There are many deleterious effects of chemical pesticides. The social and eco-economical
problems such as health hazards, contamination of food and water, resurgence
of primary pest, replacement of primary pest with secondary one, not only by
the destruction of natural enemies but also by changing pest behavior, dispersal,
development and fecundity are well recognized. Realizing these problems, recent
sustainable forest management policy and the certification schemes of forests
and forest products have given paramount importance to the environmental aspects
of pest management and emphasis for the non-chemical pest control methods (Kneeshaw
et al., 2000; Mihajlovich, 2001).
Entomopathogenic fungi could provide environmentally benign alternatives to
chemical pesticides due to their environmentally friendly characteristics. Successful
application of Beauveria bassiana against several economically important
pest insects and its wide distribution has rekindled interest to use it in non-chemical
pest control programmes (Arcas et al., 1999).
Beauveria bassiana has more than 707 host insect species (Li,
1988) belonging to 521 genera and 149 families of 15 orders. It is found
globally in alpine soil, heathland, peat bogs, soils with savannah type vegetation,
forest and cultivated soils, sand blows and dunes, desert soils, running water,
rhizoplane of peat bog plants, the rhizosphere of clover, dead bark, nests,
feathers and droppings of free-living birds (Zimmermann,
2007) and phylloplanes of various plants species (Meyling
and Eilenberg, 2006).
Despite having a wide spectrum of insect host species and cosmopolitan distribution,
studies have shown that virulence of B. bassiana isolates tends to be
host specific (Goettel et al., 1990; Vestergaard
et al., 2004). Isolates from different sources have variable degrees
of pathogenicity against an insect species and vice versa (Cottrell
and Shapiro-Ilan, 2003; Shah and Pell, 2003; Castrillo
et al., 2004). It is therefore necessary to screen isolates for pathogenicity
vis-à-vis virulence against a target insect species to find out the most
infective one for further applications.
The tiger moth, Atteva sciodoxa is a serious perennial pest in Eurycoma
longifolia plantations, a commercial medicinal plant species in Malaysia
and the South East Asian region. The plant is commonly known as tongkat Ali
in Malaysia and pasak bumi in Indonesia. Atteva sciodoxa attacks young
apical shoots, resulting in stunted plant growth and also mortality of plant
in case of severe infestation (Abood et al., 2008).
Eurycoma longifolia roots and leaves are widely consumed for medicinal
purposes and it is thus essential that pest control methods do not involve undesirable
This study was conducted to (1) screen the most infective isolate of B. bassiana against A. sciodoxa; (2) compare B. bassiana vegetative growth, germination and sporulation on the host cadavers; (3) estimate the median effective concentrations and times of the most infective isolate(s) and assess effects of fungal infection on the food consumption of A. sciodoxa.
MATERIALS AND METHODS
This study was conducted at Faculty of Forestry, Universiti Putra Malaysia, Malaysia during 2007-08.
Fungal Isolates and Conditions
Seven B. bassiana isolates from various insect species and geographic
origins (Table 1) were bioassayed against third instar A.
sciodoxa larvae at 27±2°C and 75±5% relative humidity
with 12 h photoperiod.
Isolates of B. bassiana screened against A. sciodoxa
at 27±2°C and 75±5% relative humidity with 12 h photoperiod
Cultures and Inocula
Isolates FS-11, F-1 and F-8 were obtained from Malaysian Palm Oil Board
(MPOB) while Bba-Sl1, Bba-Sl2 and Bba-Sl3 from the Entomology Laboratory, Faculty
of Forestry, Universiti Putra Malaysia. The isolates were maintained at 4°C
on Potato Dextrose Agar (PDA). Isolate Bba-Pp was isolated from Pteroma pendula
collected from a teak plantation in Sungai Buluh, Selangor, Malaysia.
Prior to bioassay, the isolates were passaged through A. sciodoxa. Fifteen 3rd instar larvae were exposed to two-week old cultures of each isolate for 10 min. The inoculated larvae were reared on E. longifolia for seven days. Dead larvae were separated daily and kept in a humidity chamber in 9.0 cm Petri dish with moistened Whatman® filter paper and sealed with parafilm for sporulation. Upon sporulation the cadavers were kept as nuclei cultures. A selective medium comprising PDA (Difco Laboratories, Augsburg, Germany) enriched with 0.5% Yeast Extract (YE) to which were added 0.05% streptomycin sulphate and 0.03% chloramphenicol, (Sigma Aldrich Chemie GmbH, Steinheim, Germany) was used to isolate inocula from cadavers. Five evenly spaced spots were marked in each Petri dish and were inoculated using sterilized inoculating needle. The inoculated Petri dishes were incubated at 27°C in darkness. After 24 h of incubation, the germinating spots were transferred singly on PDA+0.5%YE and incubated for two weeks at 27°C. Conidia were harvested from these cultures and used for further sub-cultures. These sub-cultures were maintained on PDA+YE at 27°C in darkness.
Conidia were harvested by scraping with a sterile glass rod two-week old
cultures mixed with 20 mL of 0.02% aqueous Tween 80 in PDA plates. To prepare
a homogenous suspension the mixture was transferred into test tubes and vortexed
at a speed of 3000 rpm for 5 min using IKA® MS 3-Digital Vortex
(IKA Werke GmbH and Co KG, Staufen, Germany). The aliquot was filtered twice
using cheesecloth. A dilution of 1:100 was prepared and the appropriate serial
concentrations of conidia were determined using Hirschman® Neubauer
improved haemocytometer (Hirschmann Laborgeräte, Germany).
Stock culture of A. sciodoxa was bred on E. longifolia in the laboratory at 27±2°C with 75±5% relative humidity and 12 h photoperiod. A total of 50 third instar larvae selected at random were inoculated at a concentration of 5x107 conidia mL-1 per isolate along with E. longifolia leaves using Preval® TLC sprayer (Precision Valve Corporation, NY, USA). Eight mililitre of conidial suspension was used per treatment spray while only aqueous 0.02% Tween 80 was used for the control set. The inoculated larvae were confined to cylindrical containers (11 cm diameterx8 cm) lined with moist filter paper for 24 h and then transferred to new containers provided with fresh untreated leaves. Survival of the larvae was monitored daily and moribund larvae were placed in humidity chambers for mycelial growth and sporulation.
Based on the screening results, two superior isolates, Bba-Pp and FS-11, were used to estimate the median Effective Concentrations (EC50) and median Effective Times (ET50). The inocula of Bba-Pp and FS-11 were obtained from fresh cultures as previously described and concentrations of 1x106, 5x106, 1x107 and 5x107 conidia mL-1 were prepared. A total of 50 third instar larvae per concentration were randomly selected from the stock culture and inoculated with different concentration levels as described earlier.
Sporulation and Viability
Fungal growth on two-week old cadavers was used to determine the number
of conidia mg-1 cadaver body weight and their percentage germination.
Ten milligrams of cadaver along with fungal growth was mixed in 10 mL of 0.02%
aqueous Tween 80. The mixture was vortexed for 5 min to provide a homogenous
suspension. The aliquot was filtered twice using cheesecloth. Concentration
of the conidia was determined in 1:10 time diluted suspension using Neubauer
improved haemocytometer and the number of conidia was calculated for each mg
cadaver body weight.
A 0.1 mL suspension from a 1:10 dilution was pipetted on PDA+YE plate and spread evenly using a cell spreader. The inoculated plates were incubated at 27°C in darkness for 24 h. The germinating and non-germinating conidia were stained with lactophenol cotton blue. A rectangular piece of each PDA isolate (1.5x2) cm2 was mounted on microscope slide and examined for conidial germination using a compound microscope (400x).
Effect of Fungal Infection on Food Consumption
Effect of infection by Bba-Pp and FS-11 on food consumption of A. sciodoxa
was assessed using a gravimetric method as described by Waldbauer
(1968) for concentrations of 1x106, 5x106 and 1x107
conidia mL-1. For each treatment, two sets of E. longifolia
leaves of equal weight were provided: one to feed the larvae while the other
was a control to obtain the equivalent oven-dry weight of the exposed leaf.
After 24 h, the remaining leaves in each set were removed and oven-dried. Daily
mean food intake per larva was calculated by subtracting the oven-dry weight
of the remaining leaf from oven-dry weight of equal weight obtained from blank
set, divided by the number of larvae in the respective treatment.
Experimental Designs and Statistical Analysis
The trials were set up in completely randomized design with five replications.
The percent corrected mortality over time was calculated as described by Abbott
(1925). Overall degree of pathogenicity was analyzed by 1-Way ANOVA, while
differences among isolates and concentrations were analyzed by Tukeys
Median effective concentrations (EC50 and EC99) and median effective time (ET50) of Bba-Pp and FS-11 were calculated on the basis of seven days of post-inoculation using Probit Programme Version 1.5 (US Environmental Protection Agency). The intercepts and slopes were calculated by Linear Regression analysis and further subjected to χ2 test. Pearsons Linear Correlation was used to determine the correlation between log10 transformed inocula and total food consumed over seven days. The effect of inoculum concentrations on food consumption was further quantified by calculating R2 values from linear regression analysis.
Screening of the Isolates
All isolates tested were found to be pathogenic against A. sciodoxa.
However, the infectivity among isolates was highly variable. The earliest mortality
was recorded on day three after inoculation (DAI). On day seven after inoculation,
an overall highly significant (F7, 32: 377.88; p<0.01) larval
mortality was observed. Bba-Pp caused the highest mortality of 100% while Bba-Sl3
caused the lowest of 24.9±2.1 (Table 2). The differences
between Bba-Pp and FS-11; FS-11 and F-8 were significant (Tukeys HSD;
p = 0.05; CV: 7.6), while the differences among F-1, Bba-Sl1 and Bba-SL2 were
not significant. The mortality caused by Bba-Pp was 1.2 to 4.0 times greater
as compared to other six tested isolates.
The mean survival time after exposure also varied among the isolates. The shortest
survival time of 50% larval population was 3.6 days in Bba-Pp treatment and
this was 0.24 times that observed with Bba-Sl3 (15.3 days).
||Mortality and median effective times of B. bassiana
isolates at 5x107 conidia mL-1 against 3rd instar
A. sciodoxa at 27±2°C and 75±5% relative humidity
with 12 h photoperiod
|Means within columns with the same letter are not significantly
different (p = 0.05, Tukeys HSD)
The time taken by Bba-Pp to kill 50% population of A. sciodoxa was 0.24
to 0.88 times greater than the other six tested isolates. The intercept and
slope values for median effective time of Bba-Pp were -9.89±1.97 and
7.70±0.98, respectively whereas these were ranged from -2.05±5.70
to -5.87±2.89 and between 2.75±2.63 and 4.75±1.35 for other
tested isolates. The intercept and greater slope value of Bba-Pp indicates good
potential for rapid high mortality (knockdown) effect as compared to the other
The results of screening bioassay further indicated that with the progression
of post-inoculation time, inter- and intra-isolate mortality rates also varied.
In all isolates, the mortality rate increased to a certain time after inoculation.
The highest mortality rate was within 24 to 48 h after first occurrence of mortality.
This indicates maximum mortality occurred between day 4 and 5 after inoculation.
The highest mortality rate for Bba-Sl2, FS-11, Bba-Sl3, Bba-Pp, F-8, F-1 and
Bba-Sl1 between any two successive days was 55.5, 68.8, 74.0, 91.2, 286.1, 289.6
and 311.0%. It was observed that there was a relatively higher rate of increment
in mortality, i.e., 77.5 and 106.7% between days four and five after inoculation
in Bba-Sl2 and Bba-Sl3. This was likely due to low initial mortality of 6.0%
in these isolates at 4 DAI as compared to Bba-Pp which resulted in 62.4% mortality.
The mean mortality at 4 DAI for Bba-Pp was 7.8 and 10.4 times of Bba-Sl2 and
Bba-Sl3, respectively. This suggests that the temporal mortality increment rates
with each isolate is high for those isolates which produced low early mortality,
such as Bba-Sl2 and Bba-Sl3 (between 4 to 5 DAI). The overall mortality level
remained high in isolates with high early mortality.
The trend and level of isolate infectivity remained the same even after 7 DAI. At 12 DAI, highly significant (F7, 32: 176.53; p<0.01) difference in survival of A. sciodoxa was recorded. The lowest larval survival was zero percent in Bba-Pp while the highest among test isolates was 68% in Bba-Sl3 (Fig. 1). In the control, survival was 96% and the mortality recorded was non-infectious because no fungal growth appeared on the cadavers. The differences in mortality observed between Bba-Sl1 and Bba-Sl2, Bba-Sl2 and Bba-Sl3 as well as F-1 and Bba-Sl1 were not significant ((Tukey HSD, p = 0.05; CV: 10.50).
From the overall results of infectivity, the isolates can be classified into three groups:
||Isolates of high infectivity with a mean survival of inoculated
A. sciodoxa between 0% to 20% (Bba-Pp, FS-11)
||Isolates of intermediate infectivity with mean survival between
30 and 50% (F-1, F-8)
||Isolates with low infectivity with greater than 50% mean survival
(Bba-Sl1, Bba-Sl2, Bba-Sl3)
||Mean percent survival (±SE) of A. sciodoxa at
12 DAI with B. bassiana isolates at 27±2°C and 75±5%
relative humidity with 12 h photoperiod
||Median effective concentration values (105 conidia
mL-1) of Bba-Pp and FS-11 against A. sciodoxa at 27±2°C
and 75±5% relative humidity with 12 h photoperiod
Median Effective Concentrations (EC50) and Times (ET50)
The results showed considerable differences in both EC50 and
EC99 values between Bba-Pp and FS-11. The difference between the
two isolates was more profound at EC99 (Table 3).
The EC50 of Bba-Pp showed that it required only 0.26 times inocula
for the same level of effect as compared to FS-11 (9.89x105 conidia
mL-1 and 38.45x105 conidia mL-1) and for 99%
mortality Bba-Pp required 2.87x107 conidia mL-1 which
was 0.02 times that of FS-11 (130.97x107 conidia mL-1).
The estimated values of intercept of Bba-Pp and FS-11 were -4.53±1.81
and -1.05±1.18, while the estimated slope values were 1.59±0.28
and 0.92±0.17. The values of intercept and slope of Bba-Pp signifies
a higher level of initial mortality as well as a steep mortality curve as compared
to that of FS-11.
On day 7 after inoculation, a highly significant (F8, 36: 309.86; p<0.01) inter-concentration effect, both in Bba-Pp and FS-11 was found. Isolate Bba-Pp at the concentration of 1x107 conidia mL-1 caused significantly (Tukeys HSD; p = 0.05; CV: 8.7) higher mortality than 5x107 conidia mL-1 of FS-11. The difference in mortality between 1x106 conidia mL-1 of Bba-Pp and that of 5x106 conidia ml-1of FS-11 was not significant. Similarly, the difference in mortality between Bba-Pp at the concentration of 5x106 conidia mL-1 and FS-11 at the concentration of 5x107 conidia mL-1 was not significant. Isolate Bba-Pp at the concentrations of 1x106, 5x106, 1x107 and 5x107 conidia mL-1 caused 1.9, 1.4, 1.6 and 1.2 times greater mortality than that of the corresponding concentrations of FS-11 (Table 4). Within the isolates, the difference among concentrations was also highly significant (Bba-Pp F4, 20: 729.96; p<0.01; FS-11 F4, 20: 270.01; p<0.01) at 7 DAI. There was no significant difference in mean survival of A. sciodoxa when inoculated with 1x107 and 5x107 conidia mL-1 of Bba-Pp (Tukeys HSD; p = 0.05; CV: 6.5), while the difference between concentrations of 5x106 and 1x107 conidia mL-1 of FS-11 was not significant (Tukeys HSD; p = 0.05; CV: 8.4).
The median effective time (ET50), of inoculated larvae varied with
the concentration of inocula for both isolates. The ET50 values of
Bba-Pp ranged between 3.6 and 7.0 days for concentrations of 5x107 to
1x106 conidia mL-1 while that of FS-11 were between 4.1
to 10.3 days. Bba-Pp at a concentration of 5x107 conidia mL-1
took 0.88 times the required time to kill 50% of the inoculated larvae
compared to that of FS-11.
||Median effective time for different concentrations (Conidia
mL-1) of Bba-Pp and FS-11 against A. sciodoxa at 27±2°C
and 75±5% relative humidity with 12 h photoperiod
Means within columns
with the same letter are not significantly different (p = 0.05, Tukeys
HSD); NA: Larval mortality was less than 50% at 7 DAI
to mycelial appearance (h) (±SE) of B. bassiana isolates
on A. sciodoxa cadavers at 27±2°C and 90±5% relative
humidity with 12 h photoperiod
within columns with the same letter are not significantly different (p
= 0.05, Tukeys HSD)
The difference in ET50 values was more evident at low concentrations
than the higher concentrations. Bba-Pp at a concentration of 1x107
conidia mL-1 required almost the same duration for 50% mortality
of inoculated larvae as that of FS-11 at a concentration of 5x107
conidia mL-1 (4.4 days and 4.1 days). An inoculum of 1x106
conidia mL-1 of Bba-Pp needed 0.68 times duration to kill 50% inoculated
larvae that of FS-11 at the same inoculum level (Table 4).
Vegetative Fungal Growth and Sporulation
Mycelial appearance and time to mycelial appearance on cadavers are important
parameters to assess larval death by fungal infection (Moorhouse
et al., 1993; Vandenberg, 1996) and to determine
the growth rate of the fungus. The results indicated that the time to mycelial
appearance on cadavers is correlated to infectivity potential of the isolate
as well as the time taken to death following inoculation. Mycelia of Bba-Pp
appeared on 24 h old cadavers when larvae died at 3 DAI. This mycelial appearance
of Bba-Pp was 0.79, 0.61, 56 and 0.52 times faster than that of FS-11, F-8,
F-1 and Bba-Sl1, respectively. At 4 DAI, the time to mycelial appearance of
isolates Bba-Sl2 and Bba-Sl3 was 2.1 times that of Bba-Pp. In all the isolates,
the time to mycelial appearance on cadavers decreased as the time between inoculation
and death increased (Table 5).
The time to mycelial appearance on cadavers ranged between 19.6 to 31.8 h when larvae died at 7 DAI. Based on the time to mycelial appearance on cadavers at 7 DAI, the isolates can be classified into three groups:
||FS-11, F1 and F-8 = 24 h
||Bba-Sl1, Bba-Sl2 and Bba-Sl3 > 24 h
||Mean number and viability of conidia on two-week old cadavers
of A. sciodoxa at 27±2°C in darkness
|Means within columns with the same letter are not significantly
different; (p = 0.05, Tukeys HSD); ns: Not significant
The mean daily reduction in time to mycelial appearance on the cadavers from 3 DAI to 7 DAI for Bba-Pp, FS-11, F-8, F-1 and Bba-Sl1 were 3.7, 4.2, 7.8, 8.9 and 9.3%, respectively while the daily mean reduction in time to mycelial appearance on the cadavers from 4 DAI to 7 DAI for Bba-Sl2 and Bba-Sl3 was 9.8 and 9.0%. There was an apparent positive correlation between mortality caused by the isolate and time to mycelial appearance on cadavers.
The isolates which caused high early mortality, such as Bba-Pp and FS-11, appeared earlier on the cadavers as compared to isolates with low and delayed mortality, such as Bba-Sl1, Bba-Sl2 and Bba-Sl3. The rate of mean daily reduction in time to mycelial appearance on cadavers in low virulent isolates was relatively greater as compared to high virulent isolates. This observation may be attributed to the longer initial time taken for first mycelial appearance (4 DAI, 48 h) by low virulent isolates. The high virulent isolates appeared in less time (3 DAI, 24 h) therefore mean daily rate of reduction in mycelial appearance time was low.
Production of conidia by different isolates of B. bassiana may also
affect the degree of pathogenicity because certain amount of inoculum is required
to kill the target insect species. As such, conidial production, apart from
abiotic and biotic factors, depended on the B. bassiana isolate (Luz
et al., 1999); this was also exhibited in the present study, whereby,
a highly significant (F6, 28: 235.6; p<0.01) difference in conidial
production was observed in the seven tested B. bassiana isolates. Isolate
Bba-Pp was the most prolific isolate producing 150.6x105 conidia
mg-1 on two-week old cadavers (Table 6). The differences
between F-8 and Bba-Sl2; Bba-Sl1 and Bba-Sl2; F-1 and Bba-Sl3 were not significant
(Tukeys HSD, p=0.05; CV: 14.20). Isolate Bba-Pp produced 1.4 to 12.3 times
greater number of conidia mg-1 of cadaver as compared to other tested
isolates. The production of conidia followed the trend of mortality and vegetative
growth of the isolates (Table 2 and 5).
The results indicated that pathogenicity and virulence are apparently positively
correlated to vegetative growth and sporulation of the isolate.
The results of conidial germination ranged between 94.4±0.58 and 95.8±0.58%, thus indicating high viability of the isolates (Table 6). The overall difference in conidial germination of the seven isolates was not significant (F6, 28: 0.53; p>0.05).
Effect of Fungal Infection on Food Consumption
Isolates Bba-Pp and FS-11 showed significant negative effect on chronological
food consumption of A. sciodoxa (Fig. 2). However,
there was no significant difference (p = 0.05) in food consumption within the
first two days after inoculation. Initially there was a normal increase in food
consumption in all treatments relative to the controls attributed to increase
in larval age. At 3 DAI onwards, a significant (F6, 28:17.3; p<0.01;
CV: 1.1) reduction in food consumption was recorded. At 7 DAI, there was a highly
significant (F6, 28: 1496.6; p<0.01) difference among different
concentrations of Bba-Pp, FS-11 and the control. In the control, larvae consumed
significantly (Tukeys HSD; p = 0.05; CV: 0.95) more leaves (22.8 mg dry
leaf larva-1). The highest food consumption at 7 DAI among different
concentrations of two isolates was observed in FS-11 at a concentration of 1x106
conidia mL-1 with a consumption of 4.2 mg dry leaf per larva.
||Chronological food consumption by A. sciodoxa inoculated
with Bba-Pp and FS-11 at 27±2°C and 75±5% relative humidity
with 12 h photoperiod (a = 1x106 conidia mL-1; b =
5x106 conidia mL-1; c = 1x107 conidia mL-1)
||Cumulative mortality (±SE) and amount of food ingested
(mg dry leaf per larva) (±SE) by A. sciodoxa inoculated with
different concentrations of Bba-Pp and FS-11 at 27±2°C and 75±5%
relative humidity with 12 h photoperiod (a = 1x106; b = 5x106;
c = 1x107)
The difference in food consumption between larvae inoculated with 5x106
and 1x107 conidia mL-1 of Bba-Pp was not significant.
Similarly, the difference in food consumption between larvae inoculated with
Bba-Pp at a concentration of 1x106 conidia mL-1 and FS-11
at concentrations of 5x106, 1x107 conidia mL-1
were not significant.
The overall cumulative food consumed in different treatments over 7 DAI varied
highly significantly (F6, 28: 1570.7; p<0.01). Significantly,
more food (131.3 mg dry leaf larva-1) was consumed in the control
(Fig. 3). The greatest (72.5%) reduction in food consumption
was in Bba-Pp at 1x107 conidia mL-1 while the lowest was
in FS-11 (55.8%) at 1x106 conidia mL-1 as compared to
the control. There was significant difference in food consumption between Bba-Pp
at concentration of 1x106 conidia mL-1 and FS-11 at 1x106
and 5x106 conidia mL-1. Similarly there was no significant
(Tukeys HSD, p = 0.05; CV 3.8) difference between
concentrations of 5x106 and 1x107 conidia mL-1
of Bba-Pp. A strong linear negative correlation was present between fungal inoculation
and food consumption for both Bba-Pp and FS-11, with a correlation coefficient
value of -0.99 for both.
The regression equations for Bba-Pp and FS-11 were y = 115.52-1.25x and y=114.75-9.58x, respectively. The coefficient values of correlation and regression analyses of both Bba-Pp and FS-11 showed that isolates were equally effective in reducing food consumption by A. sciodoxa.
Pathogenicity is a complex phenomenon that starts with the germination and
generally percutaneous penetration of fungal germ tube and culminates in the
death of the host. Apart from conidial germination and penetration, in vivo
fungal growth, production of metabolites and toxins are important for pathogenicity.
The germination of conidia may be affected by ambient climatic conditions especially
cuticular temperature and moisture content (Keller and Zimmermann,
1989; Fuxa, 1995), availability of food on cuticle
surface (free amino acids and a carbon source) (Smith and
Grula, 1981), age of the insect host and presence of inhibitory compounds
on cuticular surface (Hajek and St. Leger, 1994). The
successful accomplishment of penetration of the germ tube is pre-requisite for
pathogenicity. Besides activities during the course of penetration of the host
cuticle, the pathogenicity and virulence of B. bassiana also depends
on physiological characteristics of a fungal isolate and its potential of production
of biochemical compounds within the insect host body (Bidochka
and Khachatourians, 1990; Gupta et al., 1994).
The present pathogenicity results show that exogenous and endogenous conditions
are conducive for Atteva sciodoxa-Beauveria bassiana pathogenic interactions.
The occurrence of mortality on day three following inoculation reflects the
pathogenic potential of the isolates. Previously, Adane et
al. (1996) reported pathogenicity of different B. bassiana isolates
to Sitophilus zeamais with considerable variation in the virulence among
the isolates. The best isolates commenced mycosis as early as on day three after
inoculation with mycelia appearing on 24 to 48 h old cadavers. Our findings
also indicated that isolates that caused early mortality following inoculation
(on day three after inoculation) resulted in greater mortality than those isolates
with delayed mortality. These findings on time to mycelial appearance on cadavers
are in corroboration with that of Adane et al. (1996).
This signifies that commencement of mortality is an important criterion to assess
the degree of pathogenicity. The early mortality also likely to help prevent
crop losses by virtue of reduced food consumption.
The present variation in virulence of B. bassiana isolates may be due
to different physiological characteristics and enzyme production potential of
the isolates as reported previously by Leland et al.
(2005). This observed variation in virulence of the isolates may also be
attributed to geographic origins and insect host species. Our isolates are principally
from two geographic origins and four insect host species. The virulence of two
isolates of Japanese origin (F-1, F-8) was moderate while the isolates of Malaysian
origin gave different results. The significant variation in virulence of the
Malaysian isolates may be assigned to their distinct niche and host species.
Bba-Pp and FS-11 were isolated from the bagworms, Pteroma pendula and
Metisa plana infesting the teak, Tectona grandis and the oil palm,
Elaeis guineensis grown in forest plantations, while Bba-Sls were isolated
from the tobacco caterpillar, Spodoptera litura infesting the tobacco
plants, Nicotiana tobacum grown in an agro-forestry system. The trend
in virulence of Malaysian isolates seems to
follow the habitat of the host insect and plant species. Atteva sciodoxa
has similar habitats to P. pendula and M. plana as E. longifolia
is currently grown in plantations with T. grandis and E. guineensis.
Previously influence of geographic origins on the pathogenicity of B. bassiana
has been reported by Soper and Ward (1981) and Vandenberg
The present trend of median effective concentration (9.89x105 conidia
mL-1) and median effective time (3.6 days at inoculating concentration
of 5x107 conidia mL-1) of Bba-Pp is comparable with previous
findings with different fungal isolates and insect species. For example, Feng
and Johnson (1990), Ekesi (1999) and Sabbahi
et al. (2008a, b) obtained LC50
values of 0.57x105, 1.8x105, 7.8x105 and 5.3x105
conidia mL-1 by using other isolates of B. bassiana
against the Russian wheat aphid, Diuraphis noxia, the pod sucking bugs,
Clavigralla tomentosicollis, Lygus hesperus and L. lineolaris,
respectively. They found median effective times of 4.2, 4.1, 4.5 and 4.4 days
at concentrations of 107, 108, 108 and 108
conidia mL-1, respectively. The variation in calculated LC50
and LT50 values of different isolates of B. bassiana may be
attributed to vegetative growth, sporulation characteristics, geographic origin
and insect hosts of the isolates as previously reported by Feng
and Johnson (1990). The differences in present EC50 and ET50
values and previously reported further be explained in terms of variations in
virulence of different isolates to an insect species or variations in virulence
of single isolate to related species of the host insect. Such variations have
earlier been recorded by Khachatourians (1992) and Poprawski
et al. (2000).
In addition to mortality, B. bassiana infection may also cause a spectrum
of changes in behavior of host insect particularly in feeding. Our results showed
that B. bassiana infection decreased significantly food consumption by
A. sciodoxa. These findings are in corroboration with Tefera
and Pringle (2003) where they found reduced mean daily food consumption
associated with B. bassiana infection in the spotted borer, Chilo
partellus. The reduction in food consumption became evident three to four
days after inoculation and consequently food consumption decreased by 70 to
85%. Our study also showed significant reduction in mean daily food consumption
starting on day three after inoculation with 55.8 to 72.5% less food consumption
at different concentrations of FS-11 and Bba-Pp. The difference in the present
results of overall reduced food consumption and those of Tefera
and Pringle (2003) is likely to be due to the larval stage and inoculating
concentrations used; the food consumption decreased with increasing inoculating
concentration. They used 1x108 conidia mL-1 concentration
against second instar larvae while we used a maximum 1x107 conidia
mL-1 concentration against third instar larvae.
The effect of fungal infection on food consumption has also been reported in
other insect pests, such as B. bassiana on L. decemlineata
(Fargues et al., 1994) and C. tomentosicollis
(Ekesi, 1999), Entomophaga maimaiga on the
hairy caterpillar, Lymantria dispar (Hajek, 1989),
Metarhizium flavoviride on the desert locust, Schistocerca gregaria
(Moore et al., 1992) and Nomuraea riley
on Plathypena scabra (Thorvilson et al., 1985).
Reduction in feeding rate and ultimately cessation is likely to be due to progressive
toxic effects of enzymes, metabolites and physical injury to the host tissues
by mycelial growth. Samuels et al. (1988) and
Vey and Quiot (1989) found that metabolites of fungi
act on insect tissues, including the midgut, which adversely affects the feeding
rate. These reduced feeding rates may help to prevent crop damage even insects
are not killed by infection (Noma and Strickler, 2000).
Selection of a viable and virulent isolate is crucial for a successful microbial
control of any insect pest. The seven isolates of B. bassiana
used in this study exhibited pathogenicity against A. sciodoxa. Bba-Pp
was the most virulent with 100% mortality followed by FS-11 (83.3% mortality)
on day seven after inoculation. The results further showed that isolates which
have fast vegetative growth and high sporulation potential such as Bba-Pp and
FS-11 were more virulent. The EC50 and ET50 values of
Bba-Pp were superior to FS-11 and comparable to other virulent B. bassiana
isolates tested against other insect pests. There was a strong negative correlation
between fungal infection and food consumption. A 72.5% reduction in food consumption
due to Bba-Pp infection suggests that the isolate has the potential to reduce
plant damage. Based on these results of mortality, time to mortality and food
consumption, it is concluded that isolate Bba-Pp has good potential for control
of A. sciodoxa on E. longifolia. Finally, as isolate Bba-Pp is
native to the tropical region thus has less environmental risks associated with
field applications as compared to non-indigenous isolates.
The authors wish to acknowledge Mr. Ramli Muslim of Malaysian Palm Oil Board for providing isolates FS-11, F-1 and F-8. We also wish to acknowledge the Pakistan Forest Institute, Peshawar and Universiti Putra Malaysia for their support for Mr. Ghulam Ali Bajwa to undertake this study in partial fulfillment of his PhD programme.