasing temperatures. Means (n = 8) ± SE are shown. Asterisks indicate statistically significant difference (P = 0.05) between mycorrhizal and nonmycorrhizal roots.

Figure 3.

Arrhenius plots for hydraulic conductance of mycorrhizal (circles) and nonmycorrhizal (squares) roots exposed to decreasing temperatures. Points are the means of eight seedlings and the regression lines used to calculate activation energies (Ea) are shown.

Lp, calculated as Kr expressed on the root area basis, showed a similar response to root temperature as Kr (Fig. 4). The highest Lp was measured at 20°C and the lowest at 4°C. A significantly (P = 0.05) higher Lp was measured at all temperatures in mycorrhizal plants compared with the nonmycorrhizal root systems.

Figure 4.

Hydraulic conductivity (Lp) in mycorrhizal (circels) and nonmycorrhizal (squares) seedlings exposed to different temperatures. Means (n = 8) ± are shown. Asterisks indicate statistically significant difference (P = 0.05) between mycorrhizal and nonmycorrhizal roots.

Treatment of nonmycorrhizal roots with HgCl2 resulted in an inhibition of Kr from over 3 × 10−7 kg s−1 Mpa−1 to approx. 1.2 × 10−7 kg s−1 Mpa−1 within 45 min (Fig. 5). After the application of 50 mM ME to the root medium, Kr increased within 60 min to approx. 1.6 × 10−7 kg s−1 Mpa−1 (Fig. 4). In mycorrhizal roots, Kr declined within 45 min after adding 0.1 mM HgCl2 from 6 × 10−7 kg s−1 Mpa−1 to approx. 4.8 × 10−7 kg s−1 Mpa−1. Within 60 min following ME application, Kr increased almost to the control level (Fig. 5).

Figure 5.

Root hydraulic conductance (Kr) of mycorrhizal (circles) and nonmycorrhizal (squares) American elm seedlings treated with 0.1 mM HgCl2 (Hg) (arrows) followed by 50 mM mercaptoethanol (ME) (arrows). Means (n = 8) ± are shown.

Root hydraulic resistance increased with decreasing temperatures most rapidly in the nonmycorrhizal plants (Fig. 6). The increases in hydraulic resistance in both mycorrhizal and nonmycorrhizal roots with decreasing temperature were higher than the increase in water viscosity calculated for these temperatures (Fig. 6).

Figure 6.

Relative viscosity of water (triangles) and relative hydraulic resistance (reciprocal of Kr) of mycorrhizal (circles) and nonmycorrhizal (squares) seedlings with the values at 20°C set to 100%. Means of eight seedlings are shown for hydraulic resistance at each temperature. Asterisks indicate statistically significant difference (P = 0.05) between mycorrhizal and nonmycorrhizal roots.


VA mycorrhizas (Hardie & Leyton, 1981; Graham & Syvertsen, 1984; Bildusas et al., 1986) and ectomycorrhizas (Safir et al., 1972; Dixon et al., 1980) have lower hydraulic resistance to water uptake than uncolonized roots. The present results demonstrated that ectomycorrhizal roots exhibited higher Kr than control roots at all studied temperatures. The increase in Kr of mycorrhizal roots may be of ecological significance for plants growing in cold soils since the low rate of water flow in roots exposed to low temperatures was reported to be the principal reason for their slow growth (Wan et al., 1999). In the present study, ectomycorrhizal seedlings possessed smaller root surface area compared to the nonmycorrhizal seedlings. This is in agreement with Kothari et al. (1990) who reported that the root length in mycorrhizal maize was reduced by 31% and, in consequence, the water uptake per unit length of root was higher than in nonmycorrhizal plants. In the present study, both Kr and Lp values were greater in mycorrhizal compared with nonmycorrhizal roots at all studied temperatures (Figs 1, 2, 4) suggesting that the capacity for water flow of the root system as well as that of the individual roots was greater in mycorrhizal compared with the nonmycorrhizal plants. However, for Lp determinations, absorbing root surface area of mycorrhizal plants is probably underestimated since it does not include most of the hyphal mass.

Our present study showed that Kr in the mycorrhizal roots was more temperature-dependent than that in the nonmycorrhizal roots as evidenced by higher Ea values. Since low Ea values are usually associated with water transport through water-transporting pores (Chrispeels & Agre, 1994; Maurel, 1997), higher Ea values in mycorrhizal roots suggest that the contribution of mycorrhizae to root hydraulic conductance was largely due to other factors than the water channel transport. More evidence in support of this conclusion came from the HgCl2 experiment which showed a relatively lower degree of Kr inhibition by HgCl2 in mycorrhizal roots (< 20%) compared with that in the nonmycorrhizal roots (> 60%). Mercurial reagents are known to inhibit the water transport across membranes of plant cells (Maggio & Joly, 1995; Maurel, 1997; Wan & Zwiazek, 1999) and such inhibition is often used to verify the existence of water channels (Chrispeels & Maurel, 1994; Wan & Zwiazek, 1999).

A comparison of temperature effect on root hydraulic resistance (reciprocal of Kr) with that on water viscosity showed that, as with Wan et al. (2001), in the present study, higher water viscosity at lower temperatures could not fully explain an increase in root hydraulic resistance. However, the relationship between water viscosity and root hydraulic conductivity was significantly closer at lower temperatures in mycorrhizal roots compared with the nonmycorrhizal roots suggesting that water physical properties had a greater influence on water flow in mycorrhizal plants. In our earlier studies (Wan & Zwiazek, 1999, 2001; Kamaluddin & Zwiazek, 2001; Wan et al., 2001), we showed a strong metabolic control of root water channel functioning and its effect on root hydraulic conductivity. Water channels function is affected by a number of internal and environmental factors including salinity (Azaizeh & Steudle, 1991; Carvajal et al., 2000), low oxygen content of the root medium (Steudle & Heydt, 1997), root morphology (North & Nobel, 1993), temperature (Wan et al., 1999), mineral nutrition (Clarkson et al., 2000) and may be controlled by both metabolic and hormonal factors (Quintero et al., 1999). Our results indicate that the contribution of pore-mediated water transport to the total root water flow decreased in mycorrhizal seedlings. Since increases in root metabolic activity (Johansson et al., 1998; Kamaluddin & Zwiazek, 2001) and in the supply of nitrogen and phosphorus (Carvajal et al., 1996; Clarkson et al., 2000) enhance root water flow through their promoting effects on water channel activity, the decrease in the proportion of pore-mediated transport in mycorrhizal roots suggests that, contrary to our original hypothesis, the effect of ectomycorrhizas on root water flow was probably not due to metabolic or nutritional effects. However, more work is needed to better understand nutritional effects on root water transport. Fungal hyphae normally penetrate through the root central cortex and do not reach the endodermis which was also observed in root sections randomly prepared in the present study. Therefore, the large increase in root hydraulic conductivity of ectomycorrhizal seedlings provides additional support for the view that the major resistance to root water flow is not in the endodermal layer, but that it is spread over the living tissues of the root (Peterson & Steudle, 1993; Steudle & Peterson, 1998).

Clearly, more work will be required to determine the exact mechanisms of water flow regulation in mycorrhizal roots. Nonetheless, it is evident from the present study that the extraradical mycelia and the apoplastic hyphal net of ectomycorrhizal roots play an important role in the uptake and transport of water to the host plants and that water channel-mediated transport does not dominate flow characteristics in ectomycorrhizal elm roots.


We acknowledge research funds for this study from the Natural Sciences and Engineering Research Council in the form of a research grant to JJZ.