Abstract: Like other entomopathogenic fungi, Beauveria bassiana produces enzymes that degrade cuticle for penetration in the host at the time of infection. Chitinases are considered important enzymes for chitin hydrolysis, one of the main insect exoskeleton components. In this study, polymorphism in B. bassiana chitinase gene was analyzed by PCR-RFLP and compared with the enzymatic activity and mortality caused on adults of coffee berry borer, Hypothenemus hampei (Ferrari) (Coleoptera: Scolytidae). Thirty B. bassiana isolates obtained from different insect species and geographic origins were used. The activity of chitinases was not directly related with the mortality rate of each strain, emphasizing the hypothesis that the virulence is multifactorial. The chitinase gene analyses showed low variability between isolates, as only four isolates presented polymorphism. Therefore it was not possible to correlate the polymorphism with virulence and chitinolitic activity. Lack of association between chitinase gene polymorphism and enzyme activity suggests that the polymorphic region studied may not be involved in enzymatic activity of this gene. Further, lack of association between enzyme activity and virulence suggests that there may be other enzymes and factors that could contribute to infection ability. No association between polymorphism in chitinase gene with that of geographic region or origin was observed.
INTRODUCTION
Beauveria bassiana (Bals.) Vuill. (Hyphomycetes) is an opportunistic and cosmopolitan deuteromycete (Alves, 1998) that has been extensively used as a biocontrol agent against many insect species. This fungus has been largely used for the coffee berry borer, Hypothenemus hampei (Ferrari) (Coleoptera: Scolytidae) control in many countries around the world (Gaitan et al., 2002). The virulence analysis of 61 B. bassiana strains, against this species, have shown variable rates of mortality, ranging from 3.3 to 83.3% (Neves and Hirose, 2005). The different virulence levels may be related with the production of enzymes and toxins involved at the infection process (Tanada and Kaya, 1993; Alves, 1998).
The mode of infection is through the exoskeleton, which is the first barrier for the insect`s protection (Clarkson and Charnley, 1996). As the exoskeleton is mainly composed by proteins and chitin, proteases and chitinases should be considered the main enzymes involved in the fungi infection process (St. Leger et al., 1996; Kang et al., 1998; Nahar et al., 2004).
Although it has been demonstrated that proteases initiate cuticle degradation creating a more efficient environment for the chitinase action (St. Leger et al., 1986), in vitro studies have shown that proteases and chitinases act synergistically at the cuticle solubilization process (St. Leger et al., 1996). In spite of proteases being important virulence factors, this enzyme should not be considered as the single essential factor for the virulence, once B. bassiana strains unable on producing this enzyme, showed only a reduction in the infection ability (St. Leger and Joshi, 1997). Chitinases hydrolyse chitin at the β-1,4 glycosidic bonds liberating the N-acetylglucosamine residues (Tharanathan and Kittur, 2003). Chitinase activity during the infection process seems to be extremely important once the chitin composition in the insect exoskeleton reaches 20 to 50%, of the dry weight (Chapman, 1998).
Many studies investigating correlation between genetic polymorphisms of entomopathogenic fungi strains and geographic origin or insect host have been reported (Maurer et al., 1997; Coates et al., 2002; Gaitan et al., 2002; Muro et al., 2003; Wang et al., 2003), but there are few studies focusing on virulence candidate genes. Wang et al. (2003) verified DNA polymorphisms, using the PCR/RFLP method, at the protease (PR1) gene from B. bassiana and found a large amount of variability at this gene that was more related with the geographic origin than with the insect host.
Considering that sequence variation at virulence-related genes seems to occur more frequently than at other loci (Internal Transcribed Spacer, for instance) (Muro et al., 2003), the goal of this study was to verify the incidence of polymorphisms at the chitinase gene in 30 isolates of B. bassiana and try to correlate this polymorphisms with in vitro chitinase activity and virulence against adult coffee berry borer, H. hampei.
MATERIALS AND METHODS
Fungal Strains
Thirty B. bassiana strains (Table 1) stored
at the Entomopathogen Collection in the Department of Agronomy at State
University of Londrina were investigated in this study. Biochemical and
molecular analysis were conducted at the Biochemistry and Biotechnology
Department of the Londrina State University, Paraná, Brazil.
Fungal Growth Conditions
B. bassiana strains used in this study had been previously
bioassayed in the coffee berry borer (H. hampei) (Neves and Hirose,
2005). Strains were initially inoculated into solid culture medium (w/v:
2% agar, 1% glucose, 0.5% yeast extract, 0.158% NaNO3, 0.105%
Na2HPO4.7H2O, 0.1% KCl; 0.06% MgSO4.7H2O,
0.036% KH2PO4 and 0.05% streptomycin sulfate) (Alves,
1998), grown at 25°C, for 10 days, 12 h photofase and then inoculated
in coffee berry borer adults as described by Neves and Hirose (2005).
After conidiogenesis on insect, each strain was inoculated into solid
culture medium and conidia used for preparation of a 108 conidia
mL-1 suspension that was inoculated in liquid culture medium.
Submerged cultures were carried out at 28°C, 150 rpm by addition of
1% of the conidia suspension. The crude extract (CE) obtained by 5 days
cultures was filtrated using Whatman No.1 filter and was dialized into
12 kDa molecular cutoff membrane against 5 mM pH 5.0 acetate buffer at
4°C for 24 h.
Chitinase Assay
Chitinase activity was determined using the protocol described by
Nahar et al. (2004) with slight modifications: 0.5 mL aliquots
of CE were incubated with 1mL of 1% colloidal chitin (Practical Grade,
Sigma) (Kang et al., 1999) prepared in 50 mM acetate buffer pH
5.5 at 45°C. After 1 h, enzyme reaction was interrupted at 100°C
for 1 min. Residual chitin was separated by centrifugation at 1,000 g
for 15 min and the supernatant was used for determination of N-acetylglucosamine
using the method described by Reissig et al. (1955). One unit of
enzyme activity was defined as micrograms of N-acetylglucosamine released
per mL of CE at reaction conditions.
| Table 1: | Host and geographic origin of Beauveria bassiana isolates used to verify enzymatic activity and polymorphism in chitinase gene |
DNA Extraction
The mycelium was obtained by filtration of the liquid cultures, washed
with sterile water and dried using sterile absorbent paper. It was then
ground into a fine powder in liquid nitrogen and the DNA was extracted
from this powder using the protocol previously described by Bogo et
al. (1996). Pelleted DNA was resuspended in 60 μL of ultrapure
water and stored at 4°C. The quality and quantity of extracted DNA
samples was monitored routinely by electrophoresis of the material in
a 1% agarose gel in TBE buffer (89 mmol L-1 Tris borate, 89
mmol L-1 boric acid, 2 mmol L-1 EDTA; pH 8), stained
with ethidium bromide and observed under UV light. Quantification was
performed through comparison with known dilutions of lambda phage DNA.
PCR Reaction
A pair of primers were designed from a 1047 bp nucleotide sequence,
corresponding to the chitinase (Bbchit1) gene from B. bassiana that was
retrieved from the GenBank (Access Number AY145440) (Fang et al.,
2005). Primers were designed (Chit 1F 5´-GCTCGCAACATACCAATC-3` and Chit
1R 5´-GTCGCCAAATGTCCAATTC-3`) using the Gene Runner (Hastings Software,
Inc). The PCR reactions were carried out in a final volume of 10 μL
with the following composition: 5 ng of the DNA, 1 μL of the 10X
reaction buffer (200 mmol L-1 Tris-HCl pH 8.4; 500 mmol L-1
KCl), 200 μmol L-1 of each dNTP, 10 pmol of each primer,
1.5 mmol L-1 of MgCl2, 1U of Taq DNA polymerase.
The temperature program was composed of an initial denaturation at 95°C
for 3 min, 30 cycles at 95°C for 1 min, 55°C for 1 min, 72°C
for 2 min and a final extension at 72°C for 5 min. All reactions were
repeated at least twice, always including both negative (DNA-free) and
positive controls. The success of the reaction was confirmed by running
the product in 1% agarose gel stained with ethidium bromide and visualized
under UV light. Pearson`s correlation coefficient were calculated to study
the association between chitinase activity and virulence of all lineages
against Hypothenemus hampeii.
PCR/RFLP Analysis
Restriction enzyme screening was done using the pDRAW32 (AcaClone)
based on the same sequence used for the primer design. Nine enzymes capable
of cutting the PCR fragment into two or three fragments: BglI (518/517
bp), BseYI (899/136 bp), DrdI (593/319/123 bp), HgaI (726/216/93 bp),
FokI (776/199/60 bp), AflIII (579/344/112 bp), BtgI (546/418/71 bp), BstF5I
(763/212/60 bp) and KasI (528/414/93) were selected. Aliquots (10 μL)
of PCR products were digested for 4 h at 37°C with the exception of
BstF5I that was incubated at 65°C. Digestion reactions were carried
out in a 15 μL final volume with 1.5 μL of 10X reaction buffer
and 1.5 μL of BSA (100 μg mL-1) according to the
manufacturer`s instructions. The restriction fragments were analyzed by
electrophoresis in 1 X TBE buffer with 3% agarose gels and detected by
UV fluorescence after ethidium bromide staining. The 1 Kb plus DNA ladder
(Invitrogen) was used as a molecular size marker. The occurrence of polymorphisms
was compared with the in vitro chitinase production and virulence
against H. hampei in all of the 30 B. bassiana isolates.
RESULTS AND DISCUSSION
The highest chitinase activity (1510.8 U mL-1) was detected for isolate CB95 but which only caused 36.7% of mortality. On the other hand, the most virulent isolate, causing the highest mortality (CB102: 83.3%), presented low chitinase activity, 46.4 U mL-1. Thus the in vitro chitinase activity was not directly related to the mortality rates in H. hampei (r = -0.122) that was previously described by Neves and Hirose (2005) (Table 2).
Using the above-described PCR protocol, a single product of approximately 1.035 Kb was amplified from all of the 30 isolates of B. bassiana. This corresponded to the estimated size based on the previously GenBank deposited sequence of chitinase gene. Enzymatic digestions of this PCR product generated the expected band patterns with all restriction enzymes. Three (DrdI, FokI and KasI) of the 9 studied enzymes showed polymorphism in 4 of the 30 isolates analyzed.
Figure 1 shows the polymorphic patterns observed in these isolates: CG245, UEL04, CB47, CG138. The chitinase gene has two restriction sites for DrdI and digestion of the 1.035 Kb PCR product was expected to generate three fragments of approximately 593, 319 and 123 bp. However, the restriction of the sequence amplified from strain CG245 showed only two fragments (300 and 735 bp), suggesting the loss of a restriction site in this strain. The mortality rate caused by CG245 was 8% and can be considered one of the lowest, although the chitinase activity was considerably high: around 400 U mL-1. This data suggest that the mutation was not associated with virulence level on insects, as isolates causing high (CB102) and low (CB47) mortality rates in insects showed the same restriction pattern for this gene. Any association would be revealed by the chitinase activity and also polymorphic profile.
The PCR product of the isolates UEL04 and CB47, when digested with FokI,
showed a different pattern from that expected, with the 776 bp original
fragment digested in two with approximately 700 and 76 bp. The KasI digestion
of the CG138 PCR fragment produced Four fragments (407, 277, 255 and 96
bp). This restriction pattern suggests a new mutational site for this
enzyme as the expected 528 bp fragment was cut in two of 277 and 255 bp.
The mortality caused by UEL04 was around 40% while the strain CB47 caused
only 3% death; this observation shows a non-association between these
polymorphic profiles and the mortality rates against H. hampei.
The in vitro enzymatic activity also did not show any association
with the restriction patterns (UEL04: 200.095 U mL-1 and CB47:
77.685 U mL-1).
| Table 2: | Mortality on adult coffee berry borer (Hypothenemus hampei) (Neves and Hirose, 2005) and chitinase activity of the 30 Beauveria bassiana (2.5x107 conidia mL-1) isolates |
| NA: Not available | |
| Fig. 1: | Electrophoretic profiles of the digested chitinase gene in B. bassiana. (A) strain CG245 digested with DrdI (lane 2), (B) strain UEL04 digested with FokI (lane 2) and (C) strain CG138 digested with KasI (lane 2). Lanes 1 in all figures shows the restriction profile obtained in other strains. MM indicates the DNA molecular weight marker 1 Kb Plus DNA ladder |
Similarly, CG138 showed an intermediate mortality rate (50%) and chitinase activity (444.96 U mL-1) which lead us to infer that this mutation was not related with the ability to infect and kill the insect host. No association was observed between geographic origin or insect host and the genetic variation. The isolates UEL04 and CB47 presented the same polymorphic pattern in spite of being isolated from different insects at different geographic regions (H. hampei from Paraná and A. grandis from São Paulo, respectively). Likewise, the strain CG138 isolated from Cosmopolites sordidus showed a different restriction pattern when compared with other isolated from the same species (CB97 e CB35) and in the case of CG138 and CB35, at the same geographical region (Brazilian Northeast). The strain CG245 isolated from a Hymenoptera from the Goiás State showed a different profile from others (CB95 and 447-ESALQ) isolated from the same host. However, as this isolate was unique to this area, further sampling is needed to verify if some local evolutionary forces were influencing the genetic profile. The lack of association between geographical region or host preference and genetic variation has been previously related (Muro et al., 2003; Gaitan et al., 2002; Coates et al., 2002), using different methodologies, like the Internal Transcribed Spacer sequencing, as well as RFLP and RAPD analysis.
Wang et al. (2003) found association between the protease gene variation and microsatellite molecular markers with geographical locations of the isolates, while Maurer et al. (1997) suggests that the insect host is the major selection factor on the evolutionary process of B. bassiana. The absence of a visible association between activity of chitinase, which is one of the pathogenicity-related factor, RE polymorphisms and virulence of B. bassiana on adults of the coffee borer (H. hampei) suggests that several genes act together in the process of infection/colonization. It is possible that alterations in the chitinase nucleotide sequence are not related with the fungal efficiency in infecting insects and that the virulence, because of being defined by a combination of several factors, is controlled by other regulatory genes or post-translational modifications of the enzymes and may be some determinative virulence factors are less efficient in less virulent isolates.
ACKNOWLEDGMENTS
CAPES (Brazil) is acknowledged for ProDoc Fellowship to Dr. J.E. Garcia, D.C. Sassá acknowledges CNPq (Brazil) for a Master Fellowship. This study was financed by Fundação Araucária de Apoio ao Desenvolvimento CientÃfico e Tecnológico do Paraná, Ed. 06/2003 Prot. 5873.