Journal of Invertebrate Pathology 111 (2012) 225–236
Contents lists available at SciVerse ScienceDirect
Journal of Invertebrate Pathology
journal homepage: www.elsevier .com/ locate/ j ip
Beauveria brongniartii on white grubs attacking sugarcane in South Africa
T.A. Goble a,c,⇑,3, L. Costet d,4, I. Robene d,4, S. Nibouche d,4, R.S. Rutherford a,1, D.E. Conlong a,b,1,2,M.P. Hill c,3
a South African Sugarcane Research Institute, 170 Flanders Drive, Mount Edgecombe 4300, South Africab School of Biological and Conservation Sciences, University of KwaZulu-Natal, Pietermaritzburg Campus, John Bews Building, Scottsville 3209, South Africac Department of Zoology and Entomology, Rhodes University, P.O. Box 94, Grahamstown 6140, South Africad CIRAD, UMR PVBMT, F-97410 Saint Pierre, Réunion, France
a r t i c l e i n f o
Article history:Received 6 March 2012Accepted 22 August 2012Available online 8 September 2012
Keywords:Sugarcane/wattle agro-ecosystemBiological controlGenetic diversityPhylogenyPopulation geneticsScarabaeidae
0022-2011/$ - see front matter � 2012 Elsevier Inc. Ahttp://dx.doi.org/10.1016/j.jip.2012.08.010
⇑ Corresponding author at: South African SugarcFlanders Drive, Mount Edgecombe 4300, South Africa
E-mail addresses: [email protected] (T.A. G(L. Costet), [email protected] (I. Robene(S. Nibouche), [email protected] (R.S.sugar.org.za (D.E. Conlong), [email protected] (M.P. Hi
1 Fax: +27 31 508 7597.2 Fax: +27 31 260 2029.3 Fax: +27 46 622 8959.4 Fax: +33 262 262 499293.
a b s t r a c t
Beauveria brongniartii (Saccardo) Petch fungal infections were observed on the melolonthid Hypopholissommeri Burmeister (Coleoptera: Scarabaeidae) at two sites (Harden Heights and Canema) in the sugar-cane producing area of the northern KwaZulu-Natal Midlands of South Africa. To initially identify the dis-ease-causing fungus, 17 different fluorescently-labelled microsatellite PCR primers were used to target78 isolates of Beauveria spp. DNA. Microsatellite data resolved two distinct clusters of Beauveria isolateswhich represented the Beauveria bassiana s.s. (Balsamo) Vuillemin (17 isolates) and B. brongniartii (60 iso-lates) species groups. These groupings were supported by two gene regions, the nuclear ribosomal Inter-nal Transcribed Spacer (ITS) and the nuclear Bloc gene of which 23 exemplar Beauveria isolates wererepresented and sequenced. When microsatellite data were analysed, 26 haplotypes among 58 isolatesof B. brongniartii were distinguished. Relatively low levels of genetic diversity were detected in B. bron-gniartii and isolates were shown to be closely related. No genetic differentiation was observed betweenthe Harden Heights and Canema populations; they thus may be considered one, structured and frag-mented population over a distance of 5.5 km. Historically high levels of gene flow from swarming H. som-meri beetles is the proposed mechanism for this observed lack of genetic differentiation betweenpopulations. Microsatellite analyses also showed that B. brongniartii conidia were being cycled from arbo-real forest to subterranean sugarcane habitats and vice versa in the environment by H. sommeri life stages.This is the first record of this species of fungus infecting H. sommeri larvae and adults in South Africa.
� 2012 Elsevier Inc. All rights reserved.
1. Introduction
The South African sugar industry is a major economic contribu-tor to the country, and produces an estimated average of 2.2million tons of sugar per season, 70% of which is marketed in SouthAfrica while the remaining is exported to Africa, Asia and theMiddle East (SASA, 2011). Several damaging insect pests arerecorded from sugarcane in South Africa, which can be divided intothree ecological categories: (1) soil insects (2) sap suckers and leaffeeders and (3) stem borers (Carnegie and Conlong, 1994;
ll rights reserved.
ane Research Institute, 170. Fax: +27 31 508 7597.oble), [email protected]), [email protected]
Rutherford), des.conlong@ll).
McArthur and Leslie, 2004). Of the soil pests associated withsugarcane globally, scarabaeids (Coleoptera: Scarabaeidae) arethe dominant group both in terms of damage and distribution(Wilson, 1969; Hall, 1988; Allsopp et al., 1993; Carnegie andConlong, 1994). The term ‘‘white grub’’ is often used to describethe C-shaped, immature soil dwelling larval stage of scarabaeids.They feed on the roots and underground stools of sugarcane,reducing the vigour, yield and sugar content of the plant (Allsoppet al., 1991). McArthur and Leslie (2004) estimated a 23–55% aver-age reduction in sugarcane yield (tons cane/ha) in the KwaZulu-Natal Midlands North, depending on sugarcane variety, seasonand white grub infestation level. In Florida, USA, yield reductionsin sugarcane of up to 39% have been recorded (Raid and Cherry,1992). Chelvi et al. (2011) report a potential of 80–100% damageto sugarcane by white grubs in India. In Australia, Sallam (2011)quantified the damage caused by greyback canegrub, Dermolepidaalbohirtum Waterhouse (Coleoptera: Scarabaeidae) in monetaryterms of annual losses of up to AUD $10 million.
Sugarcane is grown in three South African provinces, the mostimportant being KwaZulu-Natal, herewith known as KZN (Maloa,
226 T.A. Goble et al. / Journal of Invertebrate Pathology 111 (2012) 225–236
2001). Within this province is a region known as the MidlandsNorth which is characterised by a higher altitude and hence coolertemperatures (frost occurs in winter) compared to the rest of theprovince (Atkinson et al., 1981; Webster et al., 2009). Since theearly 1970s in this region, sugarcane had been grown on land thatwas formerly under black wattle, Acacia mearnsii De Wild cultiva-tion (Carnegie, 1974). Two melolonthid species, Hypopholissommeri Burmeister and Schizonycha affinis Boheman (Coleoptera:Scarabaeidae), were known insect pests of black wattle in the1970s and have now become damaging to sugarcane (Carnegie,1974). H. sommeri in particular, is widely distributed in South Afri-ca, where it damages a variety of cultivated plants. Adults are gen-erally brown in colour and may cause serious defoliation of someplant species but do not attack the aerial parts of sugarcane. Thisinsect species has a 2-year life cycle and annual mass emergencesand swarming take place from mid-October to early Decemberwhen adult beetles feed, mate and oviposit (Carnegie, 1974).
The increased prevalence of H. sommeri in sugarcane in the Mid-lands North of KZN since the 1970s, has raised crop loss concerns,and has increasingly become the subject of research (Way et al.,2011). Way (1997) studied white grub distribution and density lev-els in the Midlands North and recorded an average density of 2white grubs per pit (30 cm � 30 cm � 30 cm). Since the com-mencement of the current study, as many as 20 grubs per pit (per-sonal observation) have been recorded on some farms. In 2005, thefirst incident of entomopathogenic fungal infection of white grubsin this region was observed on Sunnyside Farm (29�06055.500S,30�46059.100E; altitude: 900–1400 m), when many mycosed cadav-ers were dug up from sugarcane fields (Conlong, pers. comm.). Fun-gi are particularly important in the control of Coleoptera becausebacterial and viral pathogens are uncommon (Hajek and St. Legar,1994). This led on to the current study, and in 2010, two morefarms only 10 km away from Sunnyside, namely Harden Heightsand Canema, were found to harbour numerous mycosed larvalcadavers. At first it was suspected that Beauveria bassiana (Balsam-o) Vuillemin (Ascomycota: Hypocreales) was responsible for thedeath of white grubs in this region (Hatting, 2006). However, an-other fungal species, Beauveria brongniartii is well known as anentomopathogen of Coleoptera, being quite specific to the Melo-lonthinae and causing epizootics (Hurpin and Robert, 1972; Keller,1986; Neuveglise et al., 1997; Enkerli et al., 2001). In order to de-fine this mortality factor in the Midland North and to use it as acontrol option, it became important to identify the fungal speciescorrectly.
Molecular techniques, involving the sequencing of specific re-gions within an organisms DNA, are increasingly used to separatespecies of organisms based on their particular sequenced DNA re-gions through the construction of phylogenetic data sets (Rehnerand Buckley, 2005; Meyling, 2008; Rehner et al., 2011). Primerswhich allow the amplification of the elongation factor 1 alpha(EF1a) and Bloc (a nuclear intergenic region) gene regions havebeen developed for Beauveria spp. and these sequences are nowincreasingly used for the delineation of species in this fungal group(Rehner and Buckley, 2005; Meyling et al., 2009; Rehner et al.,2011). Further, microsatellites are powerful tools for the detectionand characterisation of specific fungal strains in the environment(Groppe et al., 1995; Enkerli et al., 2001; Steimel et al., 2004; Bre-uillin et al., 2006). According to Meyling (2008) this ‘fingerprinting’technique has recently been applied to Beauveria collections in Ja-pan, China and the Middle East.
The present study employed two molecular techniques: mul-tilocus gene sequencing and microsatellite markers to ascertainwhich fungal species was causing the infection of H. sommeri inthe Midlands North region. Furthermore, microsatellite markerswere used to investigate the genetic diversity of the most com-monly occurring fungal species found in this region.
2. Methods and materials
2.1. Collection of isolates
A total of 78 Beauveria isolates (Table 1) were considered in thisstudy. Sixty-eight of these isolates were collected from two sites(Canema and Harden Heights) in the Midlands North area of SouthAfrica. The remaining ten Beauveria isolates included in this studycame from other sites in South Africa, as well as an isolate from Re-union Island (Table 1). The Midlands North area is characterised bylarge tracks of land which are a mosaic of intensively cultivatedsugarcane and black wattle plantations. At the two field sites, fun-gal isolates were collected from infected scarabaeid larvae, pupaeand adults dug up from sugarcane fields. A standard industry-based method for sampling white grubs was applied (Way et al.,2011). Where infected larvae not showing signs of overt mycosiswere found, raster patterns on the anus were investigated to con-firm H. sommeri species identity (Sweeney, 1967). In addition, deadH. sommeri beetles attached to vegetation (mainly black wattle) atfield edges were collected at random. Thirdly, soil samples col-lected randomly from both sites were baited with the greaterwax moth, Galleria mellonella (Lepidoptera: Pyralidae) to obtainBeauveria isolates (Zimmermann, 1986). These bait-soil samplesdid not overlap with collected insects or dug pits at either siteand were evenly spaced throughout fields. Lastly, random andevenly spaced sampling of various plant surfaces (Table 1), partic-ularly the leaves of black wattle at the field edge was undertaken toisolate Beauveria spp. A leaf press was made onto an agar plate con-taining full-strength Sabouraud dextrose agar (SDA) (Strasser et al.,1996) supplemented with 50 mg/l chloramphenicol, 25 mg/l cyclo-heximide and 50 mg/l rifampicin. When white fungal colonieswere observed they were carefully removed and sub-cultured onfresh SDA plates. All Beauveria isolates were maintained on full-strength SDA supplemented with the above mentioned antibioticsand kept at 23 �C. Subsequently, single conidial preparations weremade. Fungal isolates are currently housed in the South African Na-tional Collection of Fungi at the Plant Protection Research Institute(Agricultural Research Council) in Pretoria, South Africa.
2.2. DNA extraction
Conidia from fungal isolates were inoculated into flasks con-taining a sterilized liquid growth medium (3% sucrose, 2% peptoneand 0.2% yeast extract) and placed on a rotary shaker at 180 rpmfor 72 h. The resulting mycelial mats were recovered by filtrationand stored at �80 �C for DNA extraction. Mycelia were ground witha mortar and pestle in liquid nitrogen until a fine powder re-mained. Ground mycelia (1.5 g) were re-suspended in extractionbuffer (12.5 ml) consisting of 200 mM Tris–HCl (pH 8.5), 250 mMNaCl, 25 mM EDTA and 0.5% SDS, after which phenol (pH 7.9)(8.75 ml) preheated to 60 �C was added followed by the additionof chloroform/isoamylalcohol [24:1 (v/v)] (3.75 ml). After centrifu-gation (18,900g) in a Beckman J2-21 centrifuge for 60 min at 4 �Cthe top liquid phase containing the DNA was removed. RNA wasremoved from the liquid phase through the addition of 500 ll(5 mg/ml) RNase H and incubation for 20 min at 37 �C. One partphenol was added to the mixture after incubation with the RNaseand the mixture was again centrifuged (18,900g for 20 min) at 4 �C.The liquid phase was removed and the DNA was precipitated with1:1 volumes of isopropanol followed by centrifugation. The result-ing pellet was washed with 70% (v/v) ethanol, recentrifuged, afterwhich the ethanol was aspirated and the pellet was dried. Thepellet containing the isolated DNA was then dissolved in 1 ml TEbuffer [10 mM Tris (pH 7.8) and 1 mM EDTA] and stored at�20 �C for further manipulation.
T.A. Goble et al. / Journal of Invertebrate Pathology 111 (2012) 225–236 227
2.3. PCR amplification and sequencing
The complete data set for phylogenetic analyses totalled 37 iso-lates of Beauveria and one out-group, Isaria (Ascomycota: Cordyci-pitaceae) Twenty-three Beauveria isolates found in KZN, SouthAfrica (eight B. bassiana s.s isolates and 15 B. brongniartii isolates)and the remaining isolates included 15 published, Bloc and ITS se-quences, respectively from type-strain Beauveria species (Rehneret al., 2011). For the 23 South African isolates, two gene regionswere partially amplified and sequenced: Bloc, an intergenic regionoften used as a phylogenetic marker for the genus Beauveria (Reh-ner et al., 2006) and the ribosomal internal transcribed spacer (ITS).The Bloc gene region was amplified with primers (B5.1F, B3.1R andB22U, B822L) obtained from Rehner et al. (2006). PCR amplifica-tions were performed in a total volume of 20 ll, which included10 ll of 2 � AmpliTaq Gold PCR master mix (DNA polymerase250 U (0.05 U/ll) in buffer (30 mM Tris/HCl (pH 8), 100 mM KCl;dNTPs 400 lM each; MgCl2 5 mM) (Applied Biosystems, Johannes-burg); 2 ll Q-solution (Qiagen, Johannesburg); 1 ll of (10 lM) for-ward primer; 1 ll of (10 lM) reverse primer and 1 ll genomic DNA(30 ng) the rest was made up with water. PCR was performed usingthe following thermocycling regime: denaturation at 95 �C for5 min; 33� cycles (94 �C for 30 s, 58 �C for 90 s, 72 �C for 90 s)and a final elongation step of 72 �C for 30 min. The ITS gene wasamplified and sequenced with primers ITS5 and ITS4 accordingto the specifications given in Rehner and Buckley (2005). Sequenc-ing of both forward and reverse strands of both genes was per-formed using the BigDye� Terminator v3.1 Cycle Sequencing Kit(Applied Biosystems) using 0.5 ll BigDye, 2 ll of 5 � reaction buf-fer, 2 ll primer, 30 ng of DNA template and 4.5 ll of water in a to-tal volume of 10 ll. Sequence clean up was undertaken usingBigDye X Terminator Purification Kit (Applied Biosystems) accord-ing to the manufacturer’s instructions. Sequencing took place usingan ABI 3500 Genetic Analyser. Genbank accession numbers forthese sequences are provided in Table 1.
2.4. Sequence editing and phylogenetic analyses
Unrefined sequence chromatograms were assembled and editedwith Geneious™ 5.4 Pro (Drummond et al., 2011). Multiple se-quence alignments of edited sequences were created with MAFFT(Katoh et al., 2005; Katoh and Toh, 2008) using the FFT-NS-i align-ment option. Maximum parsimony (MP) analyses were performedon individual and concatenated datasets. MP and nonparametricMP bootstrapping analyses were undertaken with PAUP�4.0b10(Swofford, 2003) with all uninformative sites and gaps excluded,under equal character weighing, using a heuristic search optionwith TBR branch swapping, and executing 1000 random-additionreplicate analyses. Heuristic MP bootstrap analyses were per-formed to assist clade support with 1000 bootstrap replicates. Todetermine whether different data partitions could be joined andanalysed together, a partition homogeneity test (PHT) was per-formed. Data were found to be congruent (>0.05) and the twogenes (Bloc and ITS) were analysed together. To determine theselection of nucleotide substitution model, implementing theAkaike information criterion (AIC) for Maximum likelihood (ML)and Bayesian (BI) analyses, jModelTest 0.1.1 (Posada, 2008) wasused. The two-gene data set was partitioned, the substitution mod-els for each partition were as follows: Bloc (TVM + G) and ITS(TIM2 + G) and the combined data set (GTR + G). ML searches and1000 ML bootstrap analyses were conducted in PAUP�4.0b10. Par-titioned Bayesian analyses were carried out in MrBayes 3.1.2(Huelsenbeck and Ronquist, 2001) and started from a random treeusing the program’s default settings for the posterior probabilitieswhich consisted of four chains, one cold and three hot and ran for3 � 106 generations. A single tree was sampled ever 100th
generation. Plots of likelihood scores, tree length and averagedSD of split frequencies against number of generations showed thatstationarity was reached within the first 10% of trees generated.Thus, the first 10% of trees generated were discarded using theburn-in function ensuring only trees generated at the stationaryphase were included to calculate the posterior probabilities.
2.5. Microsatellite DNA amplification and fragment analysis
Initially, 17 fluorescently-labelled microsatellite PCR primersobtained from Enkerli et al. (2001) (Bb1F4, Bb2A3, Bb8D6,Bb10D4); Rehner and Buckley (2003) (Ba13); Meyling et al.(2009) (Ba06, Ba08, Ba12, Ba15, Ba17, Ba18, Ba21, Ba22, Ba25,Ba27, Ba28, Ba29) were used to target 78 isolates of Beauveriaspp. DNA (Table 1). It was unclear whether the epizootics werecaused by B. bassiana or B. brongniartii so primers that would beable to discriminate both species were selected. A neighbour-join-ing analysis using DARwin™ 5.0 (Perrier and Jacquemoud-Collet,2006) was undertaken.
Secondly, the genetic diversity of 58 B. brongniartii isolates ob-tained from two sites/populations: Harden Heights (35 isolates)and Canema (23 isolates), using the 17 microsatellite loci describedabove was undertaken. Two B. brongniartii isolates were not in-cluded due to poor quality of the chromatograms for some loci.PCR amplification reactions were performed in 15 ll reaction vol-umes with 1 ll (25 ng) of genomic DNA. 7.5 ll of 2 � Qiagen multi-plexing PCR master mix (Hotstart� DNA Polymerase, buffer anddNTP mix), 1.5 ll Q-solution and 1.5 ll of primer mixtures with fluo-rescent labelling either: VIC, PET, NED, 6-FAM were also added to thereaction (Applied Biosystems). PCR thermocycling parameters wereas follows: 95 �C for 15 min, 33 � cycles (94 �C for 30 s, 58 �C for90 s, 72 �C for 90 s) and a final elongation step of 72 �C for 30 min.After amplification 1 ll of PCR product was diluted 1/150 using dis-tilled water. 1 ll of diluted PCR product was then added to 10.7 ll offormamide and 0.3 ll of molecular weight ladder GS 500 LIZ (Ap-plied Biosystems). Mixtures were denatured at 95 �C for 2 min andplaced directly onto ice. Samples were then loaded into an ABI3500 Genetic Analyser. Resulting fragment analysis was undertakenusing GeneMapper™ 3.1 software (Applied Biosystems).
2.6. Microsatellite population genetic analyses
Population genetic diversity in terms of allelic frequencies (Na),the number of effective alleles (Ne), mean gene diversity (I) andunbiased diversity (h) by population was undertaken on microsat-ellite data with GenAlEx™6.3 (Peakall and Smouse, 2006). Thenumber of effective alleles (Ne) measures the number of equallyfrequent alleles it would take to achieve a given level of gene diver-sity and allows us to compare populations where the number anddistributions of alleles are drastically different. The information in-dex (I) quantifies the expected value of information at loci. Theprogram also identifies and estimates the number of microsatellitehaplotypes and their frequencies (no. of observations) within pop-ulations. Genetic discontinuities between samples from differenthosts (G. mellonella, scarabaeid larvae and H. sommeri adults) orsites/populations (Canema, Harden Heights) were quantifiedthrough analysis of molecular variance (AMOVA) using GenAl-Ex™6.3. Non-random associations of alleles among pairs of loci(linkage disequilibrium) were tested using two methods: gameticdisequilibrium among pairs of loci was tested within each popula-tion using Fisher’s exact test implemented in Genepop™4.0 and anindex of multilocus linkage disequilibrium (rd) was estimatedwithin each population using Multilocus™2.2 (Agapow and Burt,2001). Genic differentiation (exact G test) among populations andestimated spatial structure (F-statistic) for all populations wasdetermined using Genepop™4.0 (Raymond and Rousset, 1995).
Table 1The locality, substrates and host plants of B. bassiana and B. brongniartii used for sequencing and microsatellite analyses in this study.
Strainname
Organism BlocGenbank
ITSGenbank
Locality Co-ordinates Host species/substrate Host plant Dateisolated
BB1319(BT 126)
Beauveria spp. NS NS Reunion Island No data No data No data No data
BB444 Beauveria bassiana JX110345 JX110368 Clanwilliam, Western Cape No data Soil - Galleria mellonella Aspalathus linearis No dataC2 Beauveria bassiana JX110348 JX110371 Canema Farm, Seven Oaks, KZN Midlands North 29�12032.800S
30�39055.50EHypopholis sommeri L3 Sugarcane (Saccharum,
Poaceae)06/07/2010
C5 Beauveriabrongniartii
NS NS Canema Farm, Seven Oaks, KZN Midlands North 29�12033.300S30�39052.10E
Soil – Galleria mellonella Sugarcane (Saccharum,Poaceae)
28/07/2010
C12 Beauveriabrongniartii
JX110346 JX110369 Canema Farm, Seven Oaks, KZN Midlands North 29�12034.300S30�39046.40E
Soil – Galleria mellonella Sugarcane (Saccharum,Poaceae)
28/07/2010
C13 Beauveriabrongniartii
JX110347 JX110370 Canema Farm, Seven Oaks, KZN Midlands North 29�12034.300S30�39046.40E
Soil – Galleria mellonella Sugarcane (Saccharum,Poaceae)
28/07/2010
C17 Beauveriabrongniartii
NS NS Canema Farm, Seven Oaks, KZN Midlands North 29�12033.500S30�39050.50E
Soil – Galleria mellonella Sugarcane (Saccharum,Poaceae)
28/07/2010
C34 Beauveriabrongniartii
NS NS Canema Farm, Seven Oaks, KZN Midlands North 29�12033.700S30�39048.50E
Soil – Galleria mellonella Sugarcane (Saccharum,Poaceae)
28/07/2010
C36 Beauveriabrongniartii
NS NS Canema Farm, Seven Oaks, KZN Midlands North 29�12035.500S30�39048.00E
Soil – Galleria mellonella Sugarcane (Saccharum,Poaceae)
28/07/2010
C37 Beauveriabrongniartii
JX110349 JX110372 Canema Farm, Seven Oaks, KZN Midlands North 29�12035.500S30�39048.00E
Soil – Galleria mellonella Sugarcane (Saccharum,Poaceae)
28/07/2010
C38 Beauveriabrongniartii
NS NS Canema Farm, Seven Oaks, KZN Midlands North 29�12033.700S30�39047.60E
Soil – Galleria mellonella Sugarcane (Saccharum,Poaceae)
28/07/2010
C39 Beauveriabrongniartii
NS NS Canema Farm, Seven Oaks, KZN Midlands North 29�12033.500S30�39048.60E
Soil – Galleria mellonella Sugarcane (Saccharum,Poaceae)
28/07/2010
C40 Beauveriabrongniartii
NS NS Canema Farm, Seven Oaks, KZN Midlands North 29�12033.600S30�39049.40E
Soil – Galleria mellonella Sugarcane (Saccharum,Poaceae)
28/07/2010
C40A Beauveriabrongniartii
NS NS Canema Farm, Seven Oaks, KZN Midlands North 29�12033.600S30�39049.40E
Soil – Galleria mellonella Sugarcane (Saccharum,Poaceae)
28/07/2010
CFE3B Beauveriabrongniartii
NS NS Canema Farm, Seven Oaks, KZN Midlands North 29�12009.400S30�38030.50E
Hypopholis sommeribeetle
Acacia mearnsii (Fabaceae) 08/04/2011
CFE5A Beauveriabrongniartii
NS NS Canema Farm, Seven Oaks, KZN Midlands North 29�12008.700S30�38031.20E
Hypopholis sommeribeetle
Acacia mearnsii (Fabaceae) 08/04/2011
CFE5E Beauveriabrongniartii
NS NS Canema Farm, Seven Oaks, KZN Midlands North 29�12009.700S30�38030.20E
Hypopholis sommeribeetle
Acacia mearnsii (Fabaceae) 08/04/2011
CFE5F Beauveriabrongniartii
NS NS Canema Farm, Seven Oaks, KZN Midlands North 29�12008.700S30�38031.20E
Hypopholis sommeribeetle
Acacia mearnsii (Fabaceae) 08/04/2011
C5ALB Beauveriabrongniartii
NS NS Canema Farm, Seven Oaks, KZN Midlands North 29�12034.100S30�38000.20E
Unknown beetle Acacia mearnsii (Fabaceae) 08/04/2011
CFE6A Beauveriabrongniartii
NS NS Canema Farm, Seven Oaks, KZN Midlands North 29�12009.200S30�38030.80E
Hypopholis sommeribeetle
Acacia mearnsii (Fabaceae) 08/04/2011
CFE3 Beauveriabrongniartii
NS NS Canema Farm, Seven Oaks, KZN Midlands North 29�12038.200S30�8011.40E
Hypopholis sommeribeetle
Acacia mearnsii (Fabaceae) 08/04/2011
CFE4 Beauveriabrongniartii
NS NS Canema Farm, Seven Oaks, KZN Midlands North 29�12038.200S30�38011.40E
Hypopholis sommeribeetle
Acacia mearnsii (Fabaceae) 08/04/2011
CFE13 Beauveriabrongniartii
JX110350 JX110373 Canema Farm, Seven Oaks, KZN Midlands North 29�12038.200S30�38011.40E
Hypopholis sommeribeetle
Acacia mearnsii (Fabaceae) 08/04/2011
CFE17 Beauveriabrongniartii
JX110351 JX110374 Canema Farm, Seven Oaks, KZN Midlands North 29�12038.200S30�38011.40E
Hypopholis sommeribeetle
Acacia mearnsii (Fabaceae) 08/04/2011
CFE18 Beauveriabrongniartii
NS NS Canema Farm, Seven Oaks, KZN Midlands North 29�12038.200S30�38011.40E
Hypopholis sommeribeetle
Acacia mearnsii (Fabaceae) 08/04/2011
CFE20 Beauveriabrongniartii
JX110352 JX110375 Canema Farm, Seven Oaks, KZN Midlands North 29�12038.200
S30�38011.40EHypopholis sommeribeetle
Acacia mearnsii (Fabaceae) 08/04/2011
HH2 Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14055.200S30�37054.50E
Unidentified larva Sugarcane (Saccharum,POACEAE)
21/12/2010
HH110 Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14054.900S30�37055.10E
Hypopholis sommeri larva Sugarcane (Saccharum,Poaceae)
08/02/2011
228T.A
.Goble
etal./Journal
ofInvertebrate
Pathology111
(2012)225–
236
HH112 Beauveriabrongniartii
JX110354 JX110377 Harden Heights, Dalton, KZN Midlands North 29�14055.100S30�37054.70E
unidentified larva Sugarcane (Saccharum,Poaceae)
06/12/2010
HH114 Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14054.100S30�37053.50E
Unidentified larva Sugarcane (Saccharum,Poaceae)
06/12/2010
HH115 Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14053.700S30�37053.40E
Hypopholis sommeri lar Sugarcane (Saccharum,Poaceae)
30/07/2010
HH121 Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14055.400S30�37055.00E
Unidentified larva Sugarcane (Saccharum,Poaceae)
06/12/2010
HH186 Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14052.000S30�37052.60E
Unidentified larva Sugarcane (Saccharum,Poaceae)
30/07/2010
HH219 Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14051.300S30�37052.60E
Unidentified larva Sugarcane (Saccharum,Poaceae)
30/07/2010
HH343 Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14055.700S30�37056.30E
Unidentified larva Sugarcane (Saccharum,Poaceae)
08/02/2011
HH56 Beauveriabrongniartii
JX110355 JX110378 Harden Heights, Dalton, KZN Midlands North 29�14054.100S30�37053.50E
Hypopholis sommeri lar Sugarcane (Saccharum,Poaceae)
30/07/2010
HHD30 Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14054.700S30�37055.10E
Hypopholis sommeri lar Sugarcane (Saccharum,Poaceae)
21/12/2010
HHE37 Beauveriabrongniartii
JX110360 JX110383 Harden Heights, Dalton, KZN Midlands North 29�14053.000S30�37054.70E
Unidentified larva Sugarcane (Saccharum,Poaceae)
21/12/2010
HHE32 Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14055.000S30�37055.60E
unidentified larva Sugarcane (Saccharum,Poaceae)
21/12/2010
HHWG1 Beauveriabrongniartii
JX110362 JX110385 Harden Heights, Dalton, KZN Midlands North 29�14054.200S30�37052.60E
Hypopholis sommeri lar Sugarcane (Saccharum,Poaceae)
30/07/2010
HHWG3 Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14053.800S30�37053.00E
Hypopholis sommeri lar Sugarcane (Saccharum,Poaceae)
30/07/2010
HHWT 1 Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14059.000S30�37059.90E
- Galleria mellonella Acacia mearnsii (Fabaceae) 07/03/2011
HHB1A Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14056.800S30�37055.500E
Soil – Galleria mellonel Sugarcane (Saccharum,Poaceae)
22/02/2011
HHB4A Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14055.600S30�37055.000E
Soil – Galleria mellonel Sugarcane (Saccharum,Poaceae)
22/02/2011
HHB13D Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14055.700S30�37055.400E
Soil – Galleria mellonel Sugarcane (Saccharum,Poaceae)
22/02/2011
HH B14B Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14055.900S30�37055.500E
Soil – Galleria mellonel Sugarcane (Saccharum,Poaceae)
22/02/2011
HH B16B Beauveria bassiana NS NS Harden Heights, Dalton, KZN Midlands North 29�14056.900S30�37055.800E
Soil – Galleria mellonel Sugarcane (Saccharum,Poaceae)
22/02/2011
HH B28A Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14055.500S30�37056.200E
Soil – Galleria mellonel Sugarcane (Saccharum,Poaceae)
22/02/2011
HH B29A Beauveriabrongniartii
JX110356 JX110379 Harden Heights, Dalton, KZN Midlands North 29�14055.800S30�37056.300E
Soil – Galleria mellonel Sugarcane (Saccharum,Poaceae)
22/02/2011
HH B29B Beauveriabrongniartii
JX110357 JX110380 Harden Heights, Dalton, KZN Midlands North 29�14055.800S30�37056.300E
Soil – Galleria mellonel Sugarcane (Saccharum,Poaceae)
22/02/2011
HH B30B Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14056.400S30�37056.500E
Soil – Galleria mellonel Sugarcane (Saccharum,Poaceae)
22/02/2011
HHB31E Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14056.800S30�37056.600E
Soil – Galleria mellonel Sugarcane (Saccharum,Poaceae)
22/02/2011
HH B32B Beauveriabrongniartii
JX110358 JX110381 Harden Heights, Dalton, KZN Midlands North 29�14057.100S30�37056.700E
Soil – Galleria mellonel Sugarcane (Saccharum,Poaceae)
22/02/2011
HH B32E Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14057.100S30�37056.700E
Soil – Galleria mellonel Sugarcane (Saccharum,Poaceae)
22/02/2011
HH B32F Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14057.100S30�37056.700E
Soil – Galleria mellonel Sugarcane (Saccharum,Poaceae)
22/02/2011
HH B37B Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14055.500S30�37056.200E
Soil – Galleria mellonel Sugarcane (Saccharum,Poaceae)
22/02/2011
HH B39A Beauveriabrongniartii
JX110359 JX110382 Harden Heights, Dalton, KZN Midlands North 29�14054.700S30�37056.000E
Soil – Galleria mellonel Sugarcane (Saccharum,Poaceae)
22/02/2011
(continued on next page)
T.A.G
obleet
al./Journalof
InvertebratePathology
111(2012)
225–236
229
va
va
va
va
va
la
la
la
la
la
la
la
la
la
la
la
la
la
la
la
Table 1 (continued)
Strainname
Organism BlocGenbank
ITSGenbank
Locality Co-ordinates Host species/substr e Host plant Dateisolated
HH BWL1 Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14055.300S30�37 59.100E
Black wattle leaf pr t Acacia mearnsii (Fabaceae) 08/02/2011
HH BWL2 Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14055.300S30�37 59.100E
Black wattle leaf pr t Acacia mearnsii (Fabaceae) 08/03/2011
HH BWL4 Beauveria bassiana NS NS Harden Heights, Dalton, KZN Midlands North 29�15004.600S30�38001.100E
Black wattle leaf pr t Acacia mearnsii (Fabaceae) 08/03/2011
HH BWL5 Beauveria bassiana NS NS Harden Heights, Dalton, KZN Midlands North 29�14052.800S30�37 59.100E
Black wattle leaf pr s Acacia mearnsii (Fabaceae) 08/02/2011
HH BWR1 Beauveria bassiana JX110353 JX110376 Harden Heights, Dalton, KZN Midlands North 29�14054.200S30�37 58.900E
Black wattle root pr s Acacia mearnsii (Fabaceae) 08/02/2011
HHGR1 Beauveria bassiana JX110361 JX110384 Harden Heights, Dalton, KZN Midlands North 29�14055.300S30�37 59.100E
Grass root print Grass sp. (Poaceae) 08/02/2011
HHGR2 Beauveria bassiana NS NS Harden Heights, Dalton, KZN Midlands North 29�14055.300S30�37 59.100E
Grass root print Grass sp. (Poaceae) 08/03/2011
HHGR3 Beauveria bassiana NS NS Harden Heights, Dalton, KZN Midlands North 29�15004.600S30�38001.100E
Grass leaf print Grass sp. (Poaceae) 08/03/2011
HHSC1 Beauveria bassiana NS NS Harden Heights, Dalton, KZN Midlands North 29�15004.600S30�38001.100E
Sugarcane leaf pres sugarcane (Saccharum, Poaceae) 08/03/2011
HHFE3 Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14055.300S30�37 59.100E
Hypopholis sommeribeetle
Grass sp. (Poaceae) 08/03/2011
HHFE6 Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14056.600S30�37 59.100E
Hypopholis sommeribeetle
Grass sp. (Poaceae) 08/03/2011
HHFE7 Beauveriabrongniartii
NS NS Harden Heights, Dalton, KZN Midlands North 29�14059.500S30�38000.200E
Hypopholis sommeribeetle
Acacia mearnsii (Fabaceae) 08/03/2011
HHFE8 Beauveria bassiana NS NS Harden Heights, Dalton, KZN Midlands North 29�15000.900S30�38000.500E
Hypopholis sommeribeetle
Acacia mearnsii (Fabaceae) 08/02/2011
SCHA1 Beauveria bassiana NS NS Harden Heights, Dalton, KZN Midlands North 29�17000.500S30�39024.000E
Schizonycha affinis b tle not applicable (light trap) 04/11/2010
SCHA2 Beauveriabrongniartii
JX110365 JX110388 Harden Heights, Dalton, KZN Midlands North 29�17000.500S30�39024.000E
Schizonycha affinis b tle Not applicable (light trap) 04/11/2010
S4222 Beauveria bassiana JX110364 JX110387 Sunnyside Farm, Seven Oaks, KZN MidlandsNorth
No data Hypopholis sommeri 3 Sugarcane (Saccharum,Poaceae)
28/05/2007
S1174 Beauveria bassiana JX110363 JX110386 Sunnyside Farm, Seven Oaks, KZN MidlandsNorth
No data Eldana saccharinacadaver
Sugarcane (Saccharum,Poaceae)
24/10/2003
S124 Beauveriabrongniartii
NS NS Sunnyside Farm, Seven Oaks, KZN MidlandsNorth
29�06025.200S30�46052.800E
Hypopholis sommeri upa Sugarcane (Saccharum,Poaceae)
04/02/2010
Sun A Beauveria bassiana NS NS Sunnyside Farm, Seven Oaks, KZN MidlandsNorth
29�06055.500S30�46059.100E
Hypopholis sommeri upa Sugarcane (Saccharum,Poaceae)
26/02/2010
SunB Beauveria bassiana JX110366 JX110389 Sunnyside Farm, Seven Oaks, KZN MidlandsNorth
29�06054.200S30�46058.900E
Hypopholis sommeri upa Sugarcane (Saccharum,Poaceae)
26/02/2010
Woodburn Beauveria bassiana JX110367 JX110390 Ixopo, KZN South Coast No data Scale insects(Diaspididae)
Citrus (Rutaceae) 02/02/2011
230T.A
.Goble
etal./Journal
ofInvertebrate
Pathology111
(2012)225–
236
at
in
in
in
es
es
s
ee
ee
L
p
p
p
Fig. 1. Unweighted, neighbour-joining tree in axial view, rooted on an isolate obtained from Reunion Island (BB1319-BT126), showing two distinct groupings of Beauveriaisolates obtained from two sites in the KwaZulu-Natal Midlands North. Bootstrap values >80% are shown.
T.A. Goble et al. / Journal of Invertebrate Pathology 111 (2012) 225–236 231
The within locus G test were combined in a global v2-test, usingthe Fisher method (Fisher, 1938), carried out with Genepop. Con-sidering that the existence of linkage disequilibrium among pairof loci could violate the independence assumption of the Fishermethod, we also compared haplotype frequencies between bothpopulations using a Fisher’s exact test carried out with SAS procFreq (SAS Institute, 2008).
3. Results
An unweighted neighbour-joining tree (30,000 bootstraps repli-cations) based on polymorphic microsatellite loci only and rootedon a Beauveria isolate BB1319 (BT126) from Reunion Island, re-vealed two distinct species groupings, supported by high bootstrapvalues (Fig. 1). Of the 77 Beauveria isolates included as the ingroup,17 were found to group within the smaller microsatellite (SSR)group 2 (22%), which was later confirmed as the B. bassiana s.sclade A; and the remaining 60 isolates (78%) fell within the largerSSR group 1 which represented a closely-related B. brongniartiiclade C (Fig. 2). It was the distinct separation of the data intotwo groups which prompted sequencing to determine the taxon-omy which is discussed below.
3.1. Molecular sequencing
The complete data set for phylogenetic analyses totalled 37 iso-lates of Beauveria. The two loci: Bloc and ITS consisted of 1308 and532 aligned positions, respectively. When MP was considered, gap-ped or uninformative sites were excluded, rendering a final dataset with 303 and 40 parsimony informative characters for Blocand ITS respectively. The results of the analysis produced one most
parsimonious tree with a length of 934 steps (CI: 0.525, RI: 0.779).ML analysis, under the GTR + G model, rendered a log-likelihoodscore of 9836.74 for the best tree. Finally Bayesian analyses con-verged on a stationary phase that yielded topologically identical50% majority rule consensus trees. A phylogeny of Beauveria illus-trating species relationships inferred from joint Bayesian Inference(BI) analysis of Bloc and ITS gene sequences is shown in Fig. 2.Bootstrap values from MP and ML analyses, as well as the posteriorprobabilities from Bayesian analyses, are provided (Fig. 2).
All three analyses (MP, ML and BI) yielded trees with similar treetopologies that each resolved three well-supported clades, A–C(Fig. 2). Poor replication and inclusion of the basal taxa left theremaining clades indistinguishable and unsupported. Eight of theSA isolates (those not labelled with ARSEF numbers), which initiallygroup into the smaller SSR group 2 (Fig. 1), fell within the commonB. bassiana s.s clade A in the phylogeny (Fig. 2). Clade A was rela-tively diverse as seen by the deep internal branching and groupingof isolates, despite sampling localities being close together for someisolates (Fig. 2). Beauveria kipukae and Beauveria varroae formed abiphyletic clade B that was sister to B. bassiana in this study. Fifteenof the SA isolates (no ARSEF number), which grouped into the largerSSR group 1 (Fig. 1) fell into the B. brongniartii clade C and were clo-sely related as observed by very short branch lengths (Fig. 2). Beau-veria australis and Beauveria asiatica were basal to B. brongniartii inthis study with good bootstrap and posterior probability support.Beauveria caledonica and Beauveria vermiconia grouped togetherwith well-supported bootstrap and posterior probability support.Sister to this grouping was B. sungii with no support. Unfortunately,due to poor sample inclusion, an accurate placement of the remain-ing basal Beauveria taxa: B. pseudobassiana; Beauveria amorpha andBeauveria malawiensis were not observed (Fig. 2).
Fig. 2. Phylogeny of Beauveria showing species relationships of South African isolates within the global Beauveria framework inferred from joint Bayesian Inference (BI)analysis of Bloc and ITS gene sequences. The bootstrap values above internal branches correspond to Maximum Parsimony (>65), ML (>65) and Bayesian Inference posteriorprobabilities (>95) respectively.
232 T.A. Goble et al. / Journal of Invertebrate Pathology 111 (2012) 225–236
3.2. Microsatellite genotyping
B. brongniartii isolates showed relatively low levels of geneticdiversity in this study (Table 2). Twelve of the 17 microsatellite lociused were polymorphic. Loci (Ba06, Ba08, Ba12, Ba18 and Ba21) re-vealed no genetic variation among isolates. In contrast, microsatel-lite markers Ba13, Ba22, Ba25 and especially Ba29 showed highlevels of polymorphism, despite private alleles being thoroughlyrechecked and despite them originally being cloned from B. bassi-ana. The percentage of polymorphic loci from Canema (47.06%)was lower than Harden Heights (70.59%) when rare alleles wereconsidered (Table 2). Low levels of polymorphism were seen inthe average number of alleles (Na) per locus which varied from2.29 to 2.24 per site/population, respectively and a low mean
expected heterozygosity was observed between Canema (0.25)and Harden Heights (0.24) (Table 2). Further, there was littlegenetic difference between the sites when effective alleles wereconsidered: Canema (1.71) and Harden Heights (1.63).
AMOVA analyses showed that the total genetic differentiationamong populations was 0.014 (PhiPT), and was not significant(P = 0.180) when based on permutation across the full data set, ofwhich 99% (df = 56; sum of squares = 1115.77; variance = 19.925)of the genetic variation is contained within each population, whileonly 1% (df = 1; sum of squares = 27.89; variance = 0.287) occursamong populations (table not shown). The global fixation indexwas FST = 0.03 indicating low genetic differentiation between thetwo sites/populations. However, when exact tests (G test) for pop-ulation differentiation were performed looking at the distribution
Table 2The number of alleles and their frequencies, as well as the diversity of 17 microsatellite loci from two South African populations of B. brongniartii.
Population Locus Na Nab Nec Id he
Canema BA18 23 1 1.00 0.00 0.00Canema BA15 23 1 1.00 0.00 0.00Canema BA25 23 2 1.63 0.57 0.39Canema BA17 23 1 1.00 0.00 0.00Canema BA22 23 3 2.63 1.02 0.62Canema BA21 23 1 1.00 0.00 0.00Canema BA29 23 3(1) 2.99 1.21 0.67Canema BA27 23 1 1.00 0.00 0.00Canema BA08 23 1 1.00 0.00 0.00Canema BA06 23 1 1.00 0.00 0.00Canema BA12 23 1 1.00 0.00 0.00Canema BA28 23 1 1.00 0.00 0.00Canema BA13 23 3(1) 2.46 1.08 0.59Canema Bb8D6 23 2 1.63 0.57 0.39Canema BblF4 23 5(3) 5.24 1.83 0.81Canema Bb2A3 23 2(3) 2.37 1.14 0.58Canema BblOD4 23 1(1) 1.19 0.30 0.16Mean values 2.29 ± 0.48 1.71 ± 0.28 0.45 ± 0.14 0.25 ± 0.07Harden Heights BA18 35 1 1.00 0.00 0.00Harden Heights BA15 35 1(1) 1.06 0.13 0.06Harden Heights BA25 35 2 1.82 0.64 0.45Harden Heights BA17 35 1(1) 1.06 0.13 0.06Harden Heights BA22 35 3 1.95 0.76 0.49Harden Heights BA21 35 1 1.00 0.00 0.00Harden Heights BA29 35 2(1) 2.09 0.80 0.52Harden Heights BA27 35 1(1) 1.12 0.22 0.11Harden Heights BA08 35 1 1.00 0.00 0.00Harden Heights BA06 35 1 1.00 0.00 0.00Harden Heights BA12 35 1 1.00 0.00 0.00Harden Heights BA28 35 1(1) 1.06 0.13 0.06Harden Heights BA13 35 3(2) 4.15 1.49 0.76Harden Heights Bb8D6 35 2 1.69 0.60 0.41Harden Heights BblF4 35 5 4.21 1.47 0.76Harden Heights Bb2A3 35 2(1) 1.42 0.57 0.30Harden Heights BblOD4 35 1(1) 1.06 0.13 0.06Mean values 2.24 ± 0.30 1.63 ± 0.25 0.42 ± 0.12 0.24 ± 0.07
a Number of single-conidial strains analysed.b Number of alleles per locus (incl. private alleles) p < 0.05.c Number of effective alleles – it measures the number of alleles at a locus and the equality of the allele frequencies at that locus.d Information index/same as Shannon–Weaver Index of ecology-measure of diversy.e Haploid genetic diversity-provides an indication that two individuals will be different (Nei, 1987).
T.A. Goble et al. / Journal of Invertebrate Pathology 111 (2012) 225–236 233
of alleles for all populations there was a highly significant differ-ence in allele frequencies between the populations (v2 = 53.26;df = 24; P < 0.001).
Linkage disequilibrium was significant (non-random) fornumerous pairs of loci within sites/ populations: for Canema 14out of 28 possible tests were significant. There were fewer pairsof loci showing significant linkage disequilibrium at HardenHeights where 17 out of 66 possible tests were significant. Themultilocus linkage disequilibrium index was significantly differentfrom zero in both populations. It was highest at Canema(rd = 0.196) and lower at Harden Heights (rd = 0.141). These resultsindicate that both Canema and Harden Heights have non-recom-bining population structures. The comparison of haplotype fre-quencies between both populations (Fisher’s exact test)concluded that there was no significant difference (P = 0.09), whichis consistent with the AMOVA and low FST observed. These resultssuggest that Canema and Harden Heights represent the same,structured population of B. brongniarii but the population has beenfragmented over time and space.
The highest number of alleles and private alleles came fromH. sommeri adult beetles (Na = 0.43 (private alleles = 0.22) whileboth G. mellonella (Na = 0.28(0.09) and scarabaeid larvae(Na = 0.27 (0.09) had much lower allele numbers and privatealleles. In the AMOVA analyses, the total differentiation amongpopulations was 0.099 (PhiPT), and was significant (P = 0.017)when based on permutation across the full data set, of which
90% (df = 52; sum of squares = 976.42; variance = 18.777) of thegenetic variation is contained within each population, while only10% (df = 2; sum of squares = 110.84; variance = 2.067) occurs be-tween populations (table not shown). The result indicates no ge-netic structure amongst the populations, and suggests there aresignificant differences in allele frequencies between the sourcetypes with low levels of gene flow.
Our study revealed 26 haplotypes among 58 strains ofB. brongniartii. In the Harden Heights population 12 unique haplo-types were detected, some of them represented by only one strain.Nine unique haplotypes were observed at Canema, some of themalso with only one strain. More microsatellite haplotypes wereobserved at Harden Heights (17) than at Canema (14) (Fig. 3). Fivehaplotypes (H23, H5, H10, H7 and H3) were common to bothpopulations however the frequency of occurrence of a particularhaplotype was different in both populations (Fig. 3). The mostcommon haplotype, H23 was observed ten times and occurred inthe same proportion in both populations (Fig. 3 and Table 3) whensample size was averaged. In the Harden Heights population thesecond most common haplotype was H5 which was observed sixtimes, this haplotype occurred with less frequency at Canemawhere it was observed only twice (Fig. 3 and Table 3).
Some grouped haplotypes shared a common H. sommeri adulthost (Table 3). Besides these host groupings, no other structuralpatterns could be observed, even when sequence data, ecology ortaxonomy was considered. A B. brongniartii haplotype (H6), shared
Fig. 3. The geographic location and distribution of microsatellite haplotypes of two KwaZulu-Natal populations of B. brongniartii analysed in this study.
Table 3The number of observed haplotypes, the isolates representing a particular haplotype and the host insect for each isolate obtained from two B. brongniartii populations.
Population Haplotypes Nobsa Isolateb Host
Canema 1 1 C FE5Ed H. sommeri adultd
Canema 2 1 C17 G. mellonellaCanema and Harden
Heights3 2 C12andHHB28A G mellonella only
Harden Heights 4 1 HHB30B G. mellonellaCanema and Harden
Heights5 8 C5; C38; HH186e; HH121e; HH115e; HHB32B; HHB32E; HHB32F G. mellonella, scarab larvae
Harden Heights 6 4 HH 56c; HH219e; HHFE3c; HHBWL1c H. sommeri adult, scarab larvae, leafpress
Canema and HardenHeights
7 4 C40; HHWT1; HHWG1e; HH110e G. mellonella, scarab larvae
Harden Heights 8 1 HHE37e Scarab larvaeHarden Heights 9 1 HHE32e Scarab larvaeCanema and Harden
Heights10 6 C34; C FEH3d; C FEH17d; C40A; HHB3IE; HHB39A G mellonella, H. sommeri adult
Canema 11 1 CFE H4d H. sommeri adultHarden Heights 12 1 HHFE6d H. sommeri adultCanema 13 1 C FEH18d H. sommeri adultCanema 14 1 C FE5Ad H. sommeri adultHarden Heights 15 1 HHB37B G. mellonellaCanema 16 1 C FEH20d H. sommeri adultCanema 17 2 C5ALBd;CFEH13d H. sommeri adultHarden Heights 18 1 HHB4A G. mellonellaCanema 19 1 C FE5Fd H. sommeri adultHarden Heights 20 1 HH112e Scarab larvaeHarden Heights 21 3 HH114e; HHBWL2; HHB29B Scarab larvae, leaf press, G mellonellaCanema 22 2 C13;CFE6Ad G. mellonella, H. sommeri adultCanema and Harden
Heights23 10 C36; C37; C39; CFE3Be; HHWG3e; HH2e; HH343e; HHB13D; HHB14B;
HHB29AG. mellonella, H. sommeri adult, scarablarvae
Harden Heights 24 1 HHFE7d H. sommeri adultHarden Heights 25 1 HHD30e Scarab larvaeHarden Heights 26 1 HHB1A G. mellonella
a Number of times when a haplotype observed in a population.b B. brongniartii isolates with a particular haplotype.c B. brongniartii isolates which imply fungi are cycled in the environment.d Isolate obtained from a Hypopholis sommeri adult.e Isolate obtained from scarab larvae.
234 T.A. Goble et al. / Journal of Invertebrate Pathology 111 (2012) 225–236
T.A. Goble et al. / Journal of Invertebrate Pathology 111 (2012) 225–236 235
three genetically identical strains/isolates (HH 56; HHFE3 andHHBWL1 (HH219 was not considered) (Table 3). HHFE3 wasisolated from a H. sommeri adult found hanging off a black wattleleaf in a stand adjacent to the Harden Heights sugarcane field,the same strain was found on a black wattle leaf surface (HHBWL1)and finally the same strain was found within the sugarcane fieldfrom an infected H. sommeri L3 larva (HH56) (Table 3). These re-sults indicated the B. brongniartii was cycled from arboreal foresthabitat to the subterranean sugarcane environment by the species,H. sommeri.
4. Discussion
Microsatellite analyses and multilocus gene sequence data re-vealed that the incidence of fungal entomopathogen infectionoccurring in H. sommeri in the Midlands North area of KZN in SouthAfrica is as a result of B. brongniartii. Isolates were shown to be clo-sely related to each other in this study however genetic diversitywas relatively low compared to other studies (Enkerli et al.,2001). B. brongniartii found at the two sites: Canema and HardenHeights can be considered one, structured, clonal population frag-mented over a spatial scale of 5.5 km. Perhaps the geographicalscale this study was undertaken on was too small and a greater ge-netic diversity may have been observed over a greater distance.Despite low overall genetic differentiation however, within sitesmany unique haplotypes were observed represented by only onestrain. As Enkerli et al. (2001) suggested ‘the number of identicalstrains per haplotype were low, indicating that the search for exist-ing haplotypes in the populations was not saturated’. This meansthe biodiversity of B. brongniartii in the KZN Midlands North regionmay be higher than this study indicates. Further, if more B. brongni-artii-specific microsatellite markers were used, a better discrimi-nating power (possibility a greater genetic diversity) could havebeen observed than what was seen in this study.
It may be possible that these small, statistically insignificant ge-netic differences between the two populations may be as a result ofthe inflow of private alleles carried by adult H. sommeri from othersources/areas, as observed by the significant differences in allelefrequencies when different host sources were considered. Currentgene flow appears to be very low between the sites according tothe AMOVA analysis (by host), but historically high levels of geneflow must have occurred between Canema and Harden Heights toresult in such low overall genetic differentiation. Gene flow, as a re-sult of swarming H. sommeri adults, which have been observedmany times in the Midlands North region (personal observation),provides a mechanism for the observed occurrence of common al-leles, as Enkerli et al. (2001) suggested. These authors suggestedtwo possible mechanisms which could explain the occurrence ofcommon alleles. They either co-evolved independently from eachother through mutation or they are identical by descent and distrib-uted through gene flow, as B. brongniartii can efficiently be dissem-inated by swarming cockchafers (Enkerli et al., 2001). The latter issupported by this study, as strong evidence for gene flow is pro-vided when we consider that some B. brongniartii strains, whichwere isolated from adult H. sommeri, also belong to the two mostcommon haplotypes observed in both populations. Significant link-age disequilibrium indicated that both Canema and Harden Heightshave non-recombining population structures. The clonality of B.brongniartii was strengthened when a partition homogeneity test(PHT) was performed (Xu, 2006) and data was found to be congru-ent when Bloc and ITS gene sequences were considered. This sug-gests a slower in-situ mutation mechanism (because randomrecombination and ultimately the sharing of genetic material donot take place) and strengthens the idea that the spread of geneticvariation is through gene flow via H. sommeri adults.
Our study showed that B. brongniartii is also being cycled in theenvironment as evidenced by the common haplotype (H6) whichwas found as three genetically identical strains obtained from ablack wattle leaf surface, within a H. sommeri adult and on a myco-sed H. sommeri larva. Meyling and Eilenberg (2007), in discussingthe life cycle of B. bassiana (which was inferred from collectedknowledge on the ecology of these fungi in temperate environ-ments); suggest that fungal inoculum is cycled from arboreal hab-itats to subterranean environments via insect hosts. SwarmingH. sommeri adults come into contact with B. brongniartii conidiaon the surfaces of black wattle leaves when they are roosting, feed-ing or mating in trees. These adults become infected with the fun-gus and then fly to sugarcane fields were they lay their eggs. Whenthe adults die in the soil due to fungal infection, they disseminatethe fungal strain into the field where conidia are then able to infectlarvae living in the soil and feeding on roots. Furthermore, adultsemerging from pupation may come into contact with B. brongniartiiconidia in the soil as they move through to the soil surface, and be-come infected. As death due to infection may take a few days,adults fly into adjacent black wattle stands to feed and mate; theythen die on the plant surfaces and disseminate the conidia onto theleaf surfaces. We propose this process is cyclic and works in bothdirections.
Lastly, sequencing was also able to confirm a common mistakewhich is made while evaluating the G. mellonella-bait methodwhich is often employed in the isolation of entomopathogenic fun-gi from subterranean environments (Zimmermann, 1986). One ofthe perceived limitations of the G. mellonella-bait method is theinability to distinguish between morphologically identical fungalisolates in the same sample plot, which has implications for diver-sity studies (Goble et al., 2010). This study highlighted that morethan one genetically distinct isolate can occur in the same sampleplot. Sequencing and microsatellite data showed that B. brongniart-ii isolates HHB29A and HHB29B, which were morphologicallyindistinguishable and found in the same sample plot, where in factgenetically quite dissimilar and did not share a haplotype (Fig. 2and Table 3). However, other B. brongniartii isolates (HHB32B,HHB32E and HHB32F) which were all found in the same sampleplot did share the same haplotype (Table 3). This information high-lights the importance of using molecular techniques to accuratelyidentify Beauveria species for future diversity studies, becausemorphology is subjective and simply not precise enough.
This study has broadened our knowledge of the occurrence,host range, genetic structure and ecology of B. brongniartii in theKZN Midlands North region of South Africa. To our best knowledgethis is the first record of B. brongniartii attacking the larvae, pupaeand adults of H. sommeri in South Africa. Future work will investi-gate the use of adults as disseminators of B. brongniartii as a biolog-ical control strategy in both sugarcane and black wattle. Further,obtained isolates of the fungus will be tested against H. sommeriand S. affinis to identify the strain most appropriate to formulateas a biological control agent.
Acknowledgments
We would like to thank Justin Hatting and Mike Morris for theircontinued help and support in this project. Thank you to Tom Web-ster for his discovery of the Harden Heights site as well as his con-tinued help with field work. Very special thank you goes to LaurenMartin, Robyn Jacobs and Debbie Sweby for their invaluable helpwith the molecular sequencing analyses. Our sincere thanks go toPhanor Montoya-Maya. We would also like to thank the ladies atthe Plant Protection Research Institute (PPRI) Mariette Truter, Ria-na Jacobs, Elna van der Linde and Grace Kwinda. Thank you toKeith McFarlane, Nanda Govender and the technical team at SASRI.
236 T.A. Goble et al. / Journal of Invertebrate Pathology 111 (2012) 225–236
Finally thanks go to Angela Walton, Nelson Moonsamy and theguys in the Insect Unit.
References
Agapow, P.M., Burt, A., 2001. Indices of multilocus linkage disequilibrium. Mol. Ecol.Notes 1, 101–102.
Allsopp, P.G., Bull., R.M., McGill, N.G., 1991. Effect of Antitrogus consanguineus(Blackburn) (Coleoptera: Scarabaeidae) infestations on sugar-cane yield inAustralia. Crop Prot. 10, 205–208.
Allsopp, P.G., Chandler, K.J., Samson, P.R., Story, P.G., 1993. Pests of AustralianSugarcane. Bureau of Sugar Experiment Stations, Brisbane, Australia.
Atkinson, P.R., Carnegie, A.J., Smaill, R.J., 1981. A history of the outbreaks of Eldanasaccharina Walker in Natal. Proc. S. Afr. Sugar Technol. Assoc. 55, 111–115.
Breuillin, F., Dutech, C., Robin, C., 2006. Genetic diversity of the Chestnut blightfungus Cryphonectria parasitica in four French populations assessed bymicrosatellite markers. Mycol. Res. 110, 288–296.
Carnegie, A.J.M., 1974. Sugarcane white grubs (Scarabaeoidea) and their control inSouth Africa. Proc. Intern. Soc. Sugar Technol. 15, 498–512.
Carnegie, A.J.M., Conlong, D.E., 1994. Biology, pest status and control measurerelationships of sugarcane insect pests. In: Proceedings of the Second SugarcaneEntomology Workshop, South Africa.
Chelvi, C.T., Thilagaraj, W.R., Nalini, R., 2011. Field efficacy of formulations ofmicrobial insecticide Metarhizium anisopliae (Hyphocreales: Clavicipitaceae) forthe control of sugarcane white grub Holotrichia serrata F (Coleoptera:Scarabidae). J. Biopesticides. 4 (2), 186–189.
Drummond, A.J., Ashton, B., Buxton, S., Cheung, M., Cooper, A., Duran, C., Field, M.,Heled, J., Kearse, M., Markowitz, S., Moir, R., Stones-Havas, S., Sturrock, S.,Thierer, T., Wilson, A., 2011. Geneious v5.4. <http://www.geneious.com/>.
Enkerli, J., Widmer, G., Gessler, C., Keller, S., 2001. Strain-specific microsatellitemarkers in the entomopathogenic fungus Beauveria brongniartii. Mycol. Res. 105(9), 1079–1087.
Fisher, R.A., 1938. Statistical Methods for Research Workers. Oliver and Boyd,Edinburgh, UK.
Goble, T.A., Dames, J.F., Hill, M.P., Moore, S.D., 2010. The effects of farming system,habitat type and bait type on the isolation of entomopathogenic fungi fromcitrus soils in the Eastern Cape Province, South Africa. Biocontrol 55, 399–412.
Groppe, K., Sanders, I., Wiemken, A., Boller, T., 1995. A microsatellite marker forstudying the ecology and diversity of fungal endophytes (Epichloe spp.) ingrasses. Appl. Environ. Microb. 61 (11), 3943–3949.
Hajek, A.E., St. Leger, R.J., 1994. Interactions between fungal pathogens and insecthosts. Annu. Rev. Entomol. 39, 293–322.
Hall, D., 1988. Insects and mites associated with sugarcane in Florida. FloridaEntomol. 71, 138–150.
Hatting, J., 2006. Final project report on microbial control of white grubs andsugarcane borer. South African Sugarcane Research Institute Internal Report.
Huelsenbeck, J., Ronquist, F., 2001. MrBayes: Bayesian inference of phylogenetictrees. Bioinformatics 17, 754–755.
Hurpin, B., Robert, P.H., 1972. Comparison of the activity of certain pathogens of thecockchafer Melolontha melolontha in plots of natural meadowland. J. Invertebr.Pathol. 19, 291–298.
Katoh, K., Toh, H., 2008. Recent developments in the MAFFT multiple sequencealignment program. Brief. Bioinform. 9, 276–285.
Katoh, K., Kuma, K., Toh, H., Miyata, T., 2005. MAFFT version 5: improvement inaccuracy of multiple sequence alignment (describes [ancestral versions of] theG-INS-i, L-INS-i and E-INS-i strategies). Nucl. Acids Res. 33, 511–518.
Keller, S., 1986. Quantitative ecological evaluation of the May beetle pathogen,Beauveria brongniartii, and its practical application. In: Samson, R.A., Vlak, J.M.,Peters, D. (Eds.), Fundamental and Applied Aspects of Invertebrate Pathology,Wageningen, pp. 175–181.
Maloa, M.B., 2001. Sugarcane: a case as development crop in South Africa. In: SouthAfrican Regional Poverty Network (SARPN) Conference On Land Reform andPoverty Alleviation in Southern Africa. Pretoria, South Africa.
McArthur, D.G., Leslie, G., 2004. Preliminary observations on the impact of whitegrub on sugarcane yields in the Midlands north region of the South Africansugar industry. Proc. S. Sugar Technol. Assoc. 75, 283–286.
Meyling, N.V., 2008. PCR-based methods for characterisation of entomopathogenicfungi. Deliverable 5.2 from VegQure, Department of Agriculture and EcologyFaculty of Life Sciences, University of Copenhagen, Denmark, pp. 1–14.
Meyling, N., Eilenberg, J., 2007. Ecology of the entomopathogenic fungi Beauveriabassiana and Metarhizium anisopliae in temperate agroecosystems: potential forconservation biological control. Biol. Control 43, 145–155.
Nei, M., 1987. Molecular Evolutionary Genetics. Columbia University Press, NewYork.
Neuveglise, C., Brygoo, Y., Riba, G., 1997. 28S rDNA group-I introns: a powerful toolfor identifying strains of Beauveria brongniartii. Mol. Ecol. 6, 373–381.
Peakall, R., Smouse, P.E., 2006. GENALEX 6: genetic analysis in Excel. Populationgenetic software for teaching and research. Mol. Ecol. Notes 6, 288–295.
Perrier, X., Jacquemoud-Collet, J.P., 2006. DARwin Software Program. <http://darwin.cirad.fr/>.
Posada, D., 2008. JModelTest: phylogenetic model averaging. Mol. Biol. Evol. 25 (7),1253–1256.
Raid, R.N., Cherry, R.H., 1992. Pathogenicity of Metarhizium anisopliae var. major(Metschnikoft) Sorokin to a Sugarcane Grub Ligyrus subtropicus (Blatchley)(Coleoptera: Scarabaeidae). Agric. Entomol. 9 (1), 11–16.
Raymond, M., Rousset, F., 1995. GENEPOP (version 1.2): population geneticssoftware for exact tests and ecumenicism. J. Heredity 86, 248–249.
Rehner, S.A., Buckley, E.P., 2003. Isolation and characterization of microsatellite locifor the entomopathogenic fungus Beauveria bassiana (Ascomycota:Hypocreales). Mol. Ecol. 3 (3), 409–411.
Rehner, S.A., Buckley, E.P., 2005. A Beauveria phylogeny inferred from nuclear ITSand EF1-a sequences: evidence for cryptic diversification and links to Cordycepsteleomorphs. Mycologia 97, 84–98.
Rehner, S.A., Posada, F., Buckley, E.P., Infante, F., Castillo, A., Vega, F.E., 2006.Phylogenetic origins of African and Neotropical Beauveria bassiana s.l. pathogensof the coffee berry borer, Hypothenemus hampei. J. Invert. Pathol. 93, 11–21.
Rehner, S.A., Minnis, A.M., Sung, G.-H., Luangsa-ard, J.J., Devotto, L., Humber, R.A.,2011. Phylogeny and systematics of the anamorphic entomopathogenic genusBeauveria. Mycologia 103 (5), 1055–1073.
Sallam, N., 2011. Review of current knowledge on the population dynamics ofDermolepida albohirtum (Waterhouse) (Coleoptera: Scarabaeidae). Aust. J.Entomol. 50, 300–308.
SAS Institute, 2008. SAS OnlineDoc� 9.2. SAS Institute Inc., Cary, NC, USA.SASA, 2011. The South African Sugar Industry Directory 2011/2012. South African
Sugarcane Association. Mount Edgecombe, Durban. <http://www.sasa.org.za>.Steimel, J., Engelbrecht, C.J.B., Harrington, T.C., 2004. Development and
characterization of microsatellite markers for the fungus Ceratocystisfimbriata. Mol. Ecol. Notes 4, 215–218.
Strasser, H., Forer, A., Schinner, F. 1996. Development of media for the selectiveisolation and maintenance of virulence of Beauveria brongniartii. In: Jackson,T.A., Glare, T.R. (Eds.), Microbial Control of Soil Dwelling Pests. AgResearch,Lincoln, New Zealand, pp. 125–130.
Sweeney, C., 1967. The Scarabaeoidea associated with sugar-cane in Swaziland. AnAccount of Preliminary Investigations into the Bionomics and Control.Swaziland Minist Agric Res Bull No. 16.
Swofford D.L., 2003. PAUP�: Phylogenetic Analysis using Parsimony (and othermethods) Version 4.0b10. Sinauer Associates, Sunderland, Massachusetts.
Way, M.J., 1997. Surveying white grub larvae (Scarabaeidae) in the South Africansugar industry. Proc. S. Afr. Sugar Technol. Assoc. 71, 95–96.
Way, M.J., Webster, T., Sewpersad, C., Conlong, D.E., 2011. Monitoring white grubbeetle larvae (Scarabaeidae) in sugarcane in the midlands north region ofKwaZulu-Natal. Proc. S. Afr. Sugar Technol. Assoc. 84, 314–325.
Webster, T.M., Brenchley, P.G., Conlong, D.E., 2009. Progress of the area-wideintegrated pest management plan for Eldana saccharina Walker (Lepidoptera:Pyralidae) in the midlands north region of KwaZulu-Natal. Proc. S. Afr. SugarTechnol. Assoc. 82, 471–485.
Wilson, G., 1969. White grubs as pests of sugarcane. In: Williams, J.R., Metcalfe, J.R.,Mungomery, R.W., Mates, R. (Eds.), Pests of Sugarcane. Elsevier PublishingCompany, Amsterdam, pp. 237–254.
Xu, J., 2006. Fundamentals of fungal molecular population genetic analyses. Curr.Issues Mol. Biol. 8, 75–90.
Zimmermann, G., 1986. The Galleria-bait method for detection ofentomopathogenic fungi in soil. J. Appl. Entomol. 102, 213–215.
Top Related