Epidemiological assessment of cassava mosaic disease in Burkina Faso

Abstract Surveys were conducted in 2016 and 2017 across the main cassava‐growing regions of Burkina Faso to assess the status of cassava mosaic disease (CMD) and to determine the virus strains causing the disease, using field observation and phylogenetic analysis. CMD incidence varied between regions and across years but was lowest in Hauts‐Bassins (6.0%, 2016 and 5.4%, 2017) and highest in Centre‐Sud (18.5%, 2016) and in Boucle du Mouhoun (51.7%, 2017). The lowest CMD severity was found in Est region (2.0) for both years and the highest in Sud‐Ouest region (3.3, 2016) and Centre‐Sud region (2.8, 2017). The CMD infection was primarily associated with contaminated cuttings in all regions except in Hauts‐Bassins, where whitefly‐borne infection was higher than cuttings‐borne infection in 2016. PCR screening of 687 samples coupled with sequence analysis revealed the presence of African cassava mosaic‐like (ACMV‐like) viruses and East African cassava mosaic‐like (EACMV‐like) viruses as single infections at 79.5% and 1.1%, respectively. Co‐infections of ACMV‐like and EACMV‐like viruses were detected in 19.4% of the tested samples. In addition, 86.7% of the samples positive for EACMV‐like virus were found to be positive for East African cassava mosaic Cameroon virus (EACMCMV). Phylogenetic analysis revealed the segregation of cassava mosaic geminiviruses (CMGs) from Burkina Faso into three clades specific to ACMV, African cassava mosaic Burkina Faso virus (ACMBFV), and EACMCMV, confirming the presence of these viruses. The results of this study show that EACMCMV occurrence may be more prevalent in Burkina Faso than previously thought.


| INTRODUC TI ON
Cassava (Manihot esculenta, Euphorbiaceae), which originates from Latin America, is a major source of food for more than 700 million people in tropical and subtropical developing countries and enhances food security in these countries (Ntawuruhunga et al., 2013;Patil & Fauquet, 2009;Saediman et al., 2016). Cassava is a staple food crop in the sub-Saharan region of Africa and consequently a source of income for many processors and traders (Ntawuruhunga et al., 2013). The high calorie yield per hectare (250 kcal/ha/day), drought tolerance, hardiness in stressful environments, and flexibility of harvesting time are the major advantages of this crop compared to many other crops (Byju & Suja, 2020;El-Sharkawy, 2004;Pushpalatha & Gangadharan, 2020). In Burkina Faso, cassava was introduced by farmers decades ago from Ghana and Ivory Coast (Côte d'Ivoire) (Guira et al., 2017). It has long been cultivated around vegetable gardens for domestic consumption. Formerly considered as a neglected crop, cassava has become a cash crop since the formal introduction of improved varieties from IITA in 2003 (Dabiré & Belem, 2003).
Indeed, the presence of ACMV was reported in 1995 using the triple antibody sandwich-ELISA method with cross-reacting monoclonal antibodies to ACMV (Konaté et al., 1995). The molecular features of an ACMV-like virus (ACMBFV, whose Rep protein gene and intergenic region differ from ACMV) was described and the presence of EACMV-UG variant was reported around Ouagadougou (Tiendrébéogo et al., 2009(Tiendrébéogo et al., , 2012. Since then, the real status of CMD and its epidemiological parameters such as the incidence and severity of the disease, the whitefly abundance, and the mode of infection remain unclear. To overcome this knowledge gap, we conducted for the first time georeferenced surveys in the main cassava production areas in Burkina Faso.

| Cassava mosaic disease status assessment
Surveys were conducted in 2016 and 2017 in eight major cassavagrowing regions of Burkina Faso ( Figure 1). The number of fields sampled within a region depended on the number of cassavagrowing localities and the availability of cassava fields at 3-6 months after planting (MAP). Our field sampling protocol was a modification of one previously described (Sseruwagi et al., 2004) and has been adopted by 10 countries in Central and West Africa to harmonize efforts at surveillance and monitoring of these transboundary pathogens of high economic importance. Briefly, in each field, 30 cassava plants were assessed randomly along two diagonals to form an "X" pattern. Then each selected plant was assessed visually for the presence or absence of CMD symptoms (leaf mosaic, leaf distortion, and stunted growth) and the number of whiteflies settling on the leaves, and if infected, we determined whether the mode of infection was through cuttings or whitefly transmission. The whitefly population was estimated by counting the number of whiteflies on the top five fully expanded leaves. The mode of infection in each plant was determined based on the location of the leaves with symptoms as previously described by Sseruwagi et al. (2004). According to these authors, from 3 to 6 MAP it is possible to distinguish between cutting-borne and whitefly-borne infections. Symptoms appearing only on upper leaves were taken to have resulted from whiteflytransmitted infection, whereas plants that showed symptoms either only on the lower leaves or on all leaves were taken as having been infected through cassava cuttings. CMD symptom severity was recorded using a scale from 1 (no symptoms) to 5 (very severe symptoms) (Terry, 1975). We acknowledge that the severity level depends on the variety, climate, crop management, and mainly the time at which the infection occurred. To minimize the effects of these variables on our data, we sampled fields within the same locations that were within the 3-6 MAP age. The CMD incidence was calculated

| Molecular characterization of CMGs
Total DNA was extracted from cassava leaves using the CTAB protocol as previously described (Permingeat et al., 1998). The concentration of DNA in each sample was determined using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific) and adjusted to 150 ng/μl. We previously discovered that the most problematic CMGs in smallholder cassava production in Burkina Faso were ACMV and a variant of the East African cassava mosaic virus (EACMV), the EACMV-Uganda variant (Tiendrébéogo et al., 2009). Because the current status of the incidence and severity of these two CMDcausing viruses is unknown in Burkina Faso, we surveyed the whole country and focused our surveys on ACMV and EACMV. The extracted DNA was subjected to PCR using the specific primers listed in Table 1 for the detection of ACMV-like virus (JSP001/JSP002) and EACMV-like virus (JSP001/JSP003). The samples positive for EACMV-like virus were subjected to another round of PCR using specific primers for the detection of EACMCMV (VNF031/VNF032; Table 1). The PCR mix was prepared in a final volume of 25 μl using 20.9 μl of molecular biology grade water, 2.5 µl of 10× reaction  to determine their identity. PCR products of 15 EACMCMV positive samples were also subjected to sequencing in both forward and reverse orientations to confirm their identity.

| Statistical analysis
Data analysis was performed using the R software v. 3.6.1 (R

| Phylogenetic analysis
The amplicon sequences were trimmed and assembled de novo using Geneious v. 8.1.7 (Biomatters Ltd) software. Consensus sequence obtained from forward and reverse sequences for each sample was subjected to BLASTn in NCBI for preliminary species assignment and subsequently for pairwise sequence comparison (Bao et al., 2014).
The sequences were aligned with representative isolates of begomoviruses using ClustalW alignment method in MEGA X software (Kumar et al., 2018).

| Incidence and symptom severity of CMD in 2016 and 2017
The CMD incidence in 2016 varied significantly from that observed in 2017 (p ≤ 0.001). In 2016, the overall CMD incidence across the surveyed fields in Burkina Faso was 11.3% (216/1920) and ranged from 6.0% (36/600) in Hauts-Bassins region to 18.5% (50/270) in Centre-Sud region. In 2016, the difference between the lowest incidence and the highest incidence was highly significant (p ≤ 0.001). For the 2017 survey, the overall CMD incidence was 18.9% (329/1720). The

| Adult whitefly distribution and mode of infection
Determination of whitefly counts and distribution was conducted  (Mugerwa et al., 2021). The challenge is that during our survey years, whitefly pressure was not strong enough to explain the significant differences in disease incidence observed between regions and years. Also, considering that high incidence is associated with seeding using infected cuttings, we are not able to without doubt correlate whitefly numbers with the disease incidence or severity. (3.6%, 1/28), Cascades (2.3%, 3/132) and Centre-Sud (1.5%, 1/68) but no significant difference was found between these proportions (Table 3). Of the 90 EACMV-like virus positive samples (single and mixed infections), 86.7% (78/90) tested positive for EACMCMV using the primer pair VNF031/VNF032. The relatively lower incidence recorded in Burkina Faso could be explained by the fact that the intensification of cassava production is a more recent phenomenon in the country compared to other African countries (Guira et al., 2017;Legg et al., 2006), coupled with CMD awareness and the adoption of good farming practices by farmers.

| CMGs identity confirmed by sequencing
Our results showed that cases of CMD transmitted by cassava cuttings were more prevalent as compared to cases resulting from whitefly transmission. This phenomenon appears to be widespread in sub-Saharan Africa (Chikoti et al., 2013;Mulenga et al., 2016;Mwatuni et al., 2015;Torkpo et al., 2018;Zinga et al., 2013 and EACMCMV. We propose further analysis, such as the use of specific primers for each CMG species or next-generation sequencing, to resolve the issue of the occurrence of CMG species and strains in Burkina Faso. We detected the occurrence of CMGs in symptomless samples (2.1%), which shows that the viruses can be latent in the plants without manifesting symptoms. Therefore, the use of symptomless cassava landraces as an option to manage CMD could inadvertently result in increased cutting-borne transmission because they may harbour CMGs. We propose that the use of certified virus-free cuttings for the establishment of new cassava fields will be crucial for fighting the transmission of CMD. In the absence of certified virus-free cuttings, the training of farmers on how to select healthy cuttings for the new planting season and on use of in-field diagnostic applications will be crucial to bring down the incidence or transmission of these viruses of high economic importance.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.