Effects of Benomyl and Drought on the Mycorrhizal Development and Daily Net CO₂ Uptake of a Wild Platyopuntia in a Rocky Semi‐arid Environment

Abstract

The effects of drought and the fungicide benomyl on a wild platyopuntia, Opuntia robusta Wendl., growing in a rocky semi‐arid environment were assessed. Cladode phosphorus content, cladode water potential and daily net CO₂ uptake were measured monthly in 2000 and 2001 before, during and after the summer rainy period. During 2000, the formation of new roots and new cladodes was severely suppressed in response to a prolonged drought, impairing the development of the symbiotic relationship between the arbuscular mycorrhizal (AM) fungi and the roots. Hence no effect of benomyl application was observed on daily carbon assimilation by this Crassulacean acid metabolism plant. During 2001, drought was interrupted, and new cladodes and roots were formed in response to rainfall. Benomyl was highly effective in suppressing root colonization by AM‐fungi; however, daily C assimilation was reduced by benomyl application only in October. Thus, the inhibition of AM‐fungal colonization by benomyl did not affect photosynthesis, water uptake and P uptake under prolonged drought.

Received: 18 December 2002; Returned for revision: 28 April 2003; Accepted: 1 May 2003    Published electronically: 18 June 2003

INTRODUCTION

Inselbergs or rocky environments are isolated rises above a plain, consisting of hard bedrock (Bremer and Sander, 2000). Despite their widespread presence in temperate and tropical regions, inselbergs have been largely ignored as subjects of ecosystem research (Barthlott and Porembski, 2000). Rocky environments are common in the semi‐arid environments of central Mexico and are usually medium‐sized islands in which vegetation patches are formed by groups of plants growing in crevices or depressions that trap wind‐blown organic debris, minerals and water, thereby creating patches of different size that give rise to ‘islands of fertility’ (Pimienta‐Barrios et al., 2002a). A reduced number of species grow in these places, including Crassulacean acid metabolism (CAM) plants, such as members of genera Agave and Opuntia, and some C₃ plants, such as species of the genus Yucca (Pimienta‐Barrios et al., 2002a).

Inselbergs are highly stressful environments because of the extremely low availability of water and mineral nutrients in time and space (Lüttge, 1997; Kluge and Brulfert, 2000; Porembski et al., 2000; Szarzynsky, 2000), conditions that are considered favourable for mycorrhizal activity (Smith and Read, 1997; Orcutt and Nilsen, 2000). Nevertheless, data on mycorrhizal function and interactions in an ecophysiological context are lacking for plants growing in inselbergs, especially for CAM plants (Lüttge, 1997; Kluge and Brulfert, 2000). Furthermore, CAM plants have been less studied with regard to the effects of mycorrhizal symbiosis on photosynthesis (Cui and Nobel, 1992) compared with C₃ plants (Davies et al., 1993; Syvertsen and Graham, 1999; Estrada‐Luna et al., 2000).

The use of the fungicide benomyl, a benzimidazole, to chemically exclude arbuscular‐mycorrhizal (AM) fungi in natural and cultivated vegetation is valuable for studying the role of mycorrhizal symbiosis in plant functioning in the field (Carey et al., 1992; Merryweather and Fitter, 1996; Graham and Eissenstat, 1998; Wilson et al., 2001). Benomyl has selective activity against AM‐fungi and rarely has a phytotoxic effect (Paul et al., 1989; Pedersen and Sylvia, 1997; Kahiluoto et al., 2000). In the present study, ecophysiological responses were examined for two growing seasons in a wild population of the CAM species Opuntia robusta Wendl. in a rocky environment exposed to benomyl applications at the end of a prolonged drought. It was hypothesized that the survival and development of O. robusta in a rocky semi‐arid environment would be physiologically dependent on the establishment of mycorrhizal symbiosis by AM‐fungi. Therefore, if the development of mycorrhizal symbiosis for O. robusta is restrained by benomyl, basic physiological processes such as photosynthesis will be affected compared with those in plants for which the symbiosis develops naturally, because the uptake of critical elements for photosynthesis, such as phosphorus and water, will be reduced.

MATERIALS AND METHODS

Site and plant description

Measurements were made from May 2000 to November 2001 in a rocky community near El Rayo, Zacatecas, Mexico (21°58′N, 101°35′W; 2190 m a.s.l.). The site is in the south‐western corner of the Chihuahuan Desert on highlands of the semi‐arid region of north central Mexico on a plateau known as Altiplano Potosino‐Zacatecano (Medellín‐Leal, 1982). Annual rainfall averages 500 mm and occurs primarily in the summer; the average annual temperature is 16–18 °C (Pimienta‐Barrios, 1994). Rainfall data were obtained from an official weather station maintained by the Instituto Nacional de Investigaciones Forestales y Agropecuarias near Ojuelos, Jalisco (21°52′N, 101°37′W; 2100 m a.s.l.).

Opuntia robusta var. robusta is a perennial shrubby cactus that is 0·5–1·0 m tall, with a poorly defined trunk made up of thick, orbicular, bluish‐green cladodes. It is the dominant perennial species at the study site. It grows isolated or in association with other species, resulting in a patchy distribution of the vegetation interspersed with bare rock outcrops. Vegetation at the study site is of the crassicauleous brushwood type (Rzedowski, 1978). The soil is shallow, brown, and of the haplic planosol type, with a sandy loam texture. Underneath the canopy of O. robusta, the soil has a pH of 4·5, 24 µg g^–1 extractable P and 1·1 % organic matter content, whereas soil outside the patches has a pH of 5·9, 25 µg g^–1 extractable P and 1·2 % organic matter content.

Experimental design

The experiment consisted of two treatments, a control without the fungicide benomyl (–B), and a treatment with benomyl (+B), replicated five times. Ten mature plants of O. robusta were selected at the study site, their rhizosphere area was estimated, and the plants were then randomly assigned to a treatment. Benomyl (DuPont Benlate, Willmington, DE, USA; 50 % active ingredients) was applied to the rhizosphere area of the selected plants as an aqueous suspension of 4 g l^–1 at a rate of 5 l m^–2 to give a final benomyl dosage of 20 g m^–2. The applications were made twice each month throughout the rainy seasons (May to September) of 2000 and 2001. The five control plants were irrigated with the same amount of water. The relatively high benomyl dosage was based on preliminary measurements with a lower benomyl concentration (1·4 g l^–1) that did not exert significant effects on physiological traits of treated plants; a similar high dosage of benomyl has been used to suppress AM‐fungi in Citrus roots (Graham and Eissenstat, 1998).

Gas exchange measurements

Net CO₂ uptake was measured every 2 h over 24‐h periods on 19–20 May, 28–29 Jun., 2–3 Aug., 5–6 Sep., 5–6 Oct. and 1–2 Nov. 2000, and on 22–23 Jun., 19–20 Jul., 24–25 Aug., 20–21 Sep., 18–19 Oct. and 29–30 Nov. 2001. Measurements were made on five plants per treatment using a portable photosynthesis system (LI‐6200; Li‐Cor, Lincoln, NE, USA). A 0·25 l leaf chamber was modified by replacing the distal half‐cylinder with a narrowed opening (2 cm × 4 cm) lined with a closed‐pore foam gasket that was firmly pressed against an approximately south‐west‐facing surface of a cladode. Total daily values of net CO₂ uptake were obtained by integrating the instantaneous rates over 24 h.

On the dates of gas exchange measurements and on 18–19 May 2000, the photosynthetic photon flux density (PPFD, wavelengths of 400–700 nm) on a horizontal plane was recorded hourly from sunrise to sunset using a Li‐Cor LI‐250 quantum sensor, and then integrated to obtain the total daily PPFD. Air temperature was recorded every hour using a mercury thermometer. Air relative humidity was recorded hourly using a digital humidity gauge.

Soil and plant analysis

During 2000 and 2001, soil water potential (Ψ_soil) was determined using a WP4 dew point potentiometer (Decagon Devices, Pullman, WA, USA) for one sample removed from the root zone of each of the plants under study (at a depth of 10 cm) on the same days as gas exchange was measured. Also during 2001, cladode water potential (Ψ_stem) was determined using the WP4 potentiometer for five stem samples 2·0 cm in diameter. Similar stem samples oven‐dried at 80 °C were ground to a powder; 0·5 g was digested overnight in 4 ml nitric acid before 2 ml perchloric acid was added and, after 1 h, the mixture was heated to 170 °C. Stem phosphorus (P) content was assayed colorimetrically using a vanadate–molybdate method (Wolf et al., 1991).

Mycorrhizal colonization

Fine rain‐induced roots were collected from each plant from July to September 2000 and from July to October 2001, fixed in FAA (formalin : acetic acid : ethanol : water, 10 : 5 : 50 : 35 by volume), and cut into 1·5‐cm segments. The segments were washed, cleared in 10 % (w/w) KOH, and stained with trypan blue (Phillips and Hayman, 1970). Stained segments were mounted on slides, and the percentages of root length containing hyphae, arbuscules and vesicles were assessed under a Zeiss Sinoptic microscope (Oberkochen, Germany) using the magnified intersection method (McGonigle et al., 1990).

Statistical analysis

Significances of differences between treatments for monthly daily net CO₂ uptake, the average over the measurement period, colonization of roots by AM‐fungi, cladode P content, and cladode water potential were determined using a t‐test; when data did not follow a normal distribution, the non‐parametric Mann–Whitney procedure was used. Percentage colonization was arcsine‐transformed before statistical analysis to achieve normality (Zar, 1999).

RESULTS

The region experienced a prolonged 6‐year drought beginning in 1994 and ending in 2001. Annual rainfall over the period was 317 mm in 1994, 344 mm in 1995, 372 mm in 1996, 279 mm in 1997, 345 mm in 1998, 188 mm in 1999 and 255 mm in 2000, with 470 mm being the historical annual average (1964–1984). Rainfall increased to 512 mm in 2001, interrupting the drought. During 2000, 29 % of the annual precipitation occurred in May and 61 % occurred between June and September, whereas in 2001, May received 17 % of the annual precipitation and 71 % occurred from June to September. There was only 11 mm of rainfall in June 2000, making it the driest month of the rainy season, whereas June 2001 was the wettest month, with 160 mm of rainfall (Fig. 1).

Total daily PPFD in 2000 ranged from 38 mol m^–2 d^–1 in October and November to 65 mol m^–2 d^–1 in June, and in 2001 from 33 mol m^–2 d^–1 in August to 51 mol m^–2 d^–1 in October (Table 1). The reduction due to cloudiness during the summer of 2001 was not observed in the drier, clearer summer of 2000 (Table 1). Mean day/night air temperatures showed less variation during 2000 compared with 2001. Average relative humidity was lower during the summer of 2000 than the summer of 2001. During 2000, the soil water potential rose from –35 MPa in May to –1·9 MPa in June, then decreased, and reached its maximum in October after a heavy rainfall (Table 1). During 2001, the soil water potential was highest in June and September, when measurement dates coincided with rainfall events (Table 1).

Mycorrhizal colonization

In 2000, few fine roots were formed, so the effect of benomyl on colonization by AM‐fungi could not be statistically evaluated. During 2001, the amount of fine roots increased dramatically in comparison with 2000. All the fungal structures (hyphae, arbuscules and vesicles) were present throughout the rainy season in the –B plants; in contrast, colonization of roots of +B plants by AM‐fungi was inhibited by benomyl application in 2001. The percentage of roots colonized by hyphae for the –B plants increased from July and August to September and October 2001 (Fig. 2A). For the +B plants, percentage hyphal colonization was lowest in August. The inhibition of hyphal colonization for +B plants ranged from 53 % of the percentage recorded for –B plants in July to 79 % in September 2001 (Fig. 2A). Arbuscules were absent in roots of +B plants throughout 2001, except in September, when they were reduced by 99 % compared with the control (Fig. 2B). The percentage of vesicles of AM‐fungi was similar for both treatments in July and August 2001. In September, vesicle colonization for +B plants was reduced by 97 % compared with the control, and it was absent in October (Fig. 2C).

Gas exchange

Most net CO₂ uptake by O. robusta occurred at night in 2000 and 2001, although diurnal fixation in late afternoon and early morning also contributed to total assimilation. In 2000 and 2001, the total daily assimilation for both –B and +B plants increased from early summer (June) to the middle of the summer (August), decreasing in October and November (Fig. 3A and B). Total daily assimilation was higher from June to August 2000 compared with 2001, but values from late summer (October) to early autumn (November) were higher in 2001 than in 2000. The average net daily CO₂ uptake over the study period was 279 mmol m^–2 d ^–1 for 2000 and 254 mmol m^–2 d ^–1 for 2001 (statistically non‐significant). During 2000, total daily net CO₂ uptake was statistically similar for –B and +B plants. However, during 2001, –B plants showed more total daily net CO₂ uptake than +B plants; this difference was statistically significant only in October (Fig. 3).

Stem water potential and stem phosphorous content

Benomyl application had no significant effect on Ψ_stem and cladode P content in 2000. In 2001, the lowest Ψ_stem for both –B and +B plants occurred in late spring (Table 2); the highest values occurred in July and September. During the entire period of measurement, no significant differences in Ψ_stem were observed between –B and +B plants on a particular date. Cladode P content in 2001 was also not affected by benomyl application. P content ranged from 240 µg g^–1 in May for –B to 400 µg g^–1 for both –B and +B in October (Table 2).

The population of O. robusta under study did not develop new cladodes during 2000 but did so during 2001. In November 2001, 67 new cladodes occurred per ha; 6 months later (May 2002), the number had risen to 400 new cladodes per ha.

DISCUSSION

Studies assessing the effectiveness of mycorrhizal symbiosis in the field are scarce and their findings are controversial (Fitter, 1985; Merryweather and Fitter, 1996; Smith and Read, 1997). Smith and Read (1997) cite field experiments that show that mycorrhizal colonization improves the survival of plants after short periods of exposure to drought, but have little effect over long periods, particularly in arid environments. The present findings show that the unfavourable climatic conditions during 2000, following a 6‐year drought that began in 1994, affected the development of mycorrhizae, mainly because O. robusta responded to extreme drought by arresting root development as a strategy to avoid water loss. Therefore, the availability of colonizable host tissue (fine new roots) by AM‐fungi was dramatically affected.

Rainfall created favourable conditions for growth of O. robusta during the summer and early autumn of 2001, expressed by the formation of both new cladodes and new roots. Under these conditions, AM‐fungi colonized the roots of O. robusta, and benomyl was highly efficient in suppressing root colonization by AM‐fungi. The effectiveness of benomyl in 2001 was favoured by the shallow soil at the study site and because the root system of O. robusta was confined to hollows and crevices, allowing the fungicide to readily cover the rhizosphere of the microsites explored by roots. This avoided problems of benomyl efficiency common to deep‐rooted species, in which the fungicide fails to reach the target roots (Sanders and Fitter, 1992; Jakobsen et al., 2001). The percentage suppression of root colonization by AM‐fungi for O. robusta was about 70 % during 2001. This is one of the highest levels of suppression reported, which generally range from 4 to 80 % for other species (Merryweather and Fitter, 1996; Kahiluoto et al., 2000; Smith et al., 2000; Wilson et al., 2001). Arbuscules, considered necessary for a nutritionally efficient symbiosis (Smith and Smith, 1996; Orcutt and Nilsen, 2000), were more affected by benomyl during 2001 than were hyphae or vesicles, consistent with previous findings for other species (Pedersen and Sylvia, 1997).

The suppression of root colonization by AM‐fungi did not affect photosynthesis in O. robusta in 2000. In 2001, photosynthesis was unaffected from June to September, but a significant reduction in photosynthesis occurred in October. The reduction of daily carbon assimilation by benomyl in October occurred after the time of year when colonization of fine roots by AM‐fungi was highest. Therefore, mutualistic bidirectional transfer of nutrients between the fungal structures and the host was not detected until the end of the summer. Thus, suppression of mycorrhiza in the roots of +B plants during the summer apparently becomes physiologically evident in October.

Stem P content and stem water potential for O. robusta were similar in the +B and –B plants, indicating that the suppression of root colonization did not affect the foraging for soil resources. The highest values of carbon gain observed for –B plants in October could perhaps be caused by an increased sink strength arising from additional carbon demand to sustain the process of root colonization (Wright et al., 1998), instead of the effect of an enhanced uptake of water and soil nutrients, as hypothesized. AM‐fungi colonization may require the transfer of 4–20 % of total photoassimilates (Koch and Johnson, 1984; Douds et al., 1998; Wright et al., 1998).

Unexpectedly, daily carbon gain per unit of cladode area between spring and autumn was similar in 2000 and 2001 even though the rainfall was two‐fold higher in 2001. Plants growing in nutrient‐deficient environments often show conservative physiological patterns and are less plastic both morphologically and physiologically. Even when conditions are temporarily favourable, as occurred in 2001, such plants have slow growth and low rates of photosynthesis (Grime and Hunt, 1975; Valladares et al., 2000; Pimienta‐Barrios et al., 2002b), which allow them to maintain growth and reproduction even during periods of low water availability (Grime, 1979; Chapin, 1980; Pimienta‐Barrios and Nobel, 1995).

The soil at the field site is shallow and has a high sand content, leading to quick drainage and low water‐holding capacity. Nevertheless, O. robusta could extract water from the soil even under these conditions, as the water potential of its stems in July 2001 approached that of well‐hydrated CAM plants (Goldstein and Nobel, 1991; Lüttge, 1997). AM‐fungal symbiosis can alter rates of water movement into host plants by several mechanisms, including an enhanced nutrition, increased mycorrhizal root surface area, stomatal regulation, lower resistance to water transport, hyphal water uptake and increased hydraulic conductivity (Safir et al., 1971; Cui and Nobel, 1992; Ruiz‐Lozano and Azcón, 1995; Orcutt and Nilsen, 2000). However, benomyl treatment did not affect the stem water potential of O. robusta during 2001; a large capacitance may explain the absence of an effect of the suppression of AM‐fungi on stem water potential, because transport of large volumes of water into the stem would be necessary before an appreciable change in stem water potential occurred.

Inhibition of AM‐fungal colonization of O. robusta using benomyl did not affect photosynthesis, water uptake or P uptake under dry and wet conditions, revealing that the development of this CAM species in a highly stressful environment might be regulated by other physiological strategies and mechanisms, such as avoidance, tolerance and CAM metabolism, in addition to the mycorrhizal symbiosis, particularly under prolonged drought.

ACKNOWLEDGEMENTS

We thank Luis Ortiz‐Catedral, Javier García‐Galindo, Julia Zañudo‐Hernández and Joanna Acosta‐Velásquez for field and laboratory assistance, and Pablo Torres Morán for statistical advice. CONACyT grant 28818‐B, the University of Guadalajara, and the UCLA Council on Research financially supported this project.

References

Author notes

1Departamento de Ecología, Centro Universitario de Ciencias Biológicas y Agropecuarias, Universidad de Guadalajara, 45110 Zapopan, Jalisco, México and 2Department of Organismic Biology, Ecology, and Evolution, University of California, Los Angeles, CA 90095‐1606, USA