One of the chili pepper species planted in Mexico is the habanero (Capsicum chinensis), produced in Campeche, Quintana Roo, Yucatan and Tabasco. In 2015, the production of habanero peppers in 3,055.5 t (SIAP, 2015). However, in recent years this crop has been introduced into cocoa-producing municipalities. In Mexico, wilt has been reported in chili pepper plants, attacking different types of peppers and causing losses of up to 40% in several states in Mexico (^{Silva-Rojas et al., 2009}; ^{Pérez-Moreno et al., 2005}; ^{Pérez-Acevedo et al., 2017}). Nevertheless, the presence of P. capsici is suspected in Tabasco with the wilt of habanero peppers (^{López-López et al., 2018}) based on the symptomatology, therefore the presence of P. capsici must be corroborated. ^{Montes and de los Santos (1989)} and ^{Ortiz-García (1996}) indicate the presence of P. capsici in cocoa trees, according to the criteria of ^{Tsao and Alizadeth (1988}), later pointed out by Uchida et al. (1992) as P. tropicalis, considering that the species of Phytophthora that attack both the cocoa trees and the pepper plants are phylogenetically related in clade 2 (^{Martin et al., 2012}). Likewise, ^{Donahoo and Lamour (2008}) point out that when populations are superimposed, hybridization can take place after generation F1 both species of Phytophthora; this situation was documented in vitro by ^{Hurtado-González and Lamour (2009}) with cucumber and pumpkin strains in the USA, in the same way that ^{Ortiz-García (1996)} did with pumpkin Phytophthora strains of france in France and cocoa of Mexico. Based on this, and given the importance of the cocoa tree in Tabasco, as well as the expansion of the plantation areas of the habanero pepper in the cocoa-producing area of Tabasco, it became crucial to determine the causal agent of the wilting of the habanero pepper plant in cocoa-producing areas in which the production of the habanero pepper is being incentivized in municipalities of Tabasco.

Area of Study. The area of study was an 82 km transect beginning in C-34, Huimanguillo, and ending in Acachapa and Colmena, Centro, from west to east, covering cocoa-producing areas of Huimanguillo, Cárdenas and Cunduacán in hydrological area 1 and reaching subzone 2 in Acachapa y Colmena, Centro Tabasco (Figure 1) in the vicinities of which are traditionally cocoa-producing communities with habanero pepper plantations.

Collection and isolation. From January to May 2018, 15 habanero pepper plantations in four municipalities of Tabasco were visited: Huimanguillo (1), Cárdenas (2), Cunduacán (9) and Centro (3), in which plants with symptoms of different degrees of wilt were collected. To isolate the associated microorganisms, small fragments of tissue were cut, disinfected, washed and dried in a V8-Agar culture medium and incubated at 25±1 °C in the dark for 7 days in a Prendo® incubator. The cultures were purified using the hypha tip method, following the method by ^{Ortiz-García (1996}). Identification at the genus level adhered to the keys by ^{Barnett and Hunter (1972}) for Fusarium and by ^{Erwin and Ribeiro (1996}) for Phytophthora.

Pathogenicity tests in habanero chili plants. The pathogenicity test was carried out with each microorganism isolated from wilting plants in 60-day old habanero pepper plants, Phytophthora CH132; Fusarium CH2, CH3, CH7, CH11 and CH16. Additionally, we used the P. capsici strain CPV302, from national collections. In this way, a pair of disks, 0.5 cm in diameter from young cultures, were stuck to the bottom of the base of the stem to infect it by mycelial contact, then incubated in a wet chamber. Daily supervision helped register the morphological changes in the 35 habanero plants used in this test with the seven isolations (five plants per isolation) for a period of 16 days, along with five non-inoculated plants taken as a control.

Figure 1 Transect of communities of the municipal areas of Huimanguillo, Cárdenas, Cunduacán in subzone 1, and of Centro, in subzone 2 in Tabasco, where habanero pepper plants with wilt were collected, in the Grijalva hydrological zone. 

Pathogenicity tests on cocoa fruits. Using strains CH132 (Phytophthora isolated from a habanero pepper plant), PC161.2, PC219 (P. capsici isolated from chili pepper plants used as a reference) and CPM04 (from Phytophthora isolated from Tabasco cocoa as positive controls), a pathogenicity test was carried out on cocoa fruits. The fruits used for these tests were in a stage of physiological maturity, and of the Guayaquil ecotype. After washing by immersion in 1% sodium hypochlorite for one minute, followed by another wash in three containers filled with sterile distilled water, the fruits were placed in an also sterilized plastic box, with a piece of cloth in the bottom to acclimatize them for 12 hours. Next, in a laminar flow cabinet, four disks, each one of which was 0.5 mm in diameter, were taken from the growth area of the Phytophthora cultures after five days of growth and placed on the epidermis of the fruits (without lesions and in a straight line). The boxes were covered and incubated at 25±1 °C in the dark for seven days in a Prendo^® incubator.

Morphological identification. The formation of sporangia was induced following ^{Pérez-Acevedo et al. (2017}). The isolation obtained from the habanero chili pepper (CH132) was confronted with the known types of compatibility for A1 (PVM-161.2) and A2 (CPV-219) of P. capsici. In a dish with a Green Bean-Agar medium culture, a 5 mm disk of isolation CH132 was placed 1 cm away from the reference isolation (A1 or A2). The dishes were kept at 25±1 °C in a Prendo^® incubator in the dark for 2 weeks and examined under the Cole Parmer^® compound microscope (^{Fernández-Pavía et al., 2004}). The morphological characteristics were compared with those obtained by ^{Stamps et al. (1990}).

Molecular identification. In order to extract the DNA, the protocol by ^{Leslie and Summerell (2006}) was followed, with modifications. The region of the COI (cytochrome oxidase c subunit 1) mitochondrial gene was amplified with primers COIF-1 (5’ -TCAWCWMGATGGCTTTTTTCAAC-3’) and COIR-1 (5’-RRHWACKTGACTDATRATACCA AA-3’), which amplified 727 pb. The PCR amplifications were carried out in an Eppendorf MasterCycler^® Gradient thermocycler. The condi-tions for the amplification were: 2 min at 95 °C, followed by 35 one-minute cycles at 95 °C, 1 min at 55 °C and a final extension step at 72 °C for 10 min (^{Robideau et al., 2011}). The DNA fragments were purified using the Wizard^® SV Gel kit and the PCR Clean-Up System-Promega, following the manufacturer’s instructions. The PCR products were sequenced in Macrogen (Seoul, South Korea). The sequences obtained were assembled and edited using the program PreGap y Gap. A consensus sequence was created, and they were aligned using the software ClustalX2. The identity of the sequences was first carried out using the web program Blast® blastn suite on NCBI (https://blast.ncbi. nlm.nih.gov/Blast.cgi). To confirm the identity of the isolations with ex-type sequences, a genetic analysis was carried out using MEGA^® (Molecular Evolutionary Genetics Analisys), applying the Maximum Likelihood method with 100 bootstrap repetitions. The model by ^{Tamura-Nei was used to infer the evolution (Tamura and Nei, 1993}), and Pythium oopapillum was used as an external group.

Isolations obtained. The communities of hydrological subzone 1, from which diseased habanero pepper plants were collected, were C-34, in Huimanguillo (S16), Rio Seco (S2) in Cárdenas, Miahuatlán (S11), Cumuapa (S3) and San Pedro (S7) in Cunduacán, where the isolations of Fusarium (CH2, CH3, CH7, CH11 and CH16) were found to be related. The collected plants displayed slight wilting in the foliage without necrosis in the leaves or stem. In the three sampling sites of the municipal area of Centro, plants with foliar necrosis were gathered, along with plants with other damages as indicated by ^{Pérez-Moreno et al. (2005}), as well as where the Phytophthora isolation CH132 was obtained, in Acachapa and Colmena (S13).

Pathogenicity tests in chili pepper seedlings. Six dai (days after inoculation) with the strains CH132, necrotic spots were registered on the base of the stem, along with wilting of the foliage (Figure 2 A). Sixteen days later, necrosis was found on the stem and dry leaves (Figure 2 B).

From the wilted plants and with necrosis on the stem, Phytophthora was reisolated, confirming Koch’s postulates, indicating CH132 as a pathogen and the cause of the wilting of the plant. The damages caused by wilting and necrosis on the base of the stem caused by strain CH132 (Phytophthora) was similar to the one caused by P. capsici strain CPV302. The control plants and the plants inoculated with e Fusarium strains (CH2, CH3, CH7, CH11 and CH16) showed no symptoms.

Pathogenicity tests in cocoa fruits. In the cocoa fruits inoculated with the mycelium discs of the isolated cocoa cultures (CPMO 04) four days after incubation, two of the four points of inoculation displayed the typical symptoms of the black spot (^{Ortiz-García, 1996}). However, the Phytophthora strains isolated from chili peppers from Tabasco and the strains used as a reference displayed no damage on the surface.

Morphological characterization. Isolation CH132 displayed cenocytic mycelium, caducous sporangia, with a predominance of ovoid and obpyriform. The average length and width for 50 sporangia was 73.72 µm and 37.3 µm respectively; and a length/width ratio of 2.02. Papilla 4.91 µm long and pedicels 189.8 µm long, and without chlamydospores. The oospore is spherical, plerotic, with an average length of 25.04 µm, 24.57 µm wide and 3.07 µm thick (Figure 3). The amphigynous antheridium, with an average length of 12.6 µm and 11.5 µm wide. The oogonium displayed an average length of 29.5 µm and a width of 27.8 µm. These dimensions are similar to those reported for P. capsici by ^{Stamps et al. (1990}). The type of compatibility of isolation CH132 is heterothalic, type A2.

Figure 2 Pathogenicity tests of Phytophthora strains: Control, CPV302 and CH132 (from left to right) in habanero chili pepper seedlings. Symptoms nine dai (A). Symptoms 16 dai (B). 

Molecular characterization. The DNA amplification of isolation CH132 by PCR with oligonucleotides COIF-1 and COIR-1 created a 727 pb PCR product. The sizes of the PCR products are similar to those reported by ^{Choi et al. (2015}) with the use of COI. The BLAST analysis helped determine that isolation CH132 has a similarity of 100% with the species of P. capsici (Accession numbers AY129166, MH136864, MH013474, MH013475 and HQ261267). This was confirmed with the phylogenetic analysis under the criterion of maximum likelihood, which shows the similarity of isolation CH132 with two P. capsici of Capsicun annuum isolations (Figure 4).

Figure 3 Morphological characteristics of isolation CH132. A: Papillated sporangia (≥3.5 μm), ovoid and obpyriform with a long pedicel (>20 μm). B: Spherical and plerotic oospore, smooth oogonium wall and amphigynous antheridium. Bars 20 μm. 

The detection of the strain of P. capsici represents a warning, because it is a soil inhabitant, where it can survive for several years, as well as the diversity of plants it attacks (^{Pérez-Moreno et al., 2005}; ^{Erwin and Ribeiro, 1996}). Also, the wide dispersion it may have with the use of infected seeds (^{Morales-Valenzuela et al., 2002}). The low incidence of wilted plants in the sampling sites could have been due to 100% of the chili pepper plantations established with the grafting of seedlings in a nursery, and which have been areas open to the planting of chili pepper plants for one or two year ago, therefore the incidence of the pathogen in the area can expected to be low. Wilting in chili pepper plants is not exclusive to the attack of Phytophthora, but can also be due to the attack of other pathogens, such as Fusarium or Rhizoctonia (^{Vásquez et al., 2009}; ^{Anaya-López et al., 2011}; ^{Lozano et al., 2015}; ^{Pérez-Acevedo et al., 2017}). This may be possible because in the soil there are a large diversity of microorganisms such as fungi or bacteria with the ability to cause wilt in the chili pepper crop (^{Pérez-Acevedo et al., 2017}). There are non-pathogenic species of Fusarium or special proven ways to reduce the incidence and severity of diseases. It is important to understand that widening the chili pepper crop to the cocoa-producing areas favors a scenario of overpopulation of Phytophthora in cocoa and chili peppers which, in the medium term, could undergo severe losses due to the sexual reproduction between both species (^{Hurtado-González and Lamour, 2009}), given the evidence that P. capsici is near the cocoa-producing areas, with signs of A2 compatibility and P. capsici^{Tsao and Alizadeh (1988}), proportion of A1 surpasses 95% in the plains and 70% in the highlands of Tabasco (^{Ortiz-García, 1996}).

Figure 4 Maximum likelihood phylogenetic analysis of Phytophthora capsici with other Phytophthora species of clade 2 and Pythium oopapillum external group. The maximum bootstrap (1000 repetitions) values are indicated as percentages in the points of the branch. The bar of the scale indicates 0.010 substitutions per site per branch. 

In conclusion, P. capsici was identified in habanero pepper plants in previously cocoa-producing areas (Acachapa and Colmena, Centro, Tabasco), where the isolation of Phytophthora (CH132) was pathogenic in chili pepper plants and was not the case for Fusarium isolations in the pathogenicity test. Due to this, the chili pepper production is suggested to be channeled to areas other than the cocoa-producing areas of Tabasco.