The marigold (Tagetes erecta) is an annual plant from the Asteraceae family, native to Mexico. The flowers are a source of carotenoid yellow-orange pigments, mostly xanthophylls such as the lutein (^{Deineka et al., 2007}). Besides pigments, marigold also produces metabolites with antioxidant, hepatoprotective, and nematocidal activity (^{Gopi et al., 2012}), and, peculiarly, it is used as ornament in the rituals of the Day of the Death in Mexico (^{Brandes, 1998}).

Due to the importance of marigold as a source of secondary metabolites and as component of the Mexican cultural heritage, information about microbial pathogens is required to set strategies of crop protection and prevent economical losses, especially in small-scale producers. Marigold seedlings cultivated in the field are susceptible to damping-off caused by the fungal pathogen Ceratobasidium sp. (^{Saroj et al., 2013}). Unlike damping-off, diseases caused by bacteria might not prevent flower production if emerge in the late stages of plant development, and the commercialization of diseased plants contributes to the dispersal of disease. Bacterial wilt caused by Pseudomonas solanacearum and bacterial leaf spot produced by Pseudomonas syringae pv. tagetis are among the most damaging bacterial diseases in marigold (^{Horst, 2013}). P. syringae pv. tagetis was first reported as Pseudomonas tagetis (^{Hellmers, 1955}), and causes apical chlorosis in sunflower (Helianthus annuus) and Jerusalem artichoke (Ambrosia tuberosa) (^{Gulya et al., 1982}; ^{Shane and Baumer, 1984}). Besides its detrimental effects, this pathogen has the potential to be used as a biocontrol agent as causes chlorosis and necrosis when inoculated in weeds such as the Canada thistle (Cirsium arvense) (^{Gronwald et al., 2002}) and wollyleaf bur ragweed (Ambrosia grayi) (^{Sheikh et al., 2001}). Chlorosis is likely produced by tagetitoxin, an inhibitor of the plastidial RNA polymerase (^{Mathews and Durbin, 1990}).

A novel bacterial strain that resembles Pseudomonas syringae and named LF2012 was initially isolated from diseased marigold plants by Dr. Leopoldo Fucikovsky+ (Colegio de Posgraduados). Spray inoculation of 1x10⁸ CFU mL^-1 into three plants was performed to prove the Koch’s postulates. After inoculation, the plants were maintained in a growth chamber under controlled conditions for light (approximately 300 µmoles m^-2s^-1), temperature (27 °C), photoperiod (16 h light/8 h dark), and relative humidity (75 %). This experiment was repeated three times. In this set of experiments, P. syrinage pv. tomato DC3000 was used as reference of causal agent of disease in Arabidopsis and tomato, and P. syringae pv. syringae 61 was included as non- host strain as it causes no disease in the plant species tested.

Chlorosis, accompanied by a necrotic halo, became visible five days after inoculation in the three plants tested in every experiment (Figure 1). Bacteria were re-isolated from the diseased plant, which confirms this strain is pathogenic towards marigold. Host range assays were performed in plants belonging to five different families, including Asteraceae, Amaranthaceae, Brassicaceae, Poaceae, and Solanaceae (^{Valenzuela-Soto et al., 2015}). Other phytopathogenic strains were included as controls. The results are summarized in Table 1. LF2012 only caused symptoms in marigold plants but did not in other plants, including sunflower (H. annuus) “small flower” that also belongs to Asteraceae, and it has been reported as a host of P. syringae pv. tagetis (^{Rhodehamel and Durbin, 1985}). We tested three different cultivars of sunflower, in any case LF2012 caused disease in sunflower, hence, we ruled out the possibility to name the strain as P. syringae pv. tagetis.

These results suggest that LH2012 is an adapted pathogen causing disease in marigold, but it is non-adapted to access into the other plant species. Non-adapted pathogens do not cause a reaction when inoculated by spray because they do not access into the plant. But if the pathogen is forced to invade mesophyll, it will be perceived by the plant as hazardous agent, and a strong defense response will be induced. Detection of pathogen-derived molecules triggers the non-host resistance, often characterized by hypersensitive response or cell death at the site of infiltration that restricts the pathogen’s propagation (^{Lee et al., 2017}). To confirm that LH2012 has features of non-adapted pathogen towards other species, we performed infiltrations of 1x10⁸ CFU mL^-1 of LH2012 into tobacco leaves. LF2012 triggers hypersensitivity reaction (HR) like spots (Figure 2). As expected to this reaction, it was limited to the infiltration site no matter how long the plants were incubated. This result suggests that when LH2012 is recognized as pathogen in a non-host plant when is directly inoculated in mesophyll.

Figure 1 Disease symptoms of leaf spot disease on marigold (Tagetes erecta) caused by P. syringae LF2012 (LF2012) five days after spray inoculation. Mock treated plants were sprayed with phosphate buffer (Mock). Plants were kept under a photoperiod (16 h light/8 h dark), light of approximately 300 μmoles m-2s-1, 28 °C, and relative humidity of 75%. 

The metabolic profile was performed by the BIOLOG test as a first approach to identify the species of LF2012. LF2012 was inoculated in the Biolog universal growth medium and incubated at 28 °C. The data were acquired and analyzed with the MicroLog version 4.2 data software. The profile as related to P. syringae pathovars. Furthermore, the utilization of myo-inositol, D-sorbitol, D-mannose, D-fructose, α-D-glucose, sucrose, L-fucose, and maltose as carbon sources were similar to the sources reported to P. syringae pv. tagetis (^{Rhodehamel and Durbin, 1985}). Identification as P. syringae pv. tagetis was not possible as the metabolic profile of this pathovar is not included in the available database.

Figure 2  P. syringae LF2012 causes an HR-like response in infiltrated tobacco leaves (Nicotiana taba cum) (LF2012). Control plants were infiltrated with phosphate buffer (Mock). Potted plants were kept under controlled conditions of light, temperature, and humidity. Plants were photo graphed 5 days after infiltration. 

The sequence of the 16S rRNA gene is widely used as a barcode gene, and it is used to the construction of phylogenetic relationships within prokaryotes. Specifically, in Pseudomonas has been useful to redistribute some Pseudomonas species into other genera (^{Mulet et al., 2010}). The Pseudomonas genus is divided into two lineages: aeruginosa and fluorescens. Pseudomonas syringae is classified into the fluorescens lineage, and seven genomospecies (Gs) within P. syringae are distinguishable by phylogenetic analysis (^{Marcelletti and Scortichini, 2014}).

The 16S rRNA gene of LF2012 was amplified and cloned by the following experimental procedures reported before by our workgroup (^{Valenzuela-Soto et al., 2015}). The purified plasmid was subject to capillary sequencing in an ABI PRISM 3700 DNA Analyzer (Applied Biosystems) using the T7 promoter and M13 Reverse oligonucleotides that match flanking sequences of pCR-XL-TOPO (Invitrogen). The sequence was deposited in the GeneBank under the Accession Number KP796138.1. The 16S rRNA sequence of LF2012 and the corresponding sequences from representative strains of the two Pseudomonas lineages were aligned by the ClustalW algorithm. After alignment, a phylogenetic tree was constructed by the neighbor-joining method. Both bioinformatics procedures were conducted in MEGA6 (^{Tamura et al., 2013}). The phylogenetic tree shows that the 16S rRNA sequence of LF2012 is clustered into the fluorescens lineage, and it is closely related to genomospecies Gs3, Gs6, Gs8, Gs9 (Figure 3). The sequences from the aeruginosa lineage are clearly separated from fluorescens. The genomospecies closely related to LF2012 belong to the P. syringae group (^{Mulet et al., 2010}). The reference sequence of P. syringae pv. tagetis used in our analysis (AB001449.1) is clustered with other phytopathogen strains, but P. syringae LF2012 is excluded from this subtree. The tree also shows that LF2012 is less related to other species fluorescens lineage such a P. putida and P. savastanoi. Therefore, we classified this strain as Pseudomonas syringae LF2012. Analysis of other markers is necessary to determine the relationship with P. syringae pv. tagetis or sequences of reference strains for the genomospecies 7, such as P. syringae pv. helianthi, which so far is lacking.

Figure 3 Phylogenetic tree showing the relationship of the Pseudomonas syringae LF2012 isolated from marigold plants within a subset of closely related strains and species. The accession number of every sequence used is indicated in the branch before the name. 

A macro-restriction profile was performed by digesting the genomic DNA of LF2012 and our control strains with I-CeuI. Restriction products were separated by pulse-field gel electrophoresis (PFGE) (Figure 4). The restriction pattern of LF2012 (line A, Figure 4) was like the profile of P. syringae pv. syringae 61 form the Gs1 (line B, Figure 4), but no common fragments were detected between LF2012 and P. syringae pv. glycinea (line C, Figure 4) a member of the amygdali group within the fluorescens lineage (^{Gomila et al., 2017}). Differences and similarities in these restriction patterns supports the classification of the strain as P. syringae LF2012.

In conclusion, results obtained here from biochemical profile and molecular analysis permitted identifying this bacterium as Pseudomonas syringae LF2012, a pathogen that causes leaf spot disease symptoms in marigold. It is avirulent in another Asteraceae plant and plants from other families and triggers HR-like reaction in infiltrated tobacco leaves. The whole-genome sequencing of this novel strain will help determine the accurate classification that helps in molecular methods of marigold pathogens and the identification of virulence factors that illustrate the molecular basis of the disease. Due to the relation to P. syringae pv. tagetis its potential as biocontrol of weeds might be explored.

Figure 4 Pulsed-Field Gel Electrophoresis (PFGE) macrorestriction profiles produced by P. syringae LF2012 and other phytopathogenic bacteria. Chromosomal DNA was digested with the rare-cutting endonuclease I-CeuI and sub sequently separated by PFGE. (λ) Lambda ladder size marker. PFGE patterns of: (A) Pseudomonas syringae pv. tagetis LF2012; (B) P. syringae pv. syringae 61; (C) P. syringae pv. glycinea PG4180; (D) P. syringae pv. tomato DC3000; (E) Pectobacterium cacticidum FHLGJ22, and (Y) Saccharomyces cerevisiae YPH80 chromosome; (kb) sizes in kilo base-pairs.