Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Enhanced fungal resistance in transgenic cotton expressing an endochitinase gene from Trichoderma virens

Enhanced fungal resistance in transgenic cotton expressing an endochitinase gene from Trichoderma... <h1>Introduction</h1> Microbes, including mycoparasitic fungi, represent rich sources of genes that can be utilized to genetically engineer important crops for resistance against plant pathogens and insect pests ( Lorito and Scala, 1999 ; Punja, 2001 ). The use of genes encoding insecticidal proteins from Bacillus thuringiensis to confer resistance to various insect pests has been very successful in a number of crops ( Moellenbeck et al ., 2001 ; Perlak et al ., 2001 ; Tu et al ., 2000 ). More recently, chitinase-encoding transgenes from certain fungi have been shown to improve plant defence against a broad range of fungal pathogens. The enzyme chitinase is capable of degrading chitin, a linear homopolymer of ॆ-1,4-N-acetyl- d -glucosamine residues, which constitutes 3–60% of the cell wall of most fungi ( Collinge et al ., 1993 ). In planta expression of endochitinase genes from diverse fungi, such as Trichoderma harzianum ( Lorito et al ., 1998 ) and Rhizopus oligosporus ( Terekawa et al ., 1997 ), has proven to be more effective both in terms of the level and in the spectrum of fungal pathogen resistance compared to the plant-derived chitinases. In addition to the direct action of chitinases on the cell wall of an invading pathogen, it has been suggested that the fungal cell wall fragments released as a result of extracellular chitinase activity may in turn activate a number of defensive responses in plants ( Grison et al ., 1996 ; Lorito et al ., 1998 ). Trichoderma harzianum, a mycoparasitic fungus that produces a variety of chitinolytic enzymes ( Harman et al ., 1993 ), is used as a biocontrol agent against plant diseases ( Harman, 2000 ). Chitinase genes from this fungus have been used to engineer a variety of crops, such as tobacco and potato ( Lorito et al ., 1998 ), apple ( Bolar et al ., 2000 ; Bolar et al ., 2001 ; Wong et al ., 1999 ), petunia ( Esposito et al ., 2000 ), grape ( Kikkert et al ., 2000 ), and broccoli ( Mora and Earle, 2001 ). Transgenic tobacco and potato plants expressing a 42 kDa endochitinase gene from T. harzianum were shown to be more resistant to Alternaria alternata , A. solani , Botrytis cinerea and Rhizoctonia solani ( Lorito et al ., 1998 ). In apple, high level expression of the same endochitinase had an adverse effect on plant development; however, in one line with a low level of expression of the endochitinase and with a moderate level expression of an exochitinase gene, a high degree of resistance to apple scab was observed with normal plant growth ( Bolar et al ., 2001 ). More recently, genes encoding hydrolytic enzymes from T. virens have been isolated to examine their role in mycoparasitism and antifungal activities ( Baek et al ., 1999 ; Kim et al ., 2002 ). A diverse set of chitinase and glucanase genes was isolated from T. virens and their structural features and expression patterns were characterized ( Kim et al ., 2002 ). Genetic evidence has been obtained suggesting that a 42 kDa extracellular endochitinase, common to both T. harzianum and T. virens , plays a key role in biological control of phytopathogenic fungi such as Botrytis cinerea and Rhizoctonia solani ( Baek et al ., 1999 ; Woo et al ., 1999 ). Over-expression or disruption of a gene encoding a 42 kDa endochitinase was shown to enhance or reduce, respectively, the biocontrol activity of the mycoparasite. Results from a study by Baek et al . (1999 ) suggested that the transgenic expression of T. virens genes, encoding a 42 kDa class of endochitinase enzymes, might be utilized to enhance resistance of plants to various fungal pathogens. Cotton is the most important fibre crop, grown world-wide in over 80 countries. It is the fourth largest crop in terms of economic value in the USA. Cotton seedling diseases, caused by the fungal pathogens R. solani, Pythium spp., and Fusarium spp. resulted in an estimated loss of 590 000 bales in the USA in the year 2000. In addition, other fungal pathogens ( Phymatotrichum omnivorum , Fusarium oxysporum and Verticillium dahliae ) inflict a significant financial loss for cotton producers and are difficult to control by traditional means. The present study was undertaken to examine the effectiveness of the 42 kDa endochitinase genes from T. virens in protecting cotton from fungal diseases. In addition, tobacco plants were initially transformed to test the expression of various endochitinase clones, and then these were evaluated for their resistance to A. alternata . Cotton plants transformed with one of the 42 kDa endochitinase genes from T. virens showed a high level of resistance to infection by R. solani and A. alternata . <h1>Results</h1> <h2>Regeneration of transformants and endochitinase activity in T 0 plants</h2> Plants regenerated from transformed callus lines were grown to maturity in a greenhouse. Following establishment in the soil, both tobacco and cotton plants were examined for transgenic endochitinase activity in the leaf-protein extracts ( Table 1 ). Transgenic endochitinase activity was only detected in the plants that were transformed with the Tv-ech1 cDNA construct. Plants transformed with the other two cDNA constructs ( Tv-ech2 , Tv-ech3 ) and the genomic clone construct ( Tv-ech1 g) showed endochitinase activity levels that were similar to control values. The remainder of the studies were therefore conducted with plants containing the transgene Tv-ech1 . In the case of tobacco, 21 of 47 kanamycin-resistant lines transformed with Tv-ech1 showed endochitinase activities that were higher than the untransformed control plants ( Figure 1a ). The endochitinase activities ranged from 2.3 to 202 pmole 4-MU/h/µg total protein. When leaves from the higher expressing lines were assayed for endochitinase activity prior to the flowering stage, a reduction in activity levels was observed ( Table 2 ). Ten transgenic lines with endochitinase levels ranging between 10 and 29-fold greater than the average control values were maintained in the greenhouse until maturity. All these lines were fertile and set seed. In cotton, 71 of the 82 kanamycin-resistant lines transformed with the Tv-ech1 cDNA construct showed elevated endochitinase activities ( Figure 1b ). The endochitinase values ranged from 0.1 to 56 pmole 4-MU/h/µg total protein. From these, 44 lines with endochitinase activities between 20 and 180-fold greater than average control values were maintained in the greenhouse until maturity. Twenty-nine lines were fertile and set seeds. Similar to transgenic tobacco, leaves from the high expressing cotton lines assayed at the flowering stage were found to have reductions in endochitinase activities ( Table 3 ). In order to confirm that the elevated endochitinase activity was a result of the fungal transgene expression, a gel-based fluorescence assay was conducted to detect the 42 kDa endochitinase band in the protein extracts from the transgenic lines. A 42 kDa band illustrating the presence of a transgenic endochitinase was detected in seven tobacco ( Figure 2a ) and five cotton ( Figure 2b ) lines which was absent in the untransformed controls. Because it is not possible to apply the substrate evenly to the gel, this assay only provides qualitative information regarding the size of the endochitinase protein and either the presence or absence of the enzyme activity. The results shown in Figure 2 suggest that the endochitinase activities detected using the quantitative, fluorometric method in the leaf extracts in the transgenic plants ( Figure 1 ) were due to the expression of the T. virens endochitinase gene. <h2>Molecular characterizations of transgenic plants</h2> Southern blot analysis was performed with genomic DNA extracted from nine different T 0 cotton plants. Of these, six transgenic lines showed a single-copy integration of the endochitinase transgene, while the others had two copies each ( Figure 3a ). In tobacco, Southern blot analysis was conducted with genomic DNA from T 1 plants. Transgene integration of one to as many as three copies of the endochitinase gene was observed in tobacco (results not shown). Northern blot analyses on transgenic cotton revealed the presence of T. virens endochitinase transcript in both leaf and root tissues ( Figure 3b ). <h2>Endochitinase activities in T 1 and T 2 plants</h2> Seeds obtained from three of the highest expressing lines in tobacco (lines T4, T35 and T37) were germinated, homozygous T 1 progeny were identified and grown to maturity to obtain T 2 seeds. Endochitinase activity was monitored in the leaf tissues over the T 1 and T 2 generations. As shown in Table 2 , all three lines maintained their endochitinase activity over these generations. Leaves from T 2 plants from each of the three lines were assayed for resistance to A. alternata . In the case of cotton, 15 T 1 seeds from each of the seven highest expressing T 0 lines ( Figure 1b ) were germinated and the leaves from the healthy plants were examined for endochitinase activities. From each set of T 1 plants, there were two or more that showed endochitinase activities similar to the control values and these were apparently the null segregants ( Figure 4a, b and d ). In three high expressing lines (lines H9A, H9UA and H134), there was a significant drop in endochitinase activities in the T 1 progeny compared to the T 0 parent plants (results shown only for H134, Figure 4a ). These three lines were not used for further analysis and were discarded. The remaining four lines (H121B, H32, H135 and H17) were used for more extensive analyses, as their T 1 progeny maintained either the same or higher endochitinase activities compared to their respective T 0 parents. The results from the analyses on T 0 , T 1 and T 2 progeny from line H32 are shown in Figure 4b, c , and from line H135 are shown in Figure 4d, e . Null segregants in the T 1 progeny are obvious in both lines. In line H32, some of the T 1 endochitinase values were similar to those of the T 0 parent, while others were higher than the T 0 parent. These T 1 plants with higher endochitinase levels were assumed to be homozygotes. All the T 2 seeds tested from these plants showed endochitinase activity ( Figure 4c ), thus confirming their homozygous status. In line H135, out of 14 T 1 plants tested, two were null segregants, and eight plants showed endochitinase activities approximately half the level of the T 0 parent. Four plants showed activity levels similar to the T 0 parent ( Figure 4d ). These were assumed to be homozygous and were confirmed as such based on the absence of segregation in the T 2 generation. Endochitinase activity values from one such homozygous T 1 parent and its T 2 seeds are shown in Figure 4e . Unlike line H32, where homozygous T 1 progeny showed higher levels of endochitinase compared to the T 0 parent, in line H135 the endochitinase activity levels in the homozygous T 1 progeny were similar to T 0 values. Such differences among different independent transgenic lines are not unusual and have been observed in various other transformation experiments in our laboratory. Results from the four lines H121B, H32, H135 and H17 are summarized in Table 3 . Note that the T 1 plant H135/#11 was a null segregant and showed endochitinase activity levels in the leaves and seeds that were similar to control values. Since the roots and lower parts of the hypocotyls are the potential targets for infection by soil-borne fungi, we were interested in determining the transgenic endochitinase activity levels in these tissues. Roots, hypocotyl-root junction segments and basal hypocotyl segments (1 cm-long segment excised from just above the root portion) from 3-day-old, dark-grown seedlings were examined and their endochitinase activities are presented in Table 3 . The activities in T 2 roots were found to be lower in all cases compared with the activity in the respective seeds. However, the endochitinase activity in the hypocotyl-root junction tissues and basal hypocotyl tissues was higher compared with the levels detected in the roots. These differences possibly reflect the relative promoter activities in various tissues. Endochitinase activities were also examined in T 3 seeds and these were positive for transgene expression (results not shown). <h2>Disease resistance assays with R. solani</h2> Rhizoctonia solani is a soil-borne pathogen which causes seed decay and pre-emergence damping off as well as post-emergence damping off in cotton ( Hillocks, 1992a ). This fungus also is very susceptible to chitinase activity and therefore served as an appropriate test pathogen for our investigation. Disease resistance assays were performed with T 2 generation seeds obtained from a confirmed homozygous T 1 parent in three independent transgenic lines (H121B, H32 and H135). Thirty delinted seeds from each of the transgenic lines and 60 untransformed control seeds were sown in flats divided into six lanes with approximately 1800 mL of Metromix #366, fungus-infested soil per lane. Before distributing the soil in the flats, R. solani- wheat bran mix was added to the moist soil at a concentration of 0.28 g/L and mixed thoroughly. Ten seeds were sown in each lane, and every lane was watered with 500 mL of water every other day. The fungal invasion of cotton seedlings at the soil level produces a brownish, sunken lesion on the hypocotyls due to cortical decay, eventually causing the seedlings to collapse ( Hillocks, 1992a ). One week after the seeds were sown, seedlings were removed from the soil and assigned disease severity scores as described in the Experimental procedures. Disease symptoms appeared as brown-coloured lesions at the hypocotyl–root junction. In this experiment, only seven of the 60 untransformed seedlings were completely healthy, whereas 22 seedlings died and 21 failed to germinate. Importantly, a significant number of the transgenic seedlings showed no disease symptoms ( Figure 5a ). Thirty seeds from each of the transgenic lines H135/#4, H32/#4 and H121B/#11 were sown in pathogen infested soil, and 26, 22 and 19 seedlings, respectively, germinated without obvious disease symptoms. Results presented in Figure 6a indicate that T 2 seedlings from these three transgenic cotton lines showed significant resistance to R. solani . A second, similar experiment was conducted except that the seedlings were allowed to grow in the infested soil for 2 weeks. In this experiment, disease symptoms were more severe in the case of the untransformed control, where 59 seeds either failed to germinate or died following germination. The only single surviving seedling showed a disease score of 4. Thirty seeds from each of the transgenic lines H135/#4, H32/#4 and H121B/#11 were sown in pathogen infested soil, and 20, 23 and 23 seedlings, respectively, germinated without obvious disease symptoms after 2 weeks. Results from this experiment, presented in Figure 6b , show that the transgenic lines were significantly resistant to fungal infection. A third experiment was conducted where the inoculum pressure was doubled by increasing R. solani- wheat bran concentration to 0.56 g/L of soil. Thirty control seeds and 30 T 2 seeds from transgenic line H135/#4 were used in this experiment. After 9 days, all the 21 of the germinated control seedlings died as a result of severe disease and nine seeds failed to germinate ( Figure 5b ). In the transgenic line, 15 of 28 germinated seedlings were completely healthy with no observable disease symptoms, 10 seedlings showed varying degrees of disease symptoms, three seedlings died, and two seeds failed to germinate ( Figure 5b ). The disease index value for the control was 88.33 ± 0.96 (mean ± SE). There was a significant ( P < 0.0001) reduction in the disease index value at 28.88 ± 3.64 for the transgenic line. Results obtained from these three experiments clearly demonstrate that the expression of endochitinase, Tv-ech1 , provides significant protection for cotton seedlings against damping-off due to R. solani. <h2>Disease resistance assays with A. alternata</h2> Alternaria alternata either alone or in combination with A. macrospora is responsible for leaf spot/leaf blight of cotton ( Bashan, 1994 ; Hillocks, 1992b ). These two pathogens are distributed world-wide ( Hillocks, 1992b ). Under certain conditions, such as injury created by pest attack, adverse weather conditions, moisture, or nutrient stress, these and other leaf pathogens can cause significant yield loss in cotton ( Hillocks, 1992b ). In tobacco, A. alternata causes brown spot disease where disease symptoms are characterized by the appearance of necrotic lesions often surrounded by yellow halos ( Spurr, 1991 ). In order to assess the disease resistance of cotton and tobacco plants to foliar diseases, A. alternata was used as a test pathogen. Another reason for using A. alternata for resistance assays in cotton is that, of the two Alternaria species, it is the major pathogen on G. hirsutum , the most widely grown of the four cotton species ( Bashan et al. , 1991 ). Disease resistance assays were carried out with detached leaves from the control and transgenic lines of both tobacco and cotton placed individually on wet paper towels in Petri dishes. Although the detached leaf technique is routinely used in disease resistance assays, we were interested in determining whether the transgenic endochitinase activity was maintained in the isolated leaves. The endochitinase activity as determined from isolated, individual leaves daily for 6 days was found to remain stable over the entire period ( Table 4 ). In tobacco, leaves inoculated with A. alternata began showing symptoms, in the form of lesions surrounded by yellow halos within a week. The lesion area was measured 2 weeks following inoculations. Figures 7a, b show that the transgenic lines had significantly less damage on the leaf tissue compared to the control. As described in the Experimental procedures, two experiments were carried out with cotton leaves. In the first experiment, where two agar plugs containing the pathogen were used, disease symptoms were generally observed over the majority of the leaf in the case of controls (average damaged area was 87%, Figures 5c and 8a ). However, in the leaves from transgenic lines, the mean area exhibiting symptoms was significantly less than the control ( Figures 5c and 8a ). In the second experiment with a larger number of leaf samples and only one plug of inoculum, lesion development was still extensive on the control leaves. Again, leaves from transgenic cotton lines showed a significant degree of protection compared to the controls ( Figure 8b ). These results suggest that the expression of endochitinase Tv-ech1 from T. virens also confers resistance to an important foliar pathogen in tobacco and cotton. <h1>Discussion</h1> To our knowledge, this is the first successful report of engineering cotton with a gene encoding a chitinase from a mycoparasite to enhance the resistance of this important crop plant against two fungal pathogens. The only other report in the literature on transgenic disease resistance in cotton involved the expression of a tobacco chitinase gene ( McFadden et al ., 2000 ). In that study, the expression of the plant chitinase resulted in a slight reduction in disease symptoms following Verticillium infection under greenhouse conditions. We conducted a comprehensive study in which cotton and tobacco plants were transformed with three different cDNA clones (encoding three different 42 kDa endochitinases) and one genomic clone (encoding the only endochitinase whose cDNA did in fact express in transgenic plants). Both cotton and tobacco expressed only one ( Tv-ech1 ) of the three Trichoderma virens endochitinase cDNA clones tested as determined by the enzyme activity assays on leaf extracts. This is the same cht42 clone that was isolated and characterized by Baek et al . (1999 ) and found to play an important role in the biocontrol activity of T. virens . Without conducting an extensive analysis on the transformants, it is not possible to know why the other clones did not confer chitinase activity in plants. Molecular analyses of transformed cotton plants expressing the Tv-ech1 endochitinase demonstrated single copy integration of the transgene in six of nine plants tested, while the remaining plants contained two copies each. All of the high expressing lines used in the subsequent disease resistance assays showed single copy integrations. Northern blot analyses conducted on transgenic cotton plants using RNA isolated from both leaf and root tissues of the homozygous T 2 plants revealed the presence of the T. virens endochitinase transcript. The gel-based fluorescence assay clearly showed the presence of a band (42 kDa) with endochitinase activity only in transgenic plants and not in the controls. Chitinases characterized from tobacco have molecular weights of approximately 32 and 34 kDa ( Shinshi et al ., 1987 ). In cotton, a sequenced chitinase clone was found to encode a 28.8 kDa protein ( Hudspeth et al ., 1996 ) and a purified class I chitinase had a molecular value of 31.5 kDa ( Chlan and Bourgeois, 2001 ). Thus, the elevated chitinase levels detected in both tobacco and cotton were due to the expression of the fungal endochitinase transgene. There is a report of a T. harzianum endochitinase gene expression resulting in an abnormal phenotype (reduced growth) in apple ( Bolar et al ., 2000 ). In our study, none of the transgenic plants expressing high levels of endochitinase showed obvious morphological abnormalities. This result was consistent with previous studies on tobacco, potato ( Lorito et al ., 1998 ) and broccoli ( Mora and Earle, 2001 ) where endochitinase expression did not adversely affect the plants. There was some reduction in endochitinase enzyme activity when the plants reached the flowering stage of growth. A similar reduction was observed in tomato, tobacco ( Brants, 1999 ) and broccoli ( Mora and Earle, 2001 ). The promoter used in these studies, as well as in our investigation, was the CaMV 35S promoter. In our laboratory, we have examined the expression profile of the 35S promoter using the GFP gene as a reporter system in cotton ( Sunilkumar et al ., 2002 ). We found that the expression of the reporter gene was reduced in the leaves from older plants, suggesting that reduced promoter activity may account for the lower endochitinase activity in older tissues. This is the first published investigation on cotton reporting the production of a large number (82) of transgenic lines that were subjected to systematic analysis to study the transgene expression levels in different lines and then monitor the stability of the expression in selected lines over four generations. Results shown in Figure 4 highlight the importance of monitoring the expression levels over several generations to ensure the stability of transgene expression. As stated earlier, not all the high expressing lines identified in T 0 generation maintained chitinase activity in T 1 generation. Lines H9A, H9UA, and H134 showed a significant drop in chitinase activity in T 1 generation. In the other four higher endochitinase-expressing lines, H121B, H32, H135 and H17, the expression was either at the same level or showed an increase in the homozygous progeny. The differences in expression levels over generations amongst different lines reported in this study are not unusual, and have been reported by others in several plant species ( Caligari et al ., 1993 ; De Neve et al ., 1999 ; James et al ., 2002 ; Vain et al ., 2002 ). Our results add to the growing literature indicating the importance of screening a large number of independent transformation events to identify lines with desired levels of transgene expression and then monitoring the stability of expression over several generations ( Vain et al ., 2002 ). Rhizoctonia solani causes seed rot, lesions on the hypocotyls, and root rot in cotton seedlings. This pathogen can cause disease under a range of environmental conditions, but it can be devastating in cool, wet soil where it results in pre- and post-emergence damping-off. T 2 seeds from several high endochitinase-expressing lines were subjected to infection by planting them in soil infested with R. solani . At moderate inoculum pressure (0.28 g culture/L of mix), a majority of the untransformed seedlings (98%) in infested soil died due to post-emergence infections or the seeds failing to germinate; however, > 67% of the transgenic seedlings remained healthy even after 2 weeks in the infested mix. The results from these infection assays clearly demonstrate that the transgene provided significant protection to cotton seedlings against the damping-off caused by R. solani . Even when the inoculum pressure was doubled (0.56 g/L), 15 of the 28 transgenic seedlings examined 9 days after planting were free of disease symptoms, while none of the control plants survived ( Figure 5b ). Infection due to soil-borne pathogens can occur immediately after planting, resulting in seed decay ( Hillocks, 1992a ). Since high endochitinase activity was observed in the seeds, protection against this pathogen may have existed in the seed prior to germination. The slight variations in the level of disease resistance observed among different transgenic lines could not be correlated with the levels of endochitinase activity in the respective lines. Lorito et al . (1998 ) showed that the expression of a T. harzianum endochitinase gene in tobacco and potato rendered the plants highly resistant to R. solani . Several studies have indicated that the biological control activity of Trichoderma sp. is due to the presence of hydrolytic enzymes, including chitinases ( Carsolio et al ., 1999 ; De La Cruz et al ., 1995 ; Flores et al ., 1997 ; Peterbauer et al ., 1996 ; Thrane et al ., 1997 ). Hadar et al . (1979 ) showed that T. harzianum , under greenhouse conditions, when applied in the form of wheat bran culture to R. solani- infested soil, effectively controlled damping-off in bean, tomato and eggplant seedlings. They attributed this protection to the production of extracellular ॆ-(1-3) glucanases and chitinases in the mycoparasite which caused lysis of the fungal cell walls. Benhamou et al . (1993 ) studied the ultrastructural changes and cytochemical aspects of the chitin breakdown in R. solani using bean endochitinase and found that the chitin present in the cell wall of the fungus was rapidly hydrolysed upon exposure to endochitinase, leading to cell lysis. They also suggested that the chitin present in the cell wall was accessible to endochitinase, especially in the apical zone, making R. solani more vulnerable to chitinase action alone. Microscope observations in their study provided evidence that the eventual death of R. solani cells was a result of extensive cell wall breakdown and leakage of cytoplasm. Transgenic Tv-ech1 -endochitinase, may, in part, be providing direct protection to cotton against R. solani in a similar manner. Enhanced disease resistance was also evident against the foliar pathogen A. alternata in both cotton and tobacco. The significantly lower percentage of leaf area showing necrotic symptoms in transgenic leaves indicated the effective protection offered by the endochitinase transgene in cotton. Similar improved disease resistance was also seen in the three transgenic tobacco lines. An endochitinase transgene from T. harzianum has previously been shown to provide a high degree of resistance to A. alternata in tobacco ( Lorito et al ., 1998 ). Taken together, these results confirm that endochitinase from Trichoderma spp. also are effective against the foliar pathogen A. alternata . In addition to coding for the chitinolytic activity, a second, indirect mechanism of protection may be afforded by the chitinase transgene. The degradation of fungal cell wall releases glycosidic components that in turn may elicit the induction of plant defence mechanisms ( Benhamou, 1996 ; Boller, 1987 ; Graham and Sticklen, 1994 ; Lamb et al ., 1989 ; Mauch and Staehelin, 1989 ; Roby et al ., 1988 ; Shibuya and Minami, 2001 ). Transgenic expression of T. virens endochitinase gene with a constitutive promoter ensures that chitinase activity, already present at the time of fungal invasion, will hydrolyse the chitin present in the cell wall of the invading pathogen. Chitin fragments have been shown to act as potent elicitors of a variety of defence related activities, e.g. lignification in wheat leaves ( Barber et al ., 1989 ), chitinase activity in melon ( Roby et al ., 1987 ), and ॆ-1,3 glucanase gene expression in cultured barley cells ( Kaku et al ., 1997 ). In cultured rice cells, chitin oligosaccharides induced phytoalexin synthesis ( Ren and West, 1992 ; Yamada et al ., 1993 ), generation of reactive oxygen species ( Kuchitsu et al ., 1995 ), biosynthesis of jasmonic acid ( Nojiri et al ., 1996 ), and expression of defence related genes ( Minami et al ., 1996 ; Nishizawa et al ., 1999 ; Takai et al ., 2001 ). A possible explanation for the high levels of protection observed in our study may therefore be that transgenic Trichoderma endochitinase activity in plant tissues may release compounds from the cell wall of the invading fungi, that in turn elicit in the plant a faster and more comprehensive defensive response. Experiments with purified chitinolytic enzyme-treated R. solani and A. alternata which produced a resistance response in wild-type tobacco ( Lorito et al ., 1998 ) provide support for this possibility. Thus, it is possible that the effectiveness of transgenic endochitinase is due both to its direct action on the cell wall chitin of the invading fungus and to indirect action through the release of oligosaccharide elicitors from the pathogen, resulting in the induction of the plant's natural defence response. These responses may include reinforcement of the cell walls, production of phytoalexins, enzyme inhibitors, hydroxyproline-rich glycoproteins, and various lytic enzymes ( Lamb et al ., 1989 ). A series of pathogenesis-related proteins along with chitinases and ॆ-glucanases have been shown to be induced in cotton as a result of exposure to invading pathogens ( Cui et al ., 2000 ; Dubery and Slater, 1997 ; Hill et al ., 1999 ; Liu et al ., 1995 ; McFadden et al ., 2001 ). Cotton plants are also known to produce high levels of ethylene in response to attack by Alternaria spp. ( Bashan, 1994 ). Ethylene has been shown to induce high levels of chitinase activity in cotton ( Boller et al ., 1983 ; Chlan and Bourgeois, 2001 ; Hudspeth et al ., 1996 ). Thus in a manner similar to other species, the cotton plant will also be expected to launch a combination of defensive activities in response to fungal elicitors. Future studies may reveal that the transgenic endochitinase-mediated protection mechanism in plants is quite complex. We also tested the chitinase transgenic cotton against Verticillium dahliae, Fusarium oxysporum and Thielaviopsis basicola ; however, we did not observe any improvement in resistance against these pathogens (data not shown). This may be due to the fact that the chitinase gene alone may not provide effective protection against the pathogens where chitin is either not a major constituent of the fungal cell wall or is not easily accessible to enzyme action. It may be possible to confer resistance to these pathogens by combining chitinase genes with other antifungal genes from Trichoderma spp. or some other source. Regardless, our results with R. solani and A. alternata provide the first convincing demonstration of the usefulness of a mycoparasite-derived endochitinase in conferring an effective resistance against a soil-borne and a foliar pathogen in cotton. <h1>Experimental procedures</h1> <h2>Endochitinase constructs</h2> Three cDNA clones ( Tv-ech1 , Tv-ech2 and Tv-ech3 ) and one genomic clone ( Tv-ech1 g), all encoding different 42 kDa endochitinases used in this study, were isolated from T. virens as described by Kim et al . (2002 ). Transformation vectors were constructed in the following manner. The Hin dIII- Eco RI fragment consisting of the CaMV 35S promoter with the duplicated enhancer element, gus A gene and CaMV 35S polyadenylation signal from plasmid pFF19G ( Timmermans et al ., 1990 ) was inserted into the multiple cloning site of the binary vector, pCAMBIA2300. The gus A gene was removed by Pst I digestion and the vector was re-ligated. Each endochitinase clone was inserted as a Kpn I- Bam HI fragment between the 35S promoter and 35S polyA signal to obtain the plant transformation construct. The binary vectors were mobilized into the Agrobacterium tumefaciens -disarmed helper strain LBA4404 ( Ooms et al ., 1982 ) by the heat shock method ( An et al ., 1988 ). <h2>Plant material and transformation procedures</h2> Cotton ( Gossypium hirsutum L.) cv. Coker 312 seeds were obtained from Seedco, Lubbock, TX. Tobacco ( Nicotiana tabacum L. cv. Xanthi.nc) plants were kindly provided by Dr R. Smith (Texas A&M University) and were maintained through serial propagation in culture. Hypocotyl segments of cotton were used as explants for transformation and the protocols for selection of transformants and regeneration were as described by Sunilkumar and Rathore (2001 ). Tobacco was transformed by using the leaf disc method of Horsch et al . (1985 ). Transgenic tobacco plants were regenerated from transformed leaf discs following selection on 100 mg/L kanamycin. <h2>Endochitinase assays</h2> A modified method based on the fluorescence assay described by Kuranda and Robbins (1991 ) was used to quantify transgenic endochitinase activity in tissue extracts. In this assay, endochitinase activity results in the hydrolysis of 4-methylumbelliferyl ॆ-D-N,N′,N″-triacetylchitotrioside (Sigma) to produce 4-methylumbelliferone (4-MU) which is then quantified fluorometrically to provide a measure of enzyme activity. Initial screening for transformants expressing the endochitinase transgene was performed by analysing the enzyme activities in extracts from young, growing leaves of T 0 and T 1 plants. Total protein from tobacco plants was extracted by grinding leaves in 0.1 m citrate buffer, pH 3.0. Protein extraction from cotton tissues required the use of 0.2 m boric acid-borax buffer, pH 7.6 with 0.1% (v/v) ॆ-mercaptoethanol, 0.57 m m PMSF (phenylmethanesulphonyl fluoride) and 1% (w/v) PVPP (polyvinylpolypyrrolidone). Leaf extracts from both of the species were further diluted (16-fold) with 0.1 m citrate buffer, pH 3.0, prior to the enzyme assay. One hundred microlitres of diluted protein extract were mixed with 25 µL of substrate (250 µ m ) and incubated at 30 °C for 1 h. The reaction was terminated with 1 mL of 0.2 m sodium carbonate and fluorescence was measured using a DyNA Quant™ 200 fluorometer (Hoefer). Fluorometric readings were normalized to the total protein content of the samples (Bio-Rad protein assay). The endochitinase specific activity is presented as pmole 4-MU/h/µg total protein. In the case of transgenic cotton, endochitinase assays were performed with T 0 , T 1 and T 2 leaves, individual T 2 seeds, roots, root-hypocotyl junction segments and basal hypocotyl segments from individual T 2 seedlings, and T 3 seeds. <h2>Fluorometric gel-based endochitinase assay</h2> SDS-PAGE ( Laemmli, 1970 ) was conducted with protein samples from both tobacco (20 µg) and cotton (25 µg) leaves. Detection of the endochitinase band on the gel by fluorescence method was according to Ren et al . (2000 ). <h2>Molecular analyses</h2> Southern blot analyses were performed on genomic DNA from transformed cotton and tobacco plants to confirm the presence and integration of the endochitinase transgene ( Sambrook et al ., 1989 ). A 10 µg sample of genomic DNA digested with Hin dIII was used for the analyses. For transgene expression analysis, total RNA was isolated from leaves and roots of homozygous T 2 cotton plants according to Chomczynski and Sacchi (1987 ) with minor modifications. Five micrograms of RNA from the leaves and 2 µg from the roots were used for the Northern blot analyses. Radioactively labelled Tv-ech1 -cDNA fragment was used as the probe for both analyses. <h2>Rhizoctonia solani resistance assays on cotton seedlings</h2> Cultures of R. solani (Strain J-1, obtained from Dr C. Howell, USDA-ARS) were maintained on Potato Dextrose Agar (PDA, Difco) in a 90 mm Petri dish. For the inoculum preparation, a quarter of the agar/culture was removed from a 3-day-old culture and placed on to 25 g of sterile wheat bran mixed with 20 mL water and incubated for 5 days. The colonized wheat bran was dried overnight and then blended in a Wiley mill. Two different levels of inoculum were tested, i.e. 0.28 g or 0.56 g fungus-wheat bran culture/L of soil (Metromix no. 366). For cotton seedling disease trials, seeds from untransformed Coker 312 and T 2 seeds from a homozygous parent were directly seeded into flats containing the infested mix. A growth chamber at 25 °C and 14 h:10 h photoperiod was used for germination and seedling growth. Disease symptoms and severity were recorded after removing the seedlings from soil. Disease severity scores for individual seedlings were assigned as follows: 0 = no symptoms, 1 = minor lesion or spot, 2 = enlarged lesion (< 1 inch, 2.5 cm), 3 = necrotic lesion (1 inch or greater), 4 = collar and tissue collapse, 5 = dead seedling and 6 = failed germination. Disease severity scores were used to calculate disease index as described by Powell et al . (1971 ). Data were analysed by anova . <h2>Alternaria alternata resistance assays on detached leaves</h2> Cultures of A. alternata were grown for 2 weeks on carrot agar [5% (v/v) carrot juice + 20 g/L agar] at room temperature under light. When the fungal mycelia reached the edge of the plate, a 0.5 cm diameter agar plug with mycelium was removed from a region close to the edge using a cork borer and used for inoculations . Healthy and young leaves were collected from 3 to 4-week-old T2 cotton and tobacco plants and placed on wet paper towels in Petri dishes. Agar plugs containing A. alternata cultures were placed directly on the adaxial side of the leaves and incubated in the dark. In the first experiment conducted on cotton, two agar plugs were placed on each leaf, whereas in the second experiment with cotton and in both the experiments with tobacco, only one agar plug was placed on each leaf. Leaves were photographed with a digital camera after 2 weeks and the percentage of leaf area with necrosis was determined using A lpha E ase ™ software (version 5.5, Alpha Innotech Corp.). In order to confirm that the isolated leaves used in such assays maintained transgenic endochitinase activity, a set of isolated uninfected leaves were kept under the conditions described above. Endochitinase activity was monitored in a single leaf each day for 6 consecutive days. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Plant Biotechnology Journal Wiley

Enhanced fungal resistance in transgenic cotton expressing an endochitinase gene from Trichoderma virens

Download PDF
16 pages

Loading next page...
 
/lp/wiley/enhanced-fungal-resistance-in-transgenic-cotton-expressing-an-FKPH70flFM

References (66)

Publisher
Wiley
Copyright
© 2003 Blackwell Publishing Ltd
ISSN
1467-7644
eISSN
1467-7652
DOI
10.1046/j.1467-7652.2003.00029.x
pmid
17166131
Publisher site
See Article on Publisher Site

Abstract

<h1>Introduction</h1> Microbes, including mycoparasitic fungi, represent rich sources of genes that can be utilized to genetically engineer important crops for resistance against plant pathogens and insect pests ( Lorito and Scala, 1999 ; Punja, 2001 ). The use of genes encoding insecticidal proteins from Bacillus thuringiensis to confer resistance to various insect pests has been very successful in a number of crops ( Moellenbeck et al ., 2001 ; Perlak et al ., 2001 ; Tu et al ., 2000 ). More recently, chitinase-encoding transgenes from certain fungi have been shown to improve plant defence against a broad range of fungal pathogens. The enzyme chitinase is capable of degrading chitin, a linear homopolymer of ॆ-1,4-N-acetyl- d -glucosamine residues, which constitutes 3–60% of the cell wall of most fungi ( Collinge et al ., 1993 ). In planta expression of endochitinase genes from diverse fungi, such as Trichoderma harzianum ( Lorito et al ., 1998 ) and Rhizopus oligosporus ( Terekawa et al ., 1997 ), has proven to be more effective both in terms of the level and in the spectrum of fungal pathogen resistance compared to the plant-derived chitinases. In addition to the direct action of chitinases on the cell wall of an invading pathogen, it has been suggested that the fungal cell wall fragments released as a result of extracellular chitinase activity may in turn activate a number of defensive responses in plants ( Grison et al ., 1996 ; Lorito et al ., 1998 ). Trichoderma harzianum, a mycoparasitic fungus that produces a variety of chitinolytic enzymes ( Harman et al ., 1993 ), is used as a biocontrol agent against plant diseases ( Harman, 2000 ). Chitinase genes from this fungus have been used to engineer a variety of crops, such as tobacco and potato ( Lorito et al ., 1998 ), apple ( Bolar et al ., 2000 ; Bolar et al ., 2001 ; Wong et al ., 1999 ), petunia ( Esposito et al ., 2000 ), grape ( Kikkert et al ., 2000 ), and broccoli ( Mora and Earle, 2001 ). Transgenic tobacco and potato plants expressing a 42 kDa endochitinase gene from T. harzianum were shown to be more resistant to Alternaria alternata , A. solani , Botrytis cinerea and Rhizoctonia solani ( Lorito et al ., 1998 ). In apple, high level expression of the same endochitinase had an adverse effect on plant development; however, in one line with a low level of expression of the endochitinase and with a moderate level expression of an exochitinase gene, a high degree of resistance to apple scab was observed with normal plant growth ( Bolar et al ., 2001 ). More recently, genes encoding hydrolytic enzymes from T. virens have been isolated to examine their role in mycoparasitism and antifungal activities ( Baek et al ., 1999 ; Kim et al ., 2002 ). A diverse set of chitinase and glucanase genes was isolated from T. virens and their structural features and expression patterns were characterized ( Kim et al ., 2002 ). Genetic evidence has been obtained suggesting that a 42 kDa extracellular endochitinase, common to both T. harzianum and T. virens , plays a key role in biological control of phytopathogenic fungi such as Botrytis cinerea and Rhizoctonia solani ( Baek et al ., 1999 ; Woo et al ., 1999 ). Over-expression or disruption of a gene encoding a 42 kDa endochitinase was shown to enhance or reduce, respectively, the biocontrol activity of the mycoparasite. Results from a study by Baek et al . (1999 ) suggested that the transgenic expression of T. virens genes, encoding a 42 kDa class of endochitinase enzymes, might be utilized to enhance resistance of plants to various fungal pathogens. Cotton is the most important fibre crop, grown world-wide in over 80 countries. It is the fourth largest crop in terms of economic value in the USA. Cotton seedling diseases, caused by the fungal pathogens R. solani, Pythium spp., and Fusarium spp. resulted in an estimated loss of 590 000 bales in the USA in the year 2000. In addition, other fungal pathogens ( Phymatotrichum omnivorum , Fusarium oxysporum and Verticillium dahliae ) inflict a significant financial loss for cotton producers and are difficult to control by traditional means. The present study was undertaken to examine the effectiveness of the 42 kDa endochitinase genes from T. virens in protecting cotton from fungal diseases. In addition, tobacco plants were initially transformed to test the expression of various endochitinase clones, and then these were evaluated for their resistance to A. alternata . Cotton plants transformed with one of the 42 kDa endochitinase genes from T. virens showed a high level of resistance to infection by R. solani and A. alternata . <h1>Results</h1> <h2>Regeneration of transformants and endochitinase activity in T 0 plants</h2> Plants regenerated from transformed callus lines were grown to maturity in a greenhouse. Following establishment in the soil, both tobacco and cotton plants were examined for transgenic endochitinase activity in the leaf-protein extracts ( Table 1 ). Transgenic endochitinase activity was only detected in the plants that were transformed with the Tv-ech1 cDNA construct. Plants transformed with the other two cDNA constructs ( Tv-ech2 , Tv-ech3 ) and the genomic clone construct ( Tv-ech1 g) showed endochitinase activity levels that were similar to control values. The remainder of the studies were therefore conducted with plants containing the transgene Tv-ech1 . In the case of tobacco, 21 of 47 kanamycin-resistant lines transformed with Tv-ech1 showed endochitinase activities that were higher than the untransformed control plants ( Figure 1a ). The endochitinase activities ranged from 2.3 to 202 pmole 4-MU/h/µg total protein. When leaves from the higher expressing lines were assayed for endochitinase activity prior to the flowering stage, a reduction in activity levels was observed ( Table 2 ). Ten transgenic lines with endochitinase levels ranging between 10 and 29-fold greater than the average control values were maintained in the greenhouse until maturity. All these lines were fertile and set seed. In cotton, 71 of the 82 kanamycin-resistant lines transformed with the Tv-ech1 cDNA construct showed elevated endochitinase activities ( Figure 1b ). The endochitinase values ranged from 0.1 to 56 pmole 4-MU/h/µg total protein. From these, 44 lines with endochitinase activities between 20 and 180-fold greater than average control values were maintained in the greenhouse until maturity. Twenty-nine lines were fertile and set seeds. Similar to transgenic tobacco, leaves from the high expressing cotton lines assayed at the flowering stage were found to have reductions in endochitinase activities ( Table 3 ). In order to confirm that the elevated endochitinase activity was a result of the fungal transgene expression, a gel-based fluorescence assay was conducted to detect the 42 kDa endochitinase band in the protein extracts from the transgenic lines. A 42 kDa band illustrating the presence of a transgenic endochitinase was detected in seven tobacco ( Figure 2a ) and five cotton ( Figure 2b ) lines which was absent in the untransformed controls. Because it is not possible to apply the substrate evenly to the gel, this assay only provides qualitative information regarding the size of the endochitinase protein and either the presence or absence of the enzyme activity. The results shown in Figure 2 suggest that the endochitinase activities detected using the quantitative, fluorometric method in the leaf extracts in the transgenic plants ( Figure 1 ) were due to the expression of the T. virens endochitinase gene. <h2>Molecular characterizations of transgenic plants</h2> Southern blot analysis was performed with genomic DNA extracted from nine different T 0 cotton plants. Of these, six transgenic lines showed a single-copy integration of the endochitinase transgene, while the others had two copies each ( Figure 3a ). In tobacco, Southern blot analysis was conducted with genomic DNA from T 1 plants. Transgene integration of one to as many as three copies of the endochitinase gene was observed in tobacco (results not shown). Northern blot analyses on transgenic cotton revealed the presence of T. virens endochitinase transcript in both leaf and root tissues ( Figure 3b ). <h2>Endochitinase activities in T 1 and T 2 plants</h2> Seeds obtained from three of the highest expressing lines in tobacco (lines T4, T35 and T37) were germinated, homozygous T 1 progeny were identified and grown to maturity to obtain T 2 seeds. Endochitinase activity was monitored in the leaf tissues over the T 1 and T 2 generations. As shown in Table 2 , all three lines maintained their endochitinase activity over these generations. Leaves from T 2 plants from each of the three lines were assayed for resistance to A. alternata . In the case of cotton, 15 T 1 seeds from each of the seven highest expressing T 0 lines ( Figure 1b ) were germinated and the leaves from the healthy plants were examined for endochitinase activities. From each set of T 1 plants, there were two or more that showed endochitinase activities similar to the control values and these were apparently the null segregants ( Figure 4a, b and d ). In three high expressing lines (lines H9A, H9UA and H134), there was a significant drop in endochitinase activities in the T 1 progeny compared to the T 0 parent plants (results shown only for H134, Figure 4a ). These three lines were not used for further analysis and were discarded. The remaining four lines (H121B, H32, H135 and H17) were used for more extensive analyses, as their T 1 progeny maintained either the same or higher endochitinase activities compared to their respective T 0 parents. The results from the analyses on T 0 , T 1 and T 2 progeny from line H32 are shown in Figure 4b, c , and from line H135 are shown in Figure 4d, e . Null segregants in the T 1 progeny are obvious in both lines. In line H32, some of the T 1 endochitinase values were similar to those of the T 0 parent, while others were higher than the T 0 parent. These T 1 plants with higher endochitinase levels were assumed to be homozygotes. All the T 2 seeds tested from these plants showed endochitinase activity ( Figure 4c ), thus confirming their homozygous status. In line H135, out of 14 T 1 plants tested, two were null segregants, and eight plants showed endochitinase activities approximately half the level of the T 0 parent. Four plants showed activity levels similar to the T 0 parent ( Figure 4d ). These were assumed to be homozygous and were confirmed as such based on the absence of segregation in the T 2 generation. Endochitinase activity values from one such homozygous T 1 parent and its T 2 seeds are shown in Figure 4e . Unlike line H32, where homozygous T 1 progeny showed higher levels of endochitinase compared to the T 0 parent, in line H135 the endochitinase activity levels in the homozygous T 1 progeny were similar to T 0 values. Such differences among different independent transgenic lines are not unusual and have been observed in various other transformation experiments in our laboratory. Results from the four lines H121B, H32, H135 and H17 are summarized in Table 3 . Note that the T 1 plant H135/#11 was a null segregant and showed endochitinase activity levels in the leaves and seeds that were similar to control values. Since the roots and lower parts of the hypocotyls are the potential targets for infection by soil-borne fungi, we were interested in determining the transgenic endochitinase activity levels in these tissues. Roots, hypocotyl-root junction segments and basal hypocotyl segments (1 cm-long segment excised from just above the root portion) from 3-day-old, dark-grown seedlings were examined and their endochitinase activities are presented in Table 3 . The activities in T 2 roots were found to be lower in all cases compared with the activity in the respective seeds. However, the endochitinase activity in the hypocotyl-root junction tissues and basal hypocotyl tissues was higher compared with the levels detected in the roots. These differences possibly reflect the relative promoter activities in various tissues. Endochitinase activities were also examined in T 3 seeds and these were positive for transgene expression (results not shown). <h2>Disease resistance assays with R. solani</h2> Rhizoctonia solani is a soil-borne pathogen which causes seed decay and pre-emergence damping off as well as post-emergence damping off in cotton ( Hillocks, 1992a ). This fungus also is very susceptible to chitinase activity and therefore served as an appropriate test pathogen for our investigation. Disease resistance assays were performed with T 2 generation seeds obtained from a confirmed homozygous T 1 parent in three independent transgenic lines (H121B, H32 and H135). Thirty delinted seeds from each of the transgenic lines and 60 untransformed control seeds were sown in flats divided into six lanes with approximately 1800 mL of Metromix #366, fungus-infested soil per lane. Before distributing the soil in the flats, R. solani- wheat bran mix was added to the moist soil at a concentration of 0.28 g/L and mixed thoroughly. Ten seeds were sown in each lane, and every lane was watered with 500 mL of water every other day. The fungal invasion of cotton seedlings at the soil level produces a brownish, sunken lesion on the hypocotyls due to cortical decay, eventually causing the seedlings to collapse ( Hillocks, 1992a ). One week after the seeds were sown, seedlings were removed from the soil and assigned disease severity scores as described in the Experimental procedures. Disease symptoms appeared as brown-coloured lesions at the hypocotyl–root junction. In this experiment, only seven of the 60 untransformed seedlings were completely healthy, whereas 22 seedlings died and 21 failed to germinate. Importantly, a significant number of the transgenic seedlings showed no disease symptoms ( Figure 5a ). Thirty seeds from each of the transgenic lines H135/#4, H32/#4 and H121B/#11 were sown in pathogen infested soil, and 26, 22 and 19 seedlings, respectively, germinated without obvious disease symptoms. Results presented in Figure 6a indicate that T 2 seedlings from these three transgenic cotton lines showed significant resistance to R. solani . A second, similar experiment was conducted except that the seedlings were allowed to grow in the infested soil for 2 weeks. In this experiment, disease symptoms were more severe in the case of the untransformed control, where 59 seeds either failed to germinate or died following germination. The only single surviving seedling showed a disease score of 4. Thirty seeds from each of the transgenic lines H135/#4, H32/#4 and H121B/#11 were sown in pathogen infested soil, and 20, 23 and 23 seedlings, respectively, germinated without obvious disease symptoms after 2 weeks. Results from this experiment, presented in Figure 6b , show that the transgenic lines were significantly resistant to fungal infection. A third experiment was conducted where the inoculum pressure was doubled by increasing R. solani- wheat bran concentration to 0.56 g/L of soil. Thirty control seeds and 30 T 2 seeds from transgenic line H135/#4 were used in this experiment. After 9 days, all the 21 of the germinated control seedlings died as a result of severe disease and nine seeds failed to germinate ( Figure 5b ). In the transgenic line, 15 of 28 germinated seedlings were completely healthy with no observable disease symptoms, 10 seedlings showed varying degrees of disease symptoms, three seedlings died, and two seeds failed to germinate ( Figure 5b ). The disease index value for the control was 88.33 ± 0.96 (mean ± SE). There was a significant ( P < 0.0001) reduction in the disease index value at 28.88 ± 3.64 for the transgenic line. Results obtained from these three experiments clearly demonstrate that the expression of endochitinase, Tv-ech1 , provides significant protection for cotton seedlings against damping-off due to R. solani. <h2>Disease resistance assays with A. alternata</h2> Alternaria alternata either alone or in combination with A. macrospora is responsible for leaf spot/leaf blight of cotton ( Bashan, 1994 ; Hillocks, 1992b ). These two pathogens are distributed world-wide ( Hillocks, 1992b ). Under certain conditions, such as injury created by pest attack, adverse weather conditions, moisture, or nutrient stress, these and other leaf pathogens can cause significant yield loss in cotton ( Hillocks, 1992b ). In tobacco, A. alternata causes brown spot disease where disease symptoms are characterized by the appearance of necrotic lesions often surrounded by yellow halos ( Spurr, 1991 ). In order to assess the disease resistance of cotton and tobacco plants to foliar diseases, A. alternata was used as a test pathogen. Another reason for using A. alternata for resistance assays in cotton is that, of the two Alternaria species, it is the major pathogen on G. hirsutum , the most widely grown of the four cotton species ( Bashan et al. , 1991 ). Disease resistance assays were carried out with detached leaves from the control and transgenic lines of both tobacco and cotton placed individually on wet paper towels in Petri dishes. Although the detached leaf technique is routinely used in disease resistance assays, we were interested in determining whether the transgenic endochitinase activity was maintained in the isolated leaves. The endochitinase activity as determined from isolated, individual leaves daily for 6 days was found to remain stable over the entire period ( Table 4 ). In tobacco, leaves inoculated with A. alternata began showing symptoms, in the form of lesions surrounded by yellow halos within a week. The lesion area was measured 2 weeks following inoculations. Figures 7a, b show that the transgenic lines had significantly less damage on the leaf tissue compared to the control. As described in the Experimental procedures, two experiments were carried out with cotton leaves. In the first experiment, where two agar plugs containing the pathogen were used, disease symptoms were generally observed over the majority of the leaf in the case of controls (average damaged area was 87%, Figures 5c and 8a ). However, in the leaves from transgenic lines, the mean area exhibiting symptoms was significantly less than the control ( Figures 5c and 8a ). In the second experiment with a larger number of leaf samples and only one plug of inoculum, lesion development was still extensive on the control leaves. Again, leaves from transgenic cotton lines showed a significant degree of protection compared to the controls ( Figure 8b ). These results suggest that the expression of endochitinase Tv-ech1 from T. virens also confers resistance to an important foliar pathogen in tobacco and cotton. <h1>Discussion</h1> To our knowledge, this is the first successful report of engineering cotton with a gene encoding a chitinase from a mycoparasite to enhance the resistance of this important crop plant against two fungal pathogens. The only other report in the literature on transgenic disease resistance in cotton involved the expression of a tobacco chitinase gene ( McFadden et al ., 2000 ). In that study, the expression of the plant chitinase resulted in a slight reduction in disease symptoms following Verticillium infection under greenhouse conditions. We conducted a comprehensive study in which cotton and tobacco plants were transformed with three different cDNA clones (encoding three different 42 kDa endochitinases) and one genomic clone (encoding the only endochitinase whose cDNA did in fact express in transgenic plants). Both cotton and tobacco expressed only one ( Tv-ech1 ) of the three Trichoderma virens endochitinase cDNA clones tested as determined by the enzyme activity assays on leaf extracts. This is the same cht42 clone that was isolated and characterized by Baek et al . (1999 ) and found to play an important role in the biocontrol activity of T. virens . Without conducting an extensive analysis on the transformants, it is not possible to know why the other clones did not confer chitinase activity in plants. Molecular analyses of transformed cotton plants expressing the Tv-ech1 endochitinase demonstrated single copy integration of the transgene in six of nine plants tested, while the remaining plants contained two copies each. All of the high expressing lines used in the subsequent disease resistance assays showed single copy integrations. Northern blot analyses conducted on transgenic cotton plants using RNA isolated from both leaf and root tissues of the homozygous T 2 plants revealed the presence of the T. virens endochitinase transcript. The gel-based fluorescence assay clearly showed the presence of a band (42 kDa) with endochitinase activity only in transgenic plants and not in the controls. Chitinases characterized from tobacco have molecular weights of approximately 32 and 34 kDa ( Shinshi et al ., 1987 ). In cotton, a sequenced chitinase clone was found to encode a 28.8 kDa protein ( Hudspeth et al ., 1996 ) and a purified class I chitinase had a molecular value of 31.5 kDa ( Chlan and Bourgeois, 2001 ). Thus, the elevated chitinase levels detected in both tobacco and cotton were due to the expression of the fungal endochitinase transgene. There is a report of a T. harzianum endochitinase gene expression resulting in an abnormal phenotype (reduced growth) in apple ( Bolar et al ., 2000 ). In our study, none of the transgenic plants expressing high levels of endochitinase showed obvious morphological abnormalities. This result was consistent with previous studies on tobacco, potato ( Lorito et al ., 1998 ) and broccoli ( Mora and Earle, 2001 ) where endochitinase expression did not adversely affect the plants. There was some reduction in endochitinase enzyme activity when the plants reached the flowering stage of growth. A similar reduction was observed in tomato, tobacco ( Brants, 1999 ) and broccoli ( Mora and Earle, 2001 ). The promoter used in these studies, as well as in our investigation, was the CaMV 35S promoter. In our laboratory, we have examined the expression profile of the 35S promoter using the GFP gene as a reporter system in cotton ( Sunilkumar et al ., 2002 ). We found that the expression of the reporter gene was reduced in the leaves from older plants, suggesting that reduced promoter activity may account for the lower endochitinase activity in older tissues. This is the first published investigation on cotton reporting the production of a large number (82) of transgenic lines that were subjected to systematic analysis to study the transgene expression levels in different lines and then monitor the stability of the expression in selected lines over four generations. Results shown in Figure 4 highlight the importance of monitoring the expression levels over several generations to ensure the stability of transgene expression. As stated earlier, not all the high expressing lines identified in T 0 generation maintained chitinase activity in T 1 generation. Lines H9A, H9UA, and H134 showed a significant drop in chitinase activity in T 1 generation. In the other four higher endochitinase-expressing lines, H121B, H32, H135 and H17, the expression was either at the same level or showed an increase in the homozygous progeny. The differences in expression levels over generations amongst different lines reported in this study are not unusual, and have been reported by others in several plant species ( Caligari et al ., 1993 ; De Neve et al ., 1999 ; James et al ., 2002 ; Vain et al ., 2002 ). Our results add to the growing literature indicating the importance of screening a large number of independent transformation events to identify lines with desired levels of transgene expression and then monitoring the stability of expression over several generations ( Vain et al ., 2002 ). Rhizoctonia solani causes seed rot, lesions on the hypocotyls, and root rot in cotton seedlings. This pathogen can cause disease under a range of environmental conditions, but it can be devastating in cool, wet soil where it results in pre- and post-emergence damping-off. T 2 seeds from several high endochitinase-expressing lines were subjected to infection by planting them in soil infested with R. solani . At moderate inoculum pressure (0.28 g culture/L of mix), a majority of the untransformed seedlings (98%) in infested soil died due to post-emergence infections or the seeds failing to germinate; however, > 67% of the transgenic seedlings remained healthy even after 2 weeks in the infested mix. The results from these infection assays clearly demonstrate that the transgene provided significant protection to cotton seedlings against the damping-off caused by R. solani . Even when the inoculum pressure was doubled (0.56 g/L), 15 of the 28 transgenic seedlings examined 9 days after planting were free of disease symptoms, while none of the control plants survived ( Figure 5b ). Infection due to soil-borne pathogens can occur immediately after planting, resulting in seed decay ( Hillocks, 1992a ). Since high endochitinase activity was observed in the seeds, protection against this pathogen may have existed in the seed prior to germination. The slight variations in the level of disease resistance observed among different transgenic lines could not be correlated with the levels of endochitinase activity in the respective lines. Lorito et al . (1998 ) showed that the expression of a T. harzianum endochitinase gene in tobacco and potato rendered the plants highly resistant to R. solani . Several studies have indicated that the biological control activity of Trichoderma sp. is due to the presence of hydrolytic enzymes, including chitinases ( Carsolio et al ., 1999 ; De La Cruz et al ., 1995 ; Flores et al ., 1997 ; Peterbauer et al ., 1996 ; Thrane et al ., 1997 ). Hadar et al . (1979 ) showed that T. harzianum , under greenhouse conditions, when applied in the form of wheat bran culture to R. solani- infested soil, effectively controlled damping-off in bean, tomato and eggplant seedlings. They attributed this protection to the production of extracellular ॆ-(1-3) glucanases and chitinases in the mycoparasite which caused lysis of the fungal cell walls. Benhamou et al . (1993 ) studied the ultrastructural changes and cytochemical aspects of the chitin breakdown in R. solani using bean endochitinase and found that the chitin present in the cell wall of the fungus was rapidly hydrolysed upon exposure to endochitinase, leading to cell lysis. They also suggested that the chitin present in the cell wall was accessible to endochitinase, especially in the apical zone, making R. solani more vulnerable to chitinase action alone. Microscope observations in their study provided evidence that the eventual death of R. solani cells was a result of extensive cell wall breakdown and leakage of cytoplasm. Transgenic Tv-ech1 -endochitinase, may, in part, be providing direct protection to cotton against R. solani in a similar manner. Enhanced disease resistance was also evident against the foliar pathogen A. alternata in both cotton and tobacco. The significantly lower percentage of leaf area showing necrotic symptoms in transgenic leaves indicated the effective protection offered by the endochitinase transgene in cotton. Similar improved disease resistance was also seen in the three transgenic tobacco lines. An endochitinase transgene from T. harzianum has previously been shown to provide a high degree of resistance to A. alternata in tobacco ( Lorito et al ., 1998 ). Taken together, these results confirm that endochitinase from Trichoderma spp. also are effective against the foliar pathogen A. alternata . In addition to coding for the chitinolytic activity, a second, indirect mechanism of protection may be afforded by the chitinase transgene. The degradation of fungal cell wall releases glycosidic components that in turn may elicit the induction of plant defence mechanisms ( Benhamou, 1996 ; Boller, 1987 ; Graham and Sticklen, 1994 ; Lamb et al ., 1989 ; Mauch and Staehelin, 1989 ; Roby et al ., 1988 ; Shibuya and Minami, 2001 ). Transgenic expression of T. virens endochitinase gene with a constitutive promoter ensures that chitinase activity, already present at the time of fungal invasion, will hydrolyse the chitin present in the cell wall of the invading pathogen. Chitin fragments have been shown to act as potent elicitors of a variety of defence related activities, e.g. lignification in wheat leaves ( Barber et al ., 1989 ), chitinase activity in melon ( Roby et al ., 1987 ), and ॆ-1,3 glucanase gene expression in cultured barley cells ( Kaku et al ., 1997 ). In cultured rice cells, chitin oligosaccharides induced phytoalexin synthesis ( Ren and West, 1992 ; Yamada et al ., 1993 ), generation of reactive oxygen species ( Kuchitsu et al ., 1995 ), biosynthesis of jasmonic acid ( Nojiri et al ., 1996 ), and expression of defence related genes ( Minami et al ., 1996 ; Nishizawa et al ., 1999 ; Takai et al ., 2001 ). A possible explanation for the high levels of protection observed in our study may therefore be that transgenic Trichoderma endochitinase activity in plant tissues may release compounds from the cell wall of the invading fungi, that in turn elicit in the plant a faster and more comprehensive defensive response. Experiments with purified chitinolytic enzyme-treated R. solani and A. alternata which produced a resistance response in wild-type tobacco ( Lorito et al ., 1998 ) provide support for this possibility. Thus, it is possible that the effectiveness of transgenic endochitinase is due both to its direct action on the cell wall chitin of the invading fungus and to indirect action through the release of oligosaccharide elicitors from the pathogen, resulting in the induction of the plant's natural defence response. These responses may include reinforcement of the cell walls, production of phytoalexins, enzyme inhibitors, hydroxyproline-rich glycoproteins, and various lytic enzymes ( Lamb et al ., 1989 ). A series of pathogenesis-related proteins along with chitinases and ॆ-glucanases have been shown to be induced in cotton as a result of exposure to invading pathogens ( Cui et al ., 2000 ; Dubery and Slater, 1997 ; Hill et al ., 1999 ; Liu et al ., 1995 ; McFadden et al ., 2001 ). Cotton plants are also known to produce high levels of ethylene in response to attack by Alternaria spp. ( Bashan, 1994 ). Ethylene has been shown to induce high levels of chitinase activity in cotton ( Boller et al ., 1983 ; Chlan and Bourgeois, 2001 ; Hudspeth et al ., 1996 ). Thus in a manner similar to other species, the cotton plant will also be expected to launch a combination of defensive activities in response to fungal elicitors. Future studies may reveal that the transgenic endochitinase-mediated protection mechanism in plants is quite complex. We also tested the chitinase transgenic cotton against Verticillium dahliae, Fusarium oxysporum and Thielaviopsis basicola ; however, we did not observe any improvement in resistance against these pathogens (data not shown). This may be due to the fact that the chitinase gene alone may not provide effective protection against the pathogens where chitin is either not a major constituent of the fungal cell wall or is not easily accessible to enzyme action. It may be possible to confer resistance to these pathogens by combining chitinase genes with other antifungal genes from Trichoderma spp. or some other source. Regardless, our results with R. solani and A. alternata provide the first convincing demonstration of the usefulness of a mycoparasite-derived endochitinase in conferring an effective resistance against a soil-borne and a foliar pathogen in cotton. <h1>Experimental procedures</h1> <h2>Endochitinase constructs</h2> Three cDNA clones ( Tv-ech1 , Tv-ech2 and Tv-ech3 ) and one genomic clone ( Tv-ech1 g), all encoding different 42 kDa endochitinases used in this study, were isolated from T. virens as described by Kim et al . (2002 ). Transformation vectors were constructed in the following manner. The Hin dIII- Eco RI fragment consisting of the CaMV 35S promoter with the duplicated enhancer element, gus A gene and CaMV 35S polyadenylation signal from plasmid pFF19G ( Timmermans et al ., 1990 ) was inserted into the multiple cloning site of the binary vector, pCAMBIA2300. The gus A gene was removed by Pst I digestion and the vector was re-ligated. Each endochitinase clone was inserted as a Kpn I- Bam HI fragment between the 35S promoter and 35S polyA signal to obtain the plant transformation construct. The binary vectors were mobilized into the Agrobacterium tumefaciens -disarmed helper strain LBA4404 ( Ooms et al ., 1982 ) by the heat shock method ( An et al ., 1988 ). <h2>Plant material and transformation procedures</h2> Cotton ( Gossypium hirsutum L.) cv. Coker 312 seeds were obtained from Seedco, Lubbock, TX. Tobacco ( Nicotiana tabacum L. cv. Xanthi.nc) plants were kindly provided by Dr R. Smith (Texas A&M University) and were maintained through serial propagation in culture. Hypocotyl segments of cotton were used as explants for transformation and the protocols for selection of transformants and regeneration were as described by Sunilkumar and Rathore (2001 ). Tobacco was transformed by using the leaf disc method of Horsch et al . (1985 ). Transgenic tobacco plants were regenerated from transformed leaf discs following selection on 100 mg/L kanamycin. <h2>Endochitinase assays</h2> A modified method based on the fluorescence assay described by Kuranda and Robbins (1991 ) was used to quantify transgenic endochitinase activity in tissue extracts. In this assay, endochitinase activity results in the hydrolysis of 4-methylumbelliferyl ॆ-D-N,N′,N″-triacetylchitotrioside (Sigma) to produce 4-methylumbelliferone (4-MU) which is then quantified fluorometrically to provide a measure of enzyme activity. Initial screening for transformants expressing the endochitinase transgene was performed by analysing the enzyme activities in extracts from young, growing leaves of T 0 and T 1 plants. Total protein from tobacco plants was extracted by grinding leaves in 0.1 m citrate buffer, pH 3.0. Protein extraction from cotton tissues required the use of 0.2 m boric acid-borax buffer, pH 7.6 with 0.1% (v/v) ॆ-mercaptoethanol, 0.57 m m PMSF (phenylmethanesulphonyl fluoride) and 1% (w/v) PVPP (polyvinylpolypyrrolidone). Leaf extracts from both of the species were further diluted (16-fold) with 0.1 m citrate buffer, pH 3.0, prior to the enzyme assay. One hundred microlitres of diluted protein extract were mixed with 25 µL of substrate (250 µ m ) and incubated at 30 °C for 1 h. The reaction was terminated with 1 mL of 0.2 m sodium carbonate and fluorescence was measured using a DyNA Quant™ 200 fluorometer (Hoefer). Fluorometric readings were normalized to the total protein content of the samples (Bio-Rad protein assay). The endochitinase specific activity is presented as pmole 4-MU/h/µg total protein. In the case of transgenic cotton, endochitinase assays were performed with T 0 , T 1 and T 2 leaves, individual T 2 seeds, roots, root-hypocotyl junction segments and basal hypocotyl segments from individual T 2 seedlings, and T 3 seeds. <h2>Fluorometric gel-based endochitinase assay</h2> SDS-PAGE ( Laemmli, 1970 ) was conducted with protein samples from both tobacco (20 µg) and cotton (25 µg) leaves. Detection of the endochitinase band on the gel by fluorescence method was according to Ren et al . (2000 ). <h2>Molecular analyses</h2> Southern blot analyses were performed on genomic DNA from transformed cotton and tobacco plants to confirm the presence and integration of the endochitinase transgene ( Sambrook et al ., 1989 ). A 10 µg sample of genomic DNA digested with Hin dIII was used for the analyses. For transgene expression analysis, total RNA was isolated from leaves and roots of homozygous T 2 cotton plants according to Chomczynski and Sacchi (1987 ) with minor modifications. Five micrograms of RNA from the leaves and 2 µg from the roots were used for the Northern blot analyses. Radioactively labelled Tv-ech1 -cDNA fragment was used as the probe for both analyses. <h2>Rhizoctonia solani resistance assays on cotton seedlings</h2> Cultures of R. solani (Strain J-1, obtained from Dr C. Howell, USDA-ARS) were maintained on Potato Dextrose Agar (PDA, Difco) in a 90 mm Petri dish. For the inoculum preparation, a quarter of the agar/culture was removed from a 3-day-old culture and placed on to 25 g of sterile wheat bran mixed with 20 mL water and incubated for 5 days. The colonized wheat bran was dried overnight and then blended in a Wiley mill. Two different levels of inoculum were tested, i.e. 0.28 g or 0.56 g fungus-wheat bran culture/L of soil (Metromix no. 366). For cotton seedling disease trials, seeds from untransformed Coker 312 and T 2 seeds from a homozygous parent were directly seeded into flats containing the infested mix. A growth chamber at 25 °C and 14 h:10 h photoperiod was used for germination and seedling growth. Disease symptoms and severity were recorded after removing the seedlings from soil. Disease severity scores for individual seedlings were assigned as follows: 0 = no symptoms, 1 = minor lesion or spot, 2 = enlarged lesion (< 1 inch, 2.5 cm), 3 = necrotic lesion (1 inch or greater), 4 = collar and tissue collapse, 5 = dead seedling and 6 = failed germination. Disease severity scores were used to calculate disease index as described by Powell et al . (1971 ). Data were analysed by anova . <h2>Alternaria alternata resistance assays on detached leaves</h2> Cultures of A. alternata were grown for 2 weeks on carrot agar [5% (v/v) carrot juice + 20 g/L agar] at room temperature under light. When the fungal mycelia reached the edge of the plate, a 0.5 cm diameter agar plug with mycelium was removed from a region close to the edge using a cork borer and used for inoculations . Healthy and young leaves were collected from 3 to 4-week-old T2 cotton and tobacco plants and placed on wet paper towels in Petri dishes. Agar plugs containing A. alternata cultures were placed directly on the adaxial side of the leaves and incubated in the dark. In the first experiment conducted on cotton, two agar plugs were placed on each leaf, whereas in the second experiment with cotton and in both the experiments with tobacco, only one agar plug was placed on each leaf. Leaves were photographed with a digital camera after 2 weeks and the percentage of leaf area with necrosis was determined using A lpha E ase ™ software (version 5.5, Alpha Innotech Corp.). In order to confirm that the isolated leaves used in such assays maintained transgenic endochitinase activity, a set of isolated uninfected leaves were kept under the conditions described above. Endochitinase activity was monitored in a single leaf each day for 6 consecutive days.

Journal

Plant Biotechnology Journal – Wiley

Published: Sep 1, 2003

Keywords: Alternaria alternata ; chitinase; disease-resistance; Rhizoctonia solani ; transgenic cotton; Trichoderma virens

There are no references for this article.