Cellulase production by the vesicular-arbuscular mycorrhizal fungus Glomus mosseae (Nicol. & Gerd.) Gerd. and Trappe

Size: px

Start display at page:

Download "Cellulase production by the vesicular-arbuscular mycorrhizal fungus Glomus mosseae (Nicol. & Gerd.) Gerd. and Trappe"

Percival Goodman
5 years ago
Views:

1 Nezv Phytol. (1992), 121, Cellulase production by the vesicular-arbuscular mycorrhizal fungus Glomus mosseae (Nicol. & Gerd.) Gerd. and Trappe BY J. M. GARCIA-GARRIDO, I. GARCIA-ROMERA AND J. A. OCAMPO Department of Microbiology, Estacion Experimental del Zaidin, Prof. Albareda 1, E-18008, Granada, Spain {Received 25 October 1991; accepted 13 Eebruary 1992) SUMMARY Production of endoglucanase (EC ) and exoglucanase (EC ) enzymes was studied during penetration of the host and development of the vesicular-arbuscular mycorrhizal (VAM) fungus, Glomus mosseae, in roots of lettuce {Lactuca sativa) and onion {Allium cepa). Endo- and exoglucanase activities increase in VAM plants at the beginning of entry-pomt formation and arbuscule development. No relationship was found between the number of vesicles and endo- and exoglucanase activities. Extracts from spores and external mycelium of G. mosseae had endo- and exoglucanase activities. Some of the cellulase activities detected in VAM roots were attributed to the fungus, since endoglucanase activity found m the external mycelium of G. mosseae and in mycorrhizal root extracts showed the same electrophoretic mobility. Key words: Cellulases, endoglucanases, exoglucanases, Glomus mosseae, VA mycorrhiza. INTRODUCTION The colonization of plant roots by vesiculararbuscular mycorrhizal (VAM) fungi involves the formation of intercellular hyphae, highly branched intracellular arbuscules and vesicles scattered throughout the root (Bonfante-Fasolo, 1984). Arbuscules have a thin fungal wall which is separated from the host cytoplasm by interfacial material of host origin. These observations suggest that the establishment of an intracellular symbiosis between fungus and plant roots requires penetration of the host cell by fungus. Cell wall-hydrolyzing enzymes may be involved in this process (Garcia-Romera et a/., 1990). Infection of roots by pathogenic and mutualistic microorganisms such as Rhizobium and Azospirillum appears to be mediated by cell wall-hydrolyzing enzymes (Hubbel, Morales & Umali-Garcia, 1978; Um'ali-Garcia et al, 1980; Collmer & Keen, 1986). In spite of the low production of hydrolytic enzymes by these mutualistic microorganisms (Morales, Martinez-Molina & Hubbel, 1984), endo- and exoglucanase, which belong to the cellulase system (Coughlan & Ljungdhal, 1988), seem to be involved in the dissolution of the cell wall which permits Rhizobium to enter the host cell (Verma & Zogbi, 1978; Chalifour & Benhamou, 1989). The observation that VAM fungi penetrate the plant cell wall at the site of contact (Bonfante-Fasolo, 1984), and the fact that spore extracts of Glomus mosseae contain cellulolytic enzymes (Garcia-Romera et al., 1990), indicate that cellulases may be involved in the colonization process. However, since VAM fungi have not yet been cultured axenically in the absence of plant roots, it is difficult to confirm the production of cellulolytic enzymes by VA mycorrhizas or their possible participation in the colonization of the root. The aim of this work was to determine whether the VAM fungus, G. mosseae, shows endo- and exoglucanase activities, and whether these enzymes participate in the colonization of plant roots. M.^TERI.^LS.^ND METHODS Biological material and growth conditions Plants were grown in 300 ml capacity open pots of soil collected from the province of Granada, Spain. The soil was a calcexerouic xerochrept type, ph 7-6 (for full details see Garcia-Romera & Ocampo, 1988). It was steam-sterilized and mixed 1:1 (v:v) with sterilized quartz sand. Lettuce {Lactuca sativa, cv. Romana) and onion {Allium cepa, cv. Bobasa) were used as test plants. Seeds were sow-n in

$Natural light was supplemented by Sylvania incandescent and cool-white lamps providing 400 nmol m"' s-\ 400-700 nm, with a 16-8 h light-dark cycle at 25-19 C and 5 0 % relative humidity.$

2 222. M. Garcia-Garrido and others moistened sand. After 2 weeks, seedlings were transplanted to the pots and grown under greenhouse conditions. Natural light was supplemented by Sylvania incandescent and cool-white lamps providing 400 nmol m"' s-\ nm, with a 16-8 h light-dark cycle at C and 5 0 % relative humidity. Plants were watered from below using a capillary system, and fed with a nutrient solution (Hewitt, 1952), lacking phosphate for VAM-inoculated plants. The inoculum consisted of 5 g of rhizosphere soil either from maize or from alfalfa plant pot cultures of an isolate of Glomus mosseae (Nicol. & Gerd.) Gerd. and Trappe which contained spores (15 sporocarps per g with 1-5 spores per sporocarp), mycelium and colonized root fragments. ETninoculated plants were given filtered leachings from the inoculum soil. Soil filtrate (Whatman No. 1 filter paper) from the rhizosphere of mycorrhizal plants was added to uninoculated treatment. The filtrate contained common soil microorganisms, but no propagules of G. mosseae. Sporocarps of G. mosseae were isolated by wetsieving the soil (Gerdemann, 1955). Spores were obtained by dissecting the sporocarps, and were then stored in water at 4 C for use within 1 month. Before enzymatic assay, the spores were surfacesterilized (MacDonald, 1981). External mycelium was isolated from roots of 8-week-old lettuce plants colonized with G. mosseae. The roots were washed and rinsed gently with sterilized water, and the external mycelium collected with forceps under a dissecting microscope. was homogenized in 40 ml of O'l M Tris-HCl buffer (ph 7) plus 13 g polyvinyl-polypyrrolidone (PVPP), 10 mm MgClg, 10 mm NaHCOg, 10 mm/?-mercaptoethanol, 0-15 mm phenylmethyl sulphonyl fluoride (PMSE) and 0-3% (w:v) X-100 Triton. Sodium azide (0-03 %) was added to all solutions. The liquid was filtered through several layers of cheesecloth, centrifuged at j ' for 15 min, and the pellet resuspended and washed by centrifugation with the same buffer three times. The supernatant was treated with ammonium sulphate up to 80 % of full saturation, which precipitates most of the cellulase proteins (Garcia-Garrido et al., unpublished results). The solution was kept for 5 h at 4 C and centrifuged once more as described above. The supernatant was discarded, and sediment was dissolved in a small volume of the same extractant solution and was dialyzed against several hundred volumes of the same diluted extractant solutions (1:9, v:v) for 16 h at 4 C. The samples were kept frozen until use (Garcia-Garrido et al., unpublished results). Spores and external mycelia were frozen in liquid nitrogen and finely pulverized in a mortar. The resulting powder was suspended (30 mg ml~^) in the same extractant solution as for roots. The suspension was briefly sonicated (for 1 min, 5 times at 80 W), centrifuged at 20000^ for 15 min. The pellet was resuspended and sonicated again, and washed by centrifugation with the same buffer three times. The supernatant was concentrated by ultrafiltration using PM-10 membranes (Amicon), and used as a crude enzyme extract. Mycorrhizal Fnzyme assays measurements Plants were harvested at 15, 35, 50 and 80 d. The root system was washed and rinsed three times with sterilized distilled water. Parts of the root system (2 g f. wt) from each of the five replicate groups of pots were cleared and stained (Phillips & Hayman, 1970). Material from each replicate was cut into 1 cm segments that were mixed and repeatedly subdivided to yield random samples of 40 root segments. These were mounted on slides and examined under a compound microscope at x 160 magnification. The percentage of total root length which was colonized with VAM, percentage of arbuscules, numbers of entry points (appressoria) and vesicles were measured as described by Ocampo, Martin & Hayman (1980). All assays were repeated three times with similar results; in two of them, the source of G. mosseae inoculum was from maize and, in the other, inoculum was obtained from alfalfa. Preparation of extracts for enzyme assays Roots (20 g f. wt) were frozen in liquid nitrogen and finely pulverized in a mortar. The resulting powder Endoglucanase activity (EC ) was assayed viscometrically, using carboxymethylcellulose (CMC) as the substrate. Reduction in viscosity was determined in a Cannon-Fenske viscometer (5354/2) at 37 C. Six ml of reaction mixture contained 5 ml of 0-5 % CMC in 50 mm citratephosphate buffer (ph 5) and 1 ml of root or fungus extracts. Enzyme activity was expressed on a relative activity (RA) basis (reciprocal of the time in min for 50 o loss of viscosity x 1000) (Bateman, 1963). Exoglucanase activity (EC ) was determined by measuring the release of /)-nitrophenol from p-nitrophenyl-y^-d-lactopyranoside (pnlp) (Deshpande, Eriksson & Petterson, 1984). An assay mixture containing 1-8 ml of a 1 mg mf^ solution of P N E P in 0-05 M citrate-phosphate buffer (ph 5) and 0-2 ml of root or fungus extract was incubated at 37 C for 0 to 60 min. One unit was defined as the amount of enzyme which released 1 fimo\ of pnitrophenol per h. Controls for all enzyme assays were autoclaved enzyme extracts and buffers, and 0-03 % of sodium azide was added to all reaction mixtures.

3 Cellulases of VAM Glomus mosseae 223 Polyacrylamide gel electrophoresis Non-denaturing linear-gradient electrophoresis of cellulolytic enzymes in polyacrylamide gels (4 12 %) amended with 0-1 % CMC in 0-05 M Tris-0-1 M glycine buffer (ph 8.8), was based upon the method described for pectinases by Cruickshank & Wade (1980). Gels, 16x18 cm by 15 mm thick, were prepared using a gradient gel former (LKB). The electrode tank contained the same Tris-glycine buffer (ph 8-8) as used in the gel. After pre-electrophoresis of the gel for 30 min, the wells were filled with 75 ja of root or fungus extract (175/ig protein), and 1/^10-05% bromophenol blue in gel buffer was applied to the cathodic side. Electrophoresis was subsequently carried out at 4 C and a constant current of 20 ma per gel for 5 h. The gels were incubated in 100 ml of 50 mm citrate-phosphate buffer (ph 5) at 37 C for 15 h, after which they were stained with 0-1 % Congo red for 30 min. This was followed by washing in 1 M NaCl until the bands became visible. Protein determination Protein was determined by the method of Lowry et al. (1951) with bovine serum albumin (SIGMA) as the standard. non-mycorrhizal onion plants, but at 50 d this activity declined, reaching lower levels in mycorrhizal than in non-mycorrhizal plants (Fig. 3 6). Lettuce plants showed a similar pattern of development of exoglucanase activity in mycorrhizal and non-mycorrhizal plants. This activity was high when the plants were 50 d old, but at 35 and 50 d mycorrhizal plants showed more activity (differences of 0-05 and 0-02 units mg^^ protein, respectively) Figure 1. Percentage of root length colonized in lettuce (#) and onion (O) roots inoculated with Glomus mosseae. Each value is the mean of five replicates. Eor each plant, values within a row sharing the same letter were not significantly different according to Duncan's multiplerange test [P = 0-05). Statistical treatments The results were evaluated statistically by Duncan's multiple-range test. RESULTS No fungi were detected in microscopic observations of stained roots from uninoculated controls (data not shown) Typical development of VAM colonization was observed in both onion and lettuce; 15 d after transplanting, colonization increased, reaching a peak of 50 days (Fig. 1). Vesicle development reached a maximum at 35 d in lettuce roots, and declined after 50 d (Fig. la). There was no apparent decline in entry-point number, which reached a maximum at 80 d. Arbuscule development increased until day 50. In onion roots, the number of entry-points and arbuscule development reached a maximum at 50 d (Fig. 2b). There was an increase in vesicle number, which was more noticeable in the oldest plants. Endoglucanase activity was evident in lettuce and onion roots after 15 d (Fig. 3 a, b). The pattern of enzyme production differed in the two host species. In 35-d-old lettuce plants, endoglucanase activity was higher in mycorrhizal (difference 1-52 units mg"^ protein) than in non-mycorrhizal plants, although these activities were similar after 50 d (Fig. 3 a). Endoglucanase activity was greater in 35-d-old mycorrhizal (difference 1-3 units mg"' protein) than Figure 2. Entry-points (#), vesicles (O) and arbuscules ( X ) in lettuce (a) and onion (6) roots inoculated with Glomus mosseae. Each value is the mean of five replicates. Eor each fungal structure, values within a row sharing the same letter were not significantly different according to Duncan's multiple-range test (P = 0.05).

4 224. M. Garcia-Garrido and others Figure 3. Endoglucanase activity in lettuce (a) and onion (h) roots uninoculated ( ), and inoculated (O) with Glomus mosseae. Each value is the mean of three replicates. For each plant, values within a row sharing the same letter were not significantly different according to Duncan's multiple-range test {P = 0-05) Figure 4. Exoglucanase activity in lettuce (a) and onion (b) roots uninoculated ( ), and inoculated (O) with Glomus mosseae. Each value is the mean of three replicates. For each plant, values within a row sharing the same letter were not significantly different according to Duncan's multiple-range test (P = 0-OS). Table 1. Cellulolytic activities in spores and external mycelium of Glomus mosseae Cellulolytic enzymes Specific activities (units mg"' protein) Spores External mycelium Endoglucanase E^xoglucanase 3-44 a 0-005a 3-72 a (>007a Each number is the mean of three replicates. For each enzyme, values within a row sharing the same letter were not significantly different according to Duncan's multiplerange test (P = 0 05). than non-mycorrhizal plants (Fig. 4a). Mycorrhizal onion plants showed more exoglucanase activity (difference 0-06 units mg"' protein) than non-mycorrhizal plants at 35 d; thereafter, activity declined, and was similar to that in non-mycorrhizal plants at the end of the experiment (Fig. 4b). Table 1 shows that spores and external mycelia of G. mosseae contained endo- and exoglucanase activities. Activities of the enzymes were not significantly different between spores and external mycelium. Several electrophoretic bands of endoglucanase activity were observed in 35-d-old lettuce and onion Figure 5. Non-denaturing polyacrylamide gradient gel electrophoresis of cellulase on 4 to 12 % acrylamide. Lane 1, extracts from external mycelium of Glomus mosseae; lane 2, extracts from VAM onion roots; lane 3, extracts from non-vam onion roots; lane 4, extracts from VAM lettuce roots; lane 5, extracts from non-vam lettuce roots. The gel was stained with Congo red and destained as described in Materials and Methods.

5 225 Cellulases of VAM Glomus mosseae plants. The bands in root extracts differed between mycorrhizal and non-mycorrihizal plants (Fig. 5). One band similar to external mycelium was observed in the root extracts from colonized lettuce and onion plants. This band was not apparent in the nonmycorrhizal roots. DISCUSSION endoglucanase activity can be attributed to the extramatrical phase of the VA fungus, since endoglucanase activity found in the external mycelium and in mycorrhizal root extracts showed similar electrophoretic mobility (Fig. 5). However, the presence of bands in extracts from VAM roots, which were absent in non-mycorrhizal roots or external mycelium, suggests that some of this activity may be produced by the fungus inside the root or by the root itself after induction by the fungus. Endoand exoglucanase activities in mycorrhizal plants decreased at the end of the experiment to levels similar to, or lower than, that of non-mycorrhizal plants. The growth of the VAM fungus inside the root can be controlled by the plant (Harley & Smith, 1983; Anderson, 1988), and this control may reduce the production of these cellulases by the fungus, as is the case with other enzymes (Spanu & BonfanteFasolo, 1988). The sequence of endo- and exoglucanase activities observed in the VAM association suggests that fungal cellulases may be implicated in the mycorrhizal colonization of roots. Considerable attention has been devoted to the study of the role of cellulase enzymes in physiological and pathological changes in shoots (Sexton & Roberts, 1982 ; Coughlan & Ljungdhal, 1988), but research on these enzymes in plant roots, and their role in penetration and the development of symbiotic relationships is very scarce (Dazzo & Hubbel, 1974). Our results show that endo- and exoglucanase activities increased in 35-d-old VAM plants, as compared to non-vam plants, when the fungus was in its logarithmic stage of growth. The observation of differences in vesicle production between onion and lettuce (Fig. 2) agrees with that of Lackie et al. (1987), who found differences in vesicle production by the same fungal symbiont in ditterent hosts. However, no relationship between number of vesicles and endo- and exoglucanase activities was found, although in VAM lettuce and onion plants maximum endo- and exoglucanase activities coincided with the beginning of entry-point formation and arbuscule development. The formation of entrypoints and arbuscules requires that the fungal hypha passes through the cell wall, after which an interfacial matrix is formed around it (Bonfante-Fasolo, 1984). This matrix is composed of polygalacturonic acid and cellulosic materials of host origin (Bonfante-Fasolo et al, 1990). These cytochemical observations suggest that the fungus produces hydrolytic enzymes which facihtate penetration of the cell wall without affecting host viability (Bonfante-Fasolo, 1984; Bonfante-Fasolo et al, 1990). However, difficulties exist in detecting the activity of cell-wall-degrading enzymes, since breakdown of the walls must be localized during fungal development and, consequently, difficult to detect. Indeed, endo- and exoglucanase activities were extremely low, as might be expected in a mutualistic interaction. Thus, it is difficult to demonstrate a close relationship between cellulase activities and the development of fungal structures. Cell-wall-hydrolyzing enzymes were present in non-vam roots (Figs 3 and 4) during growth and development (Byrne et al, 1975). In respect of other symbiotic associations, such as that involving Rhizobium and legumes, several authors (Dazzo & Hubbell, 1974; Verma, Jumar & Maclachlan, 1982), have discussed the possibility that cellulases (mainly endoglucanases) produced by either the plant or the bacterium, may be implicated in the process of hostwall degradation during infection. Some of the ACKNOWLEDGEMENTS The authors tbank Karen Shashok for grammatical correction of tbe text. Financial support for this study was provided by tbe Comision Interministerial de Ciencia y Tecnologia, Spain. REFERENCES ANDERSON, A. J. (1988). Mycorrhizae-host specificity and recognition. Phytopathology 78, B.ATEM.^N, D. F. (1963). Pectolytic activities of culture filtrates of Rhizoctonia solani and extract of Rhizoctonia-m'iected tissues ot bean. Phytopathology 53, BONFANTE-FASOLO, P. (1984). Anatomy and morphology of VA mycorrhizae. In: VA Mycorrhizas (Ed. by C. LI. Powell & D ' J. Bagyaraj), pp CRC Press, Boca Raton, Florida. BONFANTE-F.ASOLO, P., Vl.AN, B., PEKOTTO, S., F A C C I O, A. & KNOX, ]. p. (1990). Cellulose and pectin localization in roots of mycorrhizal Allium porrum: labelling continuity between host cell and interfacial material. Planta 180, BYRNE, H., CHRISTOU, N. V., PAL, D., VERMA, S. & M.«;- I..'\CHLAN, G. A. (1975). Purification and characterization of two cellulases from auxin treated pea epicotyls. Journal of Biological Chemistry 250, CHAI.IFOUR, F. P. & BENHAMOU, N. (1989). Indirect evidence for cellulase production by Rhizobium pea root nodules during bacteroid differentiation: cytochemical aspects of cellulase breakdown in rhizobial droplets. Canadian Journal of Microbwlogv 35, CoLLMER, A. & KEEN, N. T. (1986). The role of pectic enzymes in plant pathogenesis. Annual Reviezv of Phytopathology 24, COUGHLAN, M. P. & LJUNGDHAL, L. G. (1988). Comparative biochemistry of fungal and bacterial cellulolytic enzyme systems. In : Biochemistry and Genetics of Cellulose Degradation (Ed. by J. P. Aubert, P. Beguin & J.Millet), pp Academic Press, New York. CRUICKSHANK, R. H. & W.-VDE, G. C. (1980). Detection of pectic enzymes in pectin-acrylamide gels. Analytical Biochemistry 107, D.\zzo, F. & HUBBELL, D. H. (1974). A quantitative assay of insoluble polyvinylpyrrolidone. Plant and Soil 40, DESHPANDE, M ' V., ERIKSSON, K. E. & PETTEBSON, L. G. (1984).

6 226 y. M. Garcia-Garrido and others An assay for selective determination of exo-l,4,-/?-glucanases in a mixture of cellulolytic enzymes. Analytical Biochemistry 138, G.\RCiA-RoMERA, I., GARCI'A-GARRIDO, J. M., MARTINEZ-MOLINA, E. & OcAMPO, J. A. (1990). Possible influence of hydrolytic enzymes on vesicular arbuscular mycorrhizal infection of alfalfa. Soil Biology and Biochemistry 22, GARCIA-ROMERA, I. & OcAMPO, J. A. (1988). Effect of the herbicide MCPA on VA mycorrhizal infection and growth of Pisum sativum. Zeitsehrift fiir Pflanzenernahrung und Bodenkunde 151, GERDEMANN, J. W. (1955). Relation of a large soil-borne spore to phycomycetous mycorrhizal infections. Mycologia 47, HARLEY, J. L. & SMITH, S. E. (1983). Mycorrhizal Symbiosis. Academic Press, New York. HEWITT, E. J. (1952). Sand-water Culture Methods used in the Study of Plant Nutrition. Commonwealth Agricultural Bureaux Technical Communication No. 22. HuBBEL, D. H., MORALES, V. M. & UMALI-GARCIA, M. (1978). Pectolytic enzymes in Rhizobium. Applied and Etivironmental Microbiology 35, LACKIE, S. M., GARRIOCK, M. L., PETERSON, R. L. & BOWLEY, S. R. (1987). Influence of host plant on the morphology of the vesicular-arbuscular mycorrhizal fungus, Glomus vesiforme (Daniels and Trappe) Berch. Symbiosis 3, LowRY, O. H., RosEBROUGH, N. J., FARR, A. L. & RANDALL, R. J. (1951). Protein measurement with the Folin-phenol reagent. Journal of Biological Chemistry 193, MACDONALD, R. M. (1981). Routine production of axenic vesicular-arbuscular mycorrhizas. New Phytologist 89, MORALES, V. M., MARTINEZ-MOLINA, E. & HUBBEL, D. H. (1984). Cellulase production by Rhizobium. Plant and Soil 80, OcAMPO, J. A., MARTIN, J. & HAYMAN, D. S. (1980). Influence of plant interactions on vesicular-arbuscular mycorrhizal infection. I. Host and non-host plant grown together. New Phytologist 84, PHILLIPS, J. M. & HAYMAN, D. S. (1970). Improved procedures for clearing roots and staining parasitic and vesiculararbuscular mycorrhizal fungi for rapid assessment of infection. Transactions of the British Mycological Society 55, SEXTON, R. & ROBERTS, J. A. (1982). Cell biology of abscission. Annual Review of Plant Physiology 33, SP.WU, P. & BONEANTE-FASOLO, P. (1988). Cell-wall-bound peroxidase activity in roots of mycorrhizal Alltum porrum. New Phytologist 109, UMALI-GARCIA, M., HLBBELL, D. H., GASKINGS, M. H. & DAZZO, F. B, (1980). Association of Azospirillum with grass roots. Applied and Environmental Microbiology 39, VERMA, D. P. S., KUMAR, V. & MACLACHLAN, C.A. (1982). /?- Glucanases in higher plants: localization, potential function, and regulation. In; Cellulose and Other Natural Polymer Systems : Biogenesis, Structure and Regulation (Ed. by R. M. Brown), pp Plenum, New York. VERMA, D. P. S. & ZOGBI, V. (1979). A cooperative action of plant and Rhizobium to dissolve the host cell wall during development of root symbiosis. Plant Science Letters 13,

ENDOGLUCANASE ACTIVITY IN LETTUCE PLANTS COLONIZED WITH THE VESICULAR-ARBUSCULAR MYCORRHIZAL FUNGUS GLOMUS FASCICULATUM

Soil Bid. Biochem. Vol. 24, No. 10, pp. 955-959, 1992 Printed in Great Britain. All rights reserved 0038-0717/92 $5.00 + 0.00 Copyright 0 1992 Pergamon Press Ltd ENDOGLUCANASE ACTIVITY IN LETTUCE PLANTS