Sampling plant tissues

Samples were taken in Tabasco (October 20-24, 2014) with the technical support of the personnel from the Tabasco Local Plant Health Committee (Comité Local de Sanidad Vegetal de Tabasco - CESVETAB) for the location, visual diagnosis and collection of material, according to instructions by SENASICA. The samples were deposited in Ziploc bags and labelled with the information of the tissue collected, date and sampling site. The samples were places in coolers with cooling gels and sent to the banana molecular biotechnology lab in the Scientific Research Center (Centro de Investigación Científica) in Yucatan, where they were processed.

Detection of R. solanacearum by immunology tests

To detect R. solanacearum, the ELISA (“Enzyme-Linked Immunosorbent Assay”) (Agdia®, N.C. SRP 33900) was carried out in a 96-well tray, following the instructions of the manufacturer. Positive controls and samples of plants infected with R. solanacearum produce a blue color, whereas negative controls and samples of healthy plants produce a colorless result. The tray was visually examined to avoid false positives by possible contaminations or impurities on the tray and the spectrophotometric absorbance was measured at 655 nm (Bio-Rad Modelo 680XR). Optic densities equal to or lower than 0.09 was considered negative. To scrutinize the samples in the lab, ImmunoStrips (Agdia®, N.C. STX33900) were also used, following the instructions of the manufacturer.

DNA extraction

To enrich the R. solanacearum population in the SMSA medium, a fragment (1 cm²) of the infected plant tissue was deposited in the liquid medium and incubated for 24 h, shaking constantly at 180 rpm at 28 °C (^{Álvarez-Restrepo et al., 2008}). One aliquot of this culture (1.5 mL) was centrifuged at 5,200 X g for 5 min. The bacterial pellet was resuspended with 1 mL of the buffer CTAB (2% cetyl trimethyl ammonium bromide, 1% polyvinylpyrrolidone, 100 mM Tris-HCl, 1.4 M NaCl, 20 mM EDTA), stirring with a vortex for 30s. Later it was frozen at -20 °C for 10 min and incubated at 65 °C in a water bath for 20 min. It was tempered for 1 min at room temperature and centrifuged at 20,800 X g. The supernatant was recovered in a 1.5 mL tube for centrifuging, 500 µL of phenol:chlorophorm:isoamylic (25:24:1) were added, and stirred for 10 min at room temperature (^{CNRF, 2012}). Later, it was centrifuged for 5 min at 20,800 X g and the acqueous phase was collected. Next, the standard DNA recovery procedure was carried out using cold absolute ethanol, washings with 70% ethanol, air-drying of the pellet and resuspension in ultrapure distilled water (^{Sambrook and Russell, 2001}). The DNA purity was determined by spectrophotometry at 260nm/280nm and the integrity was determined by electrophoresis in agarose gel.

Detection of R. solanacearum by PCR

Detection was carried out by the amplification of genomic regions with the primers of 759: 5’- GTC GCC GTC AAC TCA CTT TCC - 3’ and 760: 5’- GTC GCC GTC AGC AAT GCG GAA TCG- 3 (Universal diagnosis for R. solanacearum, 280 pb), (^{Opina et al., 1997}); pehA#3-F 5’-CAG CAG AAC CCG CGC CTG ATC CAG- 3’ and pehA#6-R 5’- ATC GGA CTT GAT GCG CAG GCC GTT- 3’ (gen pga, 532pb), (^{Gillings et al., 1993}); ISRso19-F 5’- TGG GAG AGG ATG GCG GCT TT - 3’ and ISRso19-R 5’ - TGA CCC GCC TTT CGG TGT TT - 3’ (primers specific to Race 2; 1884pb), (^{Lee and Chin, 2003}). The PCR reaction mixture was prepared with 20ng of DNA, 0.1 µM of each primer, 1 U of recombinant DNA Taq polymerase (Invitrogen®), IX of the Buffer of the Taq enzyme, 200 µM of each dNTP and 1.5 mM MgCl2, in 25µL of final volume. The program in the thermocycler (Bio-Rad model T100™) for the universal diagnosis of R. solanacearum (primers 759/760) had a denaturalization cycle of 2 min at 96 °C, followed by 35 cycles of 94 °C for 45 s, alignment step at 64 °C for 20 s, and a step of 72 °C for 1 min, and finally, an extension cycle of 72 °C for 5 min. The amplification program for the other genes was identical, except for the step of alignment was 70 °C for 45 s for the gene pga, and for Race 2, alignment was 55°C for 30 s, and the extension of 72 °C for 1.5 min in each cycle. The products (10 µL of each PCR reaction) were analyzed by electrophoresis in 1.5% agarose gel (p/v) using the buffer TAE (Tris-acetate-EDTA) 1X supplemented with 3 µL of ethidium bromide (10mg/mL). The strips were visualized over a transilluminator using UV light (Gel Doc™ XR, Bio-Rad).

Isolation of R. solanacearum

Tissues (pseudostem and fruit) which were positive at immunology tests ELISA and ImmunoStrips were used for bacterial isolation. Samples were cut into small pieces with a sterilized blade, and later disinfested by washing twice with sterilized water, sodium hypochlorite at 10% (from a commercial solution at 6%) for 1 min, sterile water, ethanol for 30 s, sterile water, and it was later placed in TE buffer and left to incubate for an hour, stirring softly. A bacterial loop was placed on a semi-selective SMSA medium (formula for every liter: 1g of casamino acids, 10g of peptone, 5g of glucose, 5mg crystal violet, 100mg polymyxin β-sulfate, 25mg bacitracine, 5mg chloranphenicol, 0.5mg penicillin, 17g bacteriological agar, 50mg TZC (2,3,5-triphenyl tetrazolium chloride) and incubated at 28 °C for 72 h (^{Kelman, 1954}; ^{Cardozo et al., 2010}). The bacterial colonies were streak by crossing on a solid B-King medium and incubated at 28 °C for 48 h. The B-King medium helps to distinguish Pseudomonas spp. bacteria, since they produce fluorescence when exposed to a UV wavelength of 360 nm, whereas R. solanacearum does not fluoresce (^{King et al., 1954}; ^{Dulla and Lindow, 2009}; ^{Lamichhane and Varvaro, 2013}).

Caracterization of the sequevar

For the characterization of the sequevars, we used the primers of the Mus serie from ^{Fegan and Prior (2005)}: 5´-CGGGTCGCTGAGACGAATATC-3´ and 5´-GCCTTGTCCAGAATCCGAATG-3´ (Sequevar 4 , 351 pb); 5´-GCAGTAAAGAAACCCGGTGTT-3´ and 5´-TCTGGCGAAAGACGGGATGG-3´ (Sequevar 3, 400 pb); 5´-GCTGGCATTGCTCCCGCTCAC-3´ and 5´-TCGCTTCCGCCAAGACGC-3´ (Sequevar 4 SFR, 167 pb); 5´-CGTTCTCCTTGTCAGCGATGG-3´ and 5´-CCCGTGTGACCCCGATAGC-3´ (Sequevar 6, 221 pb). PCR was carried out in Multiplex using 20 ng of DNA, 2 U of recombinant DNA Taq polymerase (Invitrogen®), 1X of the Buffer of the Taq enzyme, 200 µM of each dNTP, 1.5 mM MgCl2 and 6 pmoles of each primer, in a final volume of 25µL. The PCR conditions were similar to those described previously, using an alignment temperature of 59 °C. Single PCRs for each sequevar were carried out in the same conditions, but using only one pair of primers in each case. In each case, 15 µL of the product of the reaction was analyzed by in gel electrophoresis.

Isolation of R. solanacearum

With the sequential use of 2 culture media, SMSA and B-King, it was possible to isolate 25 R. solanacearum strains from the 28 positive samples. The three samples from which R. solanacearum could not be recovered had the palest diagnosis lines in the ImmunoStrips, indicating that these samples contained a low bacteria count.

The SMSA medium is frequently used to isolate R. solanacearum, since it avoids the growth of many other microorganisms (^{Kelman, 1954}; ^{Álvarez-Restrepo et al., 2008}; ^{Cardozo et al., 2010}). However, even so, R. solanacearum is difficult to isolate since it has slow growth rates in vitro, while other bacteria, e.g. Pseudomonas, grow quickly and contaminate the isolations (^{Ramesh et al., 2009}; ^{Champoiseau et al., 2009}; ^{Nasim, 2011}; ^{Kalpage and De Costa, 2015}). The combination of the media SMSA and B-King helps to distinguish R. solanacearum from Pseudomonas spp. Both of these bacteria grow on the SMSA medium as mucoid cultures, with irregular, white edges, and pink to red centers. On the B-King medium, Pseudomonas spp. fluoresces, whereas R. solanacearum do not (Figure 2), helping to put them apart. Ramesh et al., (2009) point out that genus Pseudomonas is opportunist during the colonization of the host by R. solanacearum. Along with Pseudomonas, the bacteria Klebsiella, Erwinia, and other species of the genus Ralstonia, such as Ralstonia mannitolilytica, are mostly endophytic of the banana (^{Thomas et al., 2008}; ^{Ganen et al., 2009}; ^{Souza et al., 2013}). The bacteria recovered from the 3 samples that produced faint lines on the ImmunoStrips presented fluorescence on the B-King medium and did not amplify in any of the PCR diagnosis reactions for R. solanacearum. Some of these bacteria were identified by sequencing the 16S, and belonged to Pseudomonas fulva, Pseudomonas aeruginosa, and Pseudomonas citronellolis, consistent with reports by the bibliography.

Figure 2 Preselection of bacteria with similar phenotypes to R. solanacearum. Isolation on SMSA medium (panel A). Strains with mucoid characteristics, white, irregular edges and red centers were transferred on a B-King medium. Bacteria of the genus Pseudomonas fluoresce on this medium (Panel B). Non-fluorescent bacteria were selected (panel C) to continue the isolation of R. solanacearum. 

Confirmation of the identity of R. solanacearum

The identity of the R. solanacearum strains was confirmed by sequencing the amplified fragment with primers pehA#3-F, pehA#6-R (fragment of the gene pga). The 25 strains that amplified the 532pb diagnostic band were confirmed to be R. solanacearum using the BlastN analysis with the sequences obtained, and compared with the GenBank database (hit 99-100% identity with R. solanacearum). Based on the positive PCR results when using the primers ISRso19, the 25 isolated strains corresponded to Race 2.

The identification of the sequevars was first determined by PCR Multiplex, using the primers of the series Mus (^{Fegan and Prior, 2005}). According to these authors, each sample must give only one diagnostic DNA band: ~167 pb (sequevar 4 ecotype SFR), ~350 pb (sequevar 4), ~220 pb (sequevar 6), ~400 pb (sequevar 3). These DNA bands were observed among the products of the Multiplex (Figure 3A), although they were not unique DNA bands per sample. According to the Multiplex (Figure 3A), the Race 2 R. solanacearum strains genotype in sequevar 3 (lanes 11 and 13), sequevar 4 (lanes 3 and 5), sequevar 4 SFR (lane 1), and sequevar 6 (lanes 2, 4, 6, 8). Other strains produced profiles that could not be assigned to any of the sequevars (e.g. lanes 7, 10). Due to the inconsistency in results of the PCR in Multiplex, each strain was later genotyped using independent PCR for each sequevar. No Tabasco strains amplified with the pairs of primers for sequevars 3, 4 and 4 SFR (data not shown). The 25 strains amplified the diagnostic DNA band of ~220 with the pair of primers SI28, that corresponds to sequevar 6 (Figure 3B). Sequevar 6 originated in the Carribbean islands (Trinidad) and entered the continent mainland through Venezuela; in 1950, it was accidentally introduced into Costa Rica in infected banana corms. In 1960, a commercial freight ship introduced it accidentally into Honduras, and from there it spread rapidly northward to Mexico and southward to Panama (^{Sequeira, 1998}). The first report in Mexico was in 1960 in Chiapas. Currently, sequevar 6 has been reported in North America in Hawaii (^{Fegan and Prior, 2005}) and Florida (^{Hong et al., 2012}); in Central America, in Guatemala (^{Sánchez-Pérez et al., 2008}), and in South America, in Brazil (^{Alburquerque et al., 2014}). Central and South America are considered the center of origin of strains related to the Moko disease of banana (^{Sequerira, 1998}; ^{Fegan and Prior, 2005}, 2006). Other countries in the Americas have, along with sequevar 6, some of the sequevars 3, 4, 24, 25, 41, and 53 (^{Raymundo et al., 1998}; ^{Prior and Fegan. 2005}; ^{Alburquerque et al., 2014}). Studies in Mexico must be broadened to determine if there are other sequevars related to the bacterium causing Moko disease in banana.

Figure 3 Characterization of R. solanacearum bacteria isolated from banana plants; tipification of the sequevars. Panel A, PCR in Multiplex (primers of the series Mus and SI28 together). Lane 1 shows a band of ~167 pb that corresponds to expected for sequevar 4 SFR, and another of ~450 pb, not described for any sequevar; lane 2 shows a band of approximately ~220 pb that corresponds to descriptions for sequevar 6; lane 3 showed a band of ~350 pb, which corresponds to sequevar 4. The following lanes showed unexpected bands, such as ~850 pb (lanes 5 and 7) and ~1000pb (lane 10). Panel B, PCR with specific primers for sequevar 6 (primers SI28, diagnosis band of ~220 pb). In both panels the lanes correspond to: M) Marker 1 Kb Plus DNA Ladder (Invitrogen® Carlsbad, CA, USA); lanes 1-13, strains of R. solanacearum, 1) MT01, 2) MT02, 3) MT03, 4) MT04, 5) MT05, 6) MT06, 7) MT07, 8) MT08, 9) MT09, 10) MT10, 11) MT11, 12) MT12, 13) MT13. 

The taxonomic classification of the strains of the R. solanacearum species complex (RSSC) has been under constant revision and correction for over 50 years, and more frequently in the last 10 years (^{Peeters et al., 2013}; ^{Prior et al., 2016}). As the world collections increase, and along with them, the known global genetic diversity of the RSSC, discrepancies have been found in the classification results of some strains with different diagnosis tools (^{Hong et al., 2012}), and even in the classification of populations, such as in the cases of the isolated R. solanacaerun of Solanaceae and Cucurbits in the French Guyana, and tomatoes in Guadalupe (^{Álvarez-Restrepo et al., 2008}; ^{Deberdt et al., 2014}). Although the reasons for the inconsistencies are not entirely clear, some of the differences have turned into diagnosis tools. For example, many strains of sequevar 4 produce a 351 pb band with primers Mus06, but they additionally amplify a band of 167pb (^{Álvarez-Restrepo et al., 2008}), while some strains of sequevar 4 isolated in Florida in 2008 do not produce the diagnosis band of 351pb, but only the unspecific 167pb band, therefore the authors propose that the Florida strains that only amplify the 167pb band are emerging strains (^{Hong et al., 2012}).

The Multiplex to distinguish between sequevars 3, 4, and 6 within Race 2 (i.e. strains causing banana Moko) was generated by ^{Fegan and Prior (2005)} and it has been used to classify strains in Martinique (^{Wicker et al., 2007}, ²⁰⁰⁹), Colombia (^{Álvarez-Restrepo et al., 2008}), United States (^{Hong et al., 2012}), French Guyana (^{Deberdt et al., 2014}), Brazil (^{Albuquerque et al., 2014}), and others. In the reports that show electrophoresic analyses of the PCR products, we can notice unspecific bands in some lanes, alongside the diagnosis band (^{Álvarez-Restrepo et al., 2008}; ^{Hong et al., 2012}; ^{Albuquerque et al., 2014}), as observed in this investigation about Mexico. Even the analysis of one same sequevar shows discrepancies between the different research groups. Sequevar 4 amplifies 2 bands in the Moko Multiplex, although the band sizes are different between the report by ^{Álvarez-Restrepo et al., (2008)} and the one by ^{Albuquerque et al., (2014)}, despite both investigations having used agarose gels with the same concentrations (1.5% p/v).

In this investigation the diagnosis by Multiplex was ambiguous. In the case of the strains that amplified the 221pb band (sequevar 6), it was a single band (Figure 3A, lanes 2, 4, 6, 8, 12), contrary to results by ^{Albuquerque et al., (2014)} who obtained 2 bands for the strain IBSBF2661 (sequevar 6), but similar to results obtained by ^{Hong et al., (2012)}, who also obtained one band for strain 527 (sequevar 6) in the Moko Multiplex. In some Mexican strains, the diagnostic band expected for sequevars 3 (400pb, lane 11) and 4 (351pb, 167pb, lanes 1 and 3) were amplified, although other bands observed in those same lanes do not coincide with any of the reports. PCR bands were also observed that have not been reported previously in the Multiplex results for Moko strains (Figure 3A, lanes 5, 7, 10), or in other words, they are new bands. The new ~850 pb band amplifies in different strains (Figure 3A, lanes 5 and 7), suggesting they have a genetic origin, and it reflects the diversity in the populations of R. solanacearum. These results are difficult to interpret in further detail because the diagnosis primers amplify marker bands, although the function in the microorganism of those genomic regions is unknown (^{Deberdt et al., 2014}).

The presence of new and reproducible PCR products in different strains (e.g. ~850 pb band in strains MT05 and MT07) could be interpreted as an emergence of new strains of R. solanacearum in Tabasco, as proposed for the Florida strains that presented atypical results in this Multiplex diagnosis (^{Hong et al., 2012}). Due to the ambiguity of results, the individual diagnosis of each sequevar was carried out. No amplification of any strain was obtained in the diagnosis of sequevars 3 or 4 (data not shown). All the strains isolated in Tabasco in this investigation belong to sequevar 6 (Figure 3B). The recommendation by the present report is to avoid basing only in Moko Multiplex results to propose the emergence of new R. solanacearum strains. We suggest amplified products to be sequenced.

Figure 4B presents a summary of the protocol proposed from the present investigation, in order to simplify the isolation of R. solanacearum Race 2 strains. It is compared with the procedure followed by other investigations (Figure 4A). Procedure 4A is more expensive, since the diagnosis by ELISA or ImmunoStrips must be carried out on each isolated bacterial strain. In the present research, that strategy had little success. When the samples were not diagnosed before isolation, a large part of the bacteria isolated on the SMSA were not R. solanacearum. The sequential use of B-King medium after the isolation on SMSA is a quick and inexpensive option that allows the phenotypic preselection of R. solanacearum. According to the bibliography, the most reliable R. solanacearum identity diagnoses are PCR and sequencing, whether of ITS or specific genes (^{Fegan and Prior, 2005}). After confirming the identity of R. solanacearum, the collection of isolated bacteria can be characterized genetically, phenotypically, phylogenetically, or be used for other studies. We recommend performing individual PCR with the specific primers of each Moko sequevar instead of the Moko Multiplex whenever results are unclear.

Figure 4 Schematic comparison of the protocol generally used for the isolation of R. solanacearum related to the Moko disease (Panel A), and the protocol recommended in the present study (panel B). Panel A is based on ^{Álvarez- Restrepo et al. (2008)}; ^{Dööolotkeldieva and Bobusheva (2016)}; ^{Hong et al. (2012)}; ^{Sánchez-Pérez et al. (2008)}. 

Moko is a quarantine-regulated disease in Mexico and a phytosanitary campaign is established. Identified outbreaks are therefore under continuous surveillance and eradication measures. It is important to emphasize that due to regulations and for food security, it is crucial to prevent the dispersal of the pathogen by collection of the material by inexperienced personnel. For this reason, all studies of this pathogen in Mexico should be carried out with the permission of, and in collaboration with the General Plant Health Board (Dirección General de Sanidad Vegetal - DGSV), the National Food Health, Safety and Quality Service (Servicio Nacional de Sanidad, Inocuidad y Calidad Agroalimentaria - SENASICA) and the State Plant Health Committees (Comités Estatales de Sanidad Vegetal).