Emerging infectious diseases threatening food security and economies in Africa

MS-Net: a novel lightweight and precise model for plant disease identification

The rapid development of image processing technology and the improvement of computing power in recent years have made deep learning one of the main methods for plant disease identification. Currently, many neural network models have shown better performance in plant disease identification. Typically, the performance improvement of the model needs to be achieved by increasing the depth of the network. However, this also increases the computational complexity, memory requirements, and training time, which will be detrimental to the deployment of the model on mobile devices. To address this problem, a novel lightweight convolutional neural network has been proposed for plant disease detection. Skip connections are introduced into the conventional MobileNetV3 network to enrich the input features of the deep network, and the feature fusion weight parameters in the skip connections are optimized using an improved whale optimization algorithm to achieve higher classification accuracy. In addition, the bias loss substitutes the conventional cross-entropy loss to reduce the interference caused by redundant data during the learning process. The proposed model is pre-trained on the plant classification task dataset instead of using the classical ImageNet for pre-training, which further enhances the performance and robustness of the model. The constructed network achieved high performance with fewer parameters, reaching an accuracy of 99.8% on the PlantVillage dataset. Encouragingly, it also achieved a prediction accuracy of 97.8% on an apple leaf disease dataset with a complex outdoor background. The experimental results show that compared with existing advanced plant disease diagnosis models, the proposed model has fewer parameters, higher recognition accuracy, and lower complexity.

Dynamics of the rice yellow mottle disease in western Burkina Faso: Epidemic monitoring, spatio-temporal variation of viral diversity, and pathogenicity in a disease hotspot

The rice yellow mottle virus (RYMV) is a model in plant virus molecular epidemiology, with the reconstruction of historical introduction routes at the scale of the African continent. However, information on patterns of viral prevalence and viral diversity over multiple years at a local scale remains scarce, in spite of potential implications for crop protection. Here, we describe a 5-year (2015-9) monitoring of RYMV prevalence in six sites from western Burkina Faso (geographic areas of Bama, Banzon, and Karfiguela). It confirmed one irrigated site as a disease hotspot and also found one rainfed lowland (RL) site with occasional high prevalence levels. Within the studied fields, a pattern of disease aggregation was evidenced at a 5-m distance, as expected for a mechanically transmitted virus. Next, we monitored RYMV genetic diversity in the irrigated disease hotspot site, revealing a high viral diversity, with the current coexistence of various distinct genetic groups at the site scale (ca. 520 ha) and also within various specific fields (25 m side). One genetic lineage, named S1bzn, is the most recently emerged group and increased in frequency over the studied period (from 20 per cent or less in 2015-6 to more than 65 per cent in 2019). Its genome results from a recombination between two other lineages (S1wa and S1ca). Finally, experimental work revealed that three rice varieties commonly cultivated in Burkina Faso were not different in terms of resistance level, and we also found no significant effect of RYMV genetic groups on symptom expression and viral load. We found, however, that infection outcome depended on the specific RYMV isolate, with two isolates from the lineage S1bzn accumulating at the highest level at early infections. Overall, this study documents a case of high viral prevalence, high viral diversity, and co-occurrence of divergent genetic lineages at a small geographic scale. A recently emerged lineage, which comprises viral isolates inducing severe symptoms and high accumulation under controlled conditions, could be recently rising through natural selection. Following up the monitoring of RYMV diversity is required to confirm this trend and further understand the factors driving the local maintenance of viral diversity.

Genetic Structure and TALome Analysis Highlight a High Level of Diversity in Burkinabe Xanthomonas Oryzae pv. oryzae Populations

Bacterial Leaf Blight of rice (BLB) caused by Xanthomonas oryzae pv. oryzae (Xoo) is a major threat for food security in many rice growing countries including Burkina Faso, where the disease was first reported in the 1980's. In line with the intensification of rice cultivation in West-Africa, BLB incidence has been rising for the last 15 years. West-African strains of Xoo differ from their Asian counterparts as they (i) are genetically distant, (ii) belong to new races and, (iii) contain reduced repertoires of Transcription Activator Like (TAL) effector genes. In order to investigate the evolutionary dynamics of Xoo populations in Burkina Faso, 177 strains were collected from 2003 to 2018 in three regions where BLB is occurring. Multilocus VNTR Analysis (MLVA-14) targeting 10 polymorphic loci discriminated 24 haplotypes and showed that Xoo populations were structured according to their geographical localization and year of collection. Considering their major role in Xoo pathogenicity, we assessed the TAL effector repertoires of the 177 strains upon RFLP-based profiling. Surprisingly, an important diversity was revealed with up to eight different RFLP patterns. Finally, comparing neutral vs. tal effector gene diversity allowed to suggest scenarios underlying the evolutionary dynamics of Xoo populations in Burkina Faso, which is key to rationally guide the deployment of durably resistant rice varieties against BLB in the country.

Effect of sterilants and plant growth regulators in regenerating commonly used cassava cultivars at the Kenyan coast

Cassava is an important root crop whose seed system is undeveloped. Micropropagation of explants in vitro has the potential of addressing the challenge of the unavailability of healthy cassava planting materials. Therefore, the study determined the effect of sterilization and plant growth regulators on cassava explants to produce certified disease-free plants of commonly used cultivars at the coast of Kenya. The apical nodes drawn from the three cultivars of cassava, Tajirika and Kibandameno and Taita, were used as explants. The sterilant sodium hypochlorite (NaOCl) at 5, 10 and 15% and 70% ethanol for 1 and 5 min and sprayed for 20 s were tested for the effect on the explant. Similarly, the effect of BAP (6-Benzyl amino purine) and NAA (1-Naphthalene acetic acid) Plant Growth Regulators (PGRs) each at 0.5, 1 and 5 mg/L under optimal conditions of sterilization was determined. Surface sterilization using 10% NaOCl followed by spraying 70% ethanol for 20 s had 85% initiation on Tajirika whereas 5% NaOCl followed by spraying 70% ethanol for 20 s had 87% and 91% initiation in Kibandameno and Taita cultivars, respectively. In Tajirika, significantly (p < 0.05) high shooting of 68% was from 5 mg/L BAP in MS media whereas approximately 50% rooting was from either 0.5 mg/L BAP or 5 mg/L NAA in MS media. Kibandameno and Taita cultivars had approximately 50% shooting from MS media without PGRs. Kibandameno had >37% rooting from 0.5 to 5 mg/L BAP or NAA in MS media whereas Taita had approximately 50% rooting from 0 to 5 mg/L NAA in MS media. This protocol showed at least 50% success rate of initiation, shooting and rooting as a rapid multiplication regeneration of Tajirika and Kibandameno and Taita cultivar plantlets with little modification of humidity and temperatures in the growth chambers. This protocol requires validation for use in large-scale production of cassava plantlets to alleviate the inadequacy of cassava planting materials among farmers.

Nitrogen budgets and nitrogen use efficiency as agricultural performance indicators in Lake Victoria basin

Too little nitrogen (N) is a threat to crop productivity and soil fertility in sub-Saharan Africa (SSA). Nitrogen budgets (NB) and nitrogen use efficiency (NUE) are critical tools for assessing N dynamics in agriculture and have received little or no attention in the region. Data were collected from smallholder farmers clustered into two categories, farmers applying and farmers not applying N fertilizers. NB were calculated using the Coupled Human and Natural Systems (CHANS) model approach for field and farm spatial scales. The results showed spatial variabilities in NB and NUE at the field level (maize and rice) across all the catchments. At the field level, N balances were negative for the two crops in all the catchments. Similarly, at the farm gate, a deficit of −78.37 kg N ha−1 was observed, an indicator of soil N mining. NUE values at the field scale varied across the catchments for both crops, with values for maize grown without N ranging from 25.76 to 140.18%. Even with the application of mineral N at higher levels in rice fields compared to maize fields, NUE values ranged between 81.92 and 224.6%. Our study revealed that the Lake Victoria region suffers from inefficient N cycling due to depleted soil N pools and low synchrony between N input and N removal. Therefore, a challenge lies in exploiting more sustainable N sources for farmers in the region for sustainable farming systems. The NB and NUE provide critical information to agriculture stakeholders to develop environmental, agronomic, and economically viable N management solutions.

Cucurbit[7]uril-Mediated Supramolecular Bactericidal Nanoparticles: Their Assembly Process, Controlled Release, and Safe Treatment of Intractable Plant Bacterial Diseases

A safe, biocompatible, and stimuli-responsive cucurbit[7]uril-mediated supramolecular bactericidal nanoparticle was fabricated by encapsulating a highly bioactive carbazole-decorated imidazolium salt (A₁, EC₅₀ = 0.647 μg/mL against phytopathogen Xanthomonas oryzae pv oryzae) into the host cucurbit[7]uril (CB[7]), thereby leading to self-assembled topographies from microsheets (A₁) to nanospheroidal architectures (A₁@CB[7]). The assembly behaviors were elucidated by acquired single-crystal structures, ¹H NMR, ITC, and X-ray powder diffraction experiments. Complex A₁@CB[7] displayed lower phytotoxicity and could efficiently switch on its potent antibacterial ability via introducing a simple competitor 1-adamantanamine hydrochloride (AD). In vivo antibacterial trials against rice bacterial blight revealed that A₁@CB[7] could relieve the disease symptoms after being triggered by AD and provide a workable control efficiency of 42.6% at 100 μg/mL, which was superior to bismerthiazol (33.4%). These materials can provide a viable platform for fabricating diverse stimuli-responsive supramolecular bactericides for managing bacterial infections with improved safety.

Confirmation report of Bacterial Leaf Streak disease of rice caused by Xanthomonas oryzae pv. oryzicola in Senegal

Xanthomonas oryzae pv. X. oryzicola (Xoc), the causal agent of Bacterial Leaf Streak (BLS), is considered as one of the most important emerging pathogens of rice in Africa. This disease is estimated as responsible of 20 to 30% yield loss (Sileshi et Gebeyehu 2021) and has been characterized in several west African countries including Mali and Burkina Faso since 2003 and more recently in Ivory Coast (Wonni et al. 2014, Diallo et al. 2021). Presence of BLS symptoms in Senegal were reported by Trinh in 1980 but, to our knowledge, BLS occurrence has never been validated further and no strain of Xoc have ever been isolated from Senegalese rice fields. Xoc is transmitted by seeds which contribute to its spread through the rice trade (Sileshi et Gebeyehu 2021). To confirm Trinh's observations, we surveyed rice fields between 2014 and 2016 in eight different regions where rice is produced in Senegal. Typical disease symptoms characterized by yellow-brown to black translucent leaf streaks sometimes along with exudates, were detected in fields of several regions and collected. Leaf pieces were successively sanitized, rinsed in sterile water, and symptomatic fragments were ground using the Qiagen Tissue Lyser System (QIAGEN, Courtaboeuf, France). The leaf powder was diluted in 1.5 ml of sterile water and incubated for 30 minutes at room temperature. Ten μl of the suspension was streaked on semi-selective PSA medium and incubated at 28°C for 3 to 7 days. Characteristic round, convex, mucous, straw-yellow Xoc candidate colonies were purified from six individual leaf samples from three distinct sites in the northern Senegal River Valley. To confirm their identity, isolated strains were tested for pathogenicity and molecular characterization. All isolates were subjected to the multiplex PCR developed for the identification of X. oryzae pathovars (Lang et al., 2010) and revealed the same PCR profile (two amplicons of 324 and 691 base pairs) similar to that of the Xoc reference strain BLS256. Leaves of 5-week-old plants of O. sativa cv. Kitaake were infiltrated with a needleless syringe containing a bacterial suspension set at an optical density of 0.5. Upon seven days of incubation under greenhouse conditions (27 ± 1°C with a 12-hour photoperiod), all infiltrated spots (2 spots on 3 plants per isolate) developed water-soaked lesions similar to those caused by control strain BLS256, except when leaves were infiltrated with water. Symptomatic leaf tissues were ground and plated on PSA medium, resulting in colonies with typical Xanthomonas morphology that were diagnosed as Xoc by multiplex PCR typing, thus fulfilling Koch's postulate. At last, four of the isolates were subjected to gyrB sequencing upon PCR amplification using the universal primers XgyrB1F and XgyrB1R (Young et al., 2008). Analysis of 780bp partial gyrB sequences of strains S18-3-4, S23-1-12, S52-1-4 and S52-1-10 highlighted 100% identity with the gyrB sequence of strain BLS256 (Acc. No. CP003057). To our knowledge, this is the first report of BLS in Senegal which is supported by molecular characterization methods. This study validates the presence of BLS in Senegal and will serve as a basis for future efforts of rice breeding for locally adapted resistance. More studies are needed to clarify the spatial distribution and prevalence of BLS in Senegal as rice cultivation is expanding rapidly in the country.

No need to beat around the bushmeat-The role of wildlife trade and conservation initiatives in the emergence of zoonotic diseases

Wildlife species constitute a vast and uncharted reservoir of zoonotic pathogens that can pose a severe threat to global human health. Zoonoses have become increasingly impactful over the past decades, and the expanding trade in wildlife is unarguably among the most significant risk factors for their emergence. Despite several warnings from the academic community about the spillover risks associated with wildlife trade, the ongoing COVID-19 pandemic underlines that current policies on the wildlife industry are deficient. Conservation initiatives, rather than practices that attempt to eradicate zoonotic pathogens or the wild species that harbour them, could play a vital role in preventing the emergence of life-threatening zoonoses. This review explores how wildlife conservation initiatives could effectively reduce the risk of new zoonotic diseases emerging from the wildlife trade by integrating existing literature on zoonotic diseases and risk factors associated with wildlife trade. Conservation should mainly aim at reducing human-wildlife interactions in the wildlife trade by protecting wildlife habitats and providing local communities with alternative protein sources. In addition, conservation should focus on regulating the legal wildlife trade and education about disease transfer and safer hunting and butchering methods. By uniting efforts for wildlife protection and universal concern for preventing zoonotic epidemics, conservation initiatives have the potential to safeguard both biodiversity, animal welfare, and global human health security.