Bacterial halo blight of coffee crop: aggressiveness and genetic diversity of strains

Transcription

1 DOI: K.W. Macel et al. PLANT PROTECTION - Artcle Bacteral halo blght of coffee crop: aggressveness and genetc dversty of strans Karen Wolf Macel 1 *, Suzete Aparecda Lanza Destefano 1, Lus Otavo Saggon Beram 1, Irene Mara Gatt de Almeda 1, Flava Rodrgues Alves Patrco 2, Lucas Mateus Rvero Rodrgues 3, Olvero Guerrero Flho 3 1.Insttuto Bológco - Laboratóro de Bacterologa Vegetal - Campnas (SP), Brazl. 2.Insttuto Bológco - Laboratóro de Ftopatologa - Campnas (SP), Brazl. 3.Insttuto Agronômco - Centro de Café - Campnas (SP), Brazl. ABSTRACT: Bacteral halo blght, caused by Pseudomonas syrngae pv. garcae, s an mportant dsease of coffee crop occurrng n Brazl and other countres. In recent years, outbreaks of ths dsease have damaged several coffee crops n Brazl. Aggressveness and genetc dversty of 25 strans of P. s. pv. garcae, obtaned between the years 1958 and 2011, n 23 ctes of São Paulo and Mnas Geras states, as well as three strans from Kenya were evaluated n ths study. The strans were noculated on coffee seedlngs cultvar Mundo Novo, and ther genetc dversty was evaluated by ERIC-PCR, REP-PCR, and ther combnaton. All the strans were pathogenc to the coffee seedlngs; the results of pathogencty tests, n both experments, could be dvded n four aggressvness classes (hghly aggressve; aggressve; moderately aggressve and less aggressve). The Kenyan strans grouped separately from the Brazlan strans wth ERIC-PCR and the combnaton of ERIC- and REP-PCR. The Brazlan strans could be grouped n two sub-clusters, the frst ncludng the older strans, obtaned from 1958 to 1978, and the other comprsng the remanng strans. Wth a few exceptons, strans solated from 1997 to 2011, grouped manly by ther regon of orgn, were predomnantly solated from hgher alttude regons, above 800 m. Ths probably occurred because the clmatc condtons that preval n these regons, characterzed by mlder temperatures and regular ranfall, are favorable for the coffee crop and for the producton of hgh qualty coffee beverage, but can be also favorable to bacteral halo blght. Key words: Pseudomonas syrngae pv. garcae, Coffea arabca, ERIC- PCR, REP-PCR. *Correspondng author: karenwmacel@hotmal.com Receved: Jul 6, 2016 Accepted: Feb. 28,

2 Genetc dversty of P. syrngae pv. garcae Introducton Brazl s the world s largest producer and exporter of coffee. The Arabca coffee (Coffea arabca L.) s cultvated manly n Mnas Geras, and São Paulo States (Conab 2015). Bacteral halo blght, caused by Pseudomonas syrngae pv. garcae, s an mportant dsease of coffee n Brazl and outbreaks of ths dsease have occurred n ths country n recent years (Rodrgues et al. 2013; Zoccol et al. 2011). The dsease s characterzed by lesons on leaves, flowers, pn-head berres, and de-back of twgs and branches. The most characterstc symptom of the dsease s the presence of brown necrotc lesons on the leaves, surrounded by a chlorotc halo. Growth and coalescence of the lesons on the leaves cause ther drop and subsequent defolaton of branches. At the end of the rany season, the symptoms are confned to the branches that exhbt sometmes severe de-back. Bacteral halo blght also affects seedlngs n nurseres, causng lesons on leaves and de-back of the seedlngs (Costa et al. 1957). Ths dsease was frst descrbed n 1955, affectng coffee crops n the county of Garça, State of São Paulo, Brazl (Amaral et al. 1956) and was consdered of mnor mportance for approxmately 17 years after ts emergence, when only solated cases were reported (Kmura et al. 1973). From 1973 to 1975, a hgh ncdence of the dsease was observed n the state of Paraná n nurseres and coffee plantatons that were recovered from frost 4 (Kmura et al. 1973; Mohan 1976). Bacteral halo blght has spread across the country and the pathogen has been detected n the States of Paraná, São Paulo and Mnas Geras (Malavolta Junor et al. 2008). Recently, damages caused by the dsease were reported n some regons of the State of São Paulo as well as n regons wth rregular topography and hgh alttudes n the State of Mnas Geras, such as Alto Paranaíba, Trângulo Mnero and Sul de Mnas (Zoccol et al. 2011; Rodrgues et al. 2013). The rep-pcr technque has been consdered an mportant tool n phylogenetc studes of bacteral populatons and shows hgh correlaton wth other more laborous technques such as AFLP and DNA hybrdzaton (Rademaker et al. 2000). Ths technque s based on the amplfcaton of three famles of repettve DNA sequences that are present n the genome of several bacteral speces, the repettve extragenc palndromc sequence (REP), the enterobacteral repettve ntergenc consensus (ERIC) and the BOX elements (Louws 4 Mohan, S.K. (1976). Investgação sobre Pseudomonas garcae Amaral et al. em cafeeros. Congresso Braslero de Pesqusas Cafeeras, Caxambu, 4. (p. 56). Ro de Janero: IBC. et al. 1994). Rep-PCR has been frequently used to assess the genetc dversty of several plant pathogenc bactera such as Xanthomonas axonopods pv. juglands (Scortchn et al. 2001), P. syrngae pv. syrngae (Nataln et al. 2006) and studes of nter-and ntraspecfcty n strans of P. syrngae (Vcente and Roberts 2007; Çepn and Gürel 2012; Gašć et al. 2012). Although bacteral halo blght occurs n Brazl snce 1955, t s unknown f the recent outbreaks of the dsease are due to changes n the populaton of the pathogen or to ncreases n the aggressveness of new strans or to other unknown factors. Therefore, aggressveness and genetc dversty of 25 Brazlan strans of P. syrngae. pv. garcae from several coffee producng areas, obtaned between the years 1958 to 2011, were nvestgated n ths study. Materal and Methods Twenty-fve strans of P. syrngae. pv. garcae, solated from coffee crops cultvated n the States of Mnas Geras and São Paulo, Brazl, were nvestgated n ths study. Three strans from Kenya were ncluded n the analyss for comparatve purposes, snce bacteral halo blght was also reported as an mportant dsease n that country (Karu 1997). The strans were obtaned from the Phytobactera Culture Collecton of Insttuto Bológco (IBSBF) and ther geographcal orgn, year of solaton, and clmate data of the localtes of orgn are shown n Table 1. The aggressveness of the strans was evaluated n two experments carred out n September and October of 2015, n a greenhouse located at Centro Expermental Central, Insttuto Bológco, Campnas, SP, wth seedlngs of the susceptble coffee cultvar Mundo Novo. Four pars of true leaves of each seedlng were noculated wth bacteral suspensons (approxmately 10 8 CFU ml -1 ) prepared by the cultvaton of the strans n nutrent agar medum (NA) for 48 h at 28 C. The noculatons were performed by sprayng the suspensons untl runoff on leaves that were prevously punctured wth sterle entomologcal needles (four punctures per leaf). Three control plants were noculated wth sterle dstlled water. After the noculatons, the plants were kept n a most chamber for 72 h and then transferred to a greenhouse at room temperature. 97

3 K.W. Macel et al. Table 1. Strans of Pseudomonas syrngae pv. garcae used n ths study. Geographcal orgn Clmate data IBSBF 1 stran number County/ State Country Average temperature ( o C) Precptaton (mm) Alttude (m) Clmate classfcaton 4 Year of solaton 65 Jaú, SP Brazl Cfb Prajú, SP Brazl Cfa Ouro Fno, MG Brazl Cfb Campnas, SP Brazl Cfa P (= LMG ) Garça, SP Brazl Cfa (=LMG 5551) - Kenya Guaxupé, MG Brazl Cfb Crstas Paulsta, SP Brazl Cfb São João da Boa Vsta, SP Braz Cfa Serra Negra, SP Brazl Cfb Franca, SP Brazl Cfb Patrocíno, MG Brazl 21, Aw Caconde, SP Brazl Cfa Águas da PrataSP Brazl Cfb (= ICMP ) - Kenya Carmo de Mnas, MG Brazl Cfa Serra Negra SP Brazl Cfb Altnópols,SP Brazl Cfb Garça, SP Brazl Cfa São Sebastão da Grama, SP Brazl Cfb Bragança Paulsta SP Brazl Cfb Serra do Saltre Brazl Cfb Vargnha, MG Brazl Cfa Albertna, MG Brazl Cfb (= LMG 5549) - Kenya Dvnolânda, SP Brazl Cfb Andradas, MG Brazl Cfa Unaí, MG Brazl Aw IBSBF - Phytobactera Culture Collecton of Insttuto Bológco Brazl (http// 2 LMG Laboratorumvoor Mcrobology Culture Collecton, Rjksunverstset, Gent, Belgum; 3 ICMP Internatonal Collecton of Mcroorgansms from Plants, Auckland, New Zealand; p Pathovar reference stran; 4 Köppen-Geger Clmate classfcaton (Clmate.date.org. 2016); 5 Cfb - Temperate humdty clmate wth warm summer; 6 Cfa - Temperate humdty clmate wth a hot summer; 7 Aw -Tropcal clmate wth a dry wnter. The severty of the dsease was assessed by the number of lesons and the area affected by the dsease n the second and thrd pars of leaves of the seedlngs at 30 days after the noculaton. To evaluate the area affected by the dsease the leaves were photographed and, n order to be measured n the nteractve applcaton Leaf Doctor (Pethybrdge and Nelson 2015), ther photographs were modfed usng Adobe Photoshop to create a black background that surrounded all the ndvdual leaves. The nteractve applcaton Leaf Doctor was used to dstngush dseased from healthy plant tssues and to calculate the percentage of dsease severty (Pethybrdge and Nelson 2015). Two ndependent experments were carred out n a completely randomzed desgn wth four replcates, each replcate represented by one seedlng. The results were submtted to analyss of varance and the averages were compared by the Scott Knott test at the level of 5% of probablty. 98

4 Genetc dversty of P. syrngae pv. garcae In the molecular characterzaton, the amplfcaton reactons were conducted wth prmer sets: ERIC1R (5 - ATG TAA GCT CCT GGG GAT TCA C - 3 ); ERIC2 (5 - AAG TAA GTG ACT GGG GTG AGC G - 3 ); REP1R-I (5 - III ICG ICG ICA TCI GGC - 3 ); REP2-I (5 - ICG ICT TAT CIG GCC TAC - 3 ), and BOX A1R (5 CTA CGG CAA GGC GAC GCT GAC G - 3 ) (Louws et al. 1994). PCR was conducted n a MyCycle thermal cycler (Bo-Rad) n 25 µl reacton volume contanng: 100 ng of genomc DNA; 0.8 µm of ERIC prmers; 0.2 mm of deoxynucleosde trphosphates; 2.5 mm of magnesum chlorde; and 2.5 unts of Taq DNA polmerase (Fermentas) n a reacton buffer. The followng cyclng condtons were used: one cycle of ntal denaturaton at 95 C for 7 mn; 30 cycles consstng of 94 C for 1 mn; 52 C for 1 mn; 65 C for 8 mn; and a sngle fnal extenson at 65 C for 16 mn. REP-PCR reactons were performed n 25 µl volume contanng: 100 ng of genomc DNA; 0.5 µm of each prmer; 0.2 mm of deoxynucleosde trphosphates; 2.5 mm of magnesum chlorde; and 2.5 unts of Taq DNA polmerase n a reacton buffer. The amplfcaton condtons ncluded one cycle of ntal denaturaton at 95 C for 6 mn; 30 cycles consstng of 94 C for 1 mn; 40 C for 1 mn; 65 C for 8 mn; and a sngle fnal extenson at 65 C for 16 mn. Alquots of REP-PCR (8 µl) and ERIC-PCR (10 µl) products were separated by electrophoress on a 2.5% agarose gel staned wth ethdum bromde 10 mg ml -1 n TAE buffer (0.04 M Trs-acetate/0.001 M EDTA) and then vsualzed and photographed under UV lght usng a dgtal system (Alpha Innotech 2200). A dendrogram was constructed usng fragments rangng from 200 to 1400 bp, scored for presence (1) or absence (0), usng the Jaccard (S J ) coeffcent and UPGMA (Unweghted Pared Group Method) algorthm from the NTSYS-PC program, verson 1.70 (Rohlf 1992). Results and Dscusson All P. syrngae pv. garcae strans nvestgated heren were pathogenc to the coffee seedlngs, but 12 of them, obtaned from 1978 to 2011, were more aggressve. Consderng the leaf area affected by the dsease, n the frst experment the stran IBSBF 1293 was the most aggressveness, followed by the strans IBSBF 65, 1664, 2212, 3046 and 75. In the second experment, the strans IBSBF 2212 and 3046 were more aggressve followed by the strans IBSBF 1664 and By observng smultaneously the results of percentage of area affected by the dsease and number of lesons per leaves, n both experments, the solates can be dvded n four categores: HA - hghly aggressve; A aggressve; MA - moderately aggressve and LA - less aggressve. Accordng to ths results, the hgher aggressve strans were IBSBF 1293, 1664, 2212, e The aggressveness classfcaton of all the other strans are descrbed on Table 2. Symptoms of bacteral halo blght were not observed n the control treatments. Ths s the frst study n whch the severty of bacteral halo blght was assessed wth the nteractve applcaton Leaf Doctor (Pethybrdge and Nelson 2015). The use of the applcaton was somewhat laborous because all the leaves needed to be photographed, but t reduced the human error nvolved n estmatng the leaf area affected by the dsease and also made possble the experment be well documented, therefore t can be a valuable tool for assessng ths dsease. There was a sgnfcant correlaton between the number of lesons per leaf and the area affected by the dsease n the frst (R 2 = 0,776847) and second experment (R 2 = 0,699389). Although all strans nvestgated heren have been obtaned from severely affected coffee crops, some strans showed hgh and others low levels of aggressveness n the coffee seedlngs. Varatons n aggressveness of bacteral strans have already been reported n other pathosystems and are rather expected. Nataln et al. (2006) found varatons n the aggressveness of P. syrngae. pv. garcae strans obtaned from a pear (Pyrus communs) orchard. Mrk et al. (2011) observed that strans of P. cchor obtaned from lettuce (Lactuca satva), tomato (Solanum lycoperscum) and Schefflera arborcola were more aggressve to the hosts from whch they had been solated, although they were pathogenc to all hosts. In our study, the genetc clusterng of the P. syrngae pv. garcae strans showed no correlaton wth the aggressveness of the solates. In contrast, Vcente and Roberts (2007) found that rep-pcr grouped, wth some exceptons, the more aggressveness strans of P. syrngae pv. morsprunorum race 1 solated from cherry trees. It was observed for P syrngae pv. garcae that aggressveness can be lost by successve subcultures (Moraes et al. 1975). Furthermore, aggressveness can vary dependng on the bacteral strans and envronmental condtons, as demonstrated n ths paper. 99

5 K.W. Macel et al. Table 2. Aggressveness of Pseudomonas syrnge pv. garcae strans to coffee seedlngs cv. Mundo Novo, evaluated by the severty, estmated by the leaf area affected by dsease and the number of lesons per leaf. IBSBF 1 strans Percentage of leaf area affected by the dsease Number of lesons per leaf Experment 1 Experment 2 Experment 1 Experment 2 Aggressvty categores b c 3.25 a 3.06 b Aggressve b 6.58 c 2.95 a 2.00 b Aggressve c 6.73 c 2.31 b 2.00 b Aggressve c 7.63 c 2.81 b 2.87 b Aggressve 248 p 0.19 e 0.13 d 0.43 d 0.31 d Lttle aggressve d 0.93 d 0.75 c 0.87 c Moderately aggressve a b 4.18 a 3.18 b Hghly aggressve d 0.51 d 2.37 b 1.16 c Moderately aggressve e 0.11 d 0.62 d 0.43 d Lttle aggressve b b 3.56 a 3.06 b Hghly aggressve b a 3.93 a 4.31 a Hghly aggressve c 3.98 c 3.06 a 2.81 b Aggressve d 3.05 d 2.25 b 3.00 b Moderately aggressve c 5.98 c 2.68 b 3.04 b Aggressve c 6.37 c 1.87 b 1.31 c Moderately aggressve e 0.06 d 0.00 d 0.06 d Lttle aggressve e 0.55 d 0.37 d 1.43 c Lttle aggressve d 0.42 d 2.12 b 1.75 c Moderately aggressve e 0.10 d 0.12 d 0.12 d Lttle aggressve e 0.10 d 0.00 d 0.18 d Lttle aggressve d 0.46 d 1.25 c 1.56 c Moderately aggressve c c 2.22 b 2.62 b Aggressve d 0.20 d 0.93 c 1.00 c Moderately aggressve e 0.20 d 0.62 d 0.68 d Lttle aggressve d 2.09 d 1.50 c 1.75 c Moderately aggressve b a 3.37 a 3.62 a Hghly aggressve d 0.18 d 1.29 c 0.56 d Lttle aggressve e 0.06 d 0.08 d 0.12 d Lttle aggressve Control 0.00 e 0.00 d 0.00 d 0.00 d CV(%) IBSBF - Phytobactera Culture Collecton of Insttuto Bológco - Brazl. (http// 2 Means were compared by the Scott-Knott test at 5 % probablty; P Patovar reference stran. In ths study, the strans dd not show varatons n the BOX-PCR (data not shown) but exhbted varablty n the ERIC- and REP-PCR patterns. The genetc relatedness of P. syrngae pv. garcae strans, nvestgated by ERIC- and REP- PCR, yelded fngerprnts wth 20 and 24 bands rangng from 1900 bp to 80 bp and 1825 bp to 200 bp, respectvely. Bandng patterns generated by ERIC- and REP-PCR revealed a hgh genetc dversty among the strans. The dendrogram constructed wth ERIC allowed the separaton of the strans nto two major clusters wth approxmately 37% smlarty. Cluster I was represented by four strans from the State of São Paulo solated between the years 1958 and 1978, wth alttudes rangng from 524 to 684 m. Cluster II, dvded nto two subclusters, a and b, allocated the remanng strans wth approxmately 45% smlarty between them. The subcluster a was represented by the three strans from 100

6 Genetc dversty of P. syrngae pv. garcae Kenya, and the subcluster b was further subdvded nto two other subclusters, and, wth approxmately 53% smlarty. The subcluster allocated strans from the countes of Guaxupé, MG, Andradas, MG, and Vargnha, MG, (70% smlarty), whle the subcluster grouped fve strans from Mnas Geras 57% smlarty), solated from 1978 to 2011, and the remanng 12 from the State of São Paulo solated from 1998 to One stran from Albertna, MG, was allocated wth São Paulo strans, and the genetc smlarty between them probably has occurred due the exchange of vegetal materal between these ctes whch are geographcally close (Fgures 1 and 2). Bandng patterns generated by REP-PCR also revealed a hgh genetc dversty among the strans. In ths analyss the strans were separated nto two clusters: Cluster I was represented only by the IBSBF 2883 stran from Kenya, wth M M M 1,000 bp 500 bp Fgure 1. Amplfcaton of DNAs from Pseudomonas syrngae pv. garcae strans wth ERIC 1R/ERIC2 prmers. (M) Molecular weght marker 100 bp (1) IBSBF 248 P ; (2) 2999; (3) 3005; (4) 3015; (5) 3019; (6) 3022; (7) 3032; (8) 3046; (9) 249; (10) 2883 and (11) (b) (a) Garça/SP (LA) Jaú/SP (A) Prajú/SP (A) Campnas/SP (A) Ouro Fno/MG (A) Patrcíno/MG (A) Serra do Saltre/MG (A) Unaí/MG (LA) Carmo de Mnas/MG (LA) Crstas Paulsta/SP (MA) Franca/SP (HA) Serra Negra/SP (LA) Altnópols/SP (MA) São João da Boa Vsta/SP (LA) São Sebastão da Grama/SP (LA) Garça/SP (LA) Águas da Prata/SP (A) Bragança Paulsta/SP (MA) Albertna/MG (LA) Dvnoânda/SP (HA) Serra Negra/SP (HA) Caconde/SP (MA) Guaxupé/MG (HA) Andradas/MG (LA) Vargnha/MG (MA) Kenya (MA) Kenya (MA) Kenya (MA) Fgure 2. Genetc dversty of Pseudomonas syrngae pv. garcae strans. Dendrogram generated accordng to the fngerprnt of the strans usng ERIC 1R/ERIC2 prmers, based on the UPGMA method, usng Jaccard smlarty coeffcent (S J ). 101

7 K.W. Macel et al. approxmately 20% smlarty to Cluster II, whch allocated the remanng strans and was subdvded nto subclusters a and b, wth about 49% smlarty to each other. The subcluster a ncluded only fve strans wth 60% smlarty, two from Kenya (IBSBF 249 and IBSBF 3037), the pathotype stran (IBSBF 248 p ), also one stran from Guaxupé, MG, and another from Ouro Fno/MG. The subcluster b grouped the remanng 22 strans, whch could be further dvded nto two other subclusters, and. Subcluster grouped only the strans solated from Mnas Geras wth approxmately 63% smlarty, whle subcluster only those from São Paulo (smlartes rangng 60 to 93%), wth excepton of Albertna, MG, as also observed n ERIC-PCR analyss. Accordng to ths analyss, the strans from São Paulo (cluster II, subcluster b and ) were manly separated from those of Mnas Geras (cluster II, subcluster b and ) (Fgures 3,4). M M M 1,000 bp 500 bp Fgure 3. Amplfcaton of DNAs from Pseudomonas syrngae pv. garcae strans wth REP1R/REP2I prmers. (M) Molecular weght marker 100 bp; (1) IBSBF 248 P ; (2) 65; (3) 75; (4) 158; (5) 1372; (6) 1373; (7) 1664; (8) 2840; (9) 2841 and (10) (a) Garça/SP (LA) Ouro Fno/MG (A) Guaxupé/MG (HA) Kenya (MA) Kenya (MA) Patrcíno/MG (A) Carmo de Mnas/MG (LA) Serra do Saltre/MG (A) Unaí/MG (LA) Vargnha/MG (MA) Andradas/MG (LA) Jaú/SP (A) Serra Negra/SP (LA) Prajú/SP (A) Altnópols/SP (MA) Crstas Paulsta/SP (MA) Serra Negra/SP (HA) Caconde/SP (MA) Campnas/SP (A) Franca/SP (HA) São João da Boa Vsta/SP (LA) São Sebastão da Grama/SP (LA) Águas da Prata/SP (A) Garça/SP (LA) Albertna/MG (LA) Dvnoânda/SP (HA) Bragança Paulsta/SP (MA) Kenya (MA) (b) Fgure 4. Genetc dversty of Pseudomonas syrngae pv. garcae strans. Dendrogram generated accordng to the fngerprnt of the strans usng REP1R/REP2I prmers, based on the UPGMA method, usng Jaccard smlarty coeffcent (SJ). 102

8 Genetc dversty of P. syrngae pv. garcae In the combned analyss, the strans were separated n two clusters (Fgure 5). Cluster I was only composed by the three strans from Kenya, wth approxmately 37% smlarty wth the other strans of ths study. Cluster II comprsed the remanng 25 strans, dvded n two subclusters, a and b, wth approxmately 50% smlarty wth each other. The subcluster a was composed by four strans from the State of São Paulo n the perod from 1958 to 1978 (IBSBF 65, IBSBF 75, IBSBF158 and IBSBF 248 P ). The subcluster b allocated 21 strans from the States of São Paulo and Mnas Geras and was subdvded n two subclusters, and, wth approxmately 57% smlarty between them. The subcluster was composed only by strans from the state of Mnas Geras and by those from São Paulo, wth excepton of IBSBF 3032 from Albertna/MG (Fgure 5). In the combned ERIC- and REP-PCR analyss, strans from Guaxupé/MG, Vargnha, MG, and Andradas, MG, obtaned from 1997 to 2009, grouped n the same subgroup (Fgure 5). These countes belong to the same coffee producton regon n Sul de Mnas, wth alttudes varyng from 865 to 881 m, Cfb and Cfa clmate, annual average temperature of 19,7 to 20,2 o C; and 1200 to 1500 mm average annual ranfall (Table 1). The strans from São João da Boa Vsta, São Sebastão da Grama, Águas da Prata, Garça, Bragança Paulsta, Dvnolânda, all from the State of São Paulo, as well as Albertna, from Mnas Geras, grouped n smlar subgroups n the REP-PCR, and n the combned ERICand REP-PCR dendrograms. These countes, wth the excepton of Garça (IBSBF 3015), are relatvely close to each other, have alttudes varyng from 764 to 1021 m, average temperature from 18,2 to 20,1 o C, and 1,493 to 1,590 mm annual average ranfall, as well as Cfa and Cfb clmate (Table 1). Strans from Crstas Paulsta, Franca and Altnópols, that are relatvely close to each other, as well as Serra Negra and Caconde, that are more dstant, grouped n the same sub-group n the combned ERIC- and REP- PCR dendrogram. These countes belong to the same coffee producng regon, Mogana, and show average temperature rangng from 18,1 to 20,1 o C, average annual ranfall of 1,479 to 1,557 mm, Cfb to Cfa clmate classfcaton (Clmate date org. 2016) (Table 1). In ths study, the strans dd not show varatons n the BOX-PCR (data not shown) but exhbted varablty n the ERIC- and REP-PCR patterns. Isolates from the same pathovar may have dentcal REP, BOX and ERIC fngerprnts or they may dffer n some bandng patterns (Louws et al. 1994). Çepn and Gürel (2012) found that ERIC-PCR could not dfferentate strans of P. savastano pv. savastano from strans of P. syrngae. pv. tomato and P. s. pv. phaseolcola, but the combnaton of BOX-PCR wth REP-PCR allowed the groupng of Garça/SP (LA) Jaú/SP (A) Prajú/SP (A) Campnas/SP (A) Ouro Fno/MG (A) Guaxupé/MG (HA) Vargnha/MG (MA) Andradas/MG (LA) Patrcíno/MG (A) Carmo de Mnas/MG (LA) Serra do Saltre/MG (A) Unaí/MG (LA) Crstas Paulsta/SP (MA) Serra Negra/SP (HA) Caconde/SP (MA) Altnópols/SP (MA) Franca/SP (HA) Serra Negra/SP (LA) São João da Boa Vsta/SP (LA) São Sebastão da Grama/SP (LA) Águas da Prata/SP (A) Garça/SP (LA) Bragança Paulsta/SP (MA) Albertna/MG (LA) Dvnoânda/SP (HA) Kenya (MA) Kenya (MA) Kenya (MA) (b) (a) Fgure 5. Genetc dversty of Pseudomonas syrngae pv. garcae strans. Dendrogram generated accordng to the fngerprnt of the strans usng the prmers ERIC-PCR and REP-PCR, based on the UPGMA method, usng Jaccard smlarty coeffcent (S J ). 103

9 K.W. Macel et al. strans of dfferent speces and hosts n dverse clusters. Gašć et al. (2012) demonstrated that ths technque was able to dfferentate strans of P. syrngae. pv. syrngae, P. syrngae. pv. morsprunorum and P. syrngae. pv. perscae from stone fruts wth rep-pcr. On the other hand, Vcente and Roberts (2007), whle studyng strans of P. syrngae. pv. syrngae from cherry trees and other hosts and P. syrngae. pv. morsprunorum solated from cherry trees, found that strans of P. syrngae. pv. syrngae could vary n ther rep-pcr patterns, and strans from dfferent hosts grouped together, but strans from P. syrngae. pv. morsprunorum were more aggressve and less varable. Kaluzna et al. (2010) found that BOX-PCR provded a better dfferentaton of the genetc dversty of P. syrngae. pv. syrngae strans and P. syrngae. pv. morsprunorum races 1 and 2 than ERIC-PCR. In ths study, strans from Kenya grouped separately from Brazlan strans n the ERIC- and REP-PCR analyss. The low genetc smlarty observed between these strans could be related to ther geographc orgn. Karu (1997), prevously related dfferences between strans solated from Brazl and Kenya, n relaton to aggressveness and bochemcal characterstcs such as UV fluorescent pgment and bacterocn producton. In a smlar pathosystem, Ferrante and Scortchn (2010) also observed that P. syrngae pv. actndae strans, the causal agent of an outbreak of bacteral canker on Actnda chnenss (yellow kwfrut) n central Italy, grouped separately by rep-pcr from strans obtaned n other countres. In ths study, the ERIC-PCR technque allows to cluster the strans of P. syrngae pv. garcae by perod of solaton, whle wth the REP-PCR technque the strans were grouped by geographcal orgn. Smlar results were observed for Xanthomonads of tomato by Louws et al The combnaton of ERIC wth REP-PCR showed the genetc relatedness of the Brazlan strans of P. syrngae. pv. garcae. Wth some exceptons, the strans obtaned after the year of 1997 could be grouped by geographc orgn. Ths may have occurred, probably, because nurseres dstrbute coffee seedlngs n the regons where they are produced, and the exchange of seedlngs between dfferent regons may occur, but s rather nfrequent. As nfected seedlngs are the man source of spread of bacteral halo blght to new coffee crops, the strans could evolve separately n each regon. Ths s a relevant nformaton found n ths study and mples that control measures amng to reduce the dsease n coffee seedlngs nurseres can mpact the spread and management of the dsease n new coffee crops. Although some of the strans were grouped by aggressveness classfcaton, such as, n the combned analyss (ERIC/REP-PCR), the Cluster II, subcluster a were allocated three strans (IBSBF 65, 75 and 158) as aggressve, and the three strans of Cluster I (IBSBF 249, 2883 and 3037) as moderately aggressve. Even though some strans were grouped, these results could not be correlated. A study wth a major number of strans mght gve a better correlaton of aggressveness and genetc dversty of P. syrngae pv. garcae. Smlar results were observed by Scortchn et al. (2001), whle studyng strans of Xanthomonas arborcola pv. juglands from Persan walnut. The authors observed that, although the genetc smlarty of those solates was hgh, the strans grouped n three dfferent clusters related to ther Italan or Greek geographc orgn. Gutérrez-Barranquero et al. (2013) also reported that the ERIC- and REP-PCR clustered separately P. syrngae pv. syrngae strans solated from mango from strans obtaned from other hosts, and also grouped them by ther geographc orgn. Snce 1997, the majorty of the P. syrngae pv. garcae strans (80.9%) was orgnated from regons near or above 800 m alttude, consdered hgh alttude regons n Brazl. In hgher alttude regons, the clmatc condtons are very sutable to the coffee crop and are favorable to produce beverage wth hgher qualty. However, n these areas, the mlder temperatures and frequent ncdence of wnds are also favorable for bacteral halo blght. Consderng that P. syrngae pv. garcae strans solated n the last years clustered by ther regon of orgn, recent outbreaks of the dsease could be related to nurseres dstrbutng nfected seedlngs n regons wth favorable clmatc condtons, but further studes are stll necessary to valdate ths hypothess. Concluson All the Pseudomonas syrngae pv. garcae strans were pathogenc to coffee seedlngs but 12, obtaned from 1977 to 2011, were more aggressve on Coffea arabca cv. Mundo novo. 104

10 Genetc dversty of P. syrngae pv. garcae In the combned analyss of ERIC- and REP-PCR, the Brazlan strans obtaned from 1958 to 1978 grouped separately from the remanng strans and the Kenyan strans grouped separately from the Brazlan strans. Strans solated from 1997 to 2011 clustered by ther regon of orgn n the combnaton of ERIC- and REP- PCR, and were predomnately from regons above 800 m. Acknowledgments The authors are grateful to the Natonal Councl for Scentfc and Technologcal Development (CNPq) and the Coordnaton for the Improvement of Hgher Educaton Personnel (CAPES) for fnancal supports; and Dr. B. C. Trasferett for the revson of the Englsh language. References Amaral, J. F., Texera, C. G. and Pnhero, E. D. (1956). O bactéro causador da mancha aureolada do cafeero. Arquvos do Insttuto Bológco, 23, Çepn, E. and Gürel, F. (2012).Varaton n extragenc repettve DNA sequences n Pseudomonas syrngae and potental use of modfed REP prmers n the dentfcaton of closely related solates. Genetcs and Molecular Bology, 35, dx.do.org/ /s Clmate Data Org. Dados clmátcos para as cdades mundas; [accessed 2016 February 22]. PT.clmate-data.org Companha Naconal de Abastecmento (2014). Avalação da Safra Agrícola Cafeera Tercera Estmatva - Janero/2015; [accessed 2016 January 15]. uploads/arquvos/15_01_14_11_57_33_boletm_cafe_janero_2015. pdf. Costa, A. S., Amaral, J. F., Vegas, A. P. Slva, D. M., Texera, C. G. and Pnhero, E. D. (1957). Bacteral halo blght of coffee n Brazl. Phytopathologsche Zetschrft, 28, Ferrante, P. and Scortchn, M. (2010). Molecular and phenotypc features of Pseudomonas syrngae pv. actndae solated durng recent epdemcs of bacteral canker on yellow kwfrut (Actnda chnenss) n central Italy. Plant Pathology, 59, Gašć, K., Prokć, A., Ivanovć, M., Kuzmanovć, N. and Obradovć, A. (2012). Dfferentaton of Pseudomonas syrngae pathovars orgnatng from stone fruts. Pestcdes and Phytomedcne, 27, Gutérrez-Barranquero, J. A., Carrón, V. J., Murllo, J., Arrebola, E., Arnold, D. L., Cazorla, F. M. and Vcente, A. (2013). A Pseudomonas syrngae dversty survey reveals a dfferentated phylotype of the pathovar syrngae assocated wth the mango host and mango toxn producton. Phytopathology 103, Kaluzna M., Ferrante P., Sobczewsk P. and Scortchn M. (2010). Characterzaton and genetc dversty of Pseudomonas syrngae from stone fruts and hazelnut usng repettve-pcr and MLST. Journal of Plant Pathology, 92, Karu, M.G. (1997). Bochemcal and pathogenc dfferences between Kenyan and Brazlan solates of Pseudomonas syrnge pv. garcae. Plant Pathology, 46, org/ /j d x. Kmura, O., Robbs, C. F. and Rbero, R. L. D. (1973). Estudos sobre o agente da mancha aureolada do cafeero (Pseudomonas garcae Amaral et al.). Arquvos da Unversdade Federal Rural do Ro de Janero, 3, Louws, F. J., Fulbrght, D. W., Stephens, C. T. and De Brujn, F. (1994). Specfc genomc fngerprnts of phytopathogenc Xanthomonas and Pseudomonas pathovars and strans generated wth repettve sequences and PCR. Appled and Envronmental Mcrobology, 60, Malavolta Junor, V. A., Beram, L. O. S., Almeda, I. M. G., Rodrgues Neto, J. and Robbs, C. F. (2008). Bactéras ftopatogêncas assnaladas no Brasl: uma atualzação. Summa Phytopathologca, 34, Mrk, M., Aysan, Y. and Sahn, F. (2011). Characterzaton of Pseudomonas cchor solated from dfferent hosts n Turkey. Internatonal Journal of Agrculture and Bology, 13, Moraes, S. A., Sugmor, M. H., Tomazello Flho, M. and Carvalho, P. C. T. (1975). Resstênca de cafeeros a Pseudomonas garcae. Summa Phytopathologca, 1,

11 K.W. Macel et al. Nataln, E., Ross, M. P., Baronov, D. and Scortchn, M. (2006). Genetc and pathogenc dversty of Pseudomonas syrngae pv. syrngae solates assocated wth bud necross and leaf spot of pear n a sngle orchard. Journal of Plant Pathology, 88, dx.do.org/ /jpp.v Pethybrdge, S. J. and Nelson, S. C. (2015). Leaf Doctor: A new portable applcaton for quantfyng plant dsease severty. Plant Dsease, 99, Rademaker, J. L. W., Hoste, B., Louws, F J., Kersters, K., Swngs, J., Vautern, L., Vautern, P. and Brujn, F. J. (2000). Comparson of AFLP and rep-pcr genomc fngerprntng wth DNA DNA homology studes: Xanthomonas as a model system. Internatonal Journal of Systematc and Evolutonary Mcrobology, 50, org/ /J x. Rodrgues, L.M.R., Almeda, I. M. G., Patríco, F. R. A., Beram, L. O. S., Macel, K. W., Braghn, M. T. and Guerrero Flho, O. (2013). Mancha aureolada do cafeero causada por Pseudomonas syrngae pv. garcae. Boletm Técnco IAC 212. (p. 24). Campnas: Insttuto Agronômco. Rohlf, F. J. (1992). NTSYS-PC Numercal Taxonomy and Multvarate Analyss Sstem Verson Stony Brook. New York: State Unversty of New York. Scortchn, M., Marches, U. and D Prospero, P. (2001). Genetc dversty of Xanthomonas arborcola pv. juglands (synonyms: X. campestrspv. juglands;x. juglands pv. juglands) strans from dfferent geographcal areas shown by repettve polymerase chan reacton genomc fngerprntng. Journal of Phytopathology, 149, org/ /j x. Vcente, J. G. and Roberts, S. J. (2007). Dscrmnaton of Pseudomonas syrngae solates from sweet and wld cherry usng rep-pcr. European Journal of Plant Pathology, 117, Zoccol, D. M., Takatsu, A. and Uesug, C. H. (2011). Ocorrênca de mancha aureolada em cafeeros na Regão do Trângulo Mnero e Alto Paranaíba. Braganta, 70, org/ /s