Restriction analysis of genetic variability of Polish isolates of Tomato black ring virus

Several different isolates of Tomato black ring virus (TBRV) have been collected in Poland from cucumber, tomato, potato and black locust plants. Biological tests showed some differences in the range of infected plants and the type of symptoms, which was the basis for selection of seven the most biologically different TBRV isolates. According to the sequence of TBRV-MJ, several primer pairs were designed and almost the entire sequence of both genomic RNAs was amplified. The RT-PCR products derived from all tested TBRV isolates were digested by restriction enzymes. On the basis of the restriction patterns, the variable and the conserved regions of the TBRV genome were defined and the relationships between the Polish TBRV isolates established.

The variability of RNA viruses is also connected with specified parts of their genomes.Generally, the most conservative is the region encoding polymerase (Dolja & Carrington, 1992;Zaccomer et al., 1995;Koonin 1991).The coat protein gene is usually variable and virus species specific, and due to its responsibility for the specification of the viral serotype and species it is frequently used in phylogenic analyses (Dolja & Carrington, 1992;Zaccomer et al., 1995;Petrzik & Lenz, 2002;Canizares et al., 2001).Many other viral proteins, especially those of unknown function, are only rarely used in viral phylogeny due to their often one-species-specific nature and a lack of counterparts in other viral species (Zaccomer et al., 1995).
The knowledge of virus population diversity and the distribution of the variable and conserved regions within the viral genomes may be useful in sequence-specific detection of viruses, the prediction of occurrence of resistance-breaking viral phenotypes as well as in developing new methods of plant protection.
In this paper we report the genetic variability of Polish TBRV isolates on the basis of restriction analysis of almost full-length cDNA derived from both viral RNAs.

MATERIALS AND METHODS
TBRV isolates.The TBRV isolates studied originated from the following plant species: cucumber (TBRV-Cuc), tomato (TBRV-Tom), black locust (TBRV-L10, TBRV-MJ, TBRV-N1, TBRV-Pn) and potato (TBRV-Pot).TBRV-Pot was kindly supplied by Prof. M. Chrzanowska (IHAR, Radzików, Poland), the other isolates were from our collection.The biological properties of the TBRV isolates were studied by mechanical inoculation of various plant species.The host range and symptoms allowed selection of the isolates displaying the most distinct properties.All the isolates were maintained and propagated in Chenopodium quinoa in the greenhouse.
Purification of viral RNA.The TBRV isolates were propagated in C. quinoa and purified as described before (Pospieszny & Borodynko, 1999).The purified viral particles were digested with proteinase K and RNA was isolated according to the phenol/chloroform protocol (Sambrook et al., 1989), precipitated with 96% ethanol and pellet was dissolved in RNase-free water, then 2 ml of RNA solution were mixed with 10 ml of Formazol (MRC) and separated on 1% agarose non-denaturing gel.

RT-PCR amplification of RNA1 and RNA2.
The purified viral RNA was reverse transcribed using SuperScript II reverse transcriptase (Invitrogen) according to manufacturer's instructions.The primer used for the first strand synthesis was oligo d(T) 22 and the full-length cDNAs obtained were amplified using four pairs of primers designed according to the TBRV-MJ sequence (Accession number: NC 004439 and NC 004440 for RNA1 and RNA2, respectively).The first primer pair (P-128 and 1MP2) amplified the 5¢ part of RNA1 about 2900 nt in length, the second one (1MP3 and 3ter) about 3600 nt from the 3¢ part of RNA1, the third (P-128 and 2MP2) the 5¢ part of RNA2 of about 3200 nt and the last pair (2MP5 and 3ter) amplified the 3¢ part of RNA2 about 2300 nt in length (Table 1).The obtained PCR products covered the entire coding sequence of RNA2 and almost entire sequence of RNA1, excluding about 800 nt from the region coding for NTP-binding protein.The amplification of cDNA obtained from seven Polish TBRV isolates was performed using ExpandLong PCR System (Roche) according to the manufacturer's instruction.
Restriction digestion of PCR derived products.The restriction enzymes were chosen on the basis of TBRV-MJ sequence using the MapDraw program from the DNA Star package (kindly made available by Institute of Biochemistry and Biophysics, Polish Acad-emy of Sciences, Warszawa, Poland).Each RT-PCR product was digested by two different enzymes in two distinct restriction experiments and the digestion products were resolved on 2.5% NuSieve 3:1 agarose gel (BMA).The patterns obtained were analyzed in two aspects: firstly, a comparative analysis of conserved and variable regions was performed according to the bands of known size derived from TBRV-MJ and secondly, on the basis of the restriction fragments, 124 characters for phylogenetic analysis of the Polish TBRV isolates were obtained.The clustering and genetic distances assessment was performed using the program FreeTree (Pavlicek et al., 1999).The phylogeny tree was bootstraped 1000 times and constructed using the TreeView program (Page, 1996).The genomic mapping of the restriction bands obtained was established by comparison with TBRV-MJ patterns.

Analysis of viral RNA
Analysis of purified viral RNA showed two RNA bands of a size typical for TBRV.For the majority of the isolates some additional bands of small non-genomic RNA appeared (Fig. 1).The origin of these small RNAs is unclear.The RNAs expected for TBRV satellite are about 1350 nucleotides in length (Hemmer et al., 1993;Fritsch et al., 1984;1993)  observed bands are mostly smaller.We also observed that those small, non-genomic RNAs could emerge during prolonged propagation of the isolates originally devoid of them (not shown).

RT-PCR amplification of RNA1 and RNA2
In RT-PCR amplification four products were obtained for each tested TBRV isolate.Two of them correspond to the RNA1 sequence and two others to RNA2.All the products obtained were of comparable size.The putative organization of the RT-PCR products is shown in Fig. 2. The displayed numbers of nucleotides corresponding to the cleavage sites for both TBRV-MJ polyproteins have been discussed elsewhere (Jończyk et al., 2004).

Restriction digestion of PCR derived products
Restriction analyses of the RT-PCR products showed differences among the TBRV isolates.A higher diversity in the restriction patterns was observed for RNA1 and the restriction patterns were more diverse for the RT-PCR products corresponding to the genes in the 5' end proximity of both genomic RNAs.Figures 3a and 3b show examples of the restriction patterns of the isolates in com-  RdRp, RNA-dependent RNA polymerase; Cof.Prot, protease cofactor; NTB, NTP-binding protein; Prot, protease; 2a, 2a protein; MP, movement protein; CP, coat protein.The numbers presented in regular numerals indicated the nucleotides corresponding to putative cleavage sites for TBRV polyproteins and the numbers in italics and underlined correspond to the sites of primer annealing.
parison with those of TBRV-MJ for which the full-length sequence of both RNAs was already known.
The obtained restriction patterns allowed creation of a dendrogram of the relationship for the Polish TBRV isolates (Fig. 4).The re-  For details see legends to Fig. 3a.
sulting tree was not supported by high bootstrap values for the isolates from crops.Four TBRV isolates from black locust were clustered into two pairs: TBRV-Pn with TBRV-MJ and TBRV-N1 with TBRV-L10.The genetic distance (Nei and Li) within a pair was much lower (e.g.0.20879 between TBRV-MJ and TBRV-Pn) than between the pairs (e.g.0.5713 between TBRV-N1 and TBRV-MJ).The isolates from crop plants were more similar to each other than to the isolates from black locust.The most different among the isolates originating from crops was TBRV-Pot.

DISCUSSION
The Polish TBRV population is an interesting example of RNA virus diversity in a few plant species.The presented restriction analysis gave an overall view into the genetic variability of the isolates and allowed prediction of distribution of mutations within the TBRV genome.The restriction tests performed create a basis for phylogenic analyses of the Polish TBRV isolates as well as permit a prediction of their likely evolution in Poland.
Contrary to our expectation, the restriction analyses showed diverse patterns for RNA1 and quite homogenous ones for RNA2.RNA1 encodes all the proteins thought to be highly conserved (e.g., RNA-dependent RNA polymerase or protease) and RNA2 encodes a capsid and a movement protein.While the movement protein is often considered conserved (Mushegian 1994;Melcher 2000), the coat protein usually varies within species (Steinkellner et al., 1992;Petrzik & Lenz 2002;Canizares et al., 2001).In fact, all Polish TBRV isolates were serologically similar, if not identical, in the tests performed such as ELISA or double-diffusion test (Borodynko et al., 2001).
The sequences in the 5' end proximity of both RNAs, coding for a putative protease cofactor and 2a protein on RNA1 and RNA2, respectively, appeared to have more variable restriction patterns in comparison with the sequences in the 3' end proximity of the RNAs.The role of the former proteins in the TBRV life cycle is not clear.Protein 2a, similarly to 2a protein of Grapevine fanleaf virus (GFLV), might participate in replication of RNA2 (Gaire et al., 1999) and the putative protease cofactor might modulate the activity of viral protease especially for polyprotein 2 maturation (Hemmer et al., 1995).Unfortunately, the mechanisms of RNA2 replication and polyprotein 2 maturation are not completely understood for TBRV and it cannot be excluded that 2a protein and the putative protease cofactor have quite different functions in the TBRV life cycle.
Restriction analysis cannot provide information concerning amino-acid variability or their influence on the viral proteins' properties.However, it should be considered that even the silent mutations that do not influence the amino-acid sequence, can modulate RNA structure.Single nucleotide substitutions might also facilitate virus adaptation to the host translation machinery due to the correlation of the viral amino-acid codons with the ones used by the host plant (Roossinck 2002; Leisner & Neher, 2002).The tests performed allowed establishing the existence of a genetic diversity of the Polish TBRV isolates and indicated which part of their genomes probably gathered or/and maintained the diversity.
The phylogenic relationships between the Polish TBRV isolates were established on the basis of restriction fragment length polymorphism analysis.The resulting dendrogram clustered the isolates from black locust into two groups and such clustering was supported by high bootstrap values (Fig. 4).The situation was different for the isolates from crops because the very low bootstrap values prevented us from identification of the most likely relationships.
Most of the Polish TBRV isolates have some additional RNA particles associated and encapsidated with the viral RNAs.Their atypical size and also abundant amount suggested their defective rather than satellite nature.This suggestion was supported by the observation that the small RNAs originated during serial passages of a non-genomic-RNA-free isolate in C. quinoa (not shown).The isolates after passages usually displayed decreased severity.
The best method for estimation of real variation among viral genomes is sequencing.However, this is usually done for a short part of the coding sequence and is typically limited to one or sometimes two most typical viral proteins.The restriction tests performed enabled a preliminary characterization of the Polish TBRV population according to almost entire genomic sequence and indicated in which part of the genome the variability was localized.
The authors are grateful to Dr Andrzej Pałucha (Institute of Biochemistry and Biophysics, Polish Academy of Sciences, Warszawa, Poland) for his kind help in sequence analysis and for helpful discussion.

Figure 2 .
Figure 2. TBRV genome organization and location of primers and RT-PCR products.

Figure 3b .
Figure 3b.RT-PCR products of the 5' part of RNA2 digested by Hinf.

Figure 4 .
Figure 4. Dendrogram of the relationships between Polish TBRV isolates.