An analysis of viruses associated with disease in three Angolan provinces S. Matic, A.T. Pais da Cunha, Jeremy R. Thompson, Mark Tepfer

To cite this version:

S. Matic, A.T. Pais da Cunha, Jeremy R. Thompson, Mark Tepfer. An analysis of viruses associated with cassava mosaic virus disease in three Angolan provinces. Journal of Plant Pathology, Springer, 2012, 94 (2), pp.443-450. ￿10.4454/JPP.FA.2012.043￿. ￿hal-01191289￿

HAL Id: hal-01191289 https://hal.archives-ouvertes.fr/hal-01191289 Submitted on 29 May 2020

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés.

Distributed under a Creative Commons Attribution - ShareAlike| 4.0 International License 021_JPP1146SC(Matic)_443 20-07-2012 10:16 Pagina 443

Journal of Plant Pathology (2012), 94 (2), 443-450 Edizioni ETS Pisa, 2012 443

SHORT COMMUNICATION AN ANALYSIS OF VIRUSES ASSOCIATED WITH CASSAVA MOSAIC DISEASE IN THREE ANGOLAN PROVINCES

S. Matic1,2*, A.T. Pais da Cunha3,4, J.R. Thompson1,5 and M. Tepfer1,6,7

1Plant Virology Group, ICGEB Biosafety Outstation, Via Piovega 23, 31056 Ca’Tron di Roncade, Italy 2Istituto di Virologia Vegetale, CNR, Strada delle Cacce 73, 10135 Torino, Italy 3Dipartimento Territorio e Sistemi Agro-Forestali - Patologia Vegetale, Università degli Studi di Padova, Agripolis, Viale dell’Università, 16, 35020 Legnaro (PD), Italy 4Departamento de Agronomia - Instituto Superior Politécnico do Kwanza sul, Rua 12 Novembro, Sumbe, Angola 5Department of Plant Pathology and Plant-Microbe Biology, 334 Plant Science Building, Cornell University, Ithaca, NY 14853, USA 6INRA, UMR1318, Institut Jean-Pierre Bourgin, INRA-Versailles 78026 Versailles cedex, France 7INRA, UR407, Station de Pathologie Végétale, INRA-Avignon, 84140 Montfavet cedex, France *Present address: Centre for Agro-Environmental Innovation (AGROINNOVA), Università degli Studi di Torino, Via Leonardo da Vinci 44, 10095 Grugliasco (TO), Italy

SUMMARY (Briddon and Markham, 1995). The majority of bego- moviruses have bipartite genomes, termed DNA-A and Seventeen cassava plants showing strong cassava mo- DNA-B (Stanley et al., 2004). The most recent taxo- saic disease symptoms were sampled from three Angolan nomic approach recognises seven African cassava mosa- provinces. The DNA-A component of two isolates was ic geminivirus species: African cassava mosaic virus amplified by rolling circle amplification (RCA) and char- (ACMV), East African cassava mosaic virus (EACMV), acterized. The complete DNA-A of 2,778 nucleotides East African cassava mosaic Cameroon virus (EACMCV), from sample AOS showed very high similarity to African East African cassava mosaic Kenya virus (EACMKV), cassava mosaic virus (ACMV), and that of 2,799 nu- East African cassava mosaic Malawi virus (EACMMV), cleotides from sample AO7 showed high identity to East East African cassava mosaic Zanzibar virus (EACMZV), African cassava mosaic virus (EACMV), particularly to and South African cassava mosaic virus (SACMV) (Fau- EACMV-Uganda2 Severe strain. This represents the first quet et al., 2008). report of complete ACMV and EACMV DNA-A se- After the first report of the severe form of CMD in quences from Angola. PCR followed by direct sequenc- Uganda, a novel recombinant named ing of PCR products of a portion of DNA-A and DNA-B EACMV-Uganda2 Severe, which we will abbreviate as detected ACMV or EACMV in all 17 cassava plants. EACMV-UG, has been associated with the severe CMD None of the five other cassava mosaic virus species were pandemic (Deng et al., 1997; Zhou et al., 1997). found. The presence of a mixed infection of ACMV and EACMV-UG was found either singly or in mixed infec- EACMV-Uganda2 Severe strain, associated with cassava tion with ACMV (Harrison et al., 1997; Pita et al., 2001; mosaic disease pandemic, highlights the need for stricter Ntawuruhunga et al., 2007). The EACMV-UG genome controls on the importation of infected cassava stem cut- is distinguished from that of other EACMV strains by tings from neighbouring countries and further investiga- the presence of a region in the coat protein (CP) AV1 tions on the sanitary status of cassava plants in the gene of ca. 500 bp with high sequence identity to the presently unsurveyed provinces of Angola. corresponding part of the ACMV genome. Angola is one of the biggest cassava producers in Key words: Begomovirus, RCA, DNA sequencing, southern Africa. Food security of the country is influ- phylogenetic analyses. enced by the success of cassava production, which is di- rectly related with the presence of CMD and crop pro- Cassava mosaic disease (CMD) is the most important tection programmes. To our knowledge, there are only viral disease affecting cassava (Manihot esculenta two reports on viruses associated with CMD in Angola. Crantz) in Africa (Thresh et al., 1994; Legg et al., 2006). The presence of EACMV from one cassava plant in CMD is caused by several -transmitted viruses southern Angola (Okavango region) was reported by belonging to the genus Begomovirus (family Geminiviri- Berry and Rey (2001). More recently, using ELISA and dae) which have circular, single-stranded DNA genomes species-specific PCR, ACMV was determined to be the predominant species, followed by EACMV and EACMV-UG from the Cuanza Norte, Malanje and Uige Corresponding author: S. Matic Fax: +39.011.6709307 provinces of Angola (Lava Kumar et al., 2008). Despite E-mail: [email protected]; [email protected] these studies, there is still a lack of data regarding the 021_JPP1146SC(Matic)_443 20-07-2012 10:16 Pagina 444

444 Viruses associated with cassava mosaic disease in Angola Journal of Plant Pathology (2012), 94 (2), 443-450

molecular characterization and genomic variability of EcoRI and BamHI (New England Biolabs, USA). A these viruses in the country, which would greatly assist fragment of approximately 2.8 kb was amplified from in the development of control strategies. The objective all Angolan samples, the size expected for the DNA-A of this study was to analyze the presence of viruses in of . The 2.8 kb-band from sample AOS cassava plants exhibiting CMD symptoms, and to esti- (obtained with EcoRI) and AO7 (obtained with BamHI) mate the genetic diversity among the virus isolates asso- was excised and purified using the QIAquick Gel Ex- ciated with the disease in three Angolan provinces. traction Kit (Qiagen, USA). Purified products were Seventeen cassava plants showing strong CMD symp- cloned into the EcoRI or BamHI site of a previously di- toms were sampled from farms in Cabo Ledo (Bengo gested pBluescript KS (+) vector (Stratagene, USA), province), Canjala (Benguela province), and Porto-Am- then transformed into Escherichia coli DH5α cells by boim, Sumbe and Uku Seles (Cuanza Sul province) dur- electroporation. Plasmids containing the desired inserts ing January 2009 (Table 1, Fig. 1). CMD symptoms were extracted by Jetprep Plasmid Miniprep Kit were observed in all surveyed orchards. Plants were (Genomed GmbH, Germany). DNA sequences of the stunted, and leaves were mottled, distorted, and smaller. selected recombinant plasmids were obtained by auto- Samples were taken from different native and imported matic sequencing at BMR (Padova, Italy). Multiple se- cultivars, such as Precose de Angola, Mundele Paço, quence alignment of nucleotide (nt) and amino acid (aa) Ngana rico, Thiti, TMS 30142, and TMS 40125. sequences and identification of open reading frames Total nucleic acid (TNA) was extracted from 100 mg (ORFs) were done using the program AlignX (Vector of tissue (leaf or phloem scrapings) according to Foissac NTI Suite V 5.5, InforMax, USA) with the Clustal W et al. (2001). Circular DNA was amplified using the Il- algorithm (Thompson et al., 1994). Detection of possi- lustra TempliPhi 100 Amplification Kit (GE Healthcare, ble recombinants was carried out using the RDP3 pro- USA) following the manufacturer’s instructions. RCA gram (Martin et al., 2010). Phylogenetic analysis was products were digested by the restriction endonucleases performed using PAUP 4.0 (Swofford, 2001), with the

Fig. 1. A map of Angola showing the localities in Bengo, Cuanza Sul and Benguela provinces where cassava plants were sampled. 021_JPP1146SC(Matic)_443 20-07-2012 10:16 Pagina 445

Journal of Plant Pathology (2012), 94 (2), 443-450 Matic et al. 445

Table 1. Angolan cassava isolates used for molecular characterization by RCA and PCR.

ACMV EACMV Isolate Accession No. Accession No. Accession No. Accession No. Province code DNA-A DNA-B DNA-A DNA-B (AC2/AC1) (BV1/BC1) (AC2/AC1) (BC1) AO1 Bengo GU580899 AO2 Bengo GU580900 GU580925 AO3 Bengo GU580901 GU580915 GU580926 AO4 Bengo GU580911 AO5 Bengo GU580902 GU580916 GU580927 AO7 Bengo GU580903 JN941177* AO9 Bengo GU580904 AO10 Bengo GU580905 AOUN Benguela GU580910 GU580934 AOA Cuanza Sul GU580906 GU580917 GU580928 AOE Cuanza Sul GU580918 GU580912 GU580929 AOH Cuanza Sul GU580907 GU580919 GU580930 AOJ Cuanza Sul GU580908 GU580920 AOK Cuanza Sul GU580921 GU580913 GU580931 AON Cuanza Sul GU580909 GU580922 GU580932 AOO Cuanza Sul GU580923 GU580914 GU580933 AOS Cuanza Sul GU580897* GU580924 *Sequences in bold were obtained by RCA and subsequent cloning, all other sequences were obtained by PCR.

best model being selected by qModeltest (Posada, AOS clone (accession No. GU580897) revealed a very 2009). high identity to ACMV DNA-A, showing the highest nt A set of primers (ACMV21for and ACMV21rev) was identity (96%) with an ACMV isolate from Uganda designed for the detection of all known African cassava (ACMV-[UG:Mpi:56:02], accession No. AM502339). begomoviruses (Table 2). They were based on the se- The full-length genomic DNA-A of ACMV-AOS was quences of all isolates available in GenBank correspon- 2,778 nt long. It comprised six ORFs, two (AV2 and ding to nts 1,447-1,998 of DNA-A of ACMV isolate AV1) in the virion sense and four (AC3, AC2, AC1 and GU580897 as determined by BLASTx. Since these AC4) in the complementary sense that were separated primers allowed the amplification of DNA-A sequences by an intergenic region (IR) containing the conserved of all known African cassava begomoviruses, specific nona-nucleotide motif (TAATATTAC) essential for the primers were designed for the discrimination between initiation of rolling circle replication (Chatterji et al., ACMV and EACMV in DNA-B: (i) ACMVB1 and 1999; Rojas et al., 2005). ACMVB2 for ACMV, corresponding to nts 746-1,373 of The sequenced full-length DNA-A of the AO7 clone isolate AF259895; and (ii) EACMVB1 and EACMVB2 (2,799 nt, accession No. JN941177) showed high identi- for EACMV, corresponding to nts 1,343-1,969 of isolate ty to EACMV DNA-A. This sequence contained AV2 AF230375 (Table 2). and AV1 genes in the virion sense, and AC5, AC3, AC2, PCR was done with AccuPrime™ Taq DNA Poly- AC1 and AC4 genes in the complementary sense. It merase (Invitrogen, USA) according to the manufactur- showed the highest nt identity (99%) with an EACMV er’s instructions. PCR conditions were 94°C for 5 min isolate from cassava in Burkina Faso (accession No. for denaturation, followed by 35 cycles at 94°C for 1 FM877474) and an EACMV isolate from Pueraria ja- min, 55°C for 1 min and 72°C for 1 min, and a final ex- vanica from the Democratic Republic (DR) of Congo tension step of 5 min at 72°C. PCR products were (accession No. FN435279), which both belong to the analysed on a 1.2% agarose/1X TAE gel stained with severe EACMV-UG strain. Furthermore, EACMV- ethidium bromide, purified and extracted using the AO7, when compared with the reference Z83257 isolate Jetquick Gel Extraction Spin Kit (Genomed GmbH, of EACMV-UG, showed 99% nt identity in the recom- Germany) and sequenced directly at BMR (Padova, binant CP gene, which confirmed its affiliation to Italy). Sequenced PCR products were deposited in Gen- EACMV-UG. Bank with accession numbers shown in Table 1. In analysing possible recombination events, there was One RCA clone each was obtained from samples evidence of multiple breakpoints in EACMV-AO7, with AOS and AO7. BLAST analysis of the sequence of the all but one of the putative parents derived from the 021_JPP1146SC(Matic)_443 20-07-2012 10:16 Pagina 446

446 Viruses associated with cassava mosaic disease in Angola Journal of Plant Pathology (2012), 94 (2), 443-450

principal clades of EACMV. According to the RDP3 A primer pair (ACMV21for/rev) was designed from program, a region from 544 to 1,008 nts of the genome the sequences of a portion of DNA-A (AC3-AC2-AC1) could plausibly have been derived from the ACMV iso- of all African cassava mosaic begomoviruses (Table 2), late AF126800 (Fig. 2a). Interspecies genetic exchange which allowed the detection of ACMV or EACMV in is also in evidence for ACMV-AOS, in which a short all 17 Angolan samples after sequencing of the 467 bp stretch from 2,123 to 2,298 nts was found to be closest PCR products, including a portion of AC2-AC1 (Table to EACMKV isolate AJ717582 (Fig. 2b). These pro- 1). Sequencing PCR products confirmed the presence posed recombinant structures were a signature for all of ACMV in sample AOS, whereas it detected ACMV isolates within the same monophyletic group as shown in sample AO7, in which EACMV had already been in Fig. 3. found by sequencing the RCA product. In the 15 re-

Fig. 2. Evidence of recombination in isolates (a) EACMV-AO7 (accession no. JN941177) and (b) ACMV-AOS (accession No. GU580897). A simplified linearized map of each isolate is depicted with the position of the crossover nucleotides for each puta- tive recombination region (light grey boxes) indicated. The floating boxes below correspond to the minor parent donor, the name of which is placed alongside or underneath for reference. The putative major parent donor names are placed above their associat- ed fragments. The probability values for the five main methods employed by the RDP3 program are shown in the boxes with the relevant putative minor donor fragment (when relevant) above each column. Note that for ACMV-AOS no major parent was iden- tified. Given the circular structure of the genomes the donation of AF259896_CIV_EACMCV to EACMV-AO7, though depicted as fragmented is continuous through nts 2799 and 1. 021_JPP1146SC(Matic)_443 20-07-2012 10:16 Pagina 447

Journal of Plant Pathology (2012), 94 (2), 443-450 Matic et al. 447

Fig. 3. Neighbour-joining phylogenetic tree of the full-length DNA-A sequences of Angolan ACMV and EACMV virus isolates (highlighted) at the nucleotide level, compared to isolates of the other African cassava mosaic virus species. Numbers at the branches refer to the bootstrap values for 100 replicates, which if below 70 were collapsed. Distance (indicated by the bar) refers to the number of nucleotide changes per base. Three letter codes at the end of each accession reference refers to the country of origin: Angola-AGO, Burkina Faso-BFA, Cameroon-CMR, Democratic Republic of Congo-COD, Ivory Coast-CIV, Kenya-KEN, Madagascar-MDG, Malawi-MWI, Nigeria-NGA, South Africa-ZAF, Tanzania-TZA, Uganda-UGA, Zimbabwe-ZWE. Groupings for each virus species is shown by the brackets on the left, with all ‘East African’ species (EACMV, EACMCV, EACMKV, EACM- MV, EACMZ) grouped as EACMV. Relevant putative recombinant donor molecules (major parents large and bold) for EACMV- AO7 and ACMV-AOS, as determined by RDP3 (see Fig. 2) are marked with a star and hatch respectively. Cross hatches on the branches mark monophyletic groups with similar proposed recombination structures, as depicted in Fig. 2. 021_JPP1146SC(Matic)_443 20-07-2012 10:16 Pagina 448

448 Viruses associated with cassava mosaic disease in Angola Journal of Plant Pathology (2012), 94 (2), 443-450

Table 2. Primers used for characterization of Angolan cassava isolates affected by Cassava mosaic disease.

Gene target Expected size of Target Primer (5’-3’) region PCR amplicon viruses ACMV21for: Seven African (GCAGTGATGAGTTCCCCGGTGCG) sense AC3/AC2/AC1 552 bp cassava mosaic ACMV21rev: virusesa (ATTCCGCTGCGCGGCCATGGAGACC) antisense

ACMVB1: (TCGGGAGTGATACATGCGAAGGC) sense BV1/BC1 628 bp ACMV ACMVB2: (GGCTACACCAGCTACCTGAAGCT) antisense

EACMVB1: (GAGGTCCTTTCTGGATACTGAGTCGG) sense BC1 627 bp EACMV EACMVB2: (GACAATCGCGTCAGCCTTGAAGACG) antisense aAfrican cassava mosaic virus (ACMV), East African cassava mosaic virus (EACMV), East African cassava mosaic Cameroon virus (EACMCV), East African cassava mosaic Kenya virus (EACMKV), East African cassava mosaic Malawi virus (EACMMV), East African cassava mosaic Zanzibar virus (EACMZV), and South African cassava mosaic virus (SACMV).

maining Angolan samples, ACMV was detected in 11, (Cuanza Sul and Benguela) provinces of Angola, which and EACMV in four samples. None of the five other showed the greatest identity with an EACMV-UG cassava mosaic virus species were found. strain. The pandemic of severe CMD has spread from DNA-B primer pairs were designed for specific de- east and central Africa to west African countries (Ndun- tection of ACMV (ACMVB1/B2) or EACMV guru et al., 2005; Legg et al., 2006; Patil and Fauquet, (EACMVB1/B2) (Table 2). This allowed detection of 2009). DR Congo, which neighbours Angola on the DNA-B of either virus in 12 Angolan samples, which north, has already been affected by the CMD epidemic, was confirmed with direct sequencing of the PCR prod- and EACMV-UG is present there (Monde et al., 2010). ucts (Table 1). Surprisingly, the DNA-B-specific primers Furthermore, the diffusion of EACMV-UG towards failed to detect virus in five samples; probably conse- West Africa was evidenced from Burkina Faso by Tien- quent to genetic variability within the targeted DNA-B drébéogo et al. (2009). Since EACMV-AO7 from Ango- segment. la shows the highest identities with EACMV-UG from Both sets of primer pairs revealed the presence of DR Congo and Burkina Faso, it might further confirm ACMV and EACMV in single (six isolates) and mixed the route of CMD pandemic movement from East to infections (eleven isolates) (Table 1). ACMV was pre- West Africa, either by long-distance distribution dominant in single infections. In some mixed infections, through infected stem cuttings, or by whitefly vector ACMV and EACMV presence was detected only for the transmission. As EACMV-AO7 belongs to EACMV-UG DNA-A or DNA-B component, which might be evi- and shares 98-100% nt identity with other characterized dence of trans-replication between ACMV DNA-A and Angolan EACMV isolates, it suggests that all the An- EACMV DNA-B, as already reported in natural infec- golan EACMV isolates could be variants of this strain, tions (Pita et al., 2000). but this can only be determined with certainty by se- Phylogenetic analysis of nt (Fig. 3) or aa sequences quencing the CP of these isolates. Our results confirm (not shown) of the full-length DNA-A of ACMV-AOS the CMD pandemic development estimation done by and EACMV-AO7, together with all representative Legg et al. (2006), where northern Angola was predict- African cassava mosaic begomovirus sequences avail- ed to be affected by CMD. able in GenBank, grouped ACMV-AOS with other The EACMV isolates characterized in this work ACMV isolates, while EACMV-AO7 grouped within could represent recently established virus populations EACMV, and was closest to an EACMV-UG isolate originating from the recombinant EACMV-UG, while from Burkina-Faso. the EACMV isolate described by Berry and Rey (2001) Previous reports have demonstrated the presence of may represent a distinct introduction event. In contrast, ACMV and EACMV in Angola (Berry and Rey, 2001; the Angolan ACMV isolates identified here had highest Lava Kumar et al., 2008), which we have further con- identities with isolates originating from different coun- firmed in this work. We have sequenced EACMV iso- tries (Uganda, Burkina Faso, DR Congo, Nigeria, and lates from north-western (Bengo) and central-western Ivory Coast) which could have resulted from different 021_JPP1146SC(Matic)_443 20-07-2012 10:16 Pagina 449

Journal of Plant Pathology (2012), 94 (2), 443-450 Matic et al. 449

introduction events or from the fact that ACMV was in- Identification of replication specificity determinants in two troduced in Angola a long time ago during which evolu- strains of tomato leaf curl virus from New Delhi. Journal of tionary divergence occurred, causing its higher genetic Virology 73: 5481-5489. diversity. Deng D., Otim-Nape G.W., Sangare A., Ogwal S., Beachy The high co-infection rate of ACMV and EACMV- R.N., Fauquet C.M., 1997. Presence of a new virus closely UG found in this study (65%) fits well with the findings associated with cassava mosaic outbreak in Uganda. reported from neighbouring DR Congo (Monde et al., African Journal of Root and Tuber Crops 2: 23-28. 2010) and the Republic of Congo (Ntawuruhunga et al., Fauquet C.M., Briddon R.W., Brown J.K., Moriones E., Stanley 2007). It was reported that after establishment of epi- J., Zerbini M., Zhou X., 2008. Geminivirus strain demarca- tion and nomenclature. Archives of Virology 153: 783-821. demic CMD, ACMV was replaced by the more virulent EACMV-UG, and that the proportion of mixed infec- Foissac X., Svanella-Dumas L., Gentit P., Dulucq M.J., Can- dresse T., 2001. Polyvalent detection of fruit tree tricho-, tions was significantly reduced (Legg and Thresh, 2000; capillo- and foveaviruses by nested RT-PCR using degener- Legg et al., 2006). The high presence of mixed infec- ated and inosine containing primers (PDO-RT-PCR). Acta tions and the prevalence of ACMV in single infections Horticulturae 550: 37-43. we found in Angola could indicate a still recent intro- Harrison B.D., Zhou X., Otim-Nape G.W., Liu Y., Robinson duction of EACMV-UG to Angola that has not yet dis- D.J., 1997. Role of a novel type of double infection in the placed ACMV. geminivirus-induced epidemic of severe cassava mosaic in These findings highlight the urgent need for large- Uganda. Annals of Applied Biology 131: 437-448. scale field surveys of cassava affected by CMD in the Lava Kumar P., Akinbade S.A., Dixon A.G.O., Mahungu presently unsurveyed provinces of Angola, along with N.M., Mutunda M.P., Kiala D., Londa L., Legg J.P., 2008. epidemiological studies to clarify the significance of First report of the occurrence of East African cassava mosa- mixed infections between ACMV and EACMV-UG and ic virus-Uganda (EACMV-UG) in Angola. New Disease Re- the role of whitefly in the spread of CMD. The introduc- ports 18: 20. http://www.bspp.org.uk/publications/new- tion of resistant cultivars in combination with rigorous disease-reports/ndr.php?id=018020. phytosanitary measures have already been used to com- Legg J.P., Kapinga R., Teri J., Whyte J.B.A., 1999. The pan- bat the disease in other pandemic-affected countries of demic of cassava mosaic virus disease in East Africa: con- Africa (Legg et al., 1999). The control of importation of trol strategies and regional partnerships. Roots 6: 10-19. infected cassava material is of crucial importance, since Legg J.P., Thresh J.M., 2000. Cassava mosaic virus disease in the movement of infected stakes increases the risk of the East Africa: a dynamic disease in a changing environment. introduction of new cassava mosaic viruses, and new Virus Research 71: 135-149. mixed infections, thus increasing the likelihood of recom- Legg J.P., Fauquet C.M., 2004. Cassava mosaic geminiviruses bination and reassortment between existing viruses. in Africa. Plant Molecular Biology 56: 585-599. In conclusion, cassava begomovirus diversity detect- Legg J.P., Owor B., Sseruwagi P., Ndunguru J., 2006. Cassava ed so far in Angola is consistent with the diversity mosaic virus disease in East and Central Africa: epidemiol- ogy and management of a regional pandemic. Advances in known in neighbouring countries and suggests several Virus Research 67: 355-418. introduction events of ACMV and a recent introduction Martin D.P., Lemey P., Lott M., Moulton V., Posada D., of EACMV-UG. Lefeuvre P., 2010. RDP3: a flexible and fast computer pro- gram for analyzing recombination. Bioinformatics 26: 2462- 2463. ACKNOWLEDGEMENTS Monde G., Walangululu J., Winter S., Bragard C., 2010. Dual infection by cassava begomoviruses in two leguminous The authors wish to thank Dr. E. Noris and Dr. G.P. species (Fabaceae) in Yangambi, Northeastern Democratic Accotto (Istituto di Virologia Vegetale del CNR, Torino, Republic of Congo. Archives of Virology 155: 1865-1869. Italy) for final discussion and support in this work. Ndunguru J., Legg J.P., Aveling T.A.S., Thompson G., Fau- quet C.M., 2005. Molecular biodiversity of cassava bego- moviruses in Tanzania: evolution of cassava geminiviruses REFERENCES in Africa and evidence for East Africa being a center of di- versity of cassava geminiviruses. Virology Journal 2: 21. Berry S., Rey M.E.C., 2001. Molecular evidence for diverse Ntawuruhunga P., Okao-Okuja G., Bembe A., Obambi M., populations of cassava-infecting begomoviruses in south- Armand Mvila J.C., Legg J.P., 2007. Incidence and severity ern Africa. Archives of Virology 146: 1795-1802. of cassava mosaic disease in the Republic of Congo. Briddon R.W., Markham P.G., 1995. . In: Mur- African Crop Science Journal 15: 1-9. phy F.A., Fauquet C.M., Bishop D.H.L., Ghabrial S.A., Owor B., Legg J.P., Okao-Okuja G., Obonyo R., Ogenga- Jarvis A.W., Martelli G.P., Mayo M.A., Summers M.D. Latigo M.W., 2004. The effect of cassava mosaic gemi- (eds). Virus Taxonomy. Sixth Report of the ICTV, pp. 158- niviruses on symptom severity, growth and root yield of a 165. Springer-Verlag, Vienna, Austria. cassava mosaic virus disease-susceptible cultivar in Ugan- Chatterji A., Padidam M., Beachy R.N., Fauquet C.M., 1999. da. Annals of Applied Biology 145: 331-337. 021_JPP1146SC(Matic)_443 20-07-2012 10:16 Pagina 450

450 Viruses associated with cassava mosaic disease in Angola Journal of Plant Pathology (2012), 94 (2), 443-450

Patil B.L., Fauquet C.M., 2009. Cassava mosaic geminiviruses: Swofford D.L., 2001. PAUP*. Phylogenetic Analysis Using actual knowledge and perspectives. Molecular Plant Pathol- Parsimony (*and Other Methods). Version 4. Sinauer As- ogy 10: 685-701. sociates, Sunderland. Pita J.S., Fondong V.N., Sangare A., Otim-Nape G.W., Ogwal Thompson J.D., Higgins D.G., Gibson T.J., 1994. CLUSTAL S., Fauquet C.M., 2001. Recombination, pseudorecombi- W: Improving the sensitivity of progressive multiple se- nation and synergism of geminiviruses are determinant quence alignment through sequence weighting, positions- keys to the epidemic of severe cassava mosaic disease in specific gap penalties and weight matrix choice. Nucleic Uganda. Journal of General Virology 82: 655-665. Acids Research 22: 4673-4680. Posada D., 2009. Selection of models of DNA evolution with Thresh J.M., Fishpool L.D.C., Otim-Nape G.W., Fargette D., jModelTest. Methods in Molecular Biology 537: 93-112. 1994. African cassava mosaic virus disease: an underesti- Rojas A., Kvarnheden A., Marcenaro D., Valkonen J.P.T., mated and unsolved problem. Tropical Science 34: 3-14. 2005. Sequence characterization of Tomato leaf curl Tiendrébéogo F., Lefeuvre P., Hoareau M., Traoré V.S.E., Bar- Sinaloa virus and Tomato severe leaf curl virus: phylogeny ro N., Reynaud B., Traoré A.S., Konaté G., Traoré O., Lett of New World begomoviruses and detection of recombina- J.M., 2009. Occurrence of East African cassava mosaic tion. Archives of Virology 150: 1281-1299. virus-Uganda (EACMV-UG) in Burkina Faso. Plant Stanley J., Bisaro D.M., Briddon R.W., Brown J.K., Fauquet Pathology 58: 783. C.M., Harrison B.D., Rybicki E.P., Stenger D.C., 2004. Zhou X., Liu Y., Calvert L., Munoz C., Otim-Nape G.W., Geminiviridae. In: Fauquet C.M., Mayo M.A., Maniloff J., Robinson D.J., Harrison B.D., 1997. Evidence that DNA- Desselberger U., Ball L.A. (eds). Virus Taxonomy. Eighth A of a geminivirus associated with severe cassava mosaic Report of the ICTV, pp. 301-326. Elsevier/Academic disease in Uganda has arisen by interspecific recombina- Press, London, UK. tion. Journal of General Virology 78: 2101-2111.

Received February 18, 2012 Accepted April 6, 2012