Cassava Mosaic Virus Disease in Ghana: Distribution and Spread

Background: Cassava is an important staple crop in most of the tropics including Ghana. The productivity of the crop is beset with pest and disease attacks. With the emergence of virulent strains of the cassava mosaic virus (CMV), regular surveys are necessary to ascertain the prevalence of CMV and their whitey vectors in farmers’ elds to help manage CMV disease affecting the crop. Methods: Field surveys were conducted in September and October of 2015 and December 2016 to January 2017 using a harmonized sampling protocol developed by the West African Virus Epidemiology (WAVE) for root and tuber project. Three hundred and ninety-three elds were visited throughout Ghana and 11,760 cassava leaf samples examined. Whiteies were counted on 5 plants/eld. Diseased samples with varying symptoms collected were assayed using PCR and genomic sequencing. Results: Cassava mosaic disease (CMD) symptoms were recorded in about ninety-six percent (96.4%) of elds surveyed with varying severity. These symptoms included leaf mosaic, leaf distortion/twisting, malformation, liform leaves, stunting and chlorosis. Cultivars with red petiole colour were the most prevalent while those with green petiole colour were the least. No whitey was found on cultivars with purple and Green petioles while cultivars with reddish-green petioles had highest count of whiteies/plant. The Upper West and Upper East regions had the least number of whiteies/plant. Six CMV strains were detected: ACMV-Ivory Coast, ACMV-Kenya, EACMV-Cameroun, ACMV-Ghana, EACM-Cameroun virus-Ghana and EACMV-Kenya. Conclusion: ACMV-Kenya and EACMV-Kenya are being reported for the rst time. This indicates that more CMV strains are being detected in the country.


Background
Cassava is the number one root and tuber crop in Ghana (Fig. 1). The crop is cultivated predominantly in the central and southern parts of Ghana with signi cant production in the middle and northern parts of the country. The only areas of the country which do not have signi cant production of cassava are the Savannah and Upper East regions (Fig. 1). The estimated total land under cassava production is estimated around 900,000 hectares. In Ghana, over 70% of farmers engage in cassava production, and the sector contributes about 22% of Agricultural GDP (GEPA, 2017). Ghana ranks among the top ve cassava producers in Africa with an annual average production of sixteen million metric tons (GEPA, 2017).
Its cultivation and associated businesses along the value chain is a major source of employment for millions of the population. It is a staple food crop in areas where it is cultivated and it is processed into several industrial products such as cassava starch, cassava beer, spray starch, pharmaceutical raw material among several others (Koros et al., (2018): https://www.fao.org/3/x5032e/x5032e06.htm). In recent times cassava has become an important crop in the socio-cultural life and the economy of Ghana.
However, the production of this crop faces several challenges. These include dependence on low yielding varieties, poor agronomic practices on the part of farmers and incidence of pests and diseases (Koros et. al., 2018;Thresh and Cooter, 2005). There are several pests and diseases that affect cassava production but cassava mosaic disease (CMD) constitutes the most endemic disease in the country ( and).
To ascertain the current distribution of CMD and the strain diversity of the virus in Ghana it is important that periodic surveys and eld monitoring are conducted. This is to help develop strategies which can curtail the potential damage that new and more virulent strains of the virus could cause to cassava production in the country and beyond.
To this end, two nationwide surveys were conducted to provide baseline data that can be used by breeders and other stakeholders, including modelers, to identify disease hotspot and cold spots for breeding resistant varieties and the multiplication of disease-free planting materials for farmers.
Additionally, this would provide data for predicting the rate of spread of cassava mosaic virus in Ghana and West Africa.

Field Survey
Surveys were conducted throughout the regions indicated in Figure 1.They were carried out in September were kept in herbarium pressers and labeled. In addition, stem cuttings of infected plants were also collected and labeled. The leaf samples were stored in the laboratory under ambient conditions while the stem cuttings were maintained in insect proof screenhouses prior to their laboratory analysis. White y Bemisia tabaci vectors were collected using an aspirator from 5 plants/ eld and the total number counted and recorded. The white ies were preserved in 70% ethanol, labeled and preserved at the laboratory in a freezer. The diversity of cassava genotypes encountered on the elds were differentiated using the leaf petiole colour as most of the farm owners could not readily give the names of the varieties they had planted.

Polymerase chain reaction (PCR)
PCR was carried out in a 25 µl reaction consisting of the following reagents: PCR-grade water; 5 µl of 5x standard buffer; 2.5 µl 5% Tween-20; 0.25 µl of 10 mM dNTPs; 0.25 µl each of Forward and Reverse primers; 5 U Taq polymerase and 5 µl of DNA template. The list of primers used for the analysis can be found in Table 1.

Sequencing
Where the expected bands on the agarose gel were slightly different rom the expected band, recovered PCR products were prepared for sequencing. Sequencing was carried out by GENEWIZ, Inc (South Plain eld, NJ) using Applied Biosystems Big Dye version 3.1. The reactions were then run through an Applied Biosystem 3730xl DNA Analyzer for DNA sequencing.

Data Analysis
For our analysis, we made use of data-analysis software, namely a multi-dimensional data cube referred to, within the WAVE programme, as the WAVE Cube. This tool enabled us to interrogate and view our survey data from different perspectives to present our extracted results as tables and graphics.
Nucleotide sequences obtained from GENEWIZ were used as query and BLAST searched with ACMV and EACMV strains/isolate nucleotide sequences stored in NCBI GenBank database to detect their corresponding strains/isolates. Nucleotide sequence of samples that yielded above 80% identity with those of ACMV and EACMV isolates in the NCBI GenBank database repository were considered similar (Fondong et al., 2000).
Patristic distances between pairs of sequences analyses were conducted using the Maximum Composite Likelihood model (Tamura et al., 2004). This analysis involved 56 nucleotide sequences. All ambiguous positions were removed for each sequence pair (pairwise deletion option). Evolutionary analyses were  (Table 2) showing the endemic nature of CMD in Ghana.
Northern region had the highest incidence of CMD followed by Brong-Ahafo and the Upper West regions respectively (Fig. 2) while Greater Accra had the highest CMD severity followed by Upper West, Ashanti, Eastern, Western and then Brong-Ahafo regions (Fig. 3).
Fields with 50% to 75% disease incidence had the highest number of white y populations while those with disease incidence greater than 75% had the least white y population (Fig. 7). Ashanti region had the highest white y population/plant followed by Central and the Western regions respectively (Table 3).
Dominant varieties encountered were cultivars with red petioles followed by those with yellowish green petiole (Table 4). However, farmers could not identify the varieties by their actual names and most of them classi ed the plants as local varieties. Varieties with yellowish green petioles had the highest disease severity, followed by varieties with red petioles while plants with yellowish green petioles with pinkish colouration had the lowest severity (Table 4). No white ies were collected on plants with purple and Green petioles encountered during the eld surveys (Table 4).

Laboratory Diagnostics
The following strains of ACMV and EACMV indicated in the phylogenetic tree (Fig 8)  elds, but our work has gone beyond these geographical areas and numbers. The current study represents a large increase in the number of elds surveyed in recent times and gives a better picture of the extent of spread of CMD in Ghana. This result also gives indication as to the levels of productivity lost by farmers, assuming a 10% yield reduction to CMD infection on farmer elds. This nding requires that urgent steps be taken to nd a robust remedy to reduce the incidence of CMD in farmers' elds to improve cassava productivity and ensure food security because cassava alone contributes about 22% of Ghana's agricultural GDP.
The CMD symptoms observed is consistent with CMD symptoms reported by several authors including The low incidence of white ies on plants showing very high incidence of CMD (75% to 100%) suggests that the highly infected elds become unattractive to the insects and therefore the insects move on to elds that are fresher and healthier or possibly younger ). The studies by Fargette et al., (1990) have reported that the disease incidence on elds planted with infected cuttings and the relative abundance of white ies contribute differently to the incidence of the disease. Analysis of symptoms observed on infected plants and elds during our survey indicated that infections could be from infected cuttings used as planting materials as well as from white ies. These two sources have been established as key sources of infections and it is believed that both infected cuttings and white ies affected disease incidence and severity in the elds surveyed .
Generally, the white y populations encountered in the elds were low and varied among the regions; this is similar to ndings by Ntawuruhunga et al., (2007) in Congo Brazzaville and Torkpo et al., (2017) in Ghana. The northern parts of Ghana which have a savannah agro-ecology with unimodal rainfall had the lowest incidence of white ies which is similar to observations reported by Fargette et al., (1994). The northern parts of Ghana can be proposed for the establishment and maintenance of clean planting materials production elds. This is because with low incidence of white ies it is expected that elds planted with disease-free cuttings can maintain their health status for some time before they become infected (Bock, 1994;Fauquet and Fargette, 1990).

Laboratory Diagnostics
The detection of ACMV-Kenya isolate and EACM-Kenya virus is a rst report in Ghana. Previous reports have detected ACMV-Ghana and EACMV-Cameroun and others (Torkpo et al., 2017). The detection of ACMV-Kenya isolate and EACMV-Kenya for instance is a clear indication that more strains of CMV could be present already in Ghana or new strains of the CMV could be entering the country. It is therefore important that regular monitoring and surveillance are conducted to identify any potential new strains to assist the management of CMD including breeding for robust CMV-resistant cultivars thereby improving cassava productivity. The possible detection of new strains from regular monitoring and surveillance can also help modelers to predict the spread of CMD in the country and possibly in the West African sub region.

Conclusion
The study has shown the spread and distribution of CMD in Ghana. Disease symptoms regularly encountered in farmers eld included leaf mosaic, leaf distortion/twisting, malformation, liform leaves, stunting and chlorosis. Cultivars with red petiole colour were the most prevalent while those with green petiole colour were the least.   Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.     CMD symptoms on infected plants encountered during the survey. The arrows point to A; leaf mosaic/distortion, B; Filiforms and C; Stunting/leaf mosaic/distortion.

Figure 7
Mean CMD percentage incidence and mean white y number in elds surveyed over the two-year period Figure 8 Phylogenetic tree indicating the placement of cassava mosaic viral strains sequenced from CMD-infected cassava samples.