Morphological Variability and Races of Fusarium oxysporum f.sp. ciceris Associated with Chickpea (Cicer arietinum) Crops

Problem statement: Mexico is the third largest producer and exporter of chickpea (Cicer arietinum), with the states of Sinaloa and Sonora accounting for 70 and 20% of Mexico’s production, respectively. The most damaging disease affecting this species is caused by Fusarium oxysporum f.sp. Ciceris (FOC), which causes losses of up to 60% in Sonora. The objective of this study was to isolate and characterize the phenotype and genetics of FOC collected from affected chickpea plants in northwestern Mexico and to identify the abiotic factors that allow it to develop. Approach: Sampling focused on affected plants from 12 crops in Sonora and Sinaloa. Based on 355 isolated strains, using Polymerase Chain Reaction (PCR) 161 were positive for FOC. Results: Of the 161 strains, 91 were identified as races previously recorded for the Americas: Yellowing (R0 (41%), R1B/C (15%) and wilting (R5 (14%) and R6 (28%) reflecting the symptoms observed in the areas sampled. The other 70 isolates could be nonpathogenic, or could be races yet to be recorded for the Americas. Conclusion: Morphological variability in FOC was high in the main chickpea producing regions in northwestern Mexico and was not a function of the physical and chemical properties of the soil, nor of the geographic location of the cropfields. This is the first report of races of FOC in Mexico.


INTRODUCTION
Chickpeas (Cicer arietinum L.) are grown during winter under different agroclimate conditions in the northwestern states of Sinaloa, Sonora and Baja California in Mexico. Annual production is about 200,000 ton most of which is destined for the international market (Manjarrez-Sandoval et al., 2004;Padilla-Valenzuela et al., 2008). Chickpea crop yield varies depending on the cultivar, climate conditions and disease; among the latter are root rot caused by a complex of fungi such as Fusarium, Rhizoctonia, Macrophomina and Sclerotium, leaf diseases caused by Botrytis cinerea Pers., Peronospora sp., Uromyces ciceris-arietini (Grognot) Jacz. and G. Boyer and wilting caused by Fusarium oxysporum f.sp. ciceris Matuo and K. Sato (FOC) (Padilla-Valenzuela et al., 2008). FOC is quickly dispersed, attacks the root of the plant and causes hydric stress by blocking the xylem vessels and also causes chlorosis, necrosis and abscission (Di Pietro et al., 2001;Gupta et al., 2009). variability was determined using PCR-RAPD, but there was no consistent information for distinguishing between strains by origin or plant symptomology (Luna-Paez et al., 2004). In northwestern Mexico the disease is diagnosed based on the plant's symptoms. Chickpea wilt has been detected in 60% of the soils of La Costa de Hermosillo, Sonora, while in Sinaloa losses of 20% have been recorded (Manjarrez-Sandoval et al., 2004) and losses of 10 to 90% have been reported for other countries (Landa et al., 2006;Sharma and Muehlbauer, 2007).
There have been different attempts to control this disease, such as adjusting the sowing date and using resistant varieties of chickpea (Padilla-Valenzuela et al., 2008). Among the factors that favor the development of FOC are excess soil humidity, particularly clayey soil with poor drainage, along with a deficiency of potassium and excess sodium. Among the factors that contribute to pathogenesis is temperature, with 25-28°C optimal for the development of FOC (Bhatti and Kraft, 1992). However, an increase from 24-27°C is sufficient to change the resistance of some varieties of chickpea and make them susceptible to this pathogen (Landa et al., 2006).
The aim of this study was to isolate F. oxysporum f.sp. ciceris, characterize its phenotype and determine its pathogenic race and their relationship to abiotic factors where this fungus is found.

MATERIALS AND METHODS
Sampling sites: Based on the occurrence of chickpea yellowing and wilt, four locations were selected for sampling: in Sonora, La Costa de Hermosillo and Valle del Yaqui and in Sinaloa, Los Mochis and Guamúchil (Table 1). Three fields were sampled at each location with three quadrats per field. Samples were taken from plants showing signs of the disease. Ten plants were collected from each quadrat for a total of 360 chickpea root samples. Each plant was analyzed in triplicate, by placing five root sections in each Petri dish for a total of 5400 root fractions analyzed. All sampling sites were georeferenced using a geopositioning device (Garmin GPSmap 76S, Kansas, USA).
Strain isolation: FOC strains were obtained from the rhizoplane of the four most commonly grown varieties of chickpea in northwestern Mexico: Blanco Sinaloa 92, Aura, Teki Blanco and Costa 2004.
The roots of each one of the plants sampled were washed under running water, excess water was removed and the roots were cut into 1-2 cm sections. To eliminate surface tension from the roots, they were left in 50% ethyl alcohol for 30 s and then superficially disinfected by immersion in 2% sodium hypochlorite for 2 min. The chlorine was removed by washing the roots four times in sterile distilled water. Excess moisture was removed and five root sections were placed in each Petri dish which contained Potato Dextrose Agar (PDA; Difco) acidified to 1.4 with 10% tartaric acid and 500 mg L −1 chloramphenicol. These were incubated at 25°C for 5-7 day (Nelson et al., 1983;Burgess et al., 1994). Based on macro-and microscopic characteristics, colonies belonging to FOC were grown on PDA and monospore cultures were grown and then macroscopically characterized by color, appearance and growth rate after 7 day at 25°C; microscopically the size of their macroconidia, microconidia and chlamydospores was measured (Booth, 1975;Nelson et al., 1983). A control strain of FOC was used (ATCC 200784).
Abiotic factors: Using thematic and digital maps (INEGI, 2000), the climate, geology, edaphology, physiography and hydrology of the sampling sites were obtained. The physical and chemical properties of the soil were also analyzed for 36 samples made up of 10 subsamples per quadrat. Following the methodology of Castellanos et al. (2000), the following were determined: pH, electrical conductivity, nitrate, phosphate and potassium concentration, as well as organic matter content and texture.

Statistical analysis:
ANOVAs with a 5% level of significance were used to analyze the physical and 1.0 *Primers were designed by Jimenez-Gasco and Jimenez-Diaz (2003) and purchased from eurofins HW6 operon chemical properties of the soil and the microscopic characteristics of the FOC strains. When there was a significant difference (p≤0.05) between fields, means were compared with a Tukey test in the SAS (1999) statistical computing software package. In order to group the fields sampled based on their soil properties and the phenotypes of the FOC strains, data were subjected to a multivariate analysis and a principal component analysis using the NCSS software package (Hintze, 2007).

Characterization by PCR:
With the PCR of the FOC primers and amplifying a 1500 bp fragment the FOC positive strains were identified. This was followed by specific PCR for the races reported for the Americas: 0, 1B/C, 5 and 6, the first two yellowing races and the last two wilt races (Jimenez-Gasco and Jimenez-Diaz, 2003).
To determine the pathotype to which each of the strains identified as FOC belong, genomic DNA, was extracted using the commercial kit (ZR-Fungal) followed by PCR. In Table 2 the primers used to detect the FOC and pathotypes 0, 1B/C, 5 and 6 are given.

Morphological variability of FOC:
The multivariate analysis of the macro-and microscopic characteristics of the 161 FOC strains returned two large clades with a dissimilarity of 1.52 (Fig. 1). All the cropfields studied in Sonora, two from Guamúchil (G1 and G3) and one from Los Mochis (LM2) formed one clade; the remaining cropfields from Sinaloa were in the other. These two clades give rise to eight subgroups, with fields VY3 and VY1 the most similar (0.35), while the strains from G2 were most similar to the control strain. For the macroscopic characteristics, the color of the strains went from whitish (groups 1, 3, 5 and 6), brownish-white (groups 2 and 4), pinkish-white (group 7), to purplish-brownish-white (group 8).
At the microscopic level, the mean length of the macroconidia was 16.5-34.1 µm for the eight groups (Table 3), with the two extremes recorded in the fields in Sinaloa, LM3 and LM2 respectively (p≤0.05); most had 2-4 septa. Mean length for the other fields were similar to that of the control (p>0.05). Mean macroconidia width was the same as that of the control (p>0.05), with the exception of the samples from cropfield VY2 (4.2 µm). In Fig. 2, the macro-and microscopic characteristics of the FOC strains isolated are given.
Microconidia were abundant, unicellular, oval and did not vary significantly in length (p>0.05) (Table 3),although width was significantly different for the samples from cropfield CH3 (3.3 µm) (p≤0.05). Length and width means ranged from 7.2-9.8×2.5-3.3 µm. Chlamydospores had a thick wall, were spherical and had an intercalary or terminal location. Mean diameter was 4.8-8.1 µm, chlamydospore size for the eight groups was the same as that of the control (p>0.05), with the exception of cropfield LM3 (4.8 µm). At day 7 the growth rate of the strains was between 6 and 7.5 cm (p>0.05) ( Table 3).
The principal component analysis revealed that macroconidia length is the variable that most influences differentiation between the fields studied for the morphological characteristics of FOC. The first principal component explains 92% of the total variation. This is confirmed in Fig. 1 which shows that the most dissimilar field was LM2 where the greatest mean macroconidia length was recorded (Table 3).

Abiotic factors:
The physical and chemical properties of the soil all differed significantly among the cropfields (p<0.05; Table 3). pH was 6.9 (LM1 and G1) to 7.9 (VY2 and VY3). Electrical conductivity (EC) was 1.6 (LM3 and VY2) to 6.3 dS m −1 (G3). Na + concentration was below 5 mg L −1 in all cropfields, except in G3 (11.8) and CH2 (6.5). Cl − was lower than 5 meq L −1 in all cropfields, except G1 (8.4) (data not shown). The lowest nitrate concentration was recorded for field G1 (20 mg kg −1 ) and the values for all the other fields were higher than 30 mg kg −1 . In G1, G2 and G3 phosphate concentration was less than 30 mg kg −1 (the minimum acceptable for agricultural crops) and in the other fields it was higher. In all of the cropfields, potassium concentration was greater than 150 cmol kg −1 , the minimum recommended for agricultural land. Organic matter content was low in all of the cropfields from 1.1 (LM2 and CH1) to 1.8% (G1 and VY1). Soil texture ranged from clayey to loam, through silty clay and silty clay loam (Table 4). The multivariate analysis of the physical and chemical properties of the soil revealed that field G3 in Sinaloa was the most dissimilar (2.22), followed by CH2 in Sonora. The remaining fields fell into two clades: one comprised of the Sonora fields and cropfield G2 in Guamúchil, Sinaloa and the other comprised only of the remaining fields in Sinaloa (Fig. 3).
The principal components analysis applied to the soil properties returned salinity (45.8%) and texture (24.4%) as the variables with the greatest influence. In   fact, the highest EC was observed for G3 (6.3 dS m -1 ) followed by CH2 (3.8 dS m -1 ) (Table 4), the two least similar fields among all those evaluated (Fig. 3).
Thus, this is the first report of races of FOC in Mexico. The strains that did not fall into one of those previously mentioned have been stored for future study. These 70 remaining strains may belong to races previously not reported for the Americas, to new pathotypes or they may be nonpathogenic strains.

Isolation of FOC :
The symptoms of the disease caused by FOC and even though they did not isolate the pathogen, they estimated that 60% of the area sown with garbanzo in La Costa de Hermosillo was affected by it. However, Fusarium wilt and root rot can be confused with diseases caused by other pathogens in the diagnosis if one bases the latter solely on the symptoms of the plant.

Morphological variability of FOC:
According to morphological variability of FOC, our results coincide with those obtained by Trujillo et al. (2005), who also detected a high degree of phenotypic and genomic variability in the strains of Fusarium spp. responsible for wilting in carnations. Morphological and genetic variability was also high in Thanatephorus cucumeris (A.B. Frank) Donk in vineyards of Sonora, Mexico (Meza-Moller et al., 2007;Molina-Freaner et al., 2010).
Fusarium oxysporum has been reported to vary in color on the PDA growth medium (Groenewald et al., 2006). The aerial mycelium is white and can change to a variety of colors -from violet to dark purpledepending on the strain of F. oxysporum. Environmental conditions such as growth medium, light and temperature can cause pigment production in F. oxysporum (Rodrigues and Menezes, 2005).
The macro-and microscopic characteristics of the FOC strains this phenotypic variation coincides with that reported in other studies which identify F. oxysporum as one of the most variable species in its genus (Domsch et al., 1993;Dueñas et al., 2003;Groenewald et al., 2006).
Microconidia size did not varied significantly, being similar to that recorded by Domsch et al. (1993). Chlamydospores size, for the eight groups was the same as that of the control, Martinez et al. (1996) observed that the number of F. oxysporum chlamydospores increases as the culture dehydrates, which points to their function as a resting spore.

Abiotic factors:
The altitude of the chickpea growing region in northwestern Mexico is 12-75 m a.s.l. and the abiotic conditions that predominate are a dry to very dry climate, mean annual precipitation of 200 to 600 mm and a mean annual temperature of 23-25°C (INEGI, 2000). Ninety-eight percent of the area under cultivation and 92% of the world's chickpea production are in the semiarid tropics. Vujanovic et al. (2006) recorded F. oxysporum in Canada at temperatures between -5.7 and -17°C and temperatures between 24 and 27°C, along with high soil and air humidity are ideal conditions for FOC. In contrast, in northwestern Mexico under conditions of extreme aridity, FOC are widely distributed.
Recently, Karami et al. (2009) reported the morphological characteristics and pathogenicity of Synchytrium psophocarpi (Rac.) Baumann associated with false rust on winged bean, but further investigations to elucidate the influence of environmental factors on the disease and to determine host range is required.
The occurrence of FOC increases in acid pH and it is suppressed by alkalinity. Our results suggest the adaptation of FOC to neutral to slightly alkaline pH. Although the crop grows best at a pH of 6, it can grow when pH is 2 to 12. The optimal pH for FOC is lower than 4.5 (Martinez et al., 1996), though it has been isolated from soils at pH 5.3-6.3 (Vujanovic et al., 2006). The occurrence of FOC increases in acid pH and it is suppressed by alkalinity. Our results suggest the adaptation of FOC to neutral to slightly alkaline pH. Anuar et al. (2008) reported long-term applications of N decreased the soil pH from 4.2 to 3.7 and caused leaching of K to the lower depth.
Electrical conductivity (EC), crops grow quite well when it is 1.5-2 dS m −1 , growth is average between 2.1 and 4 and poor when EC is above 4 dS m −1 . Chickpea's water requirement is low (40-45 cm surface water), with an EC lower than 1 dS m −1 and a sodium absorption ratio of less than 5. Knowing the spatial losses of water beyond the root zone is important for precise management of the irrigation water (Aimrun et al., 2010). Vujanovic et al. (2006), observed a higher number of colony forming units (CFU g −1 soil) of F. oxysporum colonies in sandy soils. Chickpea crops need deep soils with a sandy loam to clay loam texture and with no salinity problems (Kukreja et al., 2005).

Pathotypes detection by PCR:
Of the four FOC races, two yellowing and two wilting pathotypes were identified. Races of FOC are geographically circumscribed: 2, 3 and 4 have only been reported for India, while 0, 1B/C, 5 and 6, are known for the Mediterranean region and the United States of America (California). In contrast, race 1A is the most widely distributed and has been observed in all three of the aforementioned regions (Jimenez-Gasco and Jimenez-Diaz, 2003).
In Sonora and Sinaloa, though these states are significant producers and exporters of chickpea, the diagnosis of the disease is based on the symptomology of the plants and on occasion plant yellowing and wilting is mistaken for root rot because the causal agent has not been isolated and identified. This means that studies are required to identify and understand the pathotypes, in order to contribute to the search for ways to control the disease.
In the Bajio region of Mexico, Luna-Paez et al. (2005) studied the distribution and genetic variability of FOC using genomic DNA polymorphisms and PCR-RAPD. However, they did not report the identity of the pathogenic races of FOC, just that there were two groups of isolates for yellowing and two for wilting, suggesting the possible presence of some of the races described for other countries.
The distribution and adaptation of the species in Fusarium are reported to be delimited by climate, geographic region, soil type, variety and cultivar (Vujanovic et al., 2006).

CONCLUSION
Morphological variability in Fusarium oxysporum f.sp. ciceris was high in the main chickpea producing regions in northwestern Mexico and was not a function of the physical and chemical properties of the soil, nor of the geographic location of the cropfields. Yellowing (0 and 1B/C) and wilting (5 and 6) races were found among the strains isolated, reflecting the symptoms observed in the areas sampled. This is the first report of FOC races in Mexico. There may be other races or nonpathogenic strains of FOC in the study area, which would require specific primers to carry out PCR on the strains yet to be identified.