Methylobacterium sp. resides in unculturable state in potato tissues in vitro and becomes culturable after induction by Pseudomonas fluorescens IMGB163


Olga Podolich, Institute of Agroecology of UAAS, Metrologichna str. 12, 03143 Kyiv, Ukraine. E-mail:


Aims:  To induce growth of endophytic bacteria residing in an unculturable state in tissues of in vitro-grown potato plantlets. To isolate and identify the induced bacteria and to localize the strains in tissues of in vitro-grown potato plantlets.

Methods and Results:  The inoculation of in vitro-grown potato plants with Pseudomonas fluorescens IMBG163 led to induction of another bacterium, a pink-pigmented facultative methylotroph that was identified as Methylobacterium sp. using phylogenetic 16S rDNA approach. Two molecular methods were used for localizing methylobacteria in potato plantlets: PCR and in situ hybridization (ISH/FISH). A PCR product specific for the Methylobacterium genus was found in DNA isolated from the surface-sterilized plantlet leaves. Presence of Methylobacterium rRNA was detected by ISH/FISH in leaves and stems of inoculated as well as axenic potato plantlets although the bacterium cannot be isolated from the axenic plants.

Conclusion: Methylobacterium sp. resides in unculturable state within tissues of in vitro-grown potato plants and becomes culturable after inoculation with P. fluorescens IMBG163.

Significance and Impact of the Study:  In order to develop endophytic biofertilizers and biocontrol agents, a detailed knowledge of the life-style of endophytes is essential. To our knowledge, this is the first report on increase of the culturability of endophytes in response to inoculation by nonpathogenic bacteria.


The pink-pigmented facultative methylotrophic bacteria (PPFMB) are indispensable inhabitants of the phyllosphere where methanol is produced (Corpe 1985). Under natural conditions, colonization of plant leaves by PPFMB seems to occur via soil particle transfer to the leaves by air (Romanovskaya et al. 2001; Lidstrom and Chistoserdova 2002). Bacteria in the genus Methylobacterium are often isolated from buds, roots and in vitro plant cultures, and they are also known as endophytic bacteria, found inside plant tissue (Corpe and Basil 1982; Wilson et al. 1999; Pirttiläet al. 2000, 2005; Sy et al. 2001; Idris et al. 2004). These bacteria evidently have an intimate association with plants, as they promote plant growth and development by production of phytohormones (Ivanova et al. 2000, 2001; Romanovskaya et al. 2001; Araujo et al. 2002; Madhaiyan et al. 2004). The genus Methylobacterium belongs to α-Proteobacteria and members of this genus are capable of growing on one carbon compounds such as methanol and methylamine, as well as on a wide range of multicarbon substrates (Corpe and Basil 1982). Associated with their carbon utilitization capabilities, methylotrophic bacteria can be efficient destructors of organic pollutants in the environment (Kayser et al. 2002; Van Aken et al. 2004).

Endophytic community of potato plants varies, and it is composed of a broad phylogenetic spectrum of bacteria: α-, β-, and γ-Proteobacteria, Flexibacter-Cytophaga-Bacteroides group, Gram-positive micro-organisms with high-G + C-content, and Planctomycetales (Garbeva et al. 2001; Krechel et al. 2002; Sessitsch et al. 2004). Bacteria of the genus Methylobacterium have not earlier been detected inside the in vitro-grown potato tissue. Here we describe isolation and localization of a bacterial isolate capable of utilizing methanol as the sole source of carbon and energy in tissues of in vitro-grown potato plants.

Materials and methods

Potato plantlets

Two potato cvs. Chervona Ruta and Nigru, provided by The Potato Institute UAAS (Kyiv Region, Ukraine), were used in this study. In vitro-grown potato plantlets (Solanum tuberosum L.), cv. Chervona Ruta, had been grown on Murashige and Skoog medium (MS) (Murashige and Skoog 1962) without phytohormones for one and a half year. The potato cv. Nigru tubers were used for introduction of sprout segments into tissue culture. All in vitro potato plants were grown in a conditioned room under a light-dark period of 16/8 h at 24°C and 150 μE m−2 s−1.

Bacterial strains and culture conditions

Rhizobacterium Pseudomonas fluorescens IMBG163, isolated earlier from soil (from collection of Institute of Molecular Biology and Genetics of NASU), was cultivated in the glycerol-peptone KB liquid medium (King et al. 1954) for 18 h at 28°C on an orbital shaker (110 rev min−1). PPFMB isolate was cultured in M9 (Miller 1972) supplemented with methanol (1·0%) medium at 28°C for 3–5 days.

Isolation of endophytic bacteria and plant inoculation

For isolation of endophytic bacteria, plant tissues were first surface-sterilized in 70%-ethanol for 1 min and in 6% calcium hypochlorite for 20 min, and rinsed three times 5 min in sterile distilled water (SDW). The plant material was crushed in mortar with a pestle, serially diluted and cultivated on KB, LB and M9 agar media mentioned above for 1–4 days.

For the inoculation of potato plants, a rhizobacterial culture of P. fluorescens was grown to the density of log 8 CFU ml−1, and a 10-ml aliquot was pelleted by centrifugation at 5000 rev min−1 for 5 min. The pellet was washed with SDW, and the centrifugation step was repeated. The bacterial cells were suspended in 10 ml SDW, and plant tissues were incubated in the bacterial suspension at room temperature for 90 min. Segments from sprout tips (3–4 cm) were disinfected as mentioned above, dried on a sterile paper after inoculation and placed on MS without hormones. Untreated plants were incubated in SDW instead of the bacterial culture.

Bacterial DNA isolation, PCR and sequencing

Bacterial DNA isolation was performed with UltraCleanTM Microbial DNA isolation kit (Mo Bio Laboratories, Carlsbad, USA). To determine the nucleotide sequence of the rrs gene (16S ribosomal RNA), a PCR product was amplified with primers pA and pH described by Edwards et al. (1989), cleaned with UltraCleanTM PCR Clean-up DNA purification kit (Mo Bio Laboratories) and cloned into vector pTZ57R/T using InsTAcloneTM PCR Cloning Kit (Fermentas, Vilnus, Lithuania). The PCR product was then sequenced with primers M13/pUC (forward and reverse) (Fermentas) by the Sanger method (Sanger et al. 1977) using Amersham (Uppsala, Sweden) sequencing kit Cy5 AutoCycle and apparatus ALF express (Pharmacia Biotech, Uppsala, Sweden). The nucleotide sequence was analysed with the basic local alignment search tool (Blast; Altschul et al. 1990) and Vector NTI 8·0 program (Infomax Inc., Des Moines, USA).

Molecular phylogenetic analysis

The 16S ribosomal RNA sequences of Methylobacterium species were retrieved from Ribosomal Database Project (RDP) (Cole et al. 2007) and aligned by using ClustalW (Thompson et al. 1994). A distance matrix was created with DNADIST of Phylip version 3·6 (Felsenstein 1989) from which the tree topology was built by the neighbour joining method in the program neighbor. The confidence for individual branches of the resulting tree was estimated by performing 1000 bootstrap replicates.

DNA isolation from potato plants and Methylobacterium-specific PCR

Total DNA was isolated from surface-sterilized roots, leaves and stems of 3–4-week-old in vitro-grown plantlets according to Aljanabi and Martinez (1997). The isolated DNA (100 ng) was amplified with Methylobacterium-specific primers MB4 (5′-CCGCGTGAGTGATGAAGG-3′, Escherichia coli positioning 409-417) and MB (5′-AGCGCCGTCGGGTAAGA-3′, E. coli positioning 1388-1371). The primer MB is highly specific for the genus Methylobacterium as it is 100% identical with 449 strains (including all Methylobacterium type strains) out of the total of 488 Methylobacterium strains present in RDP (Cole et al. 2007), and virtually no other genera. The primer MB4 is identical with 1127 out of 1531 strains in the genus Methylobacterium but also with many strains in the phylum Proteobacteria. The primers and PCR conditions were tested with Methylobacterium extorquens strain F (Pirttiläet al. 2000) as the control strain. The DNA was first denatured at 94°C for 5 min, and the first four cycles of 1 min at 94°C, 1 min at 62°C, and 3 min at 72°C were performed. The cycles were repeated four times with 60°C and 30 times with 58°C as annealing temperatures. The final elongation was carried out at 72°C for 5 min. The PCR products were separated in a 2·0% agarose gel containing 0·5 μmol l−1 ethidium bromide under electrophoresis for 90 min at 80 V and visualized under UV light.

In situ hybridization

Samples of stems, leaves, and roots of eight-week-old plantlets of cv. Chervona Ruta were taken for in situ hybridization experiments. Both PPFMB-infected and untreated plants were analysed. All samples were surface-sterilized for 1 min in 70% ethanol and for 15 min in 6% calcium hypochlorite, followed by rinsing three times in SDW. The leaf tissues were cut in pieces of 1·5 × 1·5 mm, stems were cut in 1 mm2 and roots in 1 mm2 pieces. The cuttings were fixed in 2% paraformaldehyde, 2·5% glutaraldehyde, 0·1 mol l−1 saline buffer, pH 7·4. The fixed samples were dehydrated, cleared through ethanol/t-butanol series and embedded in paraffin. Oligonucleotide probes MB and E11 (Pirttiläet al. 2000) complementary to the unique regions of 16S rRNA of Methylobacterium and eubacteria, respectively, were used for in situ hybridization. The oligonucleotides were end-labelled with fluorescein during synthesis and with digoxigenin using the DIG Oligonucleotide 3′-End Labeling Kit (Roche Applied Science, Espoo, Finland). Hybridization was performed as described by Pirttiläet al. (2000). The detection of the digoxigenin label was performed with the DIG Nucleic Acid Detection Kit (Roche Applied Science, IN, USA) and the slides were viewed under bright field illumination. For the fluorescein label, the samples were viewed under fluorescent light with fluor objectives, episcopic-fluorescence attachment EF-D, and the filter set UV-1A (Nikon, Tokyo, Japan).

DNA staining

The RNase-treated sections were stained with ethidium bromide (10 μg ml−1) for 45 min in the dark. The slides were rinsed with water, air dried in the dark and viewed immediately under fluorescent light.

Nucleotide sequence accession number

The sequence generated in this study has been deposited in the GenBank database under accession number EF583689.


Isolation and identification of PPFMB

Initially, surface-disinfected potato sprouts were inoculated with P. fluorescens IMBG163 to test the potential for growth promotion and biocontrol. The number of culturable endophytic bacteria isolated from inoculated in vitro potato plants tissue (1·2 × 104 CFU g−1 in roots; 0·4 × 102 CFU g−1 in stems) was higher compared to uninfected ones.

The inoculation led to induction of pink-pigmented bacterium that was named M1. When success of infection by the P. fluorescens IMBG163 was tested, both the P. fluorescens and M1 were isolated on the selective media from in vitro-grown plants of F1 to F5 vegetative generations, but not from the untreated plants (Fig. 1). Absence of the M1 strain was confirmed in the original inoculant culture. The M1 bacteria were characterized as pink-pigmented, Gram-negative, nonsporulating rods capable of utilizing methanol as the sole source of carbon and energy, which is a specific attribute of the genus Methylobacterium.

Figure 1.

 Isolation of the M1 strain from the in vitro potato plants. (C) Root and leaf of control plants. (Ps) Root and leaf of P. fluorescens IMBG163 inoculated plants.

The isolate M1 was further characterized by sequencing the 16S rDNA. Comparison of a specific sequence of the rrs gene with sequences deposited to GenBank suggested that the M1 isolate belongs to methylotrophic bacteria, having the highest homology to Methylobacterium radiotolerans (96–100%) and M. organophilum (90%). A detailed analysis of the sequence demonstrated that the region conserved between methylobacteria was present, and a variable region of 50 bp (nucleotides 901–951) was identical to M. radiotolerans. The sequence was 99% identical and had a seqmatch score of 0·933 with type strains of M. radiotolerans (JCM 2831, DSM 1819) in RDP (Cole et al. 2007). The M1 isolate differed from other species of methylotrophic bacteria at 3–10 positions at the border of the sequence. When a phylogenetic tree was constructed from the type strains of the Methylobacterium genus (Fig. 2), the isolate M1, named IMBG290, formed a cluster with Methylobacterium radiotolerans, supported by a bootstrap value of 862.

Figure 2.

 Phylogenetic position of the Methylobacterium sp. IMBG290, isolated from in vitro plantlets of potato. The 16S rDNA sequences of the isolate, Escherichia coli (outgroup), Methylobacterium oryzae and type strains of the Methylobacterium genus, as represented in the RDP (Cole et al. 2007), were analysed with the neighbour joining method. Values from 1000 bootstrap repeats are presented if support was >50%. The species names are followed by the GenBank accession numbers.

Analysis of in vitro-grown potato DNA by Methylobacterium-specific PCR

When the DNA of Methylobacterium sp. was amplified with primers MB and MB4, specific for the genus Methylobacterium, a product of the expected size of 1015 bp was obtained. When DNA was isolated from leaves of both uninfected and P. fluorescens-infected potato plants and amplified with the Methylobacterium-specific primers, the product was detected in the treated plants after two vegetative generations (Fig. 3). The band was not amplified from the DNA of untreated plants.

Figure 3.

 Electrophoregram of PCR products generated by primers specific for the Methylobacterium genus (MB and MB4). Products from DNA of two generations of potato plants infected with P. fluorescens IMBG163 (2; 3); untreated potato plants (4); Methylobacterium DNA (1); Mass Ruler Mix (Fermentas, Lithuania) (M).

A study of Methylobacterium location in potato tissues by ISH/FISH

In situ hybridization was performed on tissues of leaves, stems and roots of in vitro -grown potato plants. The probes MB and E11 (Pirttiläet al. 2000), specific for bacteria in the Methylobacterium genus and eubacteria, respectively, were used for hybridization. Both untreated and Methylobacterium-inoculated plants were surface-disinfected prior to processing for hybridization, in order to avoid epiphyte contamination of the samples. Methylobacterium sp. was not isolated on selective medium with methanol from the uninfected surface-disinfected plants. At the same time, the bacterium colonized all organs of inoculated in vitro-grown plants (stems – 2·5 × 104 CFU g−1; leaves – 1·3 × 104 CFU g−1; roots – 7·2 × 103 CFU g−1 of raw plant tissues).

The hybridization signals of Methylobacterium rRNA were detected in leaves of untreated plants mainly at the tip of the leaf in sponge and palisade parenchyma and xylem vessels (Fig. 4a–d). In the infected leaf tissues, the signals were present in the xylem vessels (Fig. 4e–g) and in parenchyma tissues (Fig. 4e). In the stems of infected and untreated plants the Methylobacterium rRNA was mainly detected in the parenchyma and the vascular tissues (Fig. 4h–j). In the untreated plants the signal was weaker than in infected plants (Fig. 4k). DNA staining of infected stem samples was observed in areas indicated by in situ hybridization, supporting the presence of bacteria in the genus Methylobacterium inside cells of parenchyma tissues (Fig. 4i). Specifically in the stem parenchyma, both types of hybridization signals and the DNA staining revealed biofilms (Fig. 4h–k). Based on the signals, bacteria were found at highly localized areas in the biofilm (Fig. 4h–j). In the root of the infected plants, Methylobacterium rRNA was detected in vascular (Fig. 4l) and parenchyma tissues. In the untreated plants Methylobacterium rRNA was not detected in the root tissues (Fig. 4m).

Figure 4.

In situ hybridization of in vitro potato plant tissues with both digoxigenin and fluorescence-labelled oligonucleotide probe MB. (a), (b), (c) Leaf control, digoxigenin detection. (d) Leaf control, fluorescein detection. (e) Leaf infected with M. radiotolerans, fluorescein detection. (f), (g) Leaf infected with M. radiotolerans, digoxigenin detection. (h) Stem infected with M. radiotolerans, digoxigenin detection. (i) Stem infected with M. radiotolerans, DNA staining. (j) Stem infected with M. radiotolerans, fluorescein detection. (k) Stem control, fluorescein detection. (l) Root infected with M. radiotolerans, digoxigenin detection. (m) Root control, digoxigenin detection. (arrows – highest signal intensity; x/v – xylem vessel; b – biofilm; p – parenchyma; v – vascular tissues; n – nuclear; x – xylem; c/c – centre cylinder; d/v – differential vessel). Scale bar = 50 μm.

Eubacterial rRNA was detected with the probe E11 in palisade and spongy parenchyma tissues (Fig. 5a) and in xylem vessels (Fig. 5b) of the leaves of both infected and untreated potato plants. In the stem tissues of the infected plants, the eubacterial rRNA was present in parenchyma as biofilms (Fig. 5c). A less abundant signal was present in the parenchyma tissues of untreated potato plants (Fig. 5d).

Figure 5.

In situ hybridization of in vitro potato plant tissues with both digoxigenin (a, c–f) and fluorescence-labelled (b) E11 oligonucleotide probe, specific for eubacteria. (a) Potato leaf infected with M. radiotolerans. (b) Leaf control. (c) Potato stem infected with M. radiotolerans. (d) Stem control. (e) Potato root infected with M. radiotolerans. (f) Root control. (arrows – highest signal intensity; p –parenchyma; x/v – xylem vessel; b – biofilm; x – xylem; c/c – centre cylinder). Scale bar = 50 μm.

The probe E11 hybridized to the root tissues of both untreated and infected plants, indicating presence of eubacteria in these tissues (Fig. 5e,f). In general the hybridization signal was abundant in the root tissues, particularly in the parenchyma where a signal was also observed inside the cells (Fig. 5e). The eubacterial rRNA was also present in the vascular tissues of roots (Fig. 5e). In the untreated plants, a distinct signal was observed in the root parenchyma tissues (Fig. 5f), but generally the signal was weaker than in inoculated plants.


The PPFMB isolate, named M1 and later given the reference number IMBG290 was classified as Methylobacterium sp. in the phylogenetic analysis, having the closest homology with the type strain of M. radiotolerans. This bacterial strain could not be isolated from the in vitro potato plantlets before the inoculation with P. fluorescens. This suggests that these bacteria had been residing in the potato tissues in a viable but unculturable state. The P. fluorescens infection likely provoked the PPFM bacteria to grow out of the potato tissues and to colonize the plant surface (Podolich et al. 2006, 2007). Induction of unculturable endophyte populations by various environmental factors to the limit where they become detectable by sensitive methods has been reported earlier (Reiter et al. 2002; Idris et al. 2004). Details of such induction mechanisms were previously unknown. Studies on the location of the PPFMB in potato tissues by in situ hybridization, as presented in this paper, could enlighten the background of these mechanisms.

Methylobacteria have earlier been found as endophytes in plant tissues (Leifert et al. 1994;Pirttiläet al. 2000, 2005) but the locations, infection processes and roles of these endophytes have rarely been studied. The methylotrophs are suspected to supply the plant with growth stimulating compounds or to help the plant in activation of defense system for a better adaptation to the environment. The results of the in situ hybridization experiment showed the presence of Methylobacterium rRNA in all organs (leaves, stems and roots) of the inoculated potato plants. Our results provide the first direct evidence that methylotrophs exist inside in vitro-grown potato plant tissues. These bacteria may exist as regular inhabitants of potato plants since we found bacteria in plantlets after in vitro cultivation of one and half year. In the untreated plants, the Methylobacterium rRNA was detected only in the leaves and stems, whereas the eubacterial probe E11 hybridized to the tissues of all organs of the both plant types. Therefore, the in vitro potato plants appear to have endophytic bacteria in all organs, but the Methylobacterium spp. preferably reside in the aerial parts of these plants. This is also supported by the fact that the hybridization signal was the most abundant in the stems, slightly less in the leaves and the lowest in the root tissues of Methylobacterium-inoculated potato plantlets.

As these bacteria were detected by molecular methods (PCR, in situ hybridization) and could not be isolated on selective media, they may normally exist in an unculturable state inside the potato plants. Methylobacteria can persist in natural environments by forming biofilms (Kelley et al. 2004). Biofilms are specifically often found on plant surfaces and provide a mechanism for the bacteria to survive harsh environments (Morris et al. 1997). Biofilms are also an unculturable state of the bacteria. In the in situ hybridization experiments, a distinct biofilm structure was detected in the stem parenchyma of the in vitro-grown potato plants. Therefore, a regular existence as biofilms inside the potato tissues may explain the lack of culturability of these bacteria.

Intercellular spaces and xylem vessels are the most commonly reported locations of endophytic bacteria (Reinhold-Hurek et al. 1987; Nguyen et al. 1989; James and Olivares 1997; Gyaneshwar et al. 2001; Compant et al. 2005). In the present study the Methylobacterium-specific in situ hybridization signals were mainly located in the xylem vessels of leaves and stems, but not in the root vessels of untreated potato plants. This is in agreement with the data of rice interior colonization by Serratia marcescens (Gyaneshwar et al. 2001), as well as cotton colonization by the endophyte Enterobacter asburiae (Quandt-Hallmann and Kloepper 1996). Especially in the latter model experiment the bacterial endophytes spread systemically, but not in root inner tissue after seed inoculation. In our study the absence of the methylobacteria in roots of untreated potato plants may be explained by the putative role of these bacteria in scavenging monocarbon wastes within the leaf parenchyma tissues, where these compounds are produced in a larger scale.

In general, it is considered that some bacterial endophytes do not inhabit living vegetative cells (James and Olivares 1997; Reinhold-Hurek and Hurek 1998; Gyaneshwar et al. 2001). However, immunogold labeling allowed the precise localization of endophyte E. asburiae within epidermal cells (Quandt-Hallmann and Kloepper 1996) and intact root cortical cells of cotton (Quadt-Hallmann et al. 1997). Using the in situ hybridization method, Pirttiläet al. (2000) detected species in the genus Methylobacterium in the meristem cells of Scots pine buds. In this study we observed bacterial rRNA hybridization signals inside cells of stem parenchyma of in vitro-grown potato plants inoculated with Methylobacterium sp. IMBG290. However, the cells appeared senescing, which may explain the bacterial presence in these cells.

In this study endophytic Methylobacterium sp. IMBG290 was found within all organs of in vitro-grown potato plants of two cultivars grown in an agar medium system. The strain had the highest homology to M. radiotolerans in the phylogenetic analysis. The occurrence of M. radiotolerans in hostile environments has been reported in several studies (review, e.g. Nogueira et al. 1998). The plant tissue environment is not considered an extreme niche and, on the contrary, may provide some benefits for the bacteria. The plant tissues exude methanol which can be utilized by the methylotrophs. On the other hand, the bacteria may benefit the growth of commercially important crop species such as potato, and therefore remain an important research topic in the future.


This work was financed by the Academy of Finland (project 118569), Ukrainian Academy of Agrarian Sciences (project 0106U004046) and National Academy of Sciences of Ukraine (project 0103U000338).