The study sites (Figure 1) are located in the Guiriko region, specifically in the municipality of Guiguima, Dande, and include Koundougou and Samendeni. Located in western Burkina Faso, the Guiriko region is limited to the north by the Bankui region, to the south by the Tannounyan region, to the east by the Djôrô region, and to the west by the Republic of Mali. The region is important and includes mineral, plant, and wildlife resources. With a total area of 25,479 km², representing 9.4% of the national territory, the region’s population is estimated at 2,572,600 inhabitants and a density of 102 inhabitants/km².Thus, these sites were selected due to the documented presence of leaf spot disease affecting V. paradoxa. Furthermore, as part of the major V. paradoxa-producing areas of Burkina Faso, they provided suitable locations for the collection of diseased V. paradoxa leaves. Also, Selle, located in the Bankui region, was chosen as the study site because of the significant presence of leaf spot disease on V. paradoxa s. Located in the west of the country, this region occupies a strategic position for agriculture and livestock farming, with significant fishing, forestry, and craft potential. With an area of 34,333 km² and an estimated population of 2.12 million inhabitants, the Bankui region is characterized by a relatively flat landscape, savannah areas and open forests. Woody species such as Vitellaria paradoxa, Parkiabiglobosa, Danielliaoliveri, Terminalia spp., etc. are found there [6].

Figure 1. Study sites.

The biological material consists of diseased Vitellaria paradoxa leaves exhibiting necrosis and desiccation. These symptomatic leaves were collected between October and December 2024 from the parklands of the study sites listed in Table 1. During this period, the infested leaves clearly displayed characteristic symptoms, notably dark brown and pale brown fungal spots (Figure 2).

For the collection of biological material, sampling consisted of identifying a few representative stands, surveying the entire area, and randomly collecting ten (10) diseased leaf samples per Vitellaria paradoxa, with ten (10) trees sampled per site. Direct field observations revealed that leaves were either simultaneously affected by necrosis and desiccation, or exhibited necrosis or desiccation alone. In this study, however, sampling was conducted using distinct leaf specimens. The samples obtained were pooled by site, placed in plastic bags, and labeled. They were subsequently dried and stored in kraft paper in the laboratory. These samples were then used for laboratory analyses, with forty (40) Petri dishes of 9 cm in diameter prepared for each site, at a rate of twenty (20) Petri dishes per symptom.

Table 1. Features of sites and observed symptoms of collection of Vitellaria paradoxa leaf samples.

Figure 2. Manifestation of wilt (a) and necrosis (b) on shea leaves.

2.3.1. Culture Medium, Isolation and Purification of Pathogens

Forty-two grams (42 g) of PDA powder were dissolved in one liter of distilled water and homogenized using a magnetic stirrer. The resulting solution was sterilized in an autoclave at 120˚C for 30 minutes. After cooling to approximately 60˚C under a laminar flow hood, the medium was aseptically dispensed into Petri dishes at a volume of 25 mL per plate.

Pathogen isolation was performed from Vitellaria paradoxa leaves exhibiting typical symptoms of mycoses. Symptomatic tissues were excised into fragments of approximately 1 - 2 cm in length and surface-sterilized by immersion in ethanol for 30 seconds, followed by sodium hypochlorite for 1 minute. The fragments were then rinsed three times with sterile distilled water, blotted dry between two sterile sheets of filter paper, and aseptically transferred onto PDA medium in Petri dishes (5 - 6 fragments per dish). Cultures were incubated at 22˚C under alternating cycles of 12 h near-UV light and 12 h darkness for 4 days, allowing colony development around the tissue fragments (Figure 3(a)). Subsequently, colonies were purified by transferring morphologically distinct isolates to fresh culture medium using a sterile needle. At each site, thirty (30) newly purified colonies were incubated under the previously established conditions for approximately fourteen (14) days to promote fungal sporulation (Figure 3(b)). From 14-day-old pure colonies, slide preparations were made under aseptic conditions in a vertical laminar flow hood. Fungal identification was performed by comparing the conidia observed under the microscope with descriptions provided in standard fungal identification keys [7].

Figure 3. Isolation of parasitic fungi from shea. (a) Isolation of infected organs on PDA medium; (b) purification and incubation of the isolates at 22˚C under near UV light.

2.3.2. Characterization of Pathogenic Agents

From 14-day-old pure colonies, slide preparations were made under aseptic conditions in a vertical laminar flow hood. Fungal identification was performed by comparing the conidia observed under the microscope with descriptions provided in standard fungal identification keys [7].

2.3.3. Data Processing and Analysis

Species characterization was performed using identification keys recommended by [8]-[12]. Data entry, cleaning, and organization were initially conducted using Microsoft Excel (version 2018), while the calculation of pathogen frequencies and all statistical analyses were performed using the R statistical software environment.

The field observations revealed that leaf necrosis in Vitellaria paradoxa initially appears as small, isolated lesions scattered across the leaf lamina. These lesions are generally circular to irregular in shape and exhibit a dark brown to blackish-brown coloration (Figure 4). As the disease progresses, the lesions enlarge and coalesce, resulting in extensive necrotic areas that progressively cover a substantial portion of the leaf surface. This tissue deterioration leads to the destruction of affected cells, reducing the photosynthetically active area of the leaf. In advanced stages of infection, severely damaged leaves become distorted, curl inward, and eventually undergo premature abscission. The fungal isolates recovered from necrotic Vitellariaparadoxa leaves exhibited distinct morphological characteristics that facilitated their identification at the genus and species levels. Morphological examination and Mycological analyses conducted on symptomatic leaves exhibiting necrotic lesions led to the isolation and identification of nine pathogenic fungal species belonging to six genera (Table 2) associated with the disease in the Guiriko and Bankui region of Burkina Faso. These fungal taxa identified are Pestalotiopsisguepinii,Pestalotiopsisheterocornis, Phomasorghina, Phomaexigua,Fusariumculmorum,Fusariummoniliforme,Colletotrichumgloeosporioides, Pyriculariagrisea, and Lasiodiplodiatheobromae.

Figure 4. Dark brown necroses observed on the leaves of the shea tree.

Table 2. List of pathogenic agents inventoried on shea leaf necroses.

Overall, the morphological observations summarized in Table 3 corroborated the taxonomic identification of the isolates and highlighted the diversity of fungal pathogens associated with leaf necrosis of Vitellaria paradoxa in the Guiriko and Bankui region of Burkina Faso.

Table 3. Main morphological characteristics of fungal pathogens associated with leaf necrosis.

Figure 3 presents a heatmap coupled with a hierarchical clustering analysis illustrating the distribution and relative abundance of pathogenic fungi isolated from necrotic Vitellaria paradoxa leaves in the five locations studied. The colour gradient, from light yellow to dark red, reflects a low and high frequency of occurrence of fungal species, respectively. The species identified mainly belong to four families known for their involvement in foliar diseases in tropical environments: Sporocadaceae, Didymellaceae, Nectriaceae and Botryosphaeriaceae. The analysis highlights a high spatial variability of the fungal community associated with the necrosis and drying of Vitellaria paradoxa leaves, revealing a non-uniform distribution of pathogens between the different localities.

The dendrogram of species distinguishes two major groups (Figure 5). The first includes the least common fungi, including Fusariumculmorum, Colletotrichumgloeosporioides, Pyriculariagrisea and Lasiodiplodiatheobromae. These species have a restricted distribution and are mainly observed in certain locations. Colletotrichumgloeosporioides is particularly abundant at Koundougou, while Pyriculariagrisea and Lasiodiplodiatheobromae are almost exclusively found there. This limited distribution could reflect ecological specialization, dependence on particular microclimatic conditions, or a secondary role within the disease complex studied.

The second group includes the dominant species of the fungal complex, namely Pestalotiopsisguepinii, Pestalotiopsisheterocornis, Phomasorghina and Phomaexigua. These species are the most frequent and abundant in several study sites. The observed proximity between Pestalotiopsisguepinii and Phomasorghina suggests a regular co-occurrence, probably related to similar ecological requirements or to concomitant colonization of necrotic leaf tissues. Similarly, the association between Pestalotiopsisheterocornis and Phomaexigua reflects comparable distribution patterns.

Figure 5. Heatmap and hierarchical grouping analysis of pathogenic fungi associated with shea leaf necrosis in five localities in western Burkina Faso.

The observed frequencies confirm the predominance of the genus Pestalotiopsis and Phoma in the development of Vitellaria paradoxa leaf symptoms. Phomasorghina reaches a frequency of 50% in Dande and Guiguima, compared to 20% in Selle. For its part, Pestalotiopsisheterocornis has frequencies of 50% in Dande and Samandéni and about 20% in Selle. In the latter locality, several species, including Pestalotiopsisguepinii, Phomasorghina, Phomaexigua and Fusariumculmorum, have similar frequencies (about 20%), reflecting a more diverse pathogenic composition. The dendrogram of the localities highlights three distinct geographical areas. The first is made up of Dande, which is distinguished by a strong dominance of Pestalotiopsisguepinii, associated with high abundances of Phomasorghina and Pestalotiopsisheterocornis. The second corresponds to Selle, whose pathogenic profile is close to that of Dande but with lower frequencies, which explains their similarity in the grouping analysis.

The third group includes Guiguima, Koundougou and Samandenu. Guiguima is characterized by the importance of Fusariummoniliforme and Pestalotiopsisguepinii, while Koundougou has the highest species diversity with the simultaneous presence of Fusariummoniliforme, Fusariumculmorum, Colletotrichumgloeosporioides, Pyriculariagrisea, Lasiodiplodiatheobromae, Pestalotiopsisguepinii and Phomaexigua. Samandeni is distinguished by a high abundance of Pestalotiopsisheterocornis and Phomaexigua, accompanied by a moderate presence of Fusariummoniliforme. In Koundougou, nearly all pathogens were associated with necrosis and desiccation.

Overall, the results show that species of the genera Pestalotiopsis and Phoma constitute the dominant nucleus of the pathogenic complex associated with Vitellaria paradoxa necrosis and leaf desiccation in western Burkina Faso. Conversely, the genera Fusarium, Colletotrichum, Pyricularia and Lasiodiplodia appear as secondary agents or fungi associated with whose importance varies according to local ecological conditions.

The results indicate that leaf desiccation is characterized by the appearance of pale-brown necrotic lesions, primarily initiating at the margins of the leaf blades. These lesions are irregular in shape and often expand progressively toward the internal leaf tissues (Figure 6). Mycological analysis of the affected leaf samples revealed the presence of nine (9) distinct fungal pathogens, suggesting a complex fungal community associated with the observed leaf symptoms and potentially contributing to the progression of the necrotic process. The results (Figure 7) reveal a clear spatial structuring and differential abundance of pathogenic fungi associated with leaf desiccation across the five study sites. Overall, infection frequencies varied considerably, with a small number of taxa showing broad distribution and high prevalence, while others remained localized or sporadic.

Among the characterized genera, Pestalotiopsis and Phoma were identified as the predominant taxa, being both widespread and abundant, with infestation rates reaching up to 50% across the surveyed sites. This strong prevalence across sites suggests a central role in the observed foliar necrosis. It is closely associated with the dominant group formed by P.guepinii and P.heterocornis, which were also widely distributed and particularly abundant in Dande, Samandéni, and Koundougou, indicating similar ecological requirements and frequent co-occurrence on symptomatic leaves. A second group of pathogens displayed intermediate frequencies and more restricted distributions, notably P.sorghina and F.culmorum. The former was confined to Guiguima, whereas the latter occurred mainly in Koundougou and Guiguima, suggesting a stronger dependence on local environmental conditions.

In contrast, a third group composed of less frequent species, including C.gloeosporioides, P. grisea, P.exigua, and L.theobromae, was largely restricted to Dande and Selle. Their limited occurrence and lower abundance suggest a secondary or opportunistic role in the disease complex. These patterns are consistent with the hierarchical clustering results, which separated the study area into three main assemblages: Dande and Samandeni formed a cluster dominated by Pestalotiopsis spp.; Guiguima exhibited an intermediate profile characterized by a mixture of Phoma and Fusarium species; while Koundougou grouped with Selle, showing higher fungal diversity and the presence of rarer taxa. Overall, the integrated analysis highlights the predominance of Pestalotiopsis species and Phoma in the Vitellaria paradoxa leaf necrosis complex, whereas Colletotrichum, Pyricularia, and Lasiodiplodia appear to act mainly as secondary or opportunistic pathogens influenced by local environmental conditions.

Figure 6. Illustration of pale brown necrosis or leaf desiccation of shea tree leaf.

Figure 7. Thermal map of pathogenic fungi associated with Vitellaria paradoxa leaf desiccation.

Figure 8. Mycelia of P.heterocornis (a) and L.theobromae (b).

Figure 8 shows representative examples of mycelia of P.heterocornis (a) and L.theobromae (b). Figure 9 presents representative examples of conidia of P.heterocornis (a) and L.theobromae (b) observed under the electron microscope.

Figure 9. Conidia of P.heterocornis (a) and L.theobromae (b) observed under electron microscopy.