Introduction

Due to its high commercial demand, in southeastern Mexico most farmers sow opaque black grain bean (^{FIRA, 2016}). Chiapas is one of the most important producers, in 2019, 114 612 ha were sown in this state, of which 71 978 ha were cultivated in the summer cycle, under rainfed conditions and the rest in the autumn-winter cycle, with residual moisture, the average yield in both conditions was less than 600 kg ha^-1 (^{SIAP, 2020}).

The low yield observed is a consequence of biotic, abiotic, technical and socioeconomic factors. Among the biotic ones, viral diseases stand out, particularly the golden yellow mosaic caused by the bean golden yellow mosaic virus (BGYMV), which is transmitted by whiteflies [Bemisia tabaci (Gennadius)]. This disease has been present since 1977 in Veracruz, Chiapas, Yucatán, Tamaulipas and in the Huastecas (^{Cuéllar and Morales, 2006}), both in the sowings under rainfed and residual moisture conditions and can cause up to 100% losses in yield when it occurs from the vegetative phase of the crop (^{Villar et al., 2003}). The control of BGYMV has focused on the development of resistant varieties (^{Morales and Anderson, 2001}; ^{Soler-Garzón et al., 2021}).

Among the abiotic factors are the frequent occurrence of periods of terminal and intermittent drought, which, depending on their intensity and magnitude, can cause losses of 20 to 100% of grain yield (^{Frahm et al., 2003}; ^{López et al., 2008}), as well as the sowing of bean in acidic soils, which limit plant development and yield, due to low nutrient availability, mainly of the exchangeable bases calcium, magnesium and potassium (^{Villar et al., 2003}; ^{Aye et al., 2017}).

Another limiting factor is the sowing of landrace materials that are low yielding, limited in adaptation and susceptible to diseases (^{Ugalde et al., 2014}) and varieties such as Negro Jamapa, generated in 1958 (^{Rosales et al., 2004}). This variety is sown in Chiapas and other states of the Mexican tropics because its grain is in high demand, but it is susceptible to the main viral and fungal diseases present in Chiapas and has poor adaptation to drought conditions and acidic soils (^{Rodríguez and Yoshii, 1990}; ^{Tosquy et al., 2008}; ²⁰¹⁴). Considering that, in Chiapas, most bean producers are oriented to self-consumption and sow on areas of less than 1 ha with limited economic resources, the use of improved varieties of high yield and tolerant to the limiting factors indicated represents the most viable alternative to contribute to mitigate the low productivity of the crop in the state (^{López et al., 2012}).

Currently, the Bean Program for southeastern Mexico has a group of advanced lines of black bean, generated by the National Bean Program of INIFAP, which were selected for their high yield potential, less interaction with the environment and some of them, for the presence of the molecular marker (MM) SR2 linked to the bgm-1 gene that gives resistance to BGYMV (^{Anaya et al., 2018}; ^{Tosquy et al., 2019}). The objective of this research work was to identify genotypes (lines or varieties) of black bean resistant to the golden yellow mosaic, with greater yield potential than the Negro Jamapa variety, under the environmental conditions in which the bean is sown in the state of Chiapas.

Materials and methods

The study included 14 genotypes of tropical black bean: 11 advanced lines (five derived from the Negro Papaloapan/SEN 46 cross, three from the Negro Citali/XRAV-187-3 cross and three from the Jamapa Plus/XRAV-187-3 cross), the improved varieties Negro Medellín and Verdín, included for their good productive behavior and adaptation to the tropical areas of the states of Chiapas and Veracruz (^{López et al., 2000}; ^{Tosquy et al., 2016}) and Negro Jamapa used as a control for being one of the most sown varieties in Chiapas (^{López et al., 2006}).

The field trial was established in three localities in the central zone of the state of Chiapas during the autumn-winter cycle (September-December) of 2019-2020 and was conducted in conditions of residual moisture; that is, using the moisture stored in the soil and the water of the last rains of the rainfed cycle.

The soil of the experimental site of Villa Corzo is a chromic Luvisol, of sandy crumb texture, with a pH of 4.79, classified as strongly acidic, according to the Official Mexican Standard PROY-NOM-021-RECNAT-2000, the soil of the Centro de Chiapas Experimental Field (CECECH, for its acronym in Spanish) is a pelic epileptic Vertisol, of sandy-clay crumb texture, with a pH of 5.74, classified as moderately acidic, and that of El Gavilán is a pelic calcaric vertisol, of clayey crumb texture, with a pH of 4.26, classified as strongly acidic (^{NOM 021-RECNAT-2000, 2000}). Table 1 shows the geographical location and some climatic characteristics of the evaluation localities.

The bean genotypes were sown in September 2019, at a density of 250 000 plants ha^-1, in a randomized block experimental design with three repetitions and experimental plots of three furrows of 5 m in length, where the useful plot corresponded to the complete central furrow.

During the conduct of the trial, rainfall was quantified, the incidence of golden yellow mosaic in the R8 stage was determined with the general scale of 1 to 9 for the evaluation of viral diseases, where 1 and 9 correspond to 0% and 100% incidence, respectively (^{Van and Pastor, 1987}) and the grain yield expressed in kilograms per hectare at 14% of moisture.

An analysis of variance of the variables quantified by locality and a combined analysis of the three test environments were performed, for the separation of means, the test based on the least significant difference at 5% probability of error (LSD, α= 0.05) was applied (^{SAS Institute, 1999}). Correlation analyses between the average values of golden yellow mosaic incidence and grain yield of the genotypes in each locality were also performed, to determine specifically, in which evaluation localities, the BGYMV was negatively and significantly related to bean yield.

Results and discussion

In the three test localities, a highly significant effect (p≤ 0.01) on the incidence of golden yellow mosaic was detected between treatments. With respect to grain yield, only in the CECECH locality was high significance (p≤ 0.01) detected between treatments, while in Villa Corzo and El Gavilán, the effect was significant (p≤ 0.05) (Table 2).

The golden yellow mosaic significantly reduced grain yield in the three test localities (Villa Corzo r= -0.562^*, CECECH r= -0.757^** and El Gavilán r= -0.552^*), which could be due to the fact that whiteflies and the symptoms were present from the V4 stage of the bean, which is when this disease can cause greater damage to bean plants (^{López et al., 2002}). The presence of the vector and incidence of symptoms in the early stages of the development of the plant help to ensure that the evaluation was carried out under conditions of disease pressure.

Table 3 shows that, in the three evaluation sites, Jamapa Plus/XRAV-187-3-4-1 showed the greatest damage from this disease, with incidence values of 5 in Villa Corzo and above 6 in CECECH and El Gavilán, indicating that this genotype was the most susceptible to the golden yellow mosaic. The Negro Papaloapan/SEN 46-3-2 line also showed significant damage in the CECECH locality with intermediate symptoms of the disease.

In the joint analysis of golden yellow mosaic incidence, a significant effect was detected in the locality factor and a highly significant effect in the genotype factor and in the interaction of both factors. With respect to the first factor, in the locality of El Gavilán, there was the highest incidence of this disease, with an average value of 4.14, significantly higher than those registered in CECECH and Villa Corzo, which had an average value of 3.43 and 3.19, respectively (Table 3), a value of 3 in the general scale of evaluation for viral diseases corresponds to an incidence between 11 and 25%, while one of 4 to an incidence between 26 and 40%, so these values can be considered as a low and medium incidence, respectively, since a value of 8 or 9 on the scale correspond to incidences between 91 and 100% (^{Van and Pastor, 1987}).

This could be due to the fact that, in El Gavilán, there was less moisture (83 mm of rainfall) and slightly higher average temperatures prevailed (25.1 °C) during the vegetative phase of the crop (Table 4), which favored the reproduction of the whitefly (B. tabaci) and consequently, the incidence of golden yellow mosaic increased, since the incidence of this disease is directly related to the populations of B. tabaci (^{Morales and Anderson, 2001}); likewise, the occurrence of periods of drought, accompanied by high temperatures, favor the reproduction of B. tabaci (^{Cuéllar and Morales, 2006}).

In the genotype factor, the lines Negro Citlali/XRAV-187-3-1-6, Negro Papaloapan/SEN 46-7-7, Negro Citlali/XRAV-187-3-1-5 and Jamapa Plus/XRAV-187-3-4-4, as well as the Verdín variety, were the most resistant to this viral disease, with average incidence scores between 1.67 and 3.22, significantly lower than that of the Negro Medellín and Negro Jamapa varieties, and that of the Jamapa Plus/XRAV-187-3-4-1 line, this last line obtained the highest average incidence value (6), which was significantly higher than the rest of the genotypes (Table 3).

The good reaction to the golden yellow mosaic of Negro Citlali/XRAV-187-3-1-6, Negro Citlali/XRAV-187-3-1-5 and Verdín is largely due to the fact that their plants have the co-dominant MM SR2, linked to the bgm-1 gene, which gives them genetic resistance to BGYMV (^{Urrea et al., 1996}; ^{Anaya et al., 2018}). In the case of Negro Papaloapan/SEN 46-7-7 and Jamapa Plus/XRAV-187-3-4-4, which, in a first scrutiny, were negative for the presence of MM SR2 (^{Anaya et al., 2018}), their good reaction to the golden yellow mosaic may be due to the fact that these lines contain other genes or mechanisms of resistance that contributed to a more stable resistance response across environments, which is likely, since their progenitors, Negro Papaloapan, in the case of the first line and XRAV-187-3 of the second, in addition to having the bgm-1 gene, have the major QTL of resistance to BGYMV (^{Anaya et al., 2018}).

The statistical significance of the locality x genotype interaction was due to differences in the values of incidence of golden yellow mosaic of some genotypes across the test localities, which was more noticeable in the Negro Citlali/XRAV-187-3-1-8 and Jamapa Plus/XRAV-187-3-1-2 lines, which, in CECECH and Villa Corzo, had low incidence of this disease, with average scores that varied between 2.33 and 3.33, while in El Gavilán, they obtained average scores of 4.67 and 5.0, respectively, as well as in the Negro Jamapa and Negro Medellín varieties, which had a ‘moderate’ incidence in Villa Corzo, with scores of 3.67 and 4.0, respectively, and ‘intermediate’ in CECECH and El Gavilán, with scores that ranged from 4.67 to 5.33 (^{Van and Pastor, 1987}) (Table 3).

With respect to grain yield, in each of the test localities, the outstanding group of genotypes (four in Villa Corzo, four in CECECH and eight in El Gavilán) statistically exceeded the yield of the Negro Jamapa and Negro Medellín varieties, Table 5. Of these three groups, the following lines stand out: Jamapa Plus/XRAV-187-3-4-4 and Negro Papaloapan/SEN 46-7-7, which, although lacking MM SR2 linked to the bgm-1 gene, obtained a significantly outstanding grain yield in the three test localities, the Jamapa Plus/XRAV-187-3-1-2 and Negro Citlali/XRAV-187-3-1-6 lines and the Verdín variety were also in the outstanding group in CECECH and El Gavilán (Table 5).

The fact that lines lacking MM SR2 have stood out for their higher yield under golden yellow mosaic pressure conditions, compared to lines that have this marker, highlights the importance of confirming resistance by confronting the pathogen, since although bgm-1 is the most widely used gene, MM SR2, according to current standards, is not closely linked to this gene (~3-7 cM), resulting in the selection of a few lines that have the marker, but are nevertheless susceptible to the virus, in addition to the existence of other genes and QTLs whose accumulation conditions high levels of resistance to BGYMV (^{Soler-Garzón et al., 2021}), as well as other mechanisms of resistance to the vector, such as antibiosis or antixenosis (^{Santos et al., 2020}), whose presence in the evaluated materials needs to be studied.

According to the combined analysis, yield varied significantly between localities (p≤ 0.05), genotypes (p≤ 0.01) and in the interaction of both factors (p≤ 0.01). In the same Table 5, it was observed that the highest average yield was obtained in the CECECH locality, due in large part to better conditions of soil pH (Table 1) and moisture during the reproductive phase of the crop, in which the bean had 130 mm of rainwater, of an accumulated total of 216 mm in the cycle (Table 4), which favored the filling of pods and therefore, the grain yield (^{López et al., 2008}).

In El Gavilán and Villa Corzo, the average yields obtained were statistically similar to each other, and lower than that of CECECH, mainly due to the fact that the soil where the trial was conducted in both locations is strongly acidic (Table 1), which limits the development and productivity of the crop, due to a low availability of exchangeable bases (K ⁺, Ca⁺⁺ and Mg⁺⁺) for bean nutrition and in some cases, also due to a high saturation of aluminum that some acidic soils of the regions of La Frailesca and central Chiapas may contain, which can cause toxicity to bean plants and reduced root growth (^{Villar, 2000}).

In general, the average grain yields obtained in the three localities were relatively low, due in large part to a low availability of moisture in the crop cycle (Table 4), since, for an adequate development of the bean, at least 300 mm of rainfall, well distributed during the phenological cycle, is required (^{Ruíz et al., 2013}). In the genotype factor, a group of six lines and the Verdín variety obtained average grain yields statistically similar to each other and higher than those of the Negro Medellín and Negro Jamapa varieties, in this group, the most productive were the Jamapa Plus/XRAV-187-3-4-4 and Negro Papaloapan/SEN 46-7-7 lines with average yields greater than 1 000 kg ha^-1 (Table 5).

The two varieties indicated above, along with the Jamapa Plus/XRAV-187-3-4-1 line, obtained the lowest grain yields, Table 5, in large part, because they were the most affected by the golden yellow mosaic; that is, the plants of these genotypes, on average, had a higher incidence of this disease, which, in general, was negatively related to grain yield (r= -0.699^**) (Table 3).

The highly significant effect of the interaction of both factors indicated that the productive response of some genotypes varied with the environment of evaluation, notorious examples of this behavior were observed in Negro Papaloapan/SEN 46-3-2, which, in El Gavilán, was the most yielding genotype, which is attributed to its good response in conditions of severe edaphic acidity (pH of 4.26), while in CECECH, with better soil acidity conditions for the proper development of the bean (pH of 5.74), it was one of the least productive (Table 5), mainly due to the damage caused by the golden yellow mosaic; as well as in Negro Papaloapan/SEN 46-7-10, which, in the locality of Villa Corzo, in conditions of strongly acidic soil (pH of 4.79), obtained a significantly outstanding grain yield, while in El Gavilán, under conditions of even more severe soil acidity, it was one of the least productive (Table 5), partly also because in that locality it showed a higher incidence of golden yellow mosaic than in the other two localities (Table 3).

Conclusions

In the conditions of moisture and acidic soils of the state of Chiapas, the Verdín variety represents a better crop option than the Negro Jamapa variety. The Jamapa Plus/XRAV-187-3-4-4, Negro Papaloapan/SEN 46-7-7 and Negro Citlali/XRAV-187-3-1-6 lines, and the Verdín variety were resistant to golden yellow mosaic and had significantly higher grain yield than the control variety.