Citrus tristeza closterovirus (CTV), is a phloem-limited virus transmitted by aphids in a semipersistent manner. The genome of CTV is composed of a ssRNA with two capsid proteins: CP, covering about 95% of the particle length, and a diverged coat protein (dCP), present only in one end of the particle, forming a rattlesnake structure. dCP is the product of p27 gene for which it is also postulated a function in the transmissibility by aphid vectors.

Hybridization analysis showed a p27 gene region, which exhibits different patterns with two probes derived from two biological distinct CTV isolates. In an attempt to screen whether that gene region differs in mild and severe strains, six CTV isolates belonging to different biogroups were compared for variations in their p27 gene by analysis of single-strand conformation polymorphism (SSCP). The p27 gene was reverse transcribed and amplified by PCR and thirty clones of each isolate were obtained. From each clone, two fragments of the gene were amplified by PCR: fragment (a), 459 bp long, and fragment (b), 281 bp long. Sequence variations in both gene fragments were studied by SSCP analysis. A variety of SSCP patterns was obtained from each isolate, being isolates belonging to the groups II-IV and III those with the higher and lower number of them. Moreover, SSCP analysis provided a rapid procedure to screen the genetic heterogeneity of the viral isolates reducing considerably the amount of nucleic acid sequenciation necessary to gain that knowledge.

Citrus tristeza virus (CTV), a member of the closterovirus group is a phloem-limited virus transmitted by aphids in a semipersistent manner (Berlansky et al, 1988). The natural CTV host range is limited to large, woody plants of citrus and related genera. Because of the severe damage caused by individual citrus trees that can be productive for a century, this virus causes one of the most ecomically important diseases of plants.

CTV has a positive sense RNA genome of 19,3 nucleotides (Karasev et al., 1995) encapsidated in flexuous, rod-shaped particles of about 2000 nm long (Bar-Joseph et al., 1989), with two capsid proteins: a 25 kDa coat protein (Pappu et al., 1993) covering about 95% of the particle length, and a diverged 27 kDa coat protein (Pappu et al., 1994; Febres et al., 1994) that covers one end of the particle forming a rattlesnake structure (Febres et al., 1996). Four regions of sequence conservation (domains I, II, III and IV) exist between the two proteins of the Florida T36 isolate (Febres et al., 1994). As it was suggested in beat yellow closterovirus (Agranovsky et al., 1995), the semipersistent mode of CTV transmission may be due to the ability of the assembled 27 K protein to cling tightly to the cell membranes lining the aphid´s alimentary tract. Another possibility, not necessarily alternative, may be that the p27 tail directs the closterovirus particle to the host (phloem) cell receptor. Experimentally, the bulk of p27 was found in the cell wall fraction of infected citrus leaves and it was also detectable in the soluble and membrane fractions (Febres et al., 1996).

A variety of CTV strains, differing by the symptoms induced on different host species, aphid transmissibility or capacity to interfere with other strains, has been described (Raccah et al., 1980; Ballester-Olmos et al., 1993; Van Vuuren et al., 1993). Several approaches for strain identification and differentiation have been tested in an attempt to develop a quick and specific procedure that could be applied to eradicate a severe strain in early stages of field trees infection. Nucleotide sequencing of individual genes is the most accurate procedure for strain identification, but it is expensive and time consuming for routine purposes. Single-strand conformation polymorphism (SSCP) analysis (Orita et al., 1989, Yap and McGee, 1994) enables rapid identification of DNA fragments of the same size but with sequence variations. Sequence variations as small as single base point mutations can be identified in PCR products amplified from a variety of genomic or cDNA sources (Hayashi, 1991).

SSCP analysis is performed by denaturing dsDNA in the presence of formamide and fractioning both DNA strands by electrophoresis in a non-denaturing polyacrylamide gel. Under the appropriate conditions, the electrophoretic mobility of single-strand DNA (ssDNA) is dependent not only on its length and molecular weigth, but also on its conformation. Small changes in the nucleotide sequence may alter conformation of ssDNA and, consequently, its electrophoretic mobility. SSCP analysis has been used to assign beet necrotic yellow vein virus to a known strain group (Koenig et al., 1995). More recently, using the same technique and combining two electrophoretic conditions and restriction of eight clones with Eco91I, Rubio et al. (1996) succeeded in the discrimination of 21 from the 22 CP gene clones selected from 17 CTV isolates.

On the other hand, SSCP analysis has proven to be a very appropriate technique for the study of viral populations, because it allows the separation of complex mixtures of equal molecular weight DNA with different nucleotide sequence. Enomoto et al. (1994), using SSCP analysis of 5'-terminal region of the E2 gene of hepatitis C virus (HCV) in four patients with chronic hepatitis, observed that the HCV population was composed of viral quasispecies with different E2-hypervariable regions (HVR).

Hybridization analysis of a cDNA library obtained from dsRNA of the field CTV isolate C268-2 (Concordia, Entre Ríos) with two probes (Semorile et al., 1993), allowed the detection of a p27 gene region which showed differential hybridization with a severe probe (T387) and no hybridization with a mild probe (T312). In an attempt to establish if p27 gene differs in mild and severe strains, we compared six CTV isolates with different biological properties for variations in this gene region by SSCP analysis. Moreover, we have used this technique to analyze the structure of viral RNA population in the six different CTV isolates tested.

Virus Isolates. The origins and biological characteristics of the CTV isolates used in this study are shown in Table 1

Table 1. Origins and biological properties of CTV isolates analyzed in this work.

The dsRNA of the Spanish isolates were kindly provided by Dr. P. Moreno and Dr. J. Guerri, Inst. Valenciano de Investigaciones Agrarias, Valencia, Spain. The Argentine isolates were biologically characterized by Eng. Norma Costa, EEA-INTA Concordia (Entre Ríos, Argentina)

Isolation of double-stranded RNAs (dsRNAs). CTV dsRNAs were purified from bark samples frozen with liquid nitrogen and pulverized in a mortar. Nucleic acids preparations enriched in dsRNAs were obtained by extraction with buffer-saturated phenol and fractionated by column chromatography on non-ionic cellulose (CF-11, Whatman) as previously described (Moreno et al., 1990).

SSCP analysis. SSCP analysis was performed directly on PCR products of fragments (a) and (b) of p27 gene. Usually, PCR products were extracted twice, first with phenol-cloroform and then with cloroform, precipitated and resuspended in 10 ml of bi-distilled water. A fraction of 1ml was mixed with 9 ml of the denaturing solution (950 ml/L) formamide, 20 mM EDTA and 500 mg/L bromophenol blue, heated by 10 min at 100 ^oC, and chilled on ice.

The p27 gene of six CTV isolates with different biological properties was reverse transcribed and amplified by PCR, and twenty seven-thirty clones of each isolate were obtained. In all cases, a single DNA fragment of 722 bp was produced.

From each clone, two fragments of p27 gene were amplified by PCR: one of them, 459 bp long (a), includes domains I and II, and the other, 281 bp long ( b), includes domains II, III and IV, according to the scheme showed in Figure 1. In both cases, only one PCR fragment of the expected size was obtained for each clone (Figure 2).

Nucleotide sequence variation in both gene fragments was studied by SSCP analysis. In preliminary experiments, several electrophoretical conditions (acrylamide concentration, temperature, presence or absence of glycerol, time, buffer concentration) were assayed. It was established that electrophoresis at a constant voltage of 200 V, for 3.5 h, in 12% polyacrylamide gels, at 22^oC, provided adequate resolution of the two strands of PCR products of p27(a) gene fragment. For PCR products of p27(b) gene fragment, the best discrimination of the two strands was obtained with the same conditions mentioned above, except that acrylamide gel concentration was 15% and the temperature during electrophoresis was 25^oC.

SSCP showed, for each CTV isolate analyzed and for each p27 gene fragment, a variable number of patterns and clones in each pattern (Figure 3, Table 2). In a high percentage of the clones, only two intense bands, corresponding to only one stable molecular conformation of each DNA strand, were obtained. For p27 (a) gene fragment the exceptions were: ten clones of C269-6, two clones of T385 and two clones of T300 CTV isolates, that showed three intense bands. For p27 (b) gene fragment the exceptions were: one clone of C269-6 and one clone of T388 that showed only one SSCP band; one clone of C269-6, one clone of T388 and one clone of T300, that showed three intense bands, and one clone of T388 CTV isolate, that showed four SSCP bands.

Four to nine SSCP patterns could be distinguished for fragment (a) p27 gene, depending on the CTV isolate (Table 2). T388 isolate showed the lowest number of SSCP profiles (4), whereas the C269-6 isolate exhibited the highest number of them (9) (Table 2).

On the other hand, p27(b) gene fragment (Table 2) yield a higher number of different SSCP patterns: eleven different patterns in the most variable isolate (T-388) and three distinct patterns in the less variable isolate (C268-2). The number of clones included in each of the less represented individual SSCP profiles varied from one to a maximum of seven (Table 2). From 11 to 23 clones were included in the main SSCP pattern of each p27 gene fragment of the CTV isolates analyzed, depending on the viral isolate (Table 2).

Table 2. Summary of SSCP analysis of all clones of the p27 (a) and (b) gene fragments for the six CTV isolates studied.

Table 3. Summary of SSCP analysis of all clones of p27 complete gene for the six CTV isolates analyzed.

^(a)These numbers of clones does not include the quantity of them that showed the main SSCP pattern.

The number of clones that showed sequence variation in both p27 gene fragments varied between two (T300) and nine (C269-6). Only the CTV field isolate C269-6 exhibited a higher number of those clones (Table 3). On the other hand, the number of clones that showed any sequence variation in p27 gene varied from 21, for C268-2 CTV isolate, to 7, for T-300 CTV isolate (Table 3).

The percentage of the major clone was calculated for (a) and (b) p27 gene fragments of each isolate analyzed (Table 2). The percentage of principal clone for the complete p27 gene without sequence variations in both (a) and (b) gene fragments was also estimated (Table 3). C268-2 field CTV isolate exhibited the higher value (70%) for the major clone, being the most homogeneous isolate for this gene region. On the contrary, T300 CTV isolate showed the lowest value (24%) for the major clone, being the most heterogeneous for p27 gene region among the CTV isolates analyzed. The C269-6 CTV field isolate exhibited a similar value than T300 (30%), whereas T312 isolate showed an intermediate value for the p27 majority clone percentage (47%). Percentages for the other two CTV isolates (T385 and T388) were similar (41 and 37% , respectively) and 15% higher than the lowest value (24%). According to the data observed in Tables 2 and 3, p27 (b) gene fragment showed a higher rate of sequence variation than p27 (a) fragment.

When the SSCP profiles of one representative of the majority clone for each CTV isolate were compared in a unique polyacrylamide gel, results obtained were those presented in Figure 4.

For p27 (a) gene fragment (Figure 4, left), T-385 and T-312 CTV isolates exhibited the same SSCP profile, but T300, T388, C268-2 and C269-6 CTV isolates exhibited a different SSCP profile between them and with respect to the above mentioned isolates.

For p27 (b) fragment, T300, T312 and T385 CTV isolates shared the same SSCP pattern; meanwhile, C268-2, C269-6 and T-388 CTV isolates showed a different SSCP profile (Figure 4, right). These results suggest that, when considering the complete p27 gene region, the CTV isolates belonging to the biogroups type III, IV, and V could be distinguished from the CTV isolates belonging to the biogroups type I and II, and between them, at least for the CTV genome region analyzed.

SSCP analysis was applied to p27 gene of six CTV isolates belonging to five different biogroups in an attempt to screen if that gene region differs in isolates with distinct biological properties. These CTV biogroups were defined on the basis of the symtoms induced on different greenhouse indicator plants. The isolates belonging to biogroup type I are the mildest ones and practically asymptomatic. There is an increase in the severity of the symptoms up to biogroup type V, that contains the most severe isolates causing pitting in grapefruit and sweet orange seedlings.

On the other hand, since the SSCP technique has shown to be a powerful alternative to detect the extent of variation within a virus population, it was also applied to know the intrapopulation variability for p27 gene in the CTV isolates tested.

Since the DNA conformation can be altered by several factors (i.e. electrophoresis buffer, voltage, temperature, acrylamide concentration), these factors will alter the electrophoretic mobility. According to that, the conditions to optimize the discrimination capacity of SSCP analysis were adjusted in each particular case. In this work, several experimental conditions were assayed and it was established that, for a good discrimination of SSCP profiles of p27(a) gene fragment, a lower polyacrylamide concentration (12%) and a lower temperature (22^oC) was required than for p27(b) gene fragment (15% acrylamide, 25^oC).

Different authors (Orita et al., 1989; Hayashi , 1991; Sheffield et al., 1993) have recommended to use DNA fragments from 135 to 400 bp to be able to detect single base mutations by SSCP analysis. For this reason, in this work, the p27 gene region was analyzed in two sub-regions.

For the twenty seven to thirty clones analyzed for each p27 gene fragments (a and b), more than one SSCP profile was observed in the six CTV isolates studied. This result is an indirect way of assessing the heterogeneity of the population. It was surprising that the T300 CTV isolate, that has been purified by aphid's passage and maintained under greenhouse conditions, showed a similar or higher number of SSCP profiles than field CTV isolates C269-6 and C268-2. Under field conditions, it is reasonable to suppose a higher degree of heterogeneity in viral population because successive infections by different CTV inocula could explain the presence of sequences from more than one viral variant in the same isolate. The reason for this discrepant finding could be the RT-PCR approach used to obtain the clones of p27 gene fragments. This approach may favour the amplification of some virus variants because the set of primers used was designed according to p27 gene sequence of the severe T-36 CTV isolate and three clones of cDNA library of the severe C268-2 CTV field isolate belonging to the same gene region. It could also indicate the exclusion of some virus strains by others in vivo. The results obtained, applying the SSCP analysis with the RT-PCR approach described, permit us to discern subpopulations of the CTV genome present in a proportion as low as 3%. The relative amounts of viral subpopulations detected by us using SSCP analysis, were similar to the result obtained by Enomoto et al.(1994) using the same experimental procedure. These authors could detect minor variants of E2 -HVR gene of HCV populations representing 5% of the whole population.

Sequencing studies of different p27 gene clones are now under way to determine the percentage of nucleotide exchanges among the majority clone and each of the minority ones.

Biological properties of an isolate may depend on the structure of viral RNA populations. The majority SSCP profiles obtained for the CTV isolates analyzed, which represent the predominant sequence variant observed, allowed us to discriminate four of the five biogroups defined for CTV strains, suggesting a discrimination power of 80% for this procedure. SSCP analysis, which is less time consuming than nucleotide sequencing studies, seems a convenient tool for assigning CTV isolates to a certain biogroup as well as for detecting mixed infections. To confirm this result, we are now conducting SSCP analysis of p27 gene in other CTV isolates belonging to the different biogroups mentioned above. Moreover, nucleotide sequence analysis would allow to detect changes in p27 gene sequence. These changes may be eventually related with the biological properties of the virus considering that p27 (b) gene fragment (which contains domains II, III, and IV involved in protein folding) showed the higher number of virus variants.

Acknowledgements

We are grateful to Dr. P. Moreno and Dr. J. Guerri for providing the dsRNAs of Spanish CTV isolates. This work was supported by grants of SECYT/CONICET-BID802-OC/AR PID 332 and the European Union, Contract EBBICI8CT960044.

Dr. L. Semorile belongs to the staff of Departamento de Ciencia y Tecnología, Universidad Nacional de Quilmes, Dr. O. Grau belongs to the staff of Facultad de Ciencias Exactas, Universidad Nacional de La Plata and Centro de Investigaciones en Ciencias Agrarias, INTA Castelar and Eng. N. Costa belogs to the staff of EEA-INTA Concordia, Argentina. Dr. L.Semorile and Dr. O.Grau are recipients of the Research Career award from Comisión de Investigaciones Científicas de la Provincia de Buenos Aires, Argentina.

Lic. S. Gago-Zachert is a fellow of the CONICET-Argentina.

Agranovsky, A.A., Lesemann, D.E., Maiss, E., Hull, R. and Atabekov, J.G. (1995). Rattlesnake structure of a filamentous plant RNA virus built of two capsid proteins. Proceedings of the National Academy of Sciences. USA 92: 2470-2473.

Ballester-Olmos, J.F., Pina, J.A., Carbonell, E., Moreno, P., Hermoso de Mendoza, A., Cambra, M. and Navarro, L. (1993). Biological diversity of citrus tristeza virus (CTV) isolates in Spain. Plant Pathology 42:219-229.

Ballester-Olmos, J.F., Pina, J.A., Moreno, P., Hermoso de Mendoza, A., Cambra, M. and Navarro, L. (1988). Biological characterization of different citrus tristeza virus isolates in Spain. Proc. 10th Conf. Int. Org. Citrus Virol. (IOCV) pp.28-32

Bar Joseph, M., Marcus, R, and Lee, R.F. (1989). The continuous challenge of citrus tristeza virus control. Annual Review of Phytopathology 27:291.

Beidler, L.L., Hilliard, P.R. and Rill, R.L. (1982). Ultrasensitive staining of nucleic acids with silver. Analytical Biochemistry 126:374-380.

Berlansky, R.H., Lee, R.F. and Garnsey, S.M. 1988. In situ immunofluorescence for the detection of citrus tristeza virus inclusion bodies. Plant Disease 72:820-826.

Costa, N. (1995). Estado actual de la tristeza en la Argentina. Proceedings of the . Third International Workshop of CTV and BrCA. Ed. by R. Lee, M. Rocha-Peña, C.L. Niblett, F. Ochoa, S.M. Garnsey, R.K. Yokomi, R. Lastra. pp 178-180.

Enomoto, n., Kurosaki, M., Tanaka, Y., Marumo, F. and Sato, C. (1994). Fluctuation og hepatitis C virus quasispecies in persistent infection and interferon treatment revealed by single-strand conformation polimorphism analysis. Journal of General Virology 75:1361-1369.

Febres, V.J., Ashoulin, L., Mawassi, M., Frank, A. Bar-Joseph, M., Lee, R.F. and Niblett, C.L. 1996. The p27 protein is present at one end of citrus tristeza virus particles. Phytophatology 86: 1331-1335.

Febres, V.J., Pappu, H.R., Anderson, E.J., Pappu, S.S., Lee, R.F. and Niblett, C.L. (1994) The diverged copy of the citrus tristeza virus coat protein is expressed in vivo. Virology 201:178-181.

Hayashi, K. (1991). PCR-SSCP: A simple and sensitive method for detection of mutations in the genomic DNA. PCR Methods and Applications 1:34-38.

Karasev, A.V., Boyko, V.P., Gowda, S., Nikolaeva, O.V., Hilf, M.E., Koonin, E.V., Niblett, C.L., Cline, K., Gumpf, D.J., Lee, R.F., Garnsey, S.M., Lewandowski, D.J. and Dawson, W.O. (1995) Complete sequence of the citrus tristeza virus RNA genome. Virology 208: 511-520.

Koenig, R., Lüddecke, P. and Haeberlé, A.M. (1995). Detection of beet necrotic yellow vein virus strains, variants and mixed infections by examining single-strand conformation polymorphism of immunocapture RT-PCR products. Journal of General Virology 76:2051-2055.

Moreno, P., Guerri, J. and Muñoz, N. (1990). Identification of Spanish strains of citrus tristeza virus (CTV) by analysis of double-stranded RNAs (dsRNAs). Phytopathology 80:477-482.

Moreno, P., Guerri, J., Ballester-Olmos, J. F., Fuertes-Polo, C., Albiach, R. and Martínez, M.E. (1993). Separation and interference of strains from a citrus tristeza virus isolate evidenced by biological activity and double-stranded RNA (dsRNA) analysis. Plant Pathology 42: 35-41.

Orita, M., Suzuki, Y., Sekiya, T. and Hayashi, K. (1989). Rapid and sensitive detection of point mutations and DNA polymorphisms. Genomics 5:874-879.

Pappu, H.R. Karasev, A.V., Anderson, E.J., Pappu, S.S., Hilf, M.E., Febres, V.J., Eckloff, R.M.G., McCaffery, M., Boyko, V., Gowda, S., Dolia, V.V., Koonin, E.V., Gumpf, D.J., Cline, K.C., Garnsey, S.M., Dawson, W.O., Lee, R.F. and Niblett, C.L. (1994) Nucleotide sequence and organization of eight 3' open reading frames of the citrus tristeza closterovirus genome. Virology 199:35-46..

Raccah, B., Bar-Joseph, M. and Loebenstein, G.(1980). Aphid-transmissibility of citrus tristeza virus in infected trees. Phytopathology 70:1173-1178.

Rubio, L., Ayllón, M.A., Guerri, J., Pappu, H., Niblett, C.N. and Moreno, P. (1996). Differentiation of citrus tristeza (CTV) isolates by single-strand conformation polymorphism analysis of the coat protein gene. Annals of Applied Biology 129:479-489.

Semorile, L., Dewey, R., García, M.L., Dal Bó, E., Ghiringhelli, P.D., Romanowski, V. and Grau, O. (1993) cDNA clones of CTV that discriminate severe and mild strains. In: P. Moreno, J.V. da Graça & L.W. Timmer, (Eds.), Proc. 12th Conf. Int. Org. Citrus Virol., Riverside, pp. 28-32.

Sheffield, V.C., Beck, J.S., Kwitek, A.E., Sandstrom, D.W. and Stone, E.M. (1993). The sensitivity of single-strand conformation polymorphism analysis for the detection of single base substituttion. Genomics 16:325-332.

Van Vuuren, S.P., Collins, R.P. and de Graça, J.V. (1993). Evaluation of citrus tristeza virus isolates for cross protection of grapefruit in South-Africa. Plant Disease 77: 24-28.

Yap, E.P.H. and McGee, J.O´D. (1994). Non-isotopic single-strand conformation polymorphism (SSCP) analysis of PCR products. PCR Technology: Currents Innovations. CRC Press Inc. pp 165-177.

Note: Electronic Journal of Biotechnology is not responsible if on-line references cited on manuscripts are not available any more after the date of publication.