Introduction

Upland cotton (Gossypium hirsutum L.) is one of the most leading fiber crop. In Pakistan cotton is a major crop after wheat which occupies the largest cultivated area as compared to other crops (Rehman et al. 2016). Pakistan is ranked among 4^th largest cotton producing countries worldwide (Shuli et al. 2018). Cotton crop is affected by many foliar diseases in field conditions. Historically, bacterial blight of cotton is one of the most significant cotton foliar disease caused by Xanthomonas citri pv. malvacearum (Xcm). The bacterium infect all above portion of cotton plant but cause greatest loss when the bolls are attacked. BBC produce several symptoms include water soaking lesions, leaf and fruit spots, leaf blight, angular shaped necrotic spots and boll rot (Mhedbi-Hajri et al. 2011). In favorable environmental conditions BBC can cause 20 to 50% disease losses (Arabsalmani et al. 2002; Razaghi et al. 2012). In Pakistan, during the past few years 20–37% BBC disease incidence has been reported (Hamid et al. 2012).

Pathogenic bacterial identification and classification is always complex, which requires a series of tests for conclusive identification (Alvarez 2004). Misidentification of pathogens has slowed down the efforts to combat many bacterial diseases of Pakistan. However, sequence-based techniques have replaced traditional identification methods due to their high sensitivity and reliability (Brady et al. 2013; Meng et al. 2015). DNA sequencing methods are useful for identification of both new and established pathogens (Aritua et al. 2015). Partial sequence of protein coding genes is useful for bacterial specie identification of family Enterobacteriaceae. 16SrRNA genes sequence are useful for the determination of phylogenetic relationship between distantly related species (Dauga 2002). Bacterial housekeeping genes and whole genome sequence provide reliable information for bacterial taxonomic nomenclature (Kim et al. 2014). Average nucleotide identity (ANI) is considered as most widely used next generation standard for species identification. The ANI shows identical values between homologous regions of two or more genomes (Haley et al. 2010; Chan et al. 2012; Yi et al. 2012; Grim et al. 2013). Using multi-genic and genomic analysis, many researchers have identified and characterized the BBC pathogen in several cotton growing countries of the world (Delannoy et al. 2005; Razaghi et al. 2012; Pritchard 2016), but in Pakistan the importance of this bacterial diseases on cotton crop was undervalued and not studied well. Some drastic changes in climatic conditions; increase in temperature, humidity and erratic rainfall pattern since last decade have made the BBC more prevalent every year in Pakistan.

The objective of this study was to evaluate and identify the pathogen responsible for bacterial blight of cotton in cotton growing areas of Punjab, Pakistan. Both conventional and molecular approaches were used for correct pathogen identification and species delineation. Bacterial isolates were tested on cotton plants in glasshouse under control conditions to check their ability to produce BBC symptoms.

Materials and Methods

Collection of samples from cotton growing areas

In 2016–2017 infected cotton leaf samples showing typical bacterial blight of cotton symptoms (leaf blight, angular shaped necrotic lesions) were collected from three districts of Punjab, Pakistan that included Burewala, Multan and Bahawalpur. After collection, infected samples were placed in airtight polyethylene bags with other supporting data such as sample location, host variety and sample number.

Isolation and identification of bacterial pathogen

The cotton leaf samples showing BBC symptoms were cut into pieces of 2–3 mm size. These small pieces were dipped into 70% ethanol for 30 sec followed by 2–3 dipping in sterile dd/H₂O for surface sterilization. Samples were dried on autoclave filter paper disc and placed aseptically onto nutrient agar (NA) plates. The petri dishes were incubated for 24h at 28°C to examine for bacterial growth. Different biochemical tests including Gram staining, KOH Test, Tween 80 hydrolysis Test, Fluorescent Pigment Test and Kovacs’ Oxidase Test were carried out for the confirmation of bacterial pathogen on genus level.

DNA extraction and normalization

Single bacterial colonies were obtained from overnight growth of bacteria on NA media by using pure stock culture. Pure colonies were suspended in nutrient broth and incubated for 24 h on shaker in growth chamber at 28°C. Total DNA was extracted by using Qiagen DNeasy Blood & Tissue Kit. The isolated dsDNA concentration was quantified by Nano-drop spectrophotometer. The concentration of extracted DNA was visually assessed as above on a 1% agarose gel by loading DNA sample (5 μL DNA + 2 μL loading dye) on gel electrophoresis at 100 V for 30 min. Qubit fluorometer was used for the process of dsDNA normalization. Qubit reagents and standards were prepared according to Kit protocol, which helped to calculate the actual amount of dsDNA in original samples. The samples were normalized up to 6 ng/µL for genomic library preparation.

Genomic library preparation

The normalized DNA (6 ng/µL) of four bacterial isolates (B10, Bo7, M5 and M8) were used for the preparation of genomic library. QIAGEN QIAseq FX DNA library kit protocol was used for genomic library preparation. The quality of DNA libraries were assessed with the help of Agilent Bioanalyzer 2100. The genomic libraries were quantified and normalized with the help of Pico-green and Qubit fluorometer. Next generation sequencing (NGS) protocol Illumina HiSeq 4000 was performed at the Genome Center, University of California, Davis for sequencing of genomic libraries.

Phylogenetic analysis of sequencing data

The quality of sequenced libraries was assessed. SPAdes was used for the assembly of sequenced genomes (Bankevich et al. 2012). Perl script was used to extract 16S rRNA and other housekeeping genes (gyrB, atpD, L2, leuS and rpoB) from genome sequence and phylogenetic tree was constructed using Mega 7 (Brady et al. 2013). PyANI software was used for specie assignment based on average nucleotide identity (ANI) >95% (Pritchard 2016).

Pathogenicity assay

Seed of BBC susceptible cotton variety (Acala Maxxa) were collected from U.S. National Plant Germplasm System. Cotton seeds were sown in plastic pots under greenhouse controlled conditions (26°C with >85% relative humidity and 16 h photoperiod). Bacterial isolates (B10, Bo7, M5 and M8) were revived from -40 °C by streaking on NA media and cultured for 24 h at 28°C. They were sub-cultured on nutrient broth and harvested by centrifuge after 12 h. The bacterial pellet was suspended in sterile dd/H₂O. The concentration of culture was maintained at 10⁶ cfu/mL. Spray technique was used to inoculate cotton plants at 5–7 leaf stage. Sterile ddH₂O was used as control. After inoculation plants were closely monitored for the appearance of bacterial blight symptoms. Analysis of variance (ANOVA) was used to check the variation among bacterial strains with respect to number of lesions. Least significant difference (LSD) test was used to all possible mean comparisons between the bacterial strains (Steel at al. 1996).

Results

Sampling for bacterial isolation and identification

Typical bacterial blight of cotton symptoms (water-soaked spots, necrotic lesions, yellowing and defoliation of cotton leaves) were observed during BBC samples collection (Fig. 1a). Creamy or yellow growth was observed around infected samples on nutrient agar media. Pure single colonies of bacterial pathogen were successfully grown by sub-culturing on nutrient agar media (Fig. 2a). Physiological and conventional biochemical tests including Gram staining, KOH Test, Tween 80 hydrolysis Test, Fluorescent Pigment Test and Kovacs’ Oxidase Test confirmed isolated bacteria to be Gram negative (Fig. 2b–c).

Fig. 1: (a) field symptoms of BBC during samples collection. (b) Control cotton leaves inoculated with water. (c) Infected cotton leaves showing BBC symptoms in greenhouse

Fig. 2: Image (a) showing single colonies of bacterial isolates on NA medium. (b) representing the viscous thread made during KOH test to confirm gram staining. (c) Milky white precipitation appear when bacterial colonies were subjected to Tween 80 hydrolysis test

Fig. 3: Maximum likelihood phylogenetic tree based on 16S rRNA sequences of Bo7, B10, M5 and M8 bacterial isolates showing Pantoea species and their closest phylogenetic neighbors

Genome statistics

High quality genome sequences were obtain using next generation sequence (Table 1). The BLAST searches based on whole genome sequence of four isolates B10, Bo7, M5 and M8 using NCBI (http://www.ncbi.nlm.nih.gov) revealed similarities to Pantoea species. The average genome size of four isolates B10, Bo7, M5 and M8 was ~ 4.7 Mbp with 56% genomic GC content. Moreover, average 81.75% tRNA-coding genes along with 7.25% rRNA genes were predicted in the sequence of these four B10, Bo7, M5 and M8 isolates (Table 1). All sequences were submitted to NCBI GenBank for the assignment of accession numbers (Table 1).

Phylogenetic analysis

Phylogenetic trees were constructed using 16SrRNA genes sequence and five common housekeeping genes (gyrB, Table 1: Showing samples name, genome statistics and their accession numbers

Sample	Genome size	Scaffold N50	GC (%)	CDS	Proteins	tRNA	rRNA	Accession Number
Bo7	3878409	488816	0.5647	3699	3608	82	9	VWUX00000000
B10	4986677	189135	0.5746	4892	4796	87	9	VWVO00000000
M8	4647578	341114	0.5680	4374	4300	70	4	VWUI00000000
M5	5397540	55261	0.5676	5514	5419	88	7	VWUG00000000

Fig. 6: Comparative effect of four bacterial isolates (B10, B07, M8, M5) on no. of lesions on cotton variety Acala Maxxa

atpD, L2, leuS and rpoB), which were extracted from the genome sequence of bacterial isolates B10, Bo7, M5 and M8 (Fig. 3 and 4). The phylogenetic tree of 16SrRNA made two clusters M8, Bo7 and B10 were close to P. agglomerans while M5 showed resemblance with P. dispersa (Fig. 3). As 16sRNA gene was not enough to distinguish the species, different housekeeping genes were used to identify the Pantoea species. Three clusters were formed when tree was constructed based on concatenated sequence of five housekeeping genes (Fig. 4). Isolates from Multan (M5 and M8) showed close resemblance with Pantoea anthophila and supported by 100% bootstrap values (Fig. 4). Isolates from Burewala (Bo7) and Bahawalpur (B10) were closely related to P. eucrina and P. dispersa respectively (Fig. 4). Based on initial BLAST searches on NCBI reference genomes of different Pantoea species along with four isolates (B10, Bo7, M5 and M8) from current experiment were subjected to construct heatmap based on ANI data (Fig. 5). The ANI₉₅ values revealed three groups supporting the clades from multigene analysis. Isolates from Multan (M5 and M8) had >98% ANI to P. anthophila and some unidentified Pantoea species while isolates from Burewala (Bo7) and Bahawalpur (B10) districts showed >99% ANI to P. eucrina-LMG-5346 and >98% ANI P. dispersa-NS375, respectively (Fig. 5).

Pathogenicity assay

Bacterial blight disease symptoms (water-soaked spots, necrotic lesions with yellow halo and defoliation) were noticed after 2 weeks of cotton leaves inoculation (Fig. 1c). Symptoms observed in the green house were identical to those observed at the farmer fields (Fig. 1a). The bacterial pathogen was re-isolated from the inoculated cotton leaves, which fulfilled the Koch’s postulates.

A significant difference was observed among bacterial strains in terms of lesions production on susceptible cotton leaves (Fig. 6). Among four bacterial isolates (B10, Bo7, M5 and M8) the isolate B10 produced highest number of lesions while isolate M5 produced least number of lesions on cotton leaves (Fig. 6).

Discussion

In Pakistan, the pathogen of bacterial blight of cotton was previously diagnosed by using classical methods

Fig. 4: Phylogenetic tree made by concatenated nucleotide sequences of five housekeeping genes gyrB, atpD, L2, leuS and rpoB. Maximum likelihood tree having bootstrap values after 1000 replicates are shown in percentage

Fig. 5: Heatmap based on ANI₉₅ revealed three Pantoea species; Pantoea anthophila, P. eucrina and P. dispersa

based on morphology and appearance of symptoms (Hamid et al. 2012). In the present study, the samples were collected from core cotton areas of Punjab, Pakistan. For an accurate identification, the BBC pathogen was diagnosed using morphological, molecular and pathogenicity assays. Basic morphological studies provided the preliminary validation of the pathogen based on the bacterial growth pattern, cell shape and color of bacterial colonies (Schaad et al. 2001). The growth pattern and colony color formation of BBC isolates found in this study was similar to those previously reported (Delannoy et al. 2005). The bacterial morphological features, including colonies color, cell shape are not always remain constant and can be easily influenced by environmental conditions (Brady et al. 2013). Therefore, biochemical and molecular techniques are indispensable for correct identification of bacterial genus and species. Phenotypic and biochemical properties of BBC bacterial isolates of our study were almost constant and similar to described earlier (Schaad et al. 2001; Medrano and Bell 2007). The phylogenetic tree of 16SrRNA made two clusters M8, Bo7 and B10 were close to P. agglomerans while M5 showed resemblance with P. dispersa (Fig. 3). Sequence data of multiple genes are useful for determination of phylogenetic relationships which reduce ambiguities caused by genetic recombination (Nancy et al. 2005; Brady et al. 2013). Following housekeeping genes (gyrB, atpD, L2, leuS and rpoB) of B10, Bo7, M5 and M8 isolates were used to create phylogenetic tree which formed three clusters with reference strains. Bacterial isolates M5 and M8 showed cluster with P. anthophila while other two isolates Bo7, B10 were closely related to P. eucrina and P. dispersa, respectively (Fig. 4).

The 16SrRNA and housekeeping genes (gyrB, atpD, L2, leuS and rpoB) results of Pakistani isolates were closely related to the findings of Brady et al. (2013). The 16SrRNA phylogenetic results were quite different from five housekeeping genes results. Isolates M8, Bo7 and B10 in 16SrRNA phylogenetic tree were close to P. agglomerans while in housekeeping phylogenetic tree Bo7 and B10 were close to P. eucrina and P. dispersa. Similarly, M5 in 16SrRNA phylogenetic tree was close to P. dispersa while in five housekeeping genes phylogenetic tree M5 and M8 showed resemblance with P. anthophila (Fig. 3 and 4). Sequence of 16S rRNA and other housekeeping genes is indeed a popular traditional method for species delineation. However, these methods based on single or set of conserved genes does not provide sufficient resolution at species level due to genetic discontinuities among closely related taxa (Mende et al. 2013). In recent years, the average nucleotide identity (ANI) has emerged as a robust method for bacterial species delineation (Goris et al. 2007; Richter and Rossello 2009). The finding of ANI₉₅ data of four isolates B10, Bo7, M5 and M8 were similar to housekeeping genes results (Fig. 5). Instead of Xanthomonas citri pv. malvacearum (Xcm) three Pantoea spp (P. eucrina, P. anthophila and P. dispersa) were reported from infected BBC samples (Fig. 5). There are many Pantoea species has been identified causing blight symptoms on different hosts (Kini et al. 2017; Filho et al. 2018). In order to study the isolated bacterial genomes many in silco studies were carried out. We have reported single circular chromosome with average of ~ 4.7 Mbp genome size and 56% genomic GC content (Table 1). Our genome statistical results were almost constant and similar to describe by (Palmer et al. 2016). Matsuzawa et al. (2012) found 4.8 Mbp genome size for P. agglomerans strain IG1. Similarly, P. ananatis isolated from infected onion have genome size range from 4.8 to 5.1 Mbp (De Maayer et al. 2014). Greenhouse experiment confirmed these isolates (M5, M8, Bo7 and B10) to be causal agent of BBC. The isolates from Bahwalpur region B10 produced highest number of lesions on susceptible cotton variety Acala Maxxa followed by Bo7, M8 and M5 (Fig. 6). The greenhouse BBC symptoms were similar to those reported by Sambamurty (2006).

Conclusion

This experimental study specified the importance of advance molecular techniques for pathogen identification and specie delineation. The pathogen was confirmed three Pantoea spp (Pantoea anthophila, P. eucrina and P. dispersa) instead of Xanthomonas citri pv. malvacearum (Xcm). The genome sequence of bacterial isolates (B10, Bo7, M5 and M8) has provided insight into the genomic features of these Pantoea species. Greenhouse experiment has confirmed P. anthophila, P. eucrina and P. dispersa to be responsible for BBC symptoms production.

Acknowledgments

We thank Dr. Douglas R. Cook, Brendan K. Reily, Yunpeng Gai and Amna Fayyaz for helping us in genome analysis. This work was supported by Cook’s Lab University of California, Davis and Higher Education Commission’s IRSIP program.

References

Alvarez AM (2004). Integrated approaches for detection of plant pathogenic bacteria and diagnosis of bacterial diseases. Annu Rev Phytopathol 42:339‒66

Arabsalmani M, H Rahimian, F Azad, A Qhasemi (2002). Occurrence of bacterial blight of cotton caused by Xanthomonas axonopodis pv. malvacearum in Khorasan province. In: Proceeding of the 15^th Iranian Plant Protection Congress, p:68. Razi University, Kermanshah, Iran

Aritua V, J Harrison, M Sapp, R Buruchara, J Smith, DJ Studholme (2015). Genome sequencing reveals a new lineage associated with lablab bean and genetic exchange between Xanthomonas axonopodis pv. phaseoli and Xanthomonas fuscans subsp. fuscans. Front Microbiol 6:1080

Bankevich A, S Nurk, D Antipov, A Gurevich, M Dvorkin, AS Kulikov, V Lesin, S Nikolenko, S Pham, A Prjibelski, A Pyshkin, A Sirotkin, N Vyahhi, G Tesler, MA Alekseyev, PA Pevzner (2012). SPAdes: A new genome assembly algorithm and its applications to single-cell sequencing. J Comput Biol 19:455‒477

Brady C, I Cleenwerck, S Venter, T Coutinho, P De Vos (2013). Taxonomic evaluation of the genus Enterobacter based on multi-locus sequence analysis (MLSA): Proposal to reclassify E. nimipressuralis and E. amnigenus into Lelliottia gen. nov. as Lelliottia nimipressuralis comb. nov. and Lelliottia amnigena comb. nov., respectively, E. gergoviae and E. pyrinus into Pluralibactergen.nov. as Pluralibacter gergoviae comb. nov. and Pluralibacter pyrinus comb. nov., respectively, E. cowanii, E. radicincitans, E. oryzae and E. arachidis into Kosakonia gen. nov. as Kosakonia cowanii comb. nov., Kosakonia radicincitans comb. nov., Kosakonia oryzae comb. nov. and Kosakonia arachidis comb. nov., respectively, and E. turicensis, E. helveticus and E. pulveris into Cronobacter as Cronobacter zurichensis nom. nov., Cronobacter helveticus comb. nov. and Cronobacter pulveris comb. nov., respectively, and emended description of the genera Enterobacter and Cronobacter. Syst Appl Microbiol 36:309‒319

Chan JZM, MR Halachev, NJ Loman, C Constantinidou, MJ Pallen (2012). Defining bacterial species in the genomic era: Insights from the genus Acinetobacter. BMC Microbiol 12; Article 302

Dauga C (2002). Evolution of the gyrB gene and the molecular phylogeny of Enterobacteriaceae: A model molecule for molecular systematic studies. Intl J Syst Evol Microbiol 52:531–47.

Delannoy E, B Lyon, P Marmey, A Jalloul, JL Montillet, JF Daniel, M Essenberg, M Nicole (2005). Resistance of cotton to Xanthomonas citri pv. malvacearum. Annu Rev Phytopathol 43:62‒82

De Maayer P., Chan W. Y., Blom J., Venter S. N., Duffy B., Smits T. H. M., et al. . (2012). The large universal Pantoea plasmid LPP-1 plays a major role in biological and ecological diversification. BMC Genomics 13; Article 625

Filho RC, LL Rodrigues, AG Abreu, RR Souza, PHN Rangel, RN Mello, GA Rocha, MG Cunha (2018). Detection of Pantoea agglomerans in germplasm rice accessions (Oryza sativa) in Brazil. Plant Dis 102:1‒237

Goris J, KT Konstantinidis, JA Klappenbach, T Coenye, P Vandamme, JM Tiedje (2007). DNA–DNA hybridization values and their relationship to whole-genome sequence similarities. Intl J Syst Evol Microbiol 57:81‒91

Grim CJ, ML Kotewicz, KA Power, G Gopinath, AA Franco, KG Jarvis, QQ Yan, SA Jackson, V Sathyamoorthy, L Hu, F Pagotto, C Iversen, A Lehner, R Stephan, S Fanning, BD Tall (2013). Pan-genome analysis of the emerging foodborne pathogen Cronobacter spp. suggests a species-level bidirectional divergence driven by niche adaptation. BMC Genomics 14; Article 366

Haley BJ, CJ Grim, NA Hasan, SY Choi, J Chun, TS Brettin, DC Bruce, JF Challacombe, JC Detter, CS Han, A Huq, RR Colwell (2010). Comparative genomic analysis reveals evidence of two novel Vibrio species closely related to V. cholerae. BMC Microbiol 10; Article 154

Hamid MI, MAKZ Iqbal, MU Ghazanfar, Y Iftikhar, N Akhtar (2012). Correlation of environmental conditions with bacterial blight disease of cotton (Gossypium hirsutum L.). Pak J Phytopathol 24:39‒43

Kim M, HS Oh, SC Park, J Chun (2014). Towards a taxonomic coherence between average nucleotide identity and 16S rRNA gene sequence similarity for species demarcation of prokaryotes. Intl J Syst Evol Microbiol 64:346‒351

Kini K, R Agnimonhan, O Afolabi, B Soglonou, D Silué, R Koebnik (2017). First report of a new bacterial leaf blight of rice caused by Pantoea ananatis and Pantoea stewartii in Togo. Plant Dis 101:241‒242

Matsuzawa T, K Mori, T Kadowaki, M Shimada, K Tashiro, S Kuhara, H Inagawa, GI Soma, K Takegawa (2012). Genome sequence of Pantoea agglomerans strain IG1. J Bacteriol 194:1258‒1259

Medrano EG, AA Bell (2007). Role of Pantoea agglomerans in opportunistic bacterial seed and boll rot of cotton (Gossypium hirsutum) grown in the field. J Appl Microbiol 102:134‒143

Mende D, S Sunagawa, G Zeller, P Bork (2013). Accurate and universal delineation of prokaryotic species. Nat Meth 10:881‒884

Meng X, I Bertani, P Abbruscato, P Piffanelli, D Licastro, C Wang, V Venturi (2015). Draft genome sequence of rice endophyte-associated isolate Kosakonia oryzae KO348. Genome Announc 3; Article e00594‒15

Mhedbi-Hajri N, MA Jacques, R Koebnik (2011). Adhesion mechanisms of plant-pathogenic Xanthomonadaceae. In: Advances in Experimental Medicine and Biology, Vol 715, pp:71‒90, Linke D, A Goldman (eds.). Springer, New York, USA

Nancy AM, AR Jacob, K Ryuichi, F Takema (2005). Evolutionary relationships of three new species of Enterobacteriaceae living as symbionts of aphids and other insects. Appl Environ Microbiol 71:3302‒3310

Palmer M, P de Maayer, M Poulsen, ET Steenkamp, E van Zyl, TA Coutinho, SN Venter (2016). Draft genome sequences of Pantoea agglomerans and Pantoea vagans isolates associated with termites. Stand Genomic Sci 11; Article 23

Pritchard (2016). Genomics and taxonomy in diagnostics for food security: Soft-rotting enterobacterial plant pathogens. Anal Meth 8:12–24

Razaghi A, N Hasanzadeh, A Ghasemi (2012). Characterization of Xanthomonas citri subsp. malvacearum strains in Iran. Afr J Microbiol Res 6:1165‒1170

Rehman A, L Jingdong, AA Chandio, I Hussain, SA Wagan, QUA Memon (2016). Economic perspectives of cotton crop in Pakistan: A time series analysis (1970–2015) (Part 1). J Saudi Soc Agric Sci 18:49‒54

Richter M, RM Rossello (2009). Shifting the genomic gold standard for the prokaryotic species definition. Proc Natl Acad Sci 106:19126‒19131

Sambamurty AVSS (2006). Bacterial diseases and plant galls, Chapter 9. In: A Textbook of Plant Pathology, pp:173‒220. I K International Publishing House Pvt. Ltd, New Delhi, India

Schaad NW, JB Jones, W Chum (2001). Laboratory Guide for Identification of Plant Pathogenic Bacteria, 3^rd edn. APS Press, St. Paul, Minnesota, USA

Shuli F, AH Jarwar, X Wang, L Wang, Q Ma (2018). Overview of the cotton in Pakistan and its future prospects. Pak J Agric Res 31:396‒407

Steel RG, JH Torrie, DA Deekey (1996). Principles and Procedures of Statistics. A biometrical approach 3^rd edn. McGraw Hill book Co. Inc. New York, USA

Yi H, YJ Cho, SH Yoon, SC Park, J Chun (2012). Comparative genomics of Neisseria weaveri clarifies the taxonomy of this species and identifies genetic determinants that may be associated with virulence. FEMS Microbiol 328:100‒105