1. 1 23
Microbial Ecology
ISSN 0095-3628
Volume 66
Number 3
Microb Ecol (2013) 66:608-620
DOI 10.1007/s00248-013-0249-5
Effects on Diversity of Soil Fungal
Community and Fate of an Artificially
Applied Beauveria bassiana Strain Assessed
Through 454 Pyrosequencing
Jacqueline Hirsch, Sandhya Galidevara,
Stephan Strohmeier, K. Uma Devi &
Annette Reineke
2. 1 23
Your article is protected by copyright and all
rights are held exclusively by Springer Science
+Business Media New York. This e-offprint is
for personal use only and shall not be self-
archived in electronic repositories. If you wish
to self-archive your article, please use the
accepted manuscript version for posting on
your own website. You may further deposit
the accepted manuscript version in any
repository, provided it is only made publicly
available 12 months after official publication
or later and provided acknowledgement is
given to the original source of publication
and a link is inserted to the published article
on Springer's website. The link must be
accompanied by the following text: "The final
publication is available at link.springer.com”.
3. SOIL MICROBIOLOGY
Effects on Diversity of Soil Fungal Community and Fate
of an Artificially Applied Beauveria bassiana Strain
Assessed Through 454 Pyrosequencing
Jacqueline Hirsch & Sandhya Galidevara &
Stephan Strohmeier & K. Uma Devi & Annette Reineke
Received: 22 June 2012 /Accepted: 17 May 2013 /Published online: 5 June 2013
# Springer Science+Business Media New York 2013
Abstract The entomopathogenic fungus Beauveria bassi-
ana is widely used as a biological control agent (BCA) for
insect pest control, with fungal propagules being either
incorporated into the potting media or soil or sprayed di-
rectly onto the foliage or soil. To gain a better understanding
of entomopathogenic fungal ecology when applied as a
BCA to the soil environment, a case study using tag-
encoded 454 pyrosequencing of fungal ITS sequences was
performed to assess the fate and potential effect of an
artificially applied B. bassiana strain on the diversity of soil
fungal communities in an agricultural field in India. Results
show that the overall fungal diversity was not influenced by
application of B. bassiana during the 7 weeks of investiga-
tion. Strain-specific microsatellite markers indicated both an
establishment of the applied B. bassiana strain in the treated
plot and its spread to the neighboring nontreated control
plot. These results might be important for proper risk as-
sessment of entomopathogenic fungi-based BCAs.
Introduction
Fungal entomopathogens are used worldwide as microbial
biocontrol agents (BCA) against arthropod pests [1]. Of the
~130 commercially available products based on entomopa-
thogenic fungi, about two thirds consist of conidial prepara-
tions of the two most widely studied entomopathogens: Beau-
veria bassiana (Bals.-Criv.) Vuilleman and Metarhizium ani-
sopliae (Metschn.) Sorokin (both Ascomycota: Hypocreales)
[2–4]. Fungal propagules can be incorporated into the potting
media or soil at the time of planting [5] or sprayed directly
onto the plant or soil. The entomopathogen B. bassiana is
known to infect a wide range of insects [6, 7]. It also exists as
an endophyte inside the plant or as a saprotroph in the soil [8].
While the interactions between entomopathogenic fungi and
their host insects are quite well studied [9, 10], aspects of
fungal ecology regarding putative interactions between the
entomopathogen and the soil microbiota including indigenous
fungal communities have been rarely assessed [11–13]. How-
ever, as soil fungi are involved in many key processes in soil
ecosystem functioning like decomposing organic matter or as
mycorrhizal symbionts of plants [14], any effect exerted by
the application of an entomopathogenic fungus to the structure
and diversity of indigenous fungal communities in the soil
might have important implications for various ecological pro-
cesses and functional soil biodiversity. Consequently, these
aspects should be taken into account during the process of risk
assessment required for registration of the respective entomo-
pathogenic fungal-based commercial product.
In the past, selective media were used to study the impact
of an application of entomopathogens like B. bassiana on soil
microorganisms [11]. Since many soil microorganisms are
fastidious and their morphological determination is often dif-
ficult, cultivation-independent approaches have been applied
subsequently [15, 16]. Different DNA fingerprinting techni-
ques like denaturing gradient gel electrophoresis (DGGE) or
Jacqueline Hirsch and Sandhya Galidevara Both authors contributed
equally to this work.
Electronic supplementary material The online version of this article
(doi:10.1007/s00248-013-0249-5) contains supplementary material,
which is available to authorized users.
J. Hirsch :A. Reineke (*)
Institute of Phytomedicine, Geisenheim University, Von-Lade-Str. 1,
65366 Geisenheim, Germany
e-mail: annette.reineke@hs-gm.de
URL: www.hs-geisenheim.de
S. Galidevara :K. U. Devi
Department of Botany, Andhra University, 530 003
Visakhapatnam, India
S. Strohmeier
SMS-Development, Ortsstr. 6, 69226 Nussloch, Germany
Microb Ecol (2013) 66:608–620
DOI 10.1007/s00248-013-0249-5
Author's personal copy
4. temperature gradient gel electrophoresis (TGGE) and single-
strand conformation polymorphism (SSCP) or traditional
metagenomic approaches with clone library-based techniques
have been used to define community structure of soil micro-
biota [12, 16]. However, these community profiling techni-
ques are time-consuming and costly, especially when taxo-
nomic affiliations of respective organisms are analyzed. In this
context, next-generation sequencing technologies like 454
pyrosequencing represent new, cost-efficient, and fast strate-
gies to depict microbial diversity without the need for cultur-
ing the respective organisms [17]. The internal transcribed
spacer (ITS) regions in the 18S, 5.8S, and 28S ribosomal
RNA gene cluster of fungi have been analyzed successfully
in metagenomic studies [18] and are regarded as validated
DNA barcode markers for the taxonomic classification of
fungi [19–21]. In the present study, we report the application
of multitag 454 pyrosequencing of fungal ITS-1 sequences for
characterizing the fungal community structure in an agricul-
tural field in India and for assessing both the fate and potential
effects of an artificially applied B. bassiana strain on diversity
of soil fungal communities.
Materials and Methods
Study Site, Fungal Treatment, and Sample Collection
Experiments were conducted from October to December
2010 in a cultivated agricultural field near Visakhapatnam
(Andhra Pradesh, India) with a standing crop of 4-week-old
chili plants. Chili was chosen as a crop in this study, as it is
frequently attacked by insect pests like Spodoptera litura
and Helicoverpa armigera (Lepidoptera: Noctuidae) [22]
that are, at the same time, potential targets for the applied
B. bassiana isolate [6, 23]. No naturally occurring entomo-
pathogenic fungal epizootics have been documented in the
selected field over the last 15 years and the last artificial
introduction of Beauveria sp. was done in 1996 (Ramesh
Kongara—cultivator of the field, personal communication).
In addition, no insecticides or fungicides have been applied
in the field over the last 2 years (Ramesh Kongara, personal
communication). The field is fertilized with sun-dried cow
dung. Within the field, two 50-m2
plots ~15 m apart were
selected for the experiment. One of them was treated (T)
once with B. bassiana strain ITCC 4688 (Indian Type Cul-
ture Collection, IARI, Delhi, India) and the other was left as
an untreated control (C). The experimental design was not
replicated on the given plot. Before the application of B.
bassiana to the (T) plot, seven soil cores (approximately 4×
4×15 cm depth) were collected separately every 3 m on a
27-m2
subplot (size of 9×3 m) in each plot. B. bassiana
isolate ITCC 4688 was collected originally from infected
cotton bollworm larvae (H. armigera) in Andhra Pradesh,
India [23]. Mass culture of B. bassiana strain ITCC 4688
was done in the laboratory. The starter culture was initiated
by inoculation of 1 ml of aqueous conidial suspension (107
conidia per milliliter) in 100 ml of Sabouraud's dextrose
yeast (SDY) broth medium, followed by incubation at 25±
2 °C with gentle agitation for 72 h. Subsequently, sterilized
rice bags (with a capacity of 200 g rice with 30 % moisture
and 2 % sunflower oil) were inoculated with 15 ml of a 3-
day-old blastospore suspension. The inoculated rice bags
were incubated for 14 days at 27 °C by which time the
fungus grew and sporulated extensively. Germination of
conidia was analyzed in the laboratory and was found to
be more than 90 % over an incubation time of approximate-
ly 16 h. For application of the fungus to the treated (T) plot,
200 g of rice containing sporulated B. bassiana strain ITCC
4688 was suspended in 30 l water with 2 ml Tween 80 to
give a final concentration of ~109
conidia per milliliter. The
whole suspension was dispensed manually onto the soil and
plants in the (T) plot, resulting in a concentration of approx-
imately 3×1013
conidia per 50 m2
. For assessment of effects
of this B. bassiana strain on indigenous soil fungal commu-
nity structure, soil samples were collected in each plot as
described above at weekly intervals for a duration of
7 weeks. No samples were collected in the sixth week due
to heavy rain fall. For our experiment, we used a cultivated
agricultural field. Accidentally, cow dung slurry flowed
from an adjoining cattle shed into a part of the treatment
plot. Therefore, from the fourth week after B. bassiana
application onwards, only five soil samples were taken from
the unaffected parts of the treatment plot. A total of 92 soil
samples were collected over the whole duration of the
survey from the (C) and (T) plots. The collected soil sam-
ples were transported to the laboratory (distance of ~28 km)
in an ice chest (8 °C) and stored at 4 °C (for a maximum
duration of 48 h) or frozen at −20 °C until further process-
ing. Soil parameters such as pH (6.9), organic matter
(0.74 mg/kg), and clay content (44 %) were determined
commercially (Lotus Granges India Ltd., Visakhapatnam,
India). Rainfall and temperature data for the duration of
the experiment were obtained from the local Mandal Reve-
nue Office (Anandapuram, India) and Cyclone Warning
Centre (Visakhapatnam, India) (Online Resource 1).
DNA Isolation, ITS Amplification, and Pyrosequencing
Soil samples were homogenized independently and genomic
DNA was extracted from each of the 92 samples using
PowerSoil® DNA Isolation Kit (Süd-Laborbedarf GmbH,
Gauting, Germany) according to the manufacturer's instruc-
tions. The variable region of the ITS-1 was amplified with
fungal-specific primers as described by Buée et al. [18],
which were modified for multitag 454 GS-FLX amplicon
pyrosequencing by adding a four-base library “key”
Fate of Beauveria bassiana Strain 609
Author's personal copy
5. sequence (TCAG) and a multiplex identifier (MID) tag
sequence specific to each soil sample. PCR amplifications
were set up in a total volume of 30 μl consisting of 2–6 μl of
undiluted soil DNA (~10–35 ng/μl DNA), 15 pmol primers,
and 15 μl GeNei™ Red Dye PCR Master Mix (2×) (GeNei,
Bangalore, India). The PCR reactions were performed at
94 °C for 4 min, followed by 30 cycles of 30 s at 94 °C,
55 °C for 1 min and 72 °C for 90 s, and a final elongation at
72 °C for 10 min. An aliquot of 4 μl of each amplification
product was analyzed for correct size (~400 bp) on a 1 %
agarose gel. The remaining 26 μl of PCR product was
purified for 454 pyrosequencing analysis with HiYield
PCR Clean-up/Gel Extraction Kit (Süd-Laborbedarf GmbH,
Gauting, Germany). In total, 92 fungal PCR products, tag-
encoded according to sampling date and plot, were pooled at
equimolar concentrations and 454 pyrosequencing was per-
formed commercially on a Roche (454) FLX Genome Se-
quencer (LGC Genomics GmbH, Berlin, Germany).
Sequence Editing and Taxonomy-Dependent Analysis
(MEGAN)
Clipping and sorting of 454 sequence reads by MID tags
was done by LGC Genomics GmbH (Berlin, Germany).
Individual sequences were evaluated using BLASTn
2.2.25+ with word length of 28 against NCBI nt database.
Data was imported to MEGAN version 4.64.2 (MEtaGe-
nome ANalyzer [24]) for similarity-based phylotyping.
Parameters for the Lowest Common Ancestor (LCA) as-
signment algorithm were set as follows: “min support 1”
allowing support of a taxon by a single read, “min score 35,”
“top percent 10,” “win score 0.0,” and “min complexity
0.3.”
Sequence Editing and Analysis by Operational Taxonomic
Unit Clustering
Processing, quality filtering, and clustering 454 reads into
operational taxonomic units (OTUs) was performed using
the online pipeline CLOTU [25]. Sequences with a length of
<150 bp and ambiguous bases as well as barcode and primer
sequences were trimmed by the software. Identical sequen-
ces (duplicates) were removed before clustering to reduce
redundancy in the data set. Sequences were clustered and
assigned to OTUs using the CD-HIT package implemented
in CLOTU with a threshold of 97 % sequence similarity and
with at least 75 % of sequence coverage. For taxonomic
annotation, one representative sequence from each OTU
was submitted to BLASTn for a comparison against the
NCBI nonredundant nucleotide database. All OTUs not
defined as fungi according to BLASTn search were exclud-
ed from further data analysis. Remaining OTUs per soil
sample were used for rarefaction analysis and calculation
of diversity and richness indices as described below.
Statistical Analysis of Pyrosequencing Reads
Rarefaction analysis as well as fungal diversity (Shannon)
and richness (ACE, Chao1) indices at family, species, and
OTU levels were computed using PAST [26] and EstimateS
8.20 [27], respectively. All reads that were not defined as
fungi were excluded from these computations. Singletons
were removed from the data set, except for calculation of
richness estimators as well as for rarefaction analysis. Prior
to the application of B. bassiana, similarities in fungal
diversity between the plots were assessed through a t test
(p value threshold of 0.05) based on Shannon diversity
indices (species and OTU levels) [28]. The effect (if any)
of B. bassiana application on fungal diversity and commu-
nity structure was assessed through a comparison of the
mean of Shannon diversity indices at OTU and species
levels, respectively, of samples collected in the control and
treated plots subsequent to treatment (C1–C7 versus T1–T7)
using a one-way analysis of similarities (ANOSIM) imple-
mented in PAST [26]. ANOSIM creates a test statistic of R,
which indicates if differences between samples exist. Inter-
pretation of R values is according to [29] with the following
categories: separated R>0.75, clearly different R>0.5, and
barely separable R<0.25. Prior to ANOSIM, nonmetric
multidimensional scaling (NMDS) of the samples was done
based on the Bray–Curtis dissimilarity matrix using an
algorithm implemented in PAST [26].
Microsatellite Analysis to Track the Applied B. bassiana
Strain
As the ITS-1 gene region is not suitable for strain-specific
identification of an artificially applied B. bassiana isolate,
three microsatellite (simple sequence repeats, SSR) markers
(Ba01, Ba08, and Ba13; [30]) were used in order to verify
the presence of B. bassiana isolate ITCC 4688 in the re-
spective soil samples. The allele sizes of the these SSR loci
in B. bassiana strain ITCC 4688 were previously deter-
mined as being 121, 260, and 176 bp, respectively (Reineke
et al. in preparation). To allow fluorescent labeling and
multiplexing of the PCR products, a M13(−21) tail was
placed at the 5′ end of each forward primer and a fluores-
cently labeled CY5 or IRD700 universal primer M13(−21)
was added to PCR reactions according to the method de-
scribed by Schuelke [31]. PCR amplifications were set up in
a total volume of 15 μl consisting of 90–100 ng DNA, 10×
reaction buffer with 1.5 mM MgCl2, 5 pmol of each primer,
0.5 μl 100× BSA, 0.2 mM dNTPs, and 0.5 U of Dream Taq
Polymerase (Fermentas, St. Leon-Rot, Germany). PCR reac-
tions were performed at the following conditions: 94 °C for
610 J. Hirsch et al.
Author's personal copy
6. 5 min, followed by 35 cycles of 94 °C for 30 s, 60 °C for
45 s and 72 °C for 45 s, and a final extension step at 72 °C
for 10 min. Each PCR product was checked for successful
amplification on a 2 % agarose gel and analyzed subse-
quently for size of SSR alleles via capillary electrophoresis
in a multiplex analysis on a GenomeLab GeXP DNA Ge-
netic Analysis System (Beckman Coulter GmbH, Krefeld,
Germany).
Results
Analysis of 454 Pyrosequencing Reads Based
on Taxonomy-Dependent Analysis (MEGAN)
In the 92 soil samples analyzed in the present study, a total
of ~63,000 reads were sequenced via 454 pyrosequencing.
After clipping, 29,109 reads were available for analysis with
MEGAN. Most (97 %) of the sequence reads were assigned
and only around 3 % lacked a taxonomic annotation or
showed no hits in MEGAN (Table 1). Of the assigned reads,
69 % were classified as belonging to the kingdom Fungi.
The most dominant phyla were Ascomycota and Basidio-
mycota (Table 2). Excluding singletons, 54 fungal species
were identified by collapsing the phylogenetic tree at spe-
cies level (Table 3). Among them, neither B. bassiana nor
Cordyceps sp. (the genus to which the teleomorph of B.
bassiana belongs) were detected (Table 3). However, when
the tree was collapsed at family level, in the samples col-
lected prior to treatment, four sequences were assigned to
the family Cordycipitaceae in the C0 plot while none were
detected in the T0 plot (Online Resource 2). In the samples
collected after treatment, the number of reads showing ho-
mology to Cordycipitaceae increased both in the control and
treated plots with a massive increase observed in the treated
plot in samples collected 2 weeks after treatment (T2)
(Online resource 2).
Analysis of 454 Pyrosequencing Reads Based on OTU
Clustering
After filtering and trimming of the obtained 63,000 se-
quence reads, 39,263 unique reads were clustered into a
total of 2,227 OTUs (excluding 4,000 singletons) using
software CLOTU (Table 1, Online resource 3, sheet 1).
More than half (1,164, 53 %) of OTUs showed no hits
against the NCBI database and less than 3.6 % (81) of OTUs
were not assigned to fungal taxa (Table 1). In accordance
with results from MEGAN analysis, Ascomycota was the
predominant fungal phylum represented by 75 % of detected
OTUs followed by Basidiomycota with 12 % (Table 2).
A few reads homologous to Cordyceps bassiana (tele-
omorph of B. bassiana) were detected in both the control
and treated plots prior to application of B. bassiana isolate
ITCC 4688 (Table 4, Online resource 3, sheet 2). Subse-
quent to treatment, a massive increase in number of reads
homologous to B. bassiana and Cordyceps spp. was
detected in the treatment plot in samples collected 1 and
2 weeks after treatment (T1 and T2) (Table 4, Online re-
source 3). Furthermore, from the third week onwards, B.
bassiana and Cordyceps spp. were nearly equally repre-
sented in both the control and treatment plots (Table 4,
Online resource 3, sheet 2).
The number of fungal species identified based on OTU
clustering using software CLOTU far exceeded those
detected through taxonomy-dependent analysis using
MEGAN. Species identified through CLOTU were not in-
clusive of all species detected in MEGAN analysis. For
example, reads of Fibulobasidium murrhardtense and Asco-
bolus crenulatus were detected in abundance in MEGAN
analysis, while these species were not found in data analysis
with CLOTU (Tables 3 and 4). On the opposite, B. bassiana
reads were not detected in MEGAN analysis while they
Table 1 Details of ITS-1 454 sequence reads obtained from 92 soil
samples collected at weekly intervals over 7weeks from a chili field in
India. Results of taxonomy-dependent analysis (MEGAN) and OTU-
based clustering (CLOTU) are shown
Number of reads
(MEGAN)
Number of clusters
(CLOTU)
Assigned 28,318 (97.3 %) 982 (44.0 %)
Unassigneda
483 (1.7 %) 81 (3.5 %)
No hits 308 (1 %) 1,164 (52.5 %)
Total 29,109 2,227
a
Reads were not assigned to any group, while clusters were grouped as
unclassified fungi
Table 2 Distribution of species and OTUs representing different fun-
gal phyla as assessed from ITS-1 454 sequence reads excluding single-
tons obtained from 92 soil samples before and after the application of
B. bassiana
Phylum No. species
(MEGAN)
No. OTUs
(CLOTU)
Ascomycota 35 (64.8 %) 796 (74.9 %)
Basidiomycota 9 (16.6 %) 134 (12.6 %)
Unclassified fungi 0 (0 %) 81 (7.6 %)
Blastocladiomycota 3 (5.6 %) 21 (2.0 %)
Fungi incertae sedis 1 (1.9 %) 14 (1.3 %)
Zygomycota 0 (0 %) 4 (0.4 %)
Chytridiomycota 4 (7.4 %) 6 (0.6 %)
Glomeromycota 2 (3.7 %) 7 (0.7 %)
Total 54 1,063
Fate of Beauveria bassiana Strain 611
Author's personal copy
8. were identified through OTU-based clustering (Tables 3 and
4, Online resource 3).
Assessment of Effect of B. bassiana on Soil Fungal
Community Diversity and Richness
Prior to the application of B. bassiana to the treatment plot, no
significant differences in the diversity (Shannon index) of
fungal communities between control and treatment plots were
evident, neither at the species nor at the OTU levels (t=1.96;
df=105 at species and t=0.48; df=1,669 at OTU level,
p<0.05). Samples collected at different time intervals in both
plots (control and treatment) after application of B. bassiana
to the treatment plot fitted consistently to the ordination plot in
NMDS computation (Fig. 1), also indicating the suitability of
pooling of data from each plot (C1 to C7, T1 to T7) for
comparison through ANOSIM. At species and OTU
levels, R values for diversity indices were 0.029 and
0.207, respectively (p<0.05). These values indicate that
the control and treated plots are overlapping or barely
separable regarding their assemblage of fungal commu-
nities and that there is, thus, little or no effect on fungal
community structures due to the artificial application of
B. bassiana to the treated plot.
The Chao1 and ACE mean richness estimates at all three
taxonomic levels (species, family, and OTU) were high
compared to the observed richness (Table 5). The Shannon
diversity indices of all samples ranged from 1.56 to 2.73,
2.37 to 2.99, and 5.16 to 5.69 at species, family, and OTU
levels, respectively (Table 5). Shannon diversity indices
increased by a value of 3 when the calculations were
based on OTU numbers compared to taxonomy-
dependent analysis based on MEGAN results. Rarefaction
curves revealed that the number of species, families, and
OTUs increased with the number of sequences sampled,
with none of the curves reaching saturation (Fig. 2),
indicating that further sampling would have revealed
additional fungal diversity.
Strain-Specific Identification of B. bassiana Strain ITCC
4688 Using SSR Markers
As a strain-specific identification of members of the family
Cordycipitaceae was not possible on the basis of the
obtained ITS sequences, B. bassiana ITCC 4688 strain-
specific SSR markers were amplified from the same soil
DNA samples as used for pyrosequencing. SSR profiles
typical of B. bassiana strain ITCC 4688 were not amplified
prior to the treatment in both the control and treated plots.
However, alleles with the respective size were evident in
soil samples collected in the treated plot starting from the
first week after application until the end of the experiment
7 weeks later (T1 to T7; Table 6). In DNA isolated from the
control plot, a few samples started to show minor peaks of
the respective allele size 2 weeks after B. bassiana ITCC
4688 application (C2), with a more prominent amplification
being evident during the following weeks also in samples
from the control plot (Table 6). SSR marker Ba08 amplified
alleles of the respective size in soil samples from the treat-
ment plot only 1 and 2 weeks after B. bassiana ITCC 4688
strain application, confirming previous observations on a
lower sensitivity of this marker for amplification of respec-
tive sequences from bulk soil DNA samples (Reineke et al.
in preparation). In addition, a few other alleles were
amplified from all samples, mainly in the lower molec-
ular weight range, which are likely the result of cross-
amplification of DNA of other microorganisms present
in the respective samples.
Table 3 (continued)
Fungal species C0 T0 C1 T1 C2 T2 C3 T3 C4 T4 C5 T5 C7 T7 Total
Entorrhiza aff. fineranae 2 0 0 0 3 0 0 0 3 0 0 0 4 0 12
Ganoderma lucidum 0 0 6 0 3 2 0 0 0 0 2 0 0 0 13
Fibulobasidium murrhardtense 12 21 11 16 26 17 17 24 32 117 16 26 18 10 363
Laetisaria arvalis 0 0 0 0 0 0 0 0 0 0 6 0 0 0 6
Pluteus longistriatus 0 0 0 0 0 0 2 0 0 0 0 0 0 0 2
Thanatephorus cucumeris 2 2 0 18 0 0 0 0 3 0 5 0 2 4 36
Tricholoma giganteum 2 0 0 0 0 0 0 0 0 0 0 0 0 0 2
Waitea circinata 0 0 0 0 0 2 0 0 0 0 0 0 0 0 2
Fungi incertae sedis
Endogone lactiflua 0 2 0 0 0 5 0 0 0 0 0 0 0 0 7
Glomeromycota
Glomus intraradices 0 0 0 3 0 0 0 0 0 0 0 3 0 0 6
Glomus mosseae 0 0 0 0 0 4 0 0 0 5 0 0 0 0 9
Fate of Beauveria bassiana Strain 613
Author's personal copy
10. Discussion
We assessed the composition of soil fungal communities via
tag-encoded 454 pyrosequencing to obtain insights on the
effects of artificial application of the entomopathogenic
fungus B. bassiana on indigenous fungal species present
in the soil. The plot chosen for this experiment was an
agricultural field in the tropical savannah climate zone
(Aw zone according to Köppen–Geiger climate classifica-
tion, [32]) of India, which was cultivated according to
conventional small-scale Indian farming standards and was
planted with a standing crop of chili during this experiment.
Conditions of the Aw climate zone are favorable for spread
and establishment of entomopathogenic fungi, and annual
epizootics of these fungi are known to occur regularly [32].
Moreover, we considered it to be important to perform such
a trial under managed conditions with as much practical
relevance for farmers as possible. Accordingly, in this study,
we obtained both a first insight in fungal communities
associated with this type of agricultural practice in the given
geographic location, and we were able to assess the fate and
the dynamics of spread of a fungal biocontrol agent applied
artificially to this field. As there are no true replicates of our
experimental design in the given agricultural field, our study
can be considered as a “proof-of-concept” for these partic-
ular questions.
C. bassiana (a teleomorph of B. bassiana) was detected
albeit in low concentration in both the control and treated
plots prior to artificial application of B. bassiana to the
treatment plot. After application, a remarkable increase in
the number of sequence reads homologous to B. bassiana
and C. bassiana were observed in samples from both the
control and treatment plots, with the highest number of B.
bassiana reads detected in samples from the treated plot
collected 2 weeks after application. Specific SSR markers
were amplified from the same DNA samples to confirm that
these reads belong to the applied B. bassiana strain ITCC
4688. Based on amplification of all three SSR markers, an
SSR profile identical to the applied B. bassiana strain was
observed in the treatment plot only in samples collected 1 and
2 weeks after treatment (T1 and T2). In the samples collected
subsequently in both the treatment and control plots, one (T3
to T5 and C2 to C5) or two (C7 and T7) of the three SSR
markers were not amplified. Template quantity and/or effi-
ciency of amplification of different SSR loci are reported to
affect the results of SSR profiling in samples with high se-
quence diversity [33]. Together, both the results obtained via
pyrosequencing as well as via SSR marker analysis indicate a
natural spread of the applied B. bassiana isolate from the
treatment to the control plot. As water plays an important role
in the movement of fungal pathogens [4, 34–36], we suppose
that several rainfall events from October till December 2010
Table 4 (continued)
Fungal species C0 T0 C1 T1 C2 T2 C3 T3 C4 T4 C5 T5 C7 T7
Fungi incertae sedis
Mortierella spp. 0 3 19 0 0 2 0 2 0 6 0 11 10 17
Mortierella wolfii 3 0 0 0 6 5 23 0 5 9 17 0 24 2
a
All OTUs are homologues to HQ630968.1, which is identified by the authors as Cordyceps bassiana (Online resource 3, sheet 2)
Stress = 0.062(a) Stress = 0.061(b)
Fig. 1 Nonmetric multidimensional scaling (NMDS) plot of OTU (a)-
or species (b)-based clustering of data from the fungal ITS-1 region
obtained from soil samples of a chili field treated with the
entomopathogenic fungus B. bassiana. C1 to C7 and T1 to T7 repre-
sent samples from control and treated plots, respectively, collected at
weekly intervals after treatment
Fate of Beauveria bassiana Strain 615
Author's personal copy
12. (Online Resource 1) may have favored the dispersal of B.
bassiana conidia. In addition, wind, arthropods, and agricul-
tural cultivation practices are effective dispersal mechanisms
of entomopathogenic fungal conidia [37]. We assume that
such factors favored the dispersal of B. bassiana conidia from
the treatment to the control plots and contributed to this
apparent spread of B. bassiana in the field. Such a natural
spread and establishment of B. bassiana is in agreement with
the concept of classical biological control [1]. In classical
biological control, small amounts of inoculum of entomopath-
ogens are intentionally released and are expected to naturally
increase in population density and get permanently estab-
lished. Including fungal entomopathogens in a classical bio-
logical control approach is, for sure, of interest for small-scale
farmers, where such a strategy represents a long-lasting and
cost-efficient avenue for environmentally friendly insect pest
control. However, our molecular approach does not detect
viability and virulence of the B. bassiana fungal propagules
0
5
10
15
20
25
30
35
40
45
50
No. of sequences
No.ofobservedspecies
0
5
10
15
20
25
30
35
40
45
50
1 21 41 61 81 101 121 141 161 181 201 221
1 41 81 121 161 201 241 281 321 361 401 441 481 521 561
No. of sequences
No.ofobservedfamilies
0
250
500
750
1000
1250
1500
3001200110011
No. of sequences
No.ofobservedOTUs
C0 T0 C1 T1 C2 T2 C3 T3 C4 T4 C5
T5 C7 T7
(a)
(b)
(c)
Fig. 2 Rarefaction curves
illustrating observed number of
fungal species (a), families (b),
and OTUs (c) in soil samples
collected from a chili field
treated with Beauveria
bassiana. C control plot, T
treatment plot; numbers refer to
weeks after application of B.
bassiana isolate ITCC 4688 to
the treated plot
Table 6 Amplification of strain-specific alleles of three SSR loci
(Bao1, Ba08, and Ba13) of B. bassiana strain ITCC 4688 in soil
DNA samples from a chili field in India. Presence (+) or absence (−)
of alleles of the expected size is shown. C = control plot and T =
treatment plot, and numbers refer to weeks after application of B.
bassiana isolate ITCC 4688 to the treated plot
SSR marker (size) C0 T0 C1 T1 C2 T2 C3 T3 C4 T4 C5 T5 C7 T7
Ba01 (121 bp) − − − + + + + + + + + + + +
Ba08 (260 bp)) − − − + − + − − − − − − − −
Ba13 (176 bp) − − − + + + + + + + + + − −
Fate of Beauveria bassiana Strain 617
Author's personal copy
13. in the plots. Cultivation-independent methods such as SSR
markers and 454 pyrosequencing will also amplify any DNA
from dead fungal cells or senescent conidia. A combination of
molecular methods and baiting techniques such as the Galle-
ria bait method [38] would help to detect the infectivity of the
applied fungi-based BCA.
A second goal of our study was to address the question
whether an artificial application of a microbial BCA causes
a shift in the diversity of the indigenous fungal community
in the treatment plot either indirectly due to competition for
nutrients or directly due to suppression or antibiosis. During
the 7 weeks of our investigation, no effect of B. bassiana,
applied at a concentration of 3×1013
conidia per 50 m2
, was
evident on the diversity of indigenous fungal communities.
Similar observations were reported by Schwarzenbach et al.
[39] in their assessment of the effect of an application of a B.
brongniartii-based BCA on soil fungal community struc-
tures in a controlled environment (soil microcosms). In this
study [39], only small effects on fungal community struc-
tures were evident and the authors assumed that small
effects caused by fungal BCAs to soil fungal communities
may be difficult to detect in the field due to high ecosystem
variation and fast compensation effects.
Predicted richness estimates at species, family, and OTU
levels were high (more than double at OTU level) compared
to the observed richness, indicating the presence of highly
diverse fungal communities in the selected field. Fungi
belonging to Glomeromycota and Chytridiomycota were
probably underestimated in the present study as the applied
ITS-1 primers originally have been designed for amplifica-
tion of Dikaryotic fungi [18]. Moreover, approximately
31 % of the analyzed reads in MEGAN lacked a deep
taxonomic resolution or were assigned to Cercozoa or Vir-
idiplantae. This indicates a certain nonspecificity of the
applied primers. A large proportion of the ITS-1 sequence
reads in this study were, however, found to have no hits to
database entries or were assigned to unclassified fungi.
In the present study, we used two different approaches of
analyzing our data set, i.e., taxonomy-dependent analysis using
the program MEGAN as well as taxonomy-independent anal-
ysis based on OTU clustering. Both approaches yielded con-
siderable differences regarding fungal species composi-
tion in the respective plots, which can be explained by
the different types of algorithms implemented in both
approaches. Taxonomy-dependent analysis depends on
entries in reference sequence databases such as the NCBI
taxonomy and might, thus, have poor performance if
sequences of many novel or nonculturable organisms are
present in the given data set. Taxonomy-independent algo-
rithms cluster sequences within a data set into OTUs, based on
pairwise identity cutoffs [40, 41]. Accordingly, OTU-based
approaches overcome limitations associated with taxonomy-
dependent analysis, in particular if a given sequence is not
sufficiently represented by a reference sequence in the taxon-
omy outline [42].
The SSR markers chosen for tracking the ITCC 4688
strain of B. bassiana used in this study were found useful
only when the density of the applied strain was considerably
high. For registration purposes of fungal based biocontrol
agents, any risks concerning the persistence of the applied
fungal inoculum have to be evaluated in order to assess the
organism's potential to spread and to become established in
the environment [43]. In addition, registration authorities in
the European Union require information on long-term non-
target effects such as potential competitive displacement of
soil microorganisms as well as information on the natural
background level of a particular entomopathogenic fungus.
Both requirements may be achieved by using multitag 454
pyrosequencing as obtained sequence reads give a compre-
hensive description of the fungal diversity [18, 44], and read
abundance allows a quantification of the applied fungus and
the present soil fungal community with some limitations
[45]. A more detailed insight into the dynamics and inter-
actions of entomopathogenic fungi like B. bassiana with
other microorganisms present in the soil is crucial for a
better understanding of factors influencing fungal survival
and persistence and for estimating success rates of applica-
tions of these organisms for biological insect pest control. A
combination of new molecular methods like 454 pyrose-
quencing and classical approaches like bait methods repre-
sent powerful tools to acquire a more thorough knowledge
on the ecology of fungal entomopathogens.
Acknowledgments We thank Ramesh Kongara for providing the
experimental sites and Ravi Kanth Reddy Sathi, Swapna Guntupalli,
and Suman Keerthi for field and laboratory assistance as well as Martin
Pfannkuchen and Surendra Kumar for advice in bioinformatic analysis.
We are grateful to the German Research Foundation (DFG, project
number: RE 1444/4-1) and the Department of Science and Technology
(DST, project number: INT/FRG/DFG/P-07/2008) New Delhi for fi-
nancial support.
References
1. Hajek AE, Delalibera I (2010) Fungal pathogens as classical bio-
logical control agents against arthropods. BioControl 55:147–158.
doi:10.1007/s10526-009-9253-6
2. de Faria MR, Wraight SP (2007) Mycoinsecticides and mycoacar-
icides: a comprehensive list with worldwide coverage and interna-
tional classification of formulation types. Biol Control 43:237–
256. doi:10.1016/j.biocontrol.2007.08.001
3. Jackson MA, Dunlap CA, Jaronski ST (2010) Ecological consid-
erations in producing and formulating fungal entomopathogens for
use in insect biocontrol. BioControl 55:129–145. doi:10.1007/
s10526-009-9240-y
4. Jaronski ST (2010) Ecological factors in the inundative use of
fungal entomopathogens. BioControl 55:159–185. doi:10.1007/
s10526-009-9248-3
618 J. Hirsch et al.
Author's personal copy
14. 5. Bruck DJ, Donahue KM (2007) Persistence of Metarhizium
anisopliae incorporated into soilless potting media for control
of the black vine weevil, Otiorhynchus sulcatus in container-
grown ornamentals. J Invertebr Pathol 95:146–150. doi:10.1016/
j.jip.2007.01.004
6. Zimmermann G (2007) Review on safety of the entomopathogenic
fungi Beauveria bassiana and Beauveria brongniartii. Biocontrol
Sci Tech 17:553–596. doi:10.1080/09583150701309006
7. Devi KU, Padmavathi J, Rao CUM, Khan AAP, Mohan MC
(2008) A study of host specificity in the entomopathogenic fungus
Beauveria bassiana (Hypocreales, Clavicipitaceae). Biocontrol Sci
Tech 18:975–989. doi:10.1080/09583150802450451
8. Vega FE, Posada F, Aime MC, Pava-Ripoll M, Infante F, Rehner
SA (2008) Entomopathogenic fungal endophytes. Biol Control
46:72–82. doi:10.1016/j.biocontrol.2008.01.008
9. Hajek AE, St Leger RJ (1994) Interactions between fungal patho-
gens and insect hosts. Annu Rev Entomol 39:293–322.
doi:10.1146/annurev.en.39.010194.001453
10. Roy HE, Steinkraus DC, Eilenberg J, Hajek AE, Pell JK (2006)
Bizarre interactions and endgames: entomopathogenic fungi
and their arthropod hosts. Annu Rev Entomol 51:331–357.
doi:10.1146/annurev.ento.51.110104.150941
11. Shimazu M, Maehara N, Sato H (2002) Density dynamics of the
entomopathogenic fungus, Beauveria bassiana Vuillemin
(Deuteromycotina: Hyphomycetes) introduced into forest soil,
and its influence on other soil microorganisms. Appl Entomol
Zool 37:263–269
12. Anderson IC, Cairney JWG (2004) Diversity and ecology of soil
fungal communities: increased understanding through the applica-
tion of molecular techniques. Environ Microbiol 6:769–779.
doi:10.1111/j.1462-2920.2004.00675.x
13. Meyling NV, Hajek AE (2010) Principles from community and
metapopulation ecology: application to fungal entomopathogens.
BioControl 55:39–54. doi:10.1007/s10526-009-9246-5
14. Brussaard L, Behan-Pelletier VM, Bignell DE, Brown VK,
Didden WAM, Folgarait PJ, Fragoso C, Freckman DW,
Gupta VVSR, Hattori T, Hawksworth DL, Klopatek C,
Lavelle P, Walloch D, Rusek J, Söderström B, Tiedje JM,
Virginia RA (1997) Biodiversity and ecosystem functioning
in soil. Ambio 26:563–570
15. Bridge P, Spooner B (2001) Soil fungi: diversity and detection.
Plant Soil 232:147–154
16. Kirk JL, Beaudette LA, Hart M, Moutoglis P, Khironomos JN, Lee
H, Trevors JT (2004) Methods of studying soil microbial diversity. J
Microbiol Methods 58:169–188. doi:10.1016/j.mimet.2004.04.006
17. Harkins T, Jarvie T (2007) Metagenomics analysis using the
Genome Sequencer™ FLX system. Nat Methods 4:6
18. Buée M, Reich M, Murat C, Morin E, Nilsson RH, Uroz S, Martin
F (2009) 454 pyrosequencing analyses of forest soils reveal an
unexpectedly high fungal diversity. New Phytol 184:449–456.
doi:10.1111/j.1469-8137.2009.03003.x
19. Seifert KA (2009) Progress towards DNA barcoding of fungi. Mol
Ecol Resour 9:83–89. doi:10.1111/j.1755-0998.2009.02635.x
20. Begerow D, Nilsson H, Unterseher M, Maier W (2010) Current
state and perspectives of fungal DNA barcoding and rapid identi-
fication procedures. Appl Microbiol Biotechnol 87:99–108.
doi:10.1007/s00253-010-2585-4
21. Schoch CL, Seifert KA, Huhndorf S, Robert V, Spouge JL,
Levesque CA, Chen W, Fungal Barcoding Consortium (2012)
Nuclear ribosomal internal transcribed spacer (ITS) region as a
universal DNA barcode marker for fungi. Proc Natl Acad Sci USA
109:6241–6246. doi:10.1073/pnas.1117018109
22. Tatagar MH, Mohankumar HD, Shivaprasad M, Mesta RK (2009)
Bio-efficacy of flubendiamide 20 WG against chili fruit borers,
Helicoverpa armigera (Hub.) and Spodoptera litura (Fb.).
Karnataka J Agri Sci 22:579–581
23. Devi KU, Mohan CHM, Padmavathi J, Ramesh K (2003)
Susceptibility to fungi of cotton boll worms before and after a
natural epizootic of the entomopathogenic fungus Nomuraea rileyi
(Hyphomycetes). Biocontrol Sci Tech 13:367–371. doi:10.1080/
0958315031000110373
24. Huson DH, Mitra S, Ruscheweyh HJ, Weber N, Schuster
SC (2011) Integrative analysis of environmental sequences
using MEGAN4. Genome Res 21:1552–1560. doi:10.1101/
gr.120618.111
25. Kumar S, Carlsen T, Mevik BH, Enger P, Blaalid R, Tabrizi KS,
Kauserud H (2011) CLOTU: an online application for processing
and clustering of 454 amplicon reads into OTUs followed by
taxonomic annotation. BMC Bioinforma 12:1–9. doi:10.1186/
1471-2105-12-182
26. Hammer Ø, Harper DAT, Ryan PD (2001) PAST: Paleontological
Statistics Software Package for education and data analysis.
Palaeontol Electron 4:9
27. Clowee RK (2006) EstimateS: statistical estimation of species
richness and shared species from samples. Version8, user guide
and application published at http://purl.oclc.org/estimates
28. Hutcheson K (1970) A test for comparing diversities based on the
Shannon formula. J Theor Biol 29:151–154
29. Clarke KR, Warwick RM (2005) Change in marine communities.
An approach to statistical analysis and interpretation, 2nd edn.
Plymouth Marine Laboratory, Plymouth
30. Rehner SA, Buckley EP (2003) Isolation and characterization of
microsatellite loci from the entomopathogenic fungus Beauveria
bassiana (Ascomycota: Hypocreales). Mol Ecol Notes 3:409–411.
doi:10.1046/j.1471-8286.2003.00464
31. Schuelke M (2000) An economic method for the fluorescent la-
beling of PCR fragments. Nat Biotechnol 18:233–234
32. Peel MC, Finlayson BL, McMahon TA (2007) Updated world map
of the Köppen–Geiger climate classification. Hydrol Earth Sys Sc
11:1633–1644
33. Schwarzenbach K, Widmer F, Enkerli J (2007) Cultivation-
independent analysis of fungal genotypes in soil by using simple
sequence repeat markers. Appl Environ Microbiol 73:6519–6525.
doi:10.1128/AEM.01405-07
34. McCoy CW, Storey GK, Tigano-Milani MS (1992) Environmental
factors affecting entomopathogenic fungi in the soil. Pesqui
Agropec Bras 27:107–111
35. Madden LV (1997) Effects of rain on splash dispersal of fungal
pathogens. Can J Plant Pathol 19:225–230. doi:10.1080/
07060669709500557
36. Bruck DJ, Lewis LC (2002) Rainfall and crop residue effects on
soil dispersion and Beauveria bassiana spread to corn. Appl Soil
Ecol 20:183–190
37. Meyling NV, Eilenberg J (2007) Ecology of the entomopathogenic
fungi Beauveria bassiana and Metarhizium anisopliae in temper-
ate agroecosystems: potential for conservation biological control.
Biol Control 43:145–155. doi:10.1016/j.biocontrol.2007.07.007
38. Zimmermann G (1986) The ‘Galleria bait method’ for detection of
entomopathogenic fungi in soil. J Appl Entomol 102:213–215.
doi:10.1111/j.1439-0418.1986.tb00912.x
39. Schwarzenbach K, Enkerli J, Widmer F (2009) Effects of biolog-
ical and chemical insect control agents on fungal community
structures in soil microcosms. Appl Soil Ecol 42:54–62.
doi:10.1016/j.apsoil.2009.02.001
40. Bik HM, Porazinska DL, Creer S, Caporaso JG, Knight R, Thomas
WK (2012) Sequencing our way towards understanding global eu-
karyotic biodiversity. Trends Ecol Evol 27:233–243. doi:10.1016/
j.tree.2011.11.010
41. Schloss PD, Westcott SL (2011) Assessing and improving methods
used in operational taxonomic unit-based approaches for 16S
rRNA gene sequence analysis. Appl Environ Microbiol 77:3219–
3226. doi:10.1128/aem.02810-10
Fate of Beauveria bassiana Strain 619
Author's personal copy
15. 42. Cheng L, Walker AW, Corander J (2012) Bayesian estimation of
bacterial community composition from 454 sequencing data.
Nucleic Acids Res 40:5240–5249. doi:10.1093/nar/gks227
43. Scheepmaker JWA, Butt TM (2010) Natural and released inoculum
levels of entomopathogenic fungal biocontrol agents in soil in relation
to risk assessment and in accordance with EU regulations. Biocontrol
Sci Techn 20:503–552. doi:10.1080/09583150903545035
44. Lim YW, Kim BK, Kim C, Jung HS, Kim BS, Lee JH, Chun
J (2010) Assessment of soil fungal communities using pyro-
sequencing. J Microbiol 48:284–289. doi:10.1007/s12275-010-
9369-5
45. Amend AS, Seifert KA, Bruns TD (2010) Quantifying microbial
communities with 454 pyrosequencing: does read abundance count?
Mol Ecol 19:5555–5565. doi:10.1111/j.1365-294X.2010.04898.x
620 J. Hirsch et al.
Author's personal copy
