The Cucumber mosaic virus (CMV, Cucumovirus, Bromoviridae) is one of the viruses with a wider host range that includes cucurbitaceae, solanaceae, cereals, legumes and ornamental plants. Some of the symptoms caused by CMV are mosaic, chlorosis, necrosis, foliar deformation and stunting, among others. Infected plants may also be asymptomatic (^{Palukaitis et al., 1992}). CMV has been a study model of the different plant-pathogen interaction mechanisms, and one of the 10 most economically important phytopathogenic viruses worldwide (^{Rybicki, 2015}).

Infections caused by CMV limit crop production in more than 18 states of Mexico, and in some cases losses of more than 80% have been reported (^{Robles et al., 2010}).

CMV was detected in Durango and Primo cantaloupe melon cultivars grown in the municipality of Tecomán, state of Colima, Mexico (^{Orozco et al., 1994}). Since then, no other hosts or subgroups of the virus are known nor its association with satellite RNAs (satRNA) in that location, so the objective of this study was to identify CMV isolates and their satRNA in plants in the state of Colima.

For that purpose, foliar tissue of cultivated plants and weeds with symptoms associated with CMV (i.e. mosaic, mottling, leaf filimorphism and deformation, as well as stunting, or a combination of them) was collected in the municipalities of Armería, Tecomán and Cuauhtémoc (Figures 1A and 1B). In the laboratory, the samples were washed with deionized water to remove impurities and then stored at -80 °C until their analysis.

Total RNA extraction was performed using 100 mg of tissue with the Plant RNA Purification Reagent (Invitrogen, USA), following the manufacturer’s recommendations. The precipitate with ribonucleic acid was dissolved in 32 µL of treated water with 0.01% (v/v) diethylpyrocarbonate. The concentration of nucleic acids was quantified using a NanoDrop 2000 spectrophotometer (Thermo Scientific, USA) and stored at -20 °C.

Later on, reverse transcription reactions coupled to endpoint polymerase chain reaction (RT-PCR) were performed in one step to detect: a) a fragment of approximately 300 bp of the 18S endogenous gene which indicates that the extracts were of adequate quality with the oligonucleotides described by ^{Zamboni et al. (2008}); b) one ~500 bp segment of the CMV capsid protein gene (CP) with the oligonucleotides reported by ^{Wylie et al. (1993}); and c) satRNA with the oligonucleotides reported by ^{Chen et al. (2011}) that amplify a ~390 bp fragment. All the RT-PCR reactions were performed in 5 µL volumes and consisted of a mixture of 1X reaction buffer (0.2 mM of dNTPs and 1.2 mM of MgSO₄), 0.5 µM of the corresponding oligonucleotides (sense and anti-sense), 0.6% (v/v) polyvinylpyrrolidone 40 (PVP₄₀), 0.2 µL of the enzyme stock solution included in the SuperScript^™ III One-Step RT-PCR System with Platinum Taq ^® High Fidelity kit (Invitrogen, USA) and an average concentration of 32 ng of each RNA extract.

The RT-PCRs were performed using a VeritiFast thermocycler (Applied Biosystems, USA) as follows: one 32-min cycle at 50 °C, one 2-min cycle at 94 °C, followed by 40 15-s cycles at 94 °C, 15-s cycles at 55 °C (for 18 S), 60 °C (for CMV CP) and 53 °C (for satRNA) and 15-s cycles at 68 °C; and one final 5-min cycle at 68 °C.

An extract of double-stranded RNA from Nicotiana glauca foliar tissue positive to CMV and satRNA by RT-PCR and sequencing was used as a positive control in the different trials; as negative controls, PCR grade water and one RNA extract of Catharanthus roseus foliar tissue negative to CMV and satRNA by RT-PCR were used.

The RT-PCR amplifications were analyzed by horizontal electrophoresis in 1.5 % (w/v) agarose gels in a TAE 1X buffer (tris-acetate 40 mM, Na₂EDTA•2H₂O 2 mM) at 80 V. The gel was stained with ethidium bromide (0.5 µg/mL) and exposed to UV light; the resulting image was visualized in a photo-documentation system (MF-ChemiBIS-DNR Bio-Imaging Systems, Jerusalem, Israel).

The corresponding amplicons to CMV CP and satRNA of the different isolates obtained were purified with the Wizard^® SV Gel and PCR Clean-Up System kit (Promega, USA), following the manufacturer’s instructions; the sequencing was performed in the Institute of Neurobiology’s Laboratory of Proteomics, Universidad Nacional Autónoma de México, Juriquilla campus at Querétaro (http://www.inb.unam.mx/unidades/molecularyanalitica/bios.html), using the Sanger’s method.

The sequences obtained in this study and other corresponding to the CP of different CMV subgroups (IA: U32859, U2281, D10538, Y18137, U43888, D00385, AJ276481 and AJ131623, IB: D42079, D28780, U20219 and X89652, and II: Y18138, U22822, Z12818, AB006813, AJ131620, AJ131621, L15336 and D00463) and satRNA (with necrogenic domain: D84389, D10039, U31660, D28559, DQ839133, X86421, X86425, D00542 and D10038; and without necrogenic domain: J02061, D10037, D42081, D42082, X86409, X86424, X54065 and X69136) reported worldwide, as well as the CP sequences of the Brome mosaic virus (BMV) (X58458) and the satRNA of the Peanut stunt virus (PSV) (NC_003855.1), which were used as external groups, were obtained from the GenBank-NCBI in fasta txt format and aligned with Clustal W. The phylogenetic trees were constructed using MEGA6.0 (^{Tamura et al., 2013}) and the optimal maximum likelihood criterion and the T92 substitution model (which was best suited to the sequences) with 1,000 bootstrap repetitions.

A total of 30 samples were collected: 11 of weeds: 2 of empanadilla (Commelina sp.), 2 of Cucurbita sp., 2 of amargosilla (Parthenium sp.), 2 of clover (Trifolium sp.), 1 of common mallow (Malva sp.) and 2 of white bladdermallow (Herissantia sp.); 5 of ornamental plants: 3 of periwinkle (Catharanthus roseus) and 2 of white lily (Hippeastrum sp.); and 14 of cultivated plants: 3 of papaya (Carica papaya ‘CW-3’), 3 of tomato (Solanum lycopersicum Saladet ‘Río Grande’), 3 of melon (Cucumis melo cantaloupe), 2 of chili pepper (Capsicum annuum ‘Jalapeño’) and 2 of cucumber (Cucumis sativus). From the 30 samples, 4 amplified the expected band of approximately 500 bp corresponding to a segment of the CMV coat protein (CMV-CP) (Figure 1C). Two isolates were obtained in Tecomán: one of melon (CMV-Mel) showing mosaic (19° 02’ 40” N, 103° 54’ 04.9” W), and another of periwinkle (CMV-Vin) (18° 54’ 52” N, 103° 52’ 45.7” W) from plants showing mosaic and foliar deformation (Figures 1A and 1B), all collected in Caleras. The third isolate, detected on chili pepper (CMV-Chi), was collected in the municipality of Armería (18°55’36” N, 103°58’32.5’’ W) from foliar tissue showing yellowing, mosaic and foliar deformation. The fourth isolate was detected on tomato (CMV-Tom) in plants showing mosaic and foliar distortion collected in the municipality of Cuauhtémoc (19° 19’ 42” N, 103° 36’ 08.1” W).

The products size of the CP sequencing of the four CMV isolates obtained in this study ranged from 455 to 479 bp, with 94% and 97% nucleotidic identity with other isolates of the same virus reported in the GenBank database.

Three out of the four CMV isolates used in the present study showed a lower level of variability among them but were different from the tomato isolates (CMV-Tom). A higher level of variability could be observed among isolates if the number of plant species analyzed increases. According to the phylogenetic analysis, the CMV-Tom was grouped with isolates of the IA subgroup, and more closely with a banana isolate from Colombia (U32859). CMV-Vin, CMV-Chi and CMV-Mel were grouped with the IB subgroup (Figure 1E). The subgroup I isolates are mainly found in tropical and subtropical zones, while the isolates from the subgroup II prevail in temperate zones (^{Palukaitis et al., 1992}).

Figure 1 (A) Cantaloupe melon and (B) periwinkle foliar tissue infected with CMV and satellite RNA; periwinkle foliar tis sue showing mosaic and foliar deformation. (C) Amplified product of the CMV CP, and (D) satellite RNA by end point one step RT-PCR. Lane 1: molecular weight marker of 100 bp (Promega®); Lane 2: positive control (extract of double-stranded RNA from Nicotiana glauca infected with CMV and satellite RNA); Lane 3: negative control (extract of RNA of Catharanthus roseus foliar tissue negative to CMV and satellite RNA); Lane 4: negative control (PCR grade water); Lane 5: tomato sample (CMV-Tom); Lane 6: periwinkle sample (CMV-Vin); Lane 7: chili pepper sample (CMV-Chi); Lane 8: melon sample (CMV-Mel). Phylogenetic analysis of (E) CMV isolates, and (F) satellite RNA collected in Colima compared with nucleotidic sequences obtained from the GenBank-NCBI. 

The CMV-Vin isolate was the only one containing a satRNA (Figure 1D) with an estimated size of 381 bp by sequencing and 91 % nucleotidic identity with other isolates of the world. The satellite was grouped with other isolates without necrogenic domain (X86424, X54065, X69136 and X86409) (Figure 1F). This result shows the need to study the satRNA’s biological behavior, particularly the symptoms observed when it is present, as well as its possible effect on CMV transmission by aphids, among other aspects.

Despite the low number of CMV isolates obtained and characterized in this study, our results confirmed the presence of CMV in the state of Colima and suggested that there is diversity among them. The present study suggest the need to continue analyzing a greater number of samples in order to deepen this observation.

In conclusion, three new CMV isolates are reported in the state of Colima, in chili pepper, tomato and periwinkle, the latter with a satellite RNA without necrogenic domain, as well as the presence of CMV in cantaloupe melon in the municipality of Tecomán.