Comparison of cross inoculation potential of South African avocado and mango isolates of Colletotrichum gloeosporioides

Download Comparison of cross inoculation potential of South African avocado and mango isolates of Colletotrichum gloeosporioides

Post on 21-Sep-2016




1 download

Embed Size (px)


<ul><li><p>Three hundred and eighteen Colletotrichum gloeosporioidesisolates from stem-end rot and anthracnose infected avocadosas well as from stem-end rot/anthracnose and soft brown rot onmango, were compared using fruit inoculations. Isolates couldbe categorised according to lesion size and both avocado andmango isolates produced larger lesions when inoculated ontheir own hosts. Cross-inoculation potential of these isolateswas also compared on strawberries, peppers, guavas, papayasand citrus. All isolates produced lesions on all hosts exceptcitrus. Factors such as area of origin and symptom type fromwhich original isolations were made, could not be correlatedwith lesion development on these hosts.</p><p>Key words: avocado mango Colletotrichum gloeosporio-ides inoculation study</p><p>Introduction </p><p>Colletotrichum gloeosporioides (Penz.) Penz. &amp; Penz. is a ubiquitous, proliferate and economicallyimportant pathogen causing substantial yield losses due to fruit decay and damage to vegetative parts in avariety of plant species (Freeman and Shabi 1996).Diseases caused by C. gloeosporioides include anthrac-nose, dieback, root rot, leaf spot, blossom rot and seed-ling blight on a wide range of crops including avocado,almond, peach (Freeman et al. 1998), peppers (Manand-har et al. 1995), papaya (Dickman 1994) mango (Ploetz1994), Stylosanthes spp. (Chakraborty and Jones 1993),</p><p>citrus (Timmer et al. 1994), rubber trees (Brown andSoepena 1994), passion fruit (Jeffries et al. 1990) andstrawberry (Denoyes and Baudry 1995).Isolates of C. gloeosporioides from almond, apple,</p><p>mango and avocado inoculated into detached apple,avocado, almond, mango and nectarine were shown tosuccessfully cross-infect (Freeman and Shabi 1996), butwere more virulent on the same host from which theywere originally isolated (Hayden et al. 1994). Cross-inoculation studies with C. gloeosporioides on a widerange of hosts have been reported (Maas and Howard1985; Alahakoon et al. 1992, 1994; Freeman et al.1996). Differential virulence of C. gloeosporioides iso-lated from several hosts was shown when inoculatedinto other hosts (Quimio and Quimio 1975; Freemanand Shabi 1996). Furthermore, the pathogen was shownto genetically adapt to the new host, eventually inducingthe same level of disease as isolates from the host inquestion (Alahakoon et al. 1992).Many hosts susceptible to C. gloeosporioides are</p><p>cultivated world-wide and losses where multiple hostssuch as mango, avocado, coffee, papaya and citrus aregrown in close proximity, could be enormous (Freemanet al. 1998). This is especially true in South Africa,where there are small climatic zones suited for cultiva-tion of subtropical crops, and avocados and mangoes areoften cultivated in adjacent blocks or orchards. Isolationfrequency of C. gloeosporioides from stem-end rot,anthracnose and soft brown rot was variable in avocadosand mangoes, although generally higher on mangoes(Sanders et al. 2000). Higher levels of C. gloeosporioi-des on mango trees could therefore serve as an inoculumsource for adjacently cultivated crops.</p><p>0944-5013/03/158/02-143 $15.00/0 Microbiol. Res. 158 (2003) 2 143</p><p>Microbiol. Res. (2003) 158, 143150</p><p>Comparison of cross inoculation potential of South Africanavocado and mango isolates of Colletotrichum gloeosporioides</p><p>G. M. Sanders*, L. Korsten</p><p>Department of Microbiology and Plant Pathology, Forestry and Agricultural Biotechnology Institute, University of Pretoria,Pretoria 0002, South Africa</p><p>Accepted: December 23, 2002</p><p>Abstract</p><p>Corresponding author: Dr. G. M. Sanders (Swart)e-mail :</p></li><li><p>Because of the close proximity in which crops such asavocados, mangoes, papayas and citrus are often culti-vated, it is important to determine whether the differenthosts can serve as inoculum sources for the others. Thepurpose of this study was therefore to determine thevirulence of C. gloeosporioides isolates collected dur-ing a market survey of post-harvest disease incidence onavocados and mangoes (Sanders et al. 2000) and theircross-inoculation potential each other as well as onpapayas, strawberries, peppers, guavas and citrus. </p><p>Materials and methods</p><p>Culture collection. Three hundred and eighteen mono-conidial C. gloeosporioides isolates were randomlyselected from the market survey collection described bySanders et al. (2000). Details of all isolates includinghost, symptom and stage of ripeness from which isola-tions were made were also recorded (data not shown;Swart 1999). Of these, 207 were from avocado and 111from mango (Table 1). Cultivation and maintenance ofisolates was done on oatmeal agar (20 g oatmeal, 20 gagar (Biolab), 1 l distilled water) (OA). Plates wereincubated at ambient temperature under constant mixedirradiation from near ultraviolet and daylight typefluorescent tubes (Phillips TL 40W/08RS, F40 B43 andTL 40W/33RS respectively) until sporulation occurred.Identities of isolates were confirmed using physiologi-cal and inoculation studies, molecular analysis (Swart1999) and morphology (Baxter et al. 1983). Cultureswere preserved for further use by freezing in 50% gly-cerol at 78C, as well as on potato-dextrose agar (Bio-lab) (PDA) slants and under sterile water. </p><p>Inoculation studies on Fuerte avocado fruit. Freshlyharvested untreated, unwaxed, physiologically mature,but unripe Fuerte fruit from Westfalia Estates (LimpopoProvince, South Africa) were used for plug inoculations.Prior to inoculation, all fruit were swabbed with 70%(v/v) ethanol to reduce surface contamination and left toair-dry in the laboratory. Starter cultures were preparedby incubating each C. gloeosporioides isolate for fivedays on OA as described previously. Plugs (4 mm dia-meter) were cut from actively-sporulating areas near the edge of each colony. Four 10 mm deep holes wereaseptically punched around the broadest side part ofeach fruit with a 4 mm diameter stainless steel corkborer. Three plugs, each from a different isolate, wereplaced into a hole in three replicate fruit with an OAmedium plug (control) into the fourth hole. Fruit plugsremoved from punched holes were replaced and coveredwith parafilm. Fruit were incubated upright at ambienttemperature (approximately 25C). After five days,lesion diameters were measured in two opposite direc-tions. Lesions 59 mm in size were rated as category 1,1014 mm as 2, 15 20 mm as 3, 2124 mm as 4,2529 mm as 5, 3034 mm as 6, 3539 mm as 7 and4044 mm as 8. Means were determined for all isolatesand all data were analysed with the SAS system usinganalysis of variance based on unequal subclass numbers.Pearsons product moment correlation coefficient (r)was used to correlate various factors under investiga-tion. Treatment means and ranks were compared usingDuncans multiple range test. Categories were compa-red using the chi-square test for equal proportions. Inoculation studies on Sensation mango fruit. Freshlyharvested, untreated, unwaxed, physiologically matureSensation fruit from Moria Mango Estate (LimpopoProvince, South Africa) were used for plug inoculations.Preparation of inoculum, inoculation, evaluation anddata analyses was carried out as described for avocado.Inoculation potential on different hosts. Cultures wereprepared in the same way as for avocado and mangoinoculation studies. After five days incubation, plateswere flooded with sterile distilled water (SDW) andconidia aseptically harvested using a glass rod. Conidialconcentrations were determined using a haemocyto-meter and adjusted with SDW to 1 104 conidia ml1.Strawberries (Fragaria ananassa Duchesne), guavas(Psidium guajava L.), peppers (Capsicum annuum L.),papayas (Carica papaya L.) and citrus (Citrus sinensisL. Osbeck) were selected for this study since they areoften cultivated in close proximity to avocados andmangoes. Fruit were collected from the National FreshProduce Market in Pretoria, South Africa. All fruit werethoroughly washed under running water, swabbed with70% (v/v) ethanol and left to air-dry in the laboratory.Inoculum droplets of 10 l containing ca. 104 conidia</p><p>144 Microbiol. Res. 158 (2003) 2</p><p>Table 1. Distribution and origin of Colletotrichum gloeospo-rioides isolates used in this study </p><p>Host Geographical origin Number of isolates</p><p>Mango Levubu 5Tzaneen 4Hoedspruit 6Kaapmuiden 7Letsitele Valley 26Hazyview 15Malelane 11</p><p>Total 111Avocado Louis Trichardt 39</p><p>Kwa-Zulu Natal 6Tzaneen 98Nelspruit 41Levubu 21Hazyview 2</p><p>Total 207</p></li><li><p>ml1 SDW were placed onto 2 mm deep prick woundson the surfaces of the various fruits Three isolates perfruit with three to five fruit replicates per isolate wereused (three replicates: papayas, guavas and citrus; fivereplicates: strawberries and peppers). A single prickwound receiving SDW was included as control on eachfruit replicate. Fruit were incubated at 25C in cartonslined with moist laboratory wipes and covered withplastic bags. Lesion development was monitored at five,seven and 11 days after inoculation. Isolations weremade from fruit to confirm that lesions were due toinfection by C. gloeosporioides. Sequence analysis. DNA extractions were conductedout on all isolates using the procedure described by Rae-der and Broda (1985). Representative isolates wereselected (Table 2) and compared by means of ribosomalDNA internal transcribed spacer sequence analysisusing primers ITS1 and ITS4. These primers describedby White et al. (1990) were used to amplify the inter-nally transcribed spacer region. Primers were synthe-sised by MWG Biotech (Germany). PCR reactions wereperformed in 50 l volumes, each reaction containing0.5 l template DNA, 0.6 M of each primer, 5 lrecommended 10 buffer (supplied with Taq poly-merase), 2 mM MgCl2, 100 M each of dCTP, dGTP,dATP and dTTP (Promega) and 0.5 U Taq polymerase(Promega). Twenty-five PCR cycles were performed ona Perkin-Elmer 2400 thermocycler using the followingconditions: a denaturation step of 94C for one minutefollowed by annealing at 50C for one minute andextension at 72C for 90 seconds, followed by a finalextension of 72C for 10 minutes. PCR products weresequenced in both directions using an ABI PRISM 377automated DNA sequencer (Perkin-Elmer Applied Bio-Systems). DNA sequencing reactions were carried outwith an ABI PRISM Big Dye Terminator Cycle Se-quencing Ready Reaction kit (Perkin-Elmer AppliedBioSystems) according to the manufacturers specifica-tions. Sequences of approximately 610 bp were ob-</p><p>tained for all isolates and aligned manually usingSequence Navigator version 1.0.1 (Perkin-Elmer App-lied BioSystems). Analysis of sequences was done usingPhylogenetic Analysis Parsimony (PAUP) 4.0. Missingdata was treated as a fifth character (new state). Allcharacters were given equal weight. The heuristic searchoption (based on parsimony) with random step-wiseaddition and tree-bisection-reconnection (TBR) as theswapping algorithm was used to construct the phylo-gram. The confidence limits were determined from abootstrap analysis with 1000 replications. Trees wererooted to the sequence of C. dematium. </p><p>Results</p><p>Inoculation studiesLesion sizes fitting the categories described were pro-duced by both avocado and mango isolates. Most of theisolates produced lesions that fell into category 3 andcategory 4 (Fig. 1). </p><p>Inoculation of avocado isolates on Fuerte avocado fruitDistribution of isolates in categories based on lesionformation was a left-skewed normal distribution withmost isolates producing category 3 lesions (2124 mm)(Fig. 1). No significant differences were observed inmean lesion size produced by isolates from specificproduction areas (P = 0.0083). Seven isolates did notproduce any symptoms. These isolates were from Tza-neen (stem-end rot ; overripe fruit) and Nelspruit (an-thracnose; slightly overripe fruit). Stage of fruit ripeness at which isolations were ori-</p><p>ginally made also had an effect on lesion size whenthese isolates were inoculated into avocado fruit. Iso-lates from eating-ripe fruit produced significantly largerlesions than isolates from overripe and very overripefruit (P = 0.0001) and subsequently confirmed by cor-</p><p>Microbiol. Res. 158 (2003) 2 145</p><p>Table 2. Origin of Colletotrichum isolates used in sequence analysis</p><p>Isolate number Host Symptom type Geographical origin Stage of ripeness</p><p>37 Mango Anthracnose Tzaneen Slightly overripe44 Mango Anthracnose Tzaneen Slightly overripe63 Mango Anthracnose Tzaneen Eating ripe65 Avocado Stem-end rot Nelspruit Slightly overripe67 Avocado Anthracnose Nelspruit Slightly overripe329 Avocado Stem-end rot Louis Trichardt Slightly overripe388 Avocado Stem-end rot Tzaneen Overripe408 Avocado Stem-end rot Louis Trichardt Slightly overripePPRI 61211 Cowpea Unknown Roodeplaat Not applicable</p><p>1 Colletotrichum dematium, National Collection of Fungi, Plant Protection Research Institute, Pretoria, South Africa. </p></li><li><p>relation analysis (P = 0.0238). No significant differ-ences were observed in lesion size produced betweenisolates obtained from stem-end rot or anthracnoselesions (P = 0.1198). </p><p>Inoculation of mango isolates on Sensation mango fruit Distribution of isolates in categories based on lesionformation was a left-skewed normal distribution withmost isolates producing category 2 lesions (1520 mm)(Fig. 1). No isolate produced lesions larger than 34 mm,except one which produced a lesion within the range of3539 mm. Contrary to avocado isolates, all mango iso-lates produced lesions when inoculated into mangoes. Isolates from eating-ripe fruit produced significant-</p><p>ly larger lesions than those from overripe fruit (P =0.0168). No significant differences in mean lesion sizewere observed in isolates from different productionareas (P = 0.2518). Lesion type from which original iso-lations were made, as a rule, had no effect on lesion sizein inoculated mangos. The only exception was isolates</p><p>from Hoedspruit, where isolates from soft brown rotproduced significantly smaller lesions than those fromstem end rot and anthracnose (P = 0.0351). Comparison of avocado and mango isolates and cross-infection potential. C. gloeosporioides isolates fromavocados inoculated into avocados generally resulted insignificantly larger lesions than mango isolates inocu-lated into mangoes (P = 0.0003). The mean lesion sizeproduced when avocado isolates were inoculated intoavocados was 34.44 mm compared to 29.44 mm formango isolates inoculated into mangoes. Fifty-eight percent of the mango isolates produced larger lesions onmangoes than on avocados and 3.57% did not produceany symptoms on avocados. On the other hand, 78% ofavocado isolates produced larger lesions on avocadosthan on mangoes and all produced symptoms on man-goes. Overall, avocado isolates inoculated into avocadosproduced the largest lesions (P = 0.0369) than any otherisolates on mango or avocado. Area of origin of avo-cado isolates (P = 0.3075) and stage of fruit ripenessfrom which isolations were made (P = 0.4485) had no</p><p>146 Microbiol. Res. 158 (2003) 2</p><p>Fig. 1. Distribution of Colletotrichum gloeosporioides isolates...</p></li></ul>


View more >