Abstracts
Chitinolytic activity and dry mass production were determined in culture filtrates from 17 Metarhizium anisopliae strains grown in liquid medium containing chitin as the only carbon source. The objectives were to estimate parameters such as genetic variance among strains, heritability and expected gain from selection, as well as correlations between tested traits. Wide genotypic variability was observed among strains in chitinolytic activity, permitting the exploitation of this property in selection. The high heritability suggests that progress can be made through phenotypic selection. The genotypic correlation coefficient between dry mass production and chitinolytic activity detected in the filtrates was negative (<img src="http:/img/fbpe/gmb/v21n2/pg171-1.jpg" alt="wpe1.jpg (777 bytes)" align="absmiddle">= - 0.588). One of the isolates was also investigated for variation in the two traits as a function of culture growth time. The results showed an increase in enzyme activity up to the 8th (and last) day of the experiment and a decrease in dry mass from the 4th day on.
Foi determinada a atividade quitinolítica apresentada pelos filtrados de culturas de 17 genótipos de Metarhizium anisopliae crescidos em meio líquido, tendo quitina como única fonte de carbono. O objetivo foi a estimativa de parâmetros como a variância genética, a herdabilidade e o ganho esperado na seleção. Uma grande variabilidade genotípica foi verificada na atividade quitinolítica, permitindo sua exploração no melhoramento. Os altos coeficientes de herdabilidade permitem esperar um grande progresso na seleção fenotípica. Para uma das linhagens foi também determinada a variação na atividade enzimática, em função do tempo de crescimento das culturas. Os resultados mostraram um aumento na atividade enzimática até o oitavo dia da avaliação.
Estimates of genetic parameters related to chitinase production by the entomopathogenic fungus Metarhizium anisopliae
Gilberto U.L. Braga1, Roland Vencovsky2 and Claudio L. Messias1
1Departamento de Genética e Evolução, Instituto de Biologia, Universidade Estadual de Campinas, Caixa Postal 6109, 13083-970 Campinas, SP, Brasil. Fax: (019) 788-7820. Send correspondence to G.U.L.B.
2Departamento de Genética, ESALQ/USP, Caixa Postal 83, 13400-970 Piracicaba, SP, Brasil. Fax: (019) 433-6706.
ABSTRACT
Chitinolytic activity and dry mass production were determined in culture filtrates from 17 Metarhizium anisopliae strains grown in liquid medium containing chitin as the only carbon source. The objectives were to estimate parameters such as genetic variance among strains, heritability and expected gain from selection, as well as correlations between tested traits. Wide genotypic variability was observed among strains in chitinolytic activity, permitting the exploitation of this property in selection. The high heritability suggests that progress can be made through phenotypic selection. The genotypic correlation coefficient between dry mass production and chitinolytic activity detected in the filtrates was negative (
= - 0.588). One of the isolates was also investigated for variation in the two traits as a function of culture growth time. The results showed an increase in enzyme activity up to the 8th (and last) day of the experiment and a decrease in dry mass from the 4th day on.
INTRODUCTION
Enzymes capable of hydrolyzing chitin produced by entomopathogenic fungi have been extensively studied in terms of: a) characterization and mode of action (Smith et al., 1981; St. Leger et al., 1991, 1993); b) regulatory mechanisms (Smith and Grula, 1983; St. Leger et al., 1986a; Havukkala et al., 1993; El-Sayed et al., 1993a,b,c); c) interactions with substrates (St. Leger et al., 1986b); d) correlation between enzymatic activity level and degree of virulence of various isolates (Leite, 1987; Samuels et al., 1989; El Sayed et al., 1989; Gupta et al., 1994); e) production kinetics in culture (St. Leger et al., 1986c).
Intraspecific variability of chitinolytic activity among Metarhizium anisopliae strains has been reported by several investigators (Rosato et al., 1981; St. Leger et al., 1986c; Leite, 1987; Samuels et al., 1989). However, none of these studies has used partitioning of phenotypic variance, so that parameters such as genetic variance, heritability or expected gain from selection for genetic improvement of the character were not calculated. Thus, the objective of the present study was to estimate these genetic parameters using a quantitative approach, in the belief that such estimates could be used in breeding programs which take into account, among other traits, the level of production of enzymes capable of hydrolyzing chitin. This approach is based on evidence that this character is associated, at least in part, with the degree of virulence of other fungal species such as Nomuraea rileyi (El-Sayed et al., 1989), Verticillium lecanii (Jackson et al., 1985) and Beauveria bassiana (Gupta et al., 1994).
MATERIAL AND METHODS
Metarhizium anisopliae strains, origin and maintenance
Seventeen M. anisopliae var. anisopliae strains (Table I) were obtained from the germplasm bank of the entomopathogenic fungi laboratory, State University of Campinas. Spores were obtained by growing the isolates on mininal medium (MM) (Pontecorvo et al., 1953) at 28°C for 12 days.
a Putative diploid of strain E9 (Messias and Azevedo, 1980).
Determination of viability
Conidial viability was determined by inoculating a suspension containing 106 conidia/ml onto plates containing MM. After 14 h of incubation at 28°C, a random sample of 500 conidia was examined for the percentage of germinating and non-germinating conidia.
Filtrate preparation and determination of its enzyme activity
Fifty-ml Erlenmeyer flasks containing 17 ml liquid MM with 1% chitin (w/v) replacing glucose were autoclaved at 110°C for 20 min and inoculated with 3 ml of the suspensions containing 3 x 107 conidia/ml and grown at 28°C with shaking at 150 rpm for varying periods of time. The chitin used (from crab shells, practical grade, Sigma) was sifted through a 0.149 mm mesh, and only the fraction that passed through was utilized.
After the growth period, the content of each flask was filtered through previously tared Inlab type 10 paper to retain the mycelial mass produced. The collected filtrate was divided into one-ml aliquots and stored at -60°C. For one of the isolates (strain 58), the filtrate was collected and the dry mass produced was determined between the 3rd and 8th day of growth. The filtrate of all isolates after 7 days of growth was used for comparison of the enzyme activities.
Chitinolytic activity
The chitinolytic activity of the filtrates was determined as follows: one ml of a chitin azure (Sigma) suspension, 2 mg/ml in McIlvaine's citric acid-phosphate buffer (pH 5.2), was added to each tube containing one ml of the filtrate thawed in a water bath at 37°C. The mixture was incubated at 37°C with shaking at 200 rpm. After 10 h, the content of the tube was centrifuged at 3,000 g (rav 7 cm) for 10 min, and a 100-µl aliquot of the supernatant was submitted for determination of absorbance at 575 nm. The precipitate was resuspended, and incubation was continued for additional 12 h, after which the previous procedure was repeated.
Absorbance was determined with a DU 70 Beckman spectrophotometer in an 8-mm high micro cell (Beckman). Enzyme activity is reported as the variation in supernatant absorbance during the incubation periods (10 and 22 h).
Dry mass determination
The dry mycelial mass produced by the strains in 17 ml of medium during seven days was determined by drying mycelium produced in an oven at 70°C for 72 h.
Experimental design and statistical-genetic analysis
For better environmental control, strains were evaluated in two separate experiments containing 8 and 9 treatments, respectively. The design was completely randomized with three replications (R). A common check treatment (strain 58) was included in each experiment to evaluate environmental differences and adjustment of treatment means between experiments (Pimentel Gomes and Guimarães, 1968). Since strain 58 was one of the genotypes under evaluation, it was included to contribute to genetic variability among strains in one of the experiments (Exp. 2). The scheme of the combined analysis of variance is presented in Table II. For estimation of covariances and correlations the analyses of variance scheme was applied to the sum of the variables on a plot basis, as proposed by Kempthorne (1966).
- Scheme of combined within experiments analysis of variance and covariance with expected values of mean squares (MS) and mean products (MP).
Within group variance components were estimated as follows:
a) genetic variance:
b) variance due to experimental error:
c) phenotypic variance among strain means:
To assess the correlation between traits, the following coefficients were estimated:
a) genotypic correlation among strains:
b) environmental correlation:
c) phenotypic correlation:
Estimate of error variance of any adjusted treatment mean, with C common treatments and S experiments,
was estimated through:
In the present case S = 2 and C = 1
For comparison of adjusted treatment means the following error variances were estimated:
a) contrast between two strain means within experiments:
b) contrast between two strain means of different experiments:
Estimates of additional parameters
The following additional parameters were estimated:
a) heritability on a plot basis: 
, and on a strain mean basis: 
b) expected gain from direct selection: 
with an average k = 0.96 (with small samples, k = 0.91 for selection of 3 out of 8 strains (3/8) and k = 1.00 for selection of 3 out of 9 strains (3/9) (Fisher and Yates, 1966))
c) expected gain from selection as percentage of the trait mean: 
100 where Y0 is the initial overall mean for the character.
Spearman's rank correlation
Since the chitinolytic activity of the isolates was determined using two different periods of incubation with the substrate, Spearman's coefficient of correlation was calculated to determine the coincidence of results obtained (Steel and Torrie, 1960) for strain means between periods.
RESULTS
Kinetics of chitinase production
Figure 1 shows the variation in chitinolytic activity observed in the culture filtrate of strain 58 as a function of time. Activity increased up to the 8th and last day of the study. In similar experiments, St. Leger et al. (1986c) obtained similar results for the enzymes N-acetyl-b-glucosaminidase (EC 3.2.1.30) and chitinase (EC 3.2.1.14).
- Variation in chitinolytic activity observed in the culture filtrates of strain 58. Each point represents the mean of three replications. The errors of the means were 3.998 10-3 for 10 h and 3.434 10-3 for 22 h.
Although chitinolytic activity showed successive increments, dry mass production decreased from the 4th day of growth on.
Viability
Viability of isolates was high and ranged from 95.0 to 98.6%.
Means and dispersion of the trait evaluated
Table III shows adjusted strain means of chitinolytic activity. Wide intraspecific variability in chitinase production was detected, with the highest activity being approximately 5 times larger than the smallest value when activity was determined after 10 h of incubation, and 6.5 times larger when activity was determined after 22 h. Similarly, estimated genetic variance was 3.7 times larger when activity was determined after 22 h of incubation (Table IV). Spearman's correlation coefficient between the two determinations was 0.958 (for a t-value of 12.94). Dry mass production by the different strains showed a relatively smaller but still significant variability. The highest dry mass production was 1.4 times larger than the smallest.
Adjusted means of the 17 Metarhizium anisopliae strains for chitinolytic activity and dry mass production. Error of any adjusted treatment means
Strain
Experiment
Chitinolytic activitya
Dry mass production (mg)
A
B
(58)b
1
0.0828
0.1600
17.89
63
1
0.0712 (7)c
0.1278 (8)c
18.57 (4)
E6
1
0.0453 (14)
0.0935 (13)
18.55 (6)
70
1
0.1190 (2)
0.1834 (2)
16.85 (15)
38
1
0.0294 (16)
0.0442 (16)
18.11 (7)
E9
1
0.0961 (3)
0.1631 (4)
17.60 (10)
SMC
1
0.0740 (6)
0.1535 (6)
17.93 (8)
02
1
0.0553 (10)
0.1025 (10)
17.59 (11)
29
1
0.0522 (11)
0.0795 (14)
15.56 (17)
(58)
2
0.0925 (4)
0.1695 (3)
15.97 (14)
47
2
0.0458 (13)
0.0998 (11)
18.56 (5)
P
2
0.0309 (15)
0.0695 (15)
19.01 (3)
157p
2
0.0688 (8)
0.1437 (7)
17.29 (13)
CLII
2
0.1340 (1)
0.2812 (1)
15.67 (16)
73
2
0.0785 (5)
0.1543 (5)
17.73 (9)
20
2
0.0505 (12)
0.0962 (12)
17.53 (12)
22
2
0.0261 (17)
0.0430 (17)
21.36 (1)
35
2
0.0622 (9)
0.1230 (9)
19.36 (2)
of experiments
0.0675
0.1271
17.84
of 17 strains (Y0)d
0.0663
0.1249
17.89
0.0056
0.0056
0.006
0.0065 0.0065
0.007
0.0091
0.0091
0.010
a Variation in the absorbance at 575 nm after 10 h (A) and 22 h (B) of incubation with the substrate.
b (58) unadjusted means of standard strain.
c Ranking of strain means.
d Includes adjusted means of strain 58 in Experiment 2.
Traits
Dry mass
0.19
0.16
0.02
Chitinolytic activitya
0.97323
0.95235
0.02087
Chitinolytic activityb
3.57478
3.55383
0.02095
a Variation in the absorbance at 575 nm after 10 h of incubation with the substrate.
b Variation in the absorbance at 575 nm after 22 h of incubation with the substrate.
Analysis of variance
Results of the analysis of variance of the data, with mean squares, coefficients of variation and significance of F test, are given in Table V. Results indicate significant genetic variation among strains for the character.
Table V - Mean square values of the analysis of variance for the characters and corresponding significance of F tests and coefficients of variation.
a Variation in the absorbance at 575 nm after 10 h (A) and 22 h (B) of incubation with the substrate.
** Significant at the 1% level.
Estimates of phenotypic variance and its components are shown in Table IV. Table VI shows heritability estimates ( 
and 
) and estimates of expected gain selection ( 
and % ). These values show that chitinolytic activities have a strong genetic determination. Selection among strains should therefore result in considerable improvement. In fact, on the basis of 12-h incubation time selected strains are expected to be 44.19% superior to the average performance of the sample of 17 strains. After 22 h of incubation, selection should be more efficient with an expected gain of 45.68%.
a Variation in the absorbance at 575 nm after 10 h (A) and 22 h (B) of incubation with the substrate.
Estimates of correlation coefficients
Estimates of correlation coefficients between dry mass produced and chitinolytic activity present in the culture filtrates are given in Table VII. A negative correlation was observed between characters ( = -0.588), indicating that strains with higher chitinolytic activities tended to present a lower dry mass production at the end of the seventh day of growth.
- Estimates of phenotypic (1), genotypic (2) and environmental (3) correlation coefficients between characters.
a Variation in the absorbance at 575 nm after 22 h of incubation with the substrate.
DISCUSSION
The study carried out by St. Leger et al. (1986a) on the regulation of chitinolytic enzymes in M. anisopliae showed that the most efficient inducters for chitinase and chitosanase are N-acetylglucosamine and glucosamine. These two monomers are released when culture media containing chitin are autoclaved. They have been previously identified as inducters of chitinase synthesis in Beauveria bassiana by Smith and Grula (1983).
The purpose of the culture medium, containing 1% (w/v) chitin and sterilized by autoclaving, was to induce chitinolytic activity, and the results obtained were valid, in principle, for these growth conditions. Obviously, these conditions differ from those occurring in vivo during the initial stages of infection when chitinase production may be of benefit for the virulence of the isolates.
The quantitative study of proteolytic (Braga et al., 1994) and chitinolytic activity here investigated, together with estimates of the genetic parameters of such activities, provides information for programs aiming at the improvement of these traits. It should be remembered, however, that these characters are simply components of a complex phenotypic characteristic which is the virulence presented by a given isolate against a host population under specific environmental conditions. It should be pointed out that, in the case of entomopathogenic fungi, the final objective of an improvement program could be an increase of virulence. One way to achieve this is by improving the characters related to virulence.
The methodology utilized in the determination of chitinolytic activity using chitin azure as substrate, followed by the determination of supernatant absorbance, does not permit an individual analysis of all enzymes involved in the process of chitin hydrolysis. Thus, the method evaluates the joint action of the enzymes of the chitinolytic complex produced by each isolate. According to Hackman and Goldberg (1964), the advantage of using chitin azure is the fact that this substrate is not partially hydrolyzed. Substrates such as colloidal chitins, by undergoing partial degradation during preparation, may be hydrolyzed by enzymes that do not have the ability to hydrolyze this substrate in the naturally occurring form.
The wide variability in chitinase production detected among isolates shows the adequacy of the methodology used here for this type of study. The variability in chitinolytic activity present in M. anisopliae culture filtrates had been previously detected by St. Leger et al. (1986c). With respect to chitinases, experiments in which activity is measured by determining the extension of the zone of substrate degradation around colonies grown on appropriate media (Gabriel, 1968; Hankin and Anagnostakis, 1975) appear to be unable to detect the existing range of variation (see Samuels et al., 1989).
The wide variability detected here may be attributed to the great genetic heterogeneity among strains, since these strains were not submitted to any previous process of selection for the character. Estimates of genotypic variances were the major components of the estimates of phenotypic variances, leading to considerably high heritabilities. The high value for chitinolytic activity indicates that the population can be easily improved for the trait, with the achievement of significant gains by simple phenotypic selection. Although Spearman's correlation coefficient was considerably high, the lower coefficient of variation and higher % obtained when the chitinolytic activity of the filtrates was evaluated after 22 h of incubation with the substrate indicate that strains could be selected using this period of incubation.
The fact that the correlation coefficients between the chitinolytic activity of the various isolates and their dry mass productions were negative, repeating results obtained by Braga et al. (1994) for protease production, appears to reflect the occurrence of more marked mycelial hydrolysis in strains in which culture filtrates present high proteolytic or chitinolytic activities. This fact was also observed in other genera of filamentous fungi (Reyes et al.,1977; Santamaria and Reyes, 1988). A point that should not be overlooked in the analysis of the results is the time of growth used, due to its implications in the evaluation of the traits (chitinolytic activity and dry mass production). As early as seven days, substrate particles are no longer present in any strain, with a consequent minimal interference of non-hydrolyzed particles with the gravimetric determination of dry mass. Another implication of this choice is that the strains present different growth curves due to their wide genetic variability, with some of them probably undergoing an autolysis process after 7 days. In such cases, part of the enzymatic activity detected in the filtrates may be due to the presence of chitinases originating from autolyzed hyphae.
We believe that this type of genetic analysis of characters involved in processes of host infection and colonization may be helpful in the development of programs aiming at selecting more efficient strains for biological control.
ACKNOWLEDGMENTS
This work was supported in part by the following Brazilian organizations: CNPq, FAEP, PADCT-FINEP, CNPq-Blue Ribbon. Publication supported by FAPESP.
RESUMO
Foi determinada a atividade quitinolítica apresentada pelos filtrados de culturas de 17 genótipos de Metarhizium anisopliae crescidos em meio líquido, tendo quitina como única fonte de carbono. O objetivo foi a estimativa de parâmetros como a variância genética, a herdabilidade e o ganho esperado na seleção. Uma grande variabilidade genotípica foi verificada na atividade quitinolítica, permitindo sua exploração no melhoramento. Os altos coeficientes de herdabilidade permitem esperar um grande progresso na seleção fenotípica. Para uma das linhagens foi também determinada a variação na atividade enzimática, em função do tempo de crescimento das culturas. Os resultados mostraram um aumento na atividade enzimática até o oitavo dia da avaliação.
(Received July 7, 1997)
- Braga, G.U.L., Messias, C.L. and Vencovsky, R. (1994). Estimates of genetic parameters related to protease production by Metarhizium anisopliae J. Invertebr. Pathol. 64: 6-12.
- El-Sayed, G.N., Coudron, T.A., Ignoffo, C.M. and Riba, G. (1989). Chitinolytic activity and virulence associated with native and mutant isolates of an entomopathogenic fungus, Nomuraea rileyi J. Invertebr. Pathol. 54: 394-403.
- El-Sayed, G.N., Ignoffo, C.M., Leathers, T.D. and Gupta, S.C. (1993a). Cuticular and non-cuticular substrate influence on expression of cuticle-degrading enzymes from conidia of an entomopathogenic fungus, Nomuraea rileyi. Mycopathologia 122: 79-87.
- El-Sayed, G.N., Ignoffo, C.M., Leathers, T.D. and Gupta, S.C. (1993b). Effects of cuticle source and concentration on expression of hydrolytic enzymes by an entomopathogenic fungus, Nomuraea rileyi. Mycopathologia 122: 149-152.
- El-Sayed, G.N., Ignoffo, C.M., Leathers, T.D. and Gupta, S.C. (1993c). Insect cuticle and yeast extract effects on germination, growth, and production of hydrolytic enzymes by Nomuraea rileyi Mycopathologia, 122: 143-147.
- Fisher, R.A. and Yates, F.A. (1966). Statistical Tables for Biological, Agricultural and Medical Research Oliver and Boyd, Edinburgh.
- Gabriel, B.P. (1968). Enzymatic activities of some entomophthorous fungi. J. Invertebr. Pathol. 11: 70-81.
- Gupta, S.C., Leathers, T.D., El-Sayed, G.N. and Ignoffo, C.M. (1994). Relationships among enzyme activities and virulence parameters in Beauveria bassiana infections of Galleria mellonella and Trichoplusia ni. J. Invertebr. Pathol. 64: 13-17.
- Hackman, R.H. and Goldberg, M. (1964). New substrates for use with chitinases. Anal. Biochem. 8: 397-401.
- Hankin, L. and Anagnostakis, S.L. (1975). The use of solid media for detection of enzyme production by fungi. Mycologia 67: 597-607.
- Havukkala, I., Mitamura, C., Hara, S., Hirayae, K. Nishizawa, Y. and Hibi, T. (1993). Induction and purification of Beauveria bassiana chitinolytic enzymes. J. Invertebr. Pathol. 61: 97-102.
- Jackson, C.W., Heale, J.B. and Hall, R.A. (1985). Traits associated with virulence to the aphid Macrosiphoniella sanborni in eighteen isolates of Verticillium lecanii Ann. Appl. Biol. 106: 39-48.
- Kempthorne, O. (1966). Introduction to Genetic Statistics Wiley, New York.
- Leite, B. (1987). Variabilidade natural de linhagens de Metarhizium anisopliae (Metsch.) Sorokin, quanto ŕ resistęncia ao ultravioleta, atividade enzimática e virulęncia ao vetor da doença de chagas Panstrongylus megistus Master's thesis, Universidade Estadual de Campinas, Campinas, SP.
- Messias, C.L. and Azevedo, J.L. (1980). Parasexuality in the deuteromycete Metarhizium anisopliae. Trans. Br. Mycol. Soc. 75: 473-477.
- Pimentel Gomes, F. and Guimarăes, R.F. (1968). Joint analysis of experiments in complete randomised blocks with some common treatments. Biometrics 14: 521-526.
- Pontecorvo, G., Roper, J.A., Hemmons, L.M., McDonald, K.F. and Bufton, A.W.J. (1953). The genetics of Aspergillus nidulans. Adv. Genet. 5: 141-238.
- Reyes, F., Lahoz, R. and Cornago, P. (1977). Autolysis of Neurospora crassa in different culture conditions and release of b-N-acetyl-glucosaminidase and chitinase. Trans. Br. Mycol. Soc. 68: 357-361.
- Rosato, Y.B., Messias, C.L. and Azevedo, J.L. (1981). Production of extracellular enzymes by isolates of Metarhizium anisopliae. J. Invertebr. Pathol. 38: 1-3.
- Samuels, K.D.Z., Heale, J.B. and Llewellyn, M. (1989). Characteristics relating to the pathogenicity of Metarhizium anisopliae toward Nilaparvata lugens. J. Invertebr. Pathol. 53: 25-31.
- Santamaria, F. and Reyes, F. (1988). Proteases produced during autolysis of filamentous fungi. Trans. Br. Mycol. Soc. 91: 217-220.
- Smith, R.J. and Grula, E.A. (1983). Chitinase is an inducible enzyme in Beauveria bassiana. J. Invertebr. Pathol. 42: 319-326.
- Smith, R.J., Pekrul, S. and Grula, E.A. (1981). Requirement for sequential enzymatic activities for penetration of the integument of the corn earworm (Heliothis zea). J. Invertebr. Pathol. 38: 335-344.
- Steel, R.G.D. and Torrie, J.H. (1960). Principles and Procedures of Statistics McGraw-Hill Book Company, New York.
- St. Leger, R.J., Charnley, A.K. and Cooper, R.M. (1986a). Cuticle-degrading enzymes of entomopathogenic fungi: mechanisms of interaction between pathogen enzymes and insect cuticle. J. Invertebr. Pathol. 47: 295-302.
- St. Leger, R.J., Charnley, A.K. and Cooper, R.M. (1986b). Cuticle-degrading enzymes of entomopathogenic fungi: synthesis in culture on cuticle. J. Invertebr. Pathol. 48: 85-95.
- St. Leger, R.J., Cooper, R.M. and Charnley, A.K. (1986c). Cuticle-degrading enzymes of entomopathogenic fungi: regulation of production of chitinolytic enzymes. J. Gen. Microbiol. 132: 1509-1517.
- St. Leger, R.J., Cooper, R.M. and Charnley, A.K. (1991). Characterization of chitinase and chitobiase produced by the entomopathogenic fungus Metarhizium anisopliae J. Invertebr. Pathol. 58: 415-426.
- St. Leger, R.J., Staples, R.C. and Roberts, D.W. (1993). Entomopathogenic isolates of Metarhizium anisopliae, Beauveria bassiana and Aspergillus flavus produce multiple extracellular chitinase isozymes. J. Invertebr. Pathol. 61: 81-84.
Publication Dates
-
Publication in this collection
06 Jan 1999 -
Date of issue
June 1998
History
-
Received
07 July 1997
