Occurrence and diversity of entomopathogenic fungi in cultivated and uncultivated soils in Pakistan

Abstract

The distribution of entomopathogenic fungi in various geographical areas of Punjab, Pakistan, is poorly understood. The present study was planned to explore the occurrence and diversity of entomopathogenic fungi from soils collected from cultivated and non-cultivated habitats. The detailed survey of different habitats (crop fields, fruits, vegetables, forests) was conducted to collect soil samples and the associated fungi were isolated using Galleria bait method. Among 210 soil samples, 168 fungal isolates were recovered and identified, with 98 from forests, 32 from vegetables, 30 from field crops and 8 from fruits. The major entomopathogenic fungi recovered from these samples were Beauveria bassiana, Metarhizium anisopliae, Paecilomyces lilacinus, B. brongniartii, P. chlamydosporia and Lecanicillium attenuatum. The diversity of entomopathogenic fungi was greater in soil samples from forests compared to crop fields, vegetables and fruits, respectively. The geographical attributes (altitude, longitude, latitude) greatly influenced the occurrence of entomopathogenic fungi with the highest number of isolates found from >600 m altitude, 33°–34′N latitude, and 73°74′E longitude from soil samples. The results of the surveys clearly indicated that the entomopathogenic fungi are distributed in the soils which may be used in successful IPM programs in future.

Introduction

Widely distributed and extensively studied entomopathogenic fungi belongs to two phylogenetic groups, namely Entomophthorales (Zygomycetes) and Ascomycetes (Evans 1988). Recent ancestral host reconstruction and phylogenetic dating analysis has revealed the plant-based ancestral nutritional mode (Sung et al. 2008) and the origin of entomopathogenic mode of the fungi may have arisen several times from a common saprophytic soil and leaf litter inhabiting ancestor (Spatafora & Blackwell 1993). Almost 90% of the arthropod pest species spend a part of their life cycle in the soil habitat (Kaya & Gaugler 1993) thus, soil is an important reservoir for entomopathogenic fungi and serve to control the populations of soil dwelling insect pests (Hajek 1997). Data about local isolates of entomopathogenic fungi, their diversity and distribution, is very important to conserve the native fungal species for natural control of insect pest populations within the agro-ecosystem (Hussein et al. 2010).

The pathogenic nature of the fungi became known in the early 1800s against silkworm, wiping out the silk industry in France, and the first entomopathogenic fungi Botrytis bassiana was identified as muscardine disease of silkworms (Vega et al. 2009). Later, Elie Metchnikoff (1845–1916) named this insect disease as green muscardine of wheat cockchafers, identified as Entomopthora anisopliae, and it was then mass produced by Krassilstschik (1888) for the control of sugar beet weevil. The actual number of entomopathogenic fungal species or genera is difficult to assess but according to a review there are about 90 genera and 700 species of entomopathogenic fungi (Onofre et al. 2001). While Jankevica (2004) reported 35 genera comprising of about 400 species of entomopathogenic fungi with broad host range. The entomopathogenic fungi are cosmopolitan in distribution found in various types of habitats/ecosystems (Meyling & Eilenberg 2007). These fungi have great potential as biocontrol agents and also contain different bioactive compounds (Thakur & Sandhu 2010). For instance, a group of destruxins was isolated and purified in various species of entomopathogenic fungi particularly Metarhizium anisopliae (Rao et al. 2006). Beauvericin, a secondary metabolite produced by Beauveria bassiana is a cyclohexadepsipeptide ionophore which was studied for the first time by Hamill et al. (1969) against brine shrimp. These secondary metabolites of various entomopathogenic fungi not only possess insecticidal properties but also contribute to the pathogenesis of fungal strains (Mollier et al. 1994). Presently, 170 fungal formulations have been developed from at least 12 species of fungi; the majority of them are B. bassiana (33.9%), M. anisopliae (33.9%), Isaria fumosorosea (5.8%) and B. brongniartii (4.1%) (Faria & Wraight 2007).

The comprehensive collection and isolation of naturally occurring local isolates of entomopathogenic fungi is imperative to maintain the biodiversity of biocontrol agents for the control of insect pests. There are a number of studies available on the occurrence of soil-inhabiting entomopathogenic fungi throughout the world (Meyling & Eilenberg 2007; Quesada-Moraga et al. 2007; Sun et al. 2008). The insect bait method/animal bait method/Galleria bait method (GBM) (Zimmermann 1986), Veen's medium (Milner & Lutton 1976) and Beilharz's selective medium (Beilharz et al. 1982) are generally used for the detection of naturally occurring entomopathogenic fungi in soil and certain other soil dwelling pathogens. The worldwide distribution of entomopathogenic fungi in insects from different habitats has also been reported (MacLeod 1954; Wraight et al. 1993; Oduor et al. 2000; Jankevica 2004; Barta & Cagáň 2006; Aung et al. 2008; Chen et al. 2008; Thakur & Sandhu 2010). Few studies are available on the collection and isolation of entomopathogenic nematodes from the different geographical locations of Pakistan, using the “Galleria mellonella soil trap” method (Shahina et al. 1998; Tabassum et al. 2005) but there are no previous reports indicating the natural occurrence and distribution of entomopathogenic fungi in the country. In the present studies, the biodiversity of entomopathogenic fungi was examined in the northern hilly areas of Punjab, Pakistan, which have a natural floral diversity. The specific objective of this work was to determine the occurrence and distribution of local entomopathogenic fungal species in cultivated (crop, vegetable, fruit fields) and uncultivated (forest) lands.

Materials and methods

Collection of soil samples

The soil samples were collected during a survey of different localities of the Punjab, the geographical details of the sites (Table 1) and their positions are given in Figure 1. A total of 210 soil samples were collected, of which 78 were from forests, 42 from vegetable fields, 52 from crop fields and 38 from fruit orchards. A 5 cm diameter cylindrical soil corer was used for taking the soil samples. The soil samples comprised of 400 g each from five points randomly selected from each site. These were placed in zip lock bags and stored at 4°C in the laboratory. The five samples from each locality were mixed to create a composite sample, and were analyzed not later than two months after collection. Each sample was sieved through 2 mm mesh to remove stones, roots, litter etc. The soil samples were spread over a tray under shade to remove excessive moisture in order to free them from nematode infestation and the samples were baited for fungus isolation.

Figure 1. Map of Pakistan showing ten sampling sites for the isolation of fungi from soils: 1. Lal Sohanra; 2. Chichawatni; 3. Changa Manga; 4. Faisalabad; 5. Sheikhupura; 6. Sargodha; 7. Gujranwala; 8. Rawalpindi; 9. Islamabad; 10. Murree.

Table 1. Sampling sites with geographical characteristics for the isolation of fungi from soils

Sites	Altitude (m)	Latitude	Longitude
Lal Sohanra	114	29°28′N	71°58′E
Chichawatni	159	30°53′N	72°70′E
Changa Manga	191	31°08′N	73°96′E
Sheikhupura	213	31°71′N	73°97′E
Faisalabad	184	31°30′N	73°05′E
Sargodha	193	32°10′N	72°40′E
Gujranwala	223	32°10′N	74°12′E
Rawalpindi	497	33°58′N	73°08′E
Islamabad	610	33°40′N	73°10′E
Murree	2300	33°56′N	73°28′E

Isolation and identification of fungi

The Galleria bait method was used to isolate the fungi from the soil samples (Zimmermann 1986). The dry samples were moistened with distilled water and after thorough mixing were placed in autoclaved glass tubes. Four tubes representing one sample were prepared leaving a 1 cm space at the top and lids were not tightly closed for proper aeration. The larvae of wax moth Galleria mellonella L. (Lepidoptera: Pyralidae) were collected from bees wax and were artificially reared on the appropriate diet in an incubator in the laboratory at 26 ± 2°C under constant dark conditions. Ten larvae (3^rd instar) were immersed in 56°C water for 15 s to minimize their ability to produce silk webbing in the soil (Woodring & Kaya 1988). The larvae were then added to each vial and incubated at 25°C for 12 days. The tubes were shaken up and down after every 12 h so that the larvae move from up to down and vice versa. The larvae were examined after incubation and the dead ones were removed. The dead larvae were washed for 1 min in autoclaved water and then transferred to 75% ethanol for 1 min. These larvae were surface sterilized in 3% NAClO for 3 min and again transferred to autoclaved water and 75% ethanol for 1 min. The larvae were placed on filter paper to get dry. Then the larvae were shifted to Potato Dextrose Agar (PDA) and Sabouraud Dextrose Agar (SDA) plates amended with 0.1 g/l of Streptomycin Sulphate and 0.05 g/l of Chloramphenicol media plates sealed with parafilm. These plates were incubated at 25°C till the appearance of fungal mycelium.

After the fungal growth on the media plates, the fungi were identified on the basis of morphological traits under the microscope (40 × magnification) with the help of taxonomic keys (Barnett & Hunter 1999; Domsch et al. 2007). The slides and the culture of identified entomopathogenic fungal isolates were deposited in the IPM Laboratory, Department of Agricultural Entomology, University of Agriculture, Faisalabad, Pakistan.

Pathogenicity test (Koch's postulates)

The isolated entomopathogenic fungi were sub-cultured at 25°C and 75% relative humidity with 16h illumination per day on SDA (32.5 g SDA; 7.5 g of bacto agar; 5 g yeast in one liter distilled water) plates for the production of conidia. After 14 days of incubation, the plates were kept under an aluminium foil roasting pan on the bench top to dry for one week. Then the fungal conidia were harvested by scraping the conidial layers formed on the surface of plates using a sterilized (70% ethanol) scalpel. The fungal conidia were dissolved in 0.05% Tween-80 solution and filtered through muslin cloth to remove the mycelial debris. The third larval instar of G. mellonella were immersed in the conidial suspension of the each fungal isolate for three to five seconds and were shifted to the Petri plates lined with moist filter paper sealed with parafilm. These plates were incubated at 25°C and four replicates each having ten larvae were used. The larvae treated with Tween-80 served as control and the same procedure for each isolate was repeated three times with four replications to avoid pseudo-replication. The inoculated larvae were examined after every 24 h until their death or pupation. The cadavers were placed on the wet filter paper to enhance the fungal growth and the conidiation. The fungus from each cadaver was microscopically identified and compared with the characteritics of the inoculated fungus.

Statistical analysis

The occurrence of fungi in different habitats (field crops, fruits, vegetables, forests) was compared using the chi-squared test. In the pathogenicity bioassay, the mortality of the test insect species in the control was very low (0.08%), so was not included in the analysis. The percent mortality of the cadavers from each isolate was corrected using Abbott's (1925) formula. The statistical software Minitab v13.2 was used for the analysis of the data (Minitab 2002 Software Inc., Northampton, MA, USA).

Results and discussion

Occurrence of fungi

The 210 samples of soil collected during the survey of different sites revealed great diversity in the occurrence of the fungi belonging to different genera (Table 2) and 168 out of them were positive for entomopathogenic fungi. The major genera among them were Aspergillus (28.57%), followed by Fusarium sp., Penicillum sp. and Mucor sp. (8.34%), however, the least (0.60%) occurring fungi were Neoneetria ramulariae, Phoma sp. and Verticillium dahliae. Among the habitats 98 isolates of various fungal genera from forests, 32 from vegetables, 30 from field crops and 8 from fruit orchards were found. Similar to the present work, Sun and Liu (2008) found 377 fungi belonging to 46 species and 27 genera from 425 soil samples from different geographical regions in China and among them 21 species were opportunistic, 19 were secondary colonizers and 6 species were found to be entomopathogenic. The occurrence of different fungal species greatly depends upon the type of habitats as Sun et al. (2008) reported 25 species in the field crop soils and 20 species in the orchard soils. Similarly, Pinruan et al. (2007) identified 147 fungal species from decaying palm material and among them 79 ascomycetes in 50 genera (53%), 65 anamorphic taxa in 53 genera (45%) and 3 basidiomycete species in 3 genera (2%), however, 9 ascomycetes and 5 anamorphic fungi were new to science. The opportunistic and secondary colonizers have a fast growing habit and cause infection to injured or weaker insects (Hajek et al. 1993). They may also live as non-pathogenic insect associates and they live saprophytically in the soils (Gunde-Cimerman et al. 1998). The opportunistic pathogens and secondary colonizers have been overlooked to date (Sun & Liu 2008) yet they may play a significant role in facilitating insect pathogen interaction.

Table 2. Distribution and frequency (% positive samples) of fungal species in soils of crop fields, fruits, vegetables and forests of Punjab, Pakistan

Fungal species	Sampling sites				Chi^2	P	Distribution frequency (%)	Total isolates (n)
	Crop fields	Fruits	Vegetables	Forests
Alternaria alternata	2	0	1	4	2.25	0.52	4.17	7
Alternaria tenuissima	0	0	2	2	3.76	–	2.38	4
Aspergillus flavus	2	1	4	9	4.31	0.23	9.52	16
Aspergillus fumigatus	2	1	3	7	2.42	0.48	7.74	13
Aspergillus niger	2	2	5	10	4.13	0.24	11.31	19
Beauveria bassiana	1	1	2	4	1.12	0.77	4.76	8
Beauveria brongniartii	1	0	0	1	1.37	–	1.19	2
Fuasrium oxysporum	3	1	0	7	5.06	0.16	6.55	11
Fusarium oxysporum sub. sp. lycopesisum	0	0	1	2	2.28	–	1.79	3
Lecanicillium attenuatum	0	0	0	1	1.70	–	0.60	1
Macrophomina phaseolina	1	0	2	5	3.55	0.31	4.76	8
Metarhizium anisopliae	1	0	1	3	1.69	–	2.98	5
Mucor lausanesis	1	0	1	3	1.69	–	2.98	5
Mucor varians	2	1	2	4	0.43	0.93	5.36	9
Myrothecium sp.	0	0	0	2	3.41	–	1.19	2
Neoneetria ramulariae	0	0	1	0	4.01	–	0.60	1
Paecilomyces lilacinus	1	0	1	2	0.97	–	2.38	4
Penicillium chrysogenum	1	0	1	3	1.69	–	2.98	5
Penicillium digitatum	1	0	0	2	1.96	–	1.79	3
Penicillium Italicum	1	0	1	4	2.76	0.42	3.57	6
Phoma sp.	0	0	0	1	1.70	–	0.60	1
Pochonia chlamydosporia	0	0	0	2	3.41	–	1.19	2
Pythium debaranyum	2	0	1	4	2.25	0.52	4.17	7
Rhizoctonia solani	2	1	1	5	1.50	0.68	5.36	9
Rhizopus stolonifers	2	0	0	2	2.78	–	2.38	4
Synchytrium endobioticum	1	0	0	2	1.96	–	1.79	3
Trichoderma harzianum	1	0	1	3	1.69	–	2.98	5
Trichoderma reesei	0	0	0	2	3.41	–	1.19	2
Trichoderma viride	0	0	1	1	1.86	–	1.19	2
Verticillium dahliae	0	0	0	1	1.70	–	0.60	1
Total	30	8	32	98	–	–	–	168

Insect pathogenic fungi

The distribution of entomopathogenic fungi was studied in different sites of the Punjab province, Pakistan. Entomopathogenic fungi were found to occur in 13.1% (22 of 168) of the soil samples studied. Beauveria sp. (bassiana and brongniartii) was the most frequently occurring (5.95%) compared to other fungi. M. anisopliae was the next most important distribution of entomopathogenic fungi (2.98%). Lecanicillium attenuatum was the least found fungi with (0.60%) distribution (Table 2). However, among 22 soil samples containing exclusively entomopathogenic fungi (Table 3) the Beauveria sp. was distributed in 45.45% soil samples and ranked at the top followed by M. anisopliae (22.73%), Paecilomyces lilacinus (18.18%), Pochonia chlamydosporia (9.09%) and L. attenuatum (4.55%). In the present study the collection sites greatly influenced the distribution of entomopathogenic fungi. The forest soils were contaminated with all types of mycoflora with the most frequent occurrence in these sites being Beauveria sp. The crop fields and the vegetable sites yielded all the fungi but their distribution was smaller compared to the forest soils. The soils that were collected from the fruit orchards did not contain enough entomogenous fungal distribution. The geographical parameters, similarly to the soil variables described above, greatly influenced the occurrence of entomopathogenic fungi. The coordinates (longitude, latitude) and altitude of all the localities has an impact on the entomopathogenic fungal distribution as the highest number of isolates were found above >600 m altitude, 33°34′ N latitude, and 73°74′ E, longitude (Fig. 2).

Figure 2. Effect of geographical attributes (a, altitude; b, latitude; c, longitude) on the occurrence of entomopathogenic fungi from the soils of crop fields, fruits, vegetables and forests of Punjab, Pakistan. The variables are categorized in groups indicated in the respective legend of the plots. , <200; , 200–400; , 400–600; , >600; , 29–30; , 30–31; , 31–32; , 32–33; , 33–34; , 71–72; , 72–73; , 73–74; , 74–75.

Table 3. Distribution, frequency and pathogenicity (% ± SE) of different isolates of entomopathogenic fungi against Galleria mellonella (F_5,23 = 5.36, P = 0.0034)

Fungal species	Distribution frequency (%)	Total isolates (n)	Mortality (%)	Range of mortality (%)
Beauveria bassiana	36.36	8	89.15 ± 3.21a	80.59–95.38
Beauveria brongniartii	9.09	2	66.60 ± 5.70bc	50.88–75.44
Lecanicillium attenuatum	4.55	1	70.97 ± 5.61abc	57.14–81.48
Metarhizium anisopliae	22.73	5	81.03 ± 3.11ab	72.30–87.07
Paecilomyces lilacinus	18.18	4	75.09 ± 4.04abc	63.25–81.36
Pochonia chlamydosporia	9.09	2	59.12 ± 5.11c	45.61–70.18
Total	100.00	22	–	–

Entomopathogenic fungi are important natural enemies of insect pests but their incidence in the soils of different agro-ecosystems are incompletely understood (Sun et al. 2008). Entomopathogenic fungi are widely distributed in different types of soils, as it is a conventional site for isolating them (Hajek 1997), and are found in a wide range of habitats including aquatic, forest, agriculture and pasture (Chandler et al. 1997). Ali-Shtayeh et al. (2002) found that entomopathogenic fungi were contained in 33.6% of the soil samples studied, with positive samples yielding 70 fungal isolates, belonging to 20 species from 13 genera. Klingen et al. (2002) found frequent occurrence of B. bassiana, F. merismoides, M. anisopliae and Tolypocladium cylindrosporum in a survey of the soils of northern Norway. Among the 590 samples taken in five years from natural and agricultural soils of Finland, B. bassiana was isolated from 19.8%, M. anisopliae from 15.6%, P. farinosus from 92% and P. fumosoroseus from 1% of the samples (Vanninen 1995). Furthermore, the distribution of entomopathogenic fungi is greatly influenced by the geographical location and soil types as the occurrence of M. anisopliae, a southern species in Finland indicates (Vanninen 1995). B. bassiana and P. farinosus were adversely affected by cultivation, but this was not the case for M. anisopliae. P. farinosus occurred throughout Finland, whereas B. bassiana became more common northwards and P. fumosoroseus was never isolated from intensely cultivated soils (Vanninen 1995). Our results are in conformity with Sun and Liu (2008), who isolated insect-associated fungi from 425 soil samples and the occurrence of insect pathogenic fungi was as follows: P. farinosus (19.6%), B. bassiana (14.1%) and M. anisopliae var. anisopliae (10.6%). The results presented by Medo and Cagáň (2011) further complement the present study with only B. bassiana, M. anisopliae, Isaria farinosa and I. fumosorosea detected using G. mellonella bait (GBM) and selective medium (SM) methods from 901 sampling sites during 2008 in Slovakia. Similar results have been reported by Sánchez-Peña et al. (2011) as they used Tenebrio molitor (L.) (Coleoptera: Tenebrionidae) larvae as bait and recovered B. bassiana (20%), M. anisopliae (3%) and Isaria (Paecilomyces) sp. (0.1%) from 171 samples. B. bassiana was significantly more frequent on larvae exposed to oak forest soil while M. anisopliae was more prevalent on larvae exposed to agricultural soils. Soil samples collected from 61 sites from agricultural fields, forests and a Mediterranean shrub (Nerium oleander L.) of the Alicante province of Spain exhibited the distribution of B. bassiana, M. anisopliae and L. lecanii (Asensio et al. 2003).

Sun et al. (2008) detected and identified insect-pathogenic species, such as B. bassiana, M. anisopliae var. anisopliae, and P. fumosoroseus from field crop and orchard soils but their frequency in the two agro-ecosystems differed significantly: B. bassiana and P. fumosoroseus were more frequently found in orchard soils, while M. anisopliae var. anisopliae was more frequent in field crop soils. Our results showed that the occurrence of entomopathogenic fungi was greater in forest soil as compared to other soil types. These findings are in line with that of Bidochka et al. (1998) who reported that B. bassiana was affiliated with shade and uncultivated habitats (forest) and the same fungus also occurred frequently in hedgerow soils at a Danish locality (Meyling & Eilenberg 2006b). M. anisopliae is also regarded as an agricultural species that is commonly found in exposed and regularly disturbed soil environment (Meyling & Eilenberg 2007) and it was not found in the Danish forest ecosystem (Nielsen et al. 2004). The reason of frequent occurrence of entomopathogenic fungi in the forest soils is due to these soils being less disturbed by human activity and cultural practices, such as reduced tillage regimes enhancing B. bassiana levels in the soil (Bing & Lewis 1993; Hummel et al. 2002). Moreover, the cultivation of soils exposes the fungal inocula to extreme biotic factors which destroy the occurrence of these natural enemies of insect pests (Meyling & Eilenberg 2006a). Further the application of agrochemicals, particularly fungicides, in agricultural soils (Klingen & Haukeland 2006) and crop diversification (Hooks & Johnson 2003) adversely effect the buildup of entomopathogenic inoculum in these soils.

Pathogenicity test

The pathogenicity using Koch's postulates of insect-associated fungi revealed that all the isolates were pathogenic to the larvae of G. mellonella. The mortality was significantly different among the isolates (Table 3). The pathogenicity ranged between 45.61 and 95.38% of G. mellonella larvae from the 22 entomopathogenic fungi recovered from the soil samples. B. basssiana proved to be more aggressive providing 89.15% mortality, followed by M. anisopliae with 81.03% mortality, P. lilacinus with 75.09% mortality, L. attenuatum with 70.97% mortality, B. brongniartii with 66.60% mortality and P. chlamydosporia with 59.12% mortality (Table 3). Similar to the present findings, Er et al. (2007) evaluated the pathogenicity of 13 entomopathogenic fungi belonging to the genera Paecilomyces, Tolypocladium, Beauveria, Metarhizium and Lecanicillium (= Verticillium) against fourth instar larvae of Thaumetopoea pityocampa (Schiff.) (Lep.: Thaumatopoeidae) in the laboratory and all the tested isolates were virulent ranging from 16 to 100% mortality of the larvae. Rodríguez et al. (2009) tested 50 B. bassiana (Balsamo) Vuillemin and 48 M. anisopliae (Metschn.) Sorokin isolates on Varroa destructor (Acari: Varroidae) by applying a suspension of 10⁷ conidia mL⁻¹. The isolate of M. anisopliae (Qu-M845) was found most effective which produced 85% mortality. The pathogenicity of M. anisopliae was checked against all the developmental stages of Red Palm Weevil Rhynchophorus ferrugineus which caused 80–100% mortality of both the larvae and adult weevils (Gindin et al. 2006). Likewise, Kivan (2007) applied four fungal isolates of B. bassiana (1912) and one isolate of M. anisopliae (1883) against the adult sunn bug, Eurygaster integriceps Puton, in the laboratory and mortality ranged between 40.0 and 82.5% in B. bassiana, while the mortality was 100.0% in M. anisopliae 8 days post-treatment.

Conclusion

The survey of different sites demonstrated that there is an abundant and wide distribution of insect associated fungi in soils in Pakistan. On the basis of our results, forest soils are much more highly contaminated with entomopathogenic fungi, therefore, more comprehensive surveys are needed to explore these biocontrol agents which can be an integral part of successful IPM programs against insect pests in this region.

Acknowledgments

The authors are greatly indebted to Pakistan Science Foundation, Islamabad, Pakistan for providing financial support.