1813. Properties of Monoclonal Antibodies to Potato Leafroll Luteovirus and. Their Use to Distinguish Virus Isolates Differing in Aphid Transmissibility. By P. R. ...
J. gen. Virol. (1987), 68, 1813-1821. Printed in Great Britain
1813
Key words: PLR V/aphid transmissibility/ MAbs
Properties of Monoclonal Antibodies to Potato Leafroll Luteovirus and Their Use to Distinguish Virus Isolates Differing in Aphid Transmissibility B y P. R. M A S S A L S K I t AND B. D. H A R R I S O N Scottish Crop Research Institute, Invergowrie, Dundee DD2 5DA, U.K. (Accepted 20 March 1987) SUMMARY Ten murine monoclonal antibodies (MAbs) specific for a British isolate of potato leafroll luteovirus (PLRV) were produced in large quantities in ascitic fluids and their isotypes determined. All 10 MAbs reacted in indirect ELISA with intact PLRV particles but four did not react with disrupted virus particles, suggesting that these four MAbs are specific for epitopes dependent on quaternary structure. None of the MAbs gave a precipitin reaction in immunodiffusion tests. All the MAbs reacted strongly with, but failed to differentiate, 28 readily aphid-transmissible British PLRV isolates, six of which consistently caused symptoms differing in severity in indicator hosts. Two MAbs (SCR-8 and SCR-10) reacted only weakly with two other PLRV isolates, both of which were poorly transmissible by aphids. Three MAbs (SCR-6, SCR-8 and SCR-10) reacted with groundnut rosette assistor luteovirus (GRAV), but none reacted with British isolates of three other luteoviruses: carrot red leaf, beet western yellows and barley yellow dwarf (B and F isolates). Five epitopes on PLRV particles were distinguished, of which the two that were missing in the poorly aphid-transmissible isolates of PLRV and/or were shared with G R A V were both apparently dependent on quaternary structure. INTRODUCTION Potato leafroll luteovirus (PLRV) has a worldwide distribution and is economically the most important virus affecting potato crops (Harrison, 1984). Several strains of the virus, isolated from potato, can be distinguished by differences either in the severity of symptoms induced in potato and Physalis floridana (Webb et al., 1951) and/or Montia perfoliata (Tamada et al., 1984), or in their ease of transmission by the aphid, Myzus persicae (Tamada et al., 1984). However, PLRV isolates seem in general to be antigenically very similar. For example, a polyclonal antiserum to a Japanese isolate of PLRV reacted with PLRV isolates obtained from potato in several other countries (Kojima, 1981). In addition, no spurs formed when British PLRV isolates differing in virulence or aphid transmissibility were compared in gel-diffusion precipitin tests using a polyclonal antiserum to the British stock isolate of PLRV (Tamada et al., 1984). However, the possibility remains that antigenic differences might occur between some of these isolates, such as those differing in aphid transmissibility, but be too subtle to be detected easily in tests with polyclonal antisera (Harrison & Murant, 1984). The production and use of monoclonal antibodies (MAbs) to a range of plant viruses has permitted the serological differentiation of a number of virus strains and variants which could not be readily distinguished using polyclonal antisera (Halk & de Boer, 1985). This paper describes the production of a panel of MAbs to a British isolate of PLRV, their reactions with a wide range of British PLRV isolates and some other luteoviruses, and an investigation of their potential for differentiating PLRV isolates which are not distinguishable using polyclonal antibodies. METHODS Virus isolates and viruspurification. Table 1 lists the British PLRV isolates used and their acronyms and sources. On receipt, all PLRV isolateswere transmitted from potato to P.floridana and/or potato cv. Maris Piper, in which t Dr Massalski died on 26 January 1987, shortly after this paper was completed. 0000-7688 O 1987 SGM
P. R. MASSALSKI AND B. D. HARRISON
1814
T a b l e 1. Sources o f P L R V Isolate code PLRV-A -B -C -D -E -F -G -H -I -J -K -L -M -N -O -P -Q -R -S -T -U -V -W -X -Y -Z -1 -ll -15 -30
Potato cultivar or clone from which isolate obtained Drayton Ulster Sceptre Vanessa Pentland Dell Estima Romano Arran Pilot Maris Peer Pentland Crown Pentland Ivory Pentland Javelin Croft Majestic Pentland Hawk Marls Piper Duke of York Record King Edward Desiree Pentland Squire Home Guard Up-to-Date Clone RS 1057/ab(19) Clone RS 1057/ab(20) Marls Piper - L6 Maris Piper - 16 Cara Pentland Ivory Up-to-Date Dr Mclntosh
isolates used Supplier East Craigs, Edinburgh, U.K. East Craigs, Edinburgh, U.K. East Craigs, Edinburgh, U.K. East Craigs, Edinburgh, U.K. East Craigs, Edinburgh, U.K. East Craigs, Edinburgh, U.K. East Craigs, Edinburgh, U,K. East Craigs, Edinburgh, U.K. East Craigs, Edinburgh, U.K. East Craigs, Edinburgh, U.K. East Craigs, Edinburgh, U.K. East Craigs, Edinburgh, U.K. East Craigs, Edinburgh, U.K. East Craigs, Edinburgh, U.K. East Craigs, Edinburgh, U.K. East Craigs, Edinburgh, U.K. East Craigs, Edinburgh, U.K. East Craigs, Edinburgh, U.K. East Craigs, Edinburgh, U.K. East Craigs, Edinburgh, U.K. East Craigs, Edinburgh, U.K. East Craigs, Edinburgh, U.K. SCRI (R. M. Solomon) SCRI (R. M. Solomon) SCRI (Authors) SCRI (Authors) SCRI (Tamada et al., 1984) SCRI (Tamada et al., 1984) SCRI (Tamada et al., 1984) SCRI (Tamada et al., 1984)
DAFS, DAFS, DAFS, DAFS, DAFS, DAFS, DAFS, DAFS, DAFS, DAFS, DAFS, DAFS, DAFS, DAFS, DAFS, DAFS, DAFS, DAFS, DAFS, DAFS, DAFS, DAFS,
the isolates were maintained. One isolate was transmitted by grafting and the others by M. persicae. Test plants were grown in an aphid-free glasshouse kept at 15 to 25 °C and fumigated weekly with nicotine. A British isolate of beet western yellows luteovirus (BWYV) from naturally infected shepherd's purse was obtained from D. G. A. Walkey, Wellesbourne, U.K., and two British isolates of barley yellow dwarf luteovirus (BYDV), isolates B and F (Torrance et al., 1986), cultured in oat, were obtained from L. Torrance, Harpenden, U.K. A British isolate of carrot red leaf luteovirus (CRLV; Waterhouse & Murant, 1981) cultured in chervil and a Nigerian isolate of groundnut rosette assistor luteovirus (GRAV; Reddy et al., 1985) propagated in groundnut, were provided by A. F. Murant, Invergowrie, U.K. Purified preparations of PLRV particles were obtained essentially as described by Tamada & Harrison (1980) but substituting 5% Celluclast (Novo Enzyme Products Ltd., Windsor, U.K.) for 1% Driselase. Control preparations were obtained from leaves of virus-free P.floridana, or potato cv. Maris Piper, by the same method except that the sedimentation in sucrose density gradients and all subsequent steps were omitted. Aphid- and graft-transmission. M. persicae were cultured and used to transmit PLRV isolates as described by Barker & Harrison (1985). For graft inoculation of PLRV, the shoot-tip of a virus-free potato plant, cv. Maris Piper or Cara, was removed and replaced by an infected scion, as described by Barker & Harrison (1985). Derivation, culture and selection ofhybridomas. Hybridomas secreting PLRV-specific antibodies were derived by fusion of the P3-X63-Ag8.653 murine myeloma cell line with the spleen cells of a female BALB/c mouse immunized against PLRV- 1, as follows: day 0, 50 ~tg purified particles of PLRV- 1 in an equal volume of Freund's complete adjuvant injected intraperitoneally (i.p.) and subcutaneously (s.c.); day 14, 50 ~tg virus in Freund's incomplete adjuvant i.p. and s.c. ; day 40, 50 ~tg virus in 10 mM-phosphate buffer pH 7.0, intravenously. The mouse was killed on day 44 and the spleen removed aseptically. Techniques for the subsequent production and culture of hybridomas were essentially as described by Galfr6 & Milstein (1981). Ten to 13 days after fusion, hybridoma colonies were tested for antibody production by indirect ELISA (see below) using as antigen either plate-trapped purified particles of PLRV-1, or particles of PLRV-1 trapped from infective P.floridana sap using polyclonal antibody. The MAbs were isotyped by tests on tissue culture supernatant fluids as described by Thomas et al. (1986).
M A b s to potato leafroll luteovirus
1815
Production of asciticfluids. Ascitic fluids were produced in pristane-primed BALB/c mice, and the IgG was purified as described by Thomas et al. (1986). Other virus antibodies. Polyclonal rabbit antisera to luteoviruses were provided by those named below; homologous titres (reciprocals) are given in parentheses and were obtained in gel-diffusionprecipitin tests unless otherwise specified. PLRV-1, antiserum G (512; Tamada & Harrison, 1980); BWYV (2048, titre by infectivity neutralization; J. E. Duffus, Salinas, Ca., U.S.A.); BYDV-B and BYDV-F (titres not available; L. Torrance, Harpenden, U.K.); and CRLV (512; A. F. Murant, Invergowrie). Two MAbs (371A and 372E; Martin & Stace-Smith, 1984)to a Canadian isolate of PLRV were provided by R. R. Martin, Vancouver, Canada. Preparation of immunoglobulin and conjugation with enzyme. Immunoglobulin (Ig) was precipitated from mouse ascitic fluid or from rabbit antiserum, and conjugated to alkaline phosphatase, as described by Thomas et at. (1986). ELISA. Two forms of indirect ELISA were used. For the antibody-trapped antigen form of indirect ELISA, all steps were done as described by Thomas et al. (1986) except that samples were prepared by grinding leaf tissue in 0-05 M-phosphate buffer pH 7.0, containing 10 raM-sodium diethyldithiocarbamate (extraction buffer) at 5 ml/g leaf. All tests, unless otherwise stated, used tissue culture supernatant fluids as the source of MAbs. Bound rabbit anti-mouse alkaline phosphatase conjugate (Miles Laboratories) was detected using the substrate p-nitrophenyl phosphate at 0.6 mg/ml in 1 M-diethanolamine buffer pH 9.8. Following the addition of substrate, a Titertek Multiskan photometer (Flow Laboratories) was used to record A,o5 after 90 to 120 min at 23 °C and again after subsequent overnight incubation at 5 °C. For the plate-trapped antigen form of indirect ELISA, purified PLRV-1 particles, diluted to 2 Ixg/mlin 0.05 Msodium carbonate buffer pH 9.6 (coating buffer), and similarly diluted control preparations obtained from virusfree plants, were incubated on plates overnight at 5 °C (100 ~d/well). Subsequent steps were the same as those following sample incubation in the antibody-trapped antigen form of ELISA. The double-antibody sandwich form of direct ELISA (DAS-ELISA) was done essentially as described by Tamada & Harrison (t 980) except that samples were prepared by grinding leaf tissue in extraction buffer (5 ml/g leaf). The wells of microtitre plates were coated with rabbit polyclonal anti-PLRV, anti-BWYV, anti-BYDV or anti-CP,LV Ig (1 ~tg/ml)before overnight incubation with the tissue extracts at 5 °C. After washing, antigen bound by polyclonal Ig was detected using homologous polyclonal Ig-alkaline phosphatase conjugate. Substrate was added and A4os recorded as described above. Concentrations of PLRV in tissue extracts were determined as described by Barker & Harrison (1985). Immunodiffusion tests, Agarose gel (0.5~ w/v) in 0.01 M-phosphate buffer pH 7.4, containing 0.85~ NaC1, with or without 1~ (w/v) polyethylene glycol, mol. wt. 8000 (PEG; Sigma) was used for double-diffusionprecipitin tests. Monoclonal antibodies from ascitic fluids used at a range of dilutions or from tissue culture supernatant fluids, were tested against purified particles of PLRV-1 diluted in phosphate buffer to 250 Ixg/ml. Mouse and rabbit anti-PLRV polyclonal Ig, diluted to 0.7 mg/ml, were included as antibody controls. Stability of PLRVparticles. The stability of purified particles of PLRV-1 in coating buffer at pH 9.6 or in 0.01 Mphosphate-buffered saline (PBS) at pH 7.4 was investigated by incubating 50 Ixg virus in 250 Ix buffer for 3 h at 37 °C or overnight at 5 °C. Samples were examined by electron microscopy and the remainder (200 ~tl)was layered on sucrose gradients and subjected to ultracentrifugation as described by Tamada & Harrison (1980). Gradients were fractionated and scanned by upward displacement through a u.v. absorptiometer recording at 254 nm. RESULTS
Production o f hybridomas and titres o f M A b s At the time of splenectomy, the polyclonal antiserum from the immunized mouse had a titre of at least 5 x 10-5 in indirect ELISA using antibody-trapped PLRV from P. floridana sap as antigen, and less than 5 x 10-3 using virus-flee sap. Of the 480 wells into which the fusion products were distributed, 411 contained growing colonies. As indicated by indirect ELISA using antibody-trapped antigen, 162 of 411 wells had supernatant fluids containing virusspecific antibodies and 85 wells contained antibodies specific for healthy plant components. Indirect ELISA using plate-trapped antigen indicated that 108 of 411 wells had supernatant fluids containing virus-specific antibodies and a further 32 wells contained colonies which secreted antibodies specific for components in the preparation from healthy P.floridana leaves. All supematant fluids that reacted with plate-trapped antigen in indirect ELISA also reacted with antibody-trapped antigen, but not vice versa. Five hybridomas secreting antibody detected by screening with plate-trapped antigen and five secreting antibody detected with antibodytrapped antigen were each cloned twice at limiting dilution to produce 10 stable antibody-
P. R. MASSALSKI AND B. D. HARRISON
1816
Table 2. Properties of MAbs to PLRV-1
MAb SCR-1 SCR-2 SCR-3 SCR-4 SCR-5 SCR-6 SCR-7 SCR-8 SCR-9 SCR-10 Parent myeloma (P3-X63-Ag8. 653)
Isotype IgG2a IgG2a IgG2a IgG2a IgG2a IgG2b IgG 1 IgG 1 IgG2a IgG1 -
ELISA detection system* PTA PTA PTA PTA PTA ATA ATA ATA ATA ATA
Titre of culture supernatant fluid (reciprocal) 2048t 1024 1024 1024 1024 1024 512 512 512 512
Titre of ascitic fluid 10-6t 10-s 10-6 10-s 10-5 10-5 10-2 10-4 10-4 10-4 -
Yield-IgG from ascitic fluid (mg/ml) 15.1 7.2 8.6 6.6 6-2 6.7 5.7 5.3 5-9 5-1 0-5
* Indirect ELISA system using either plate-trapped antigen (PTA) or antibody-trapped antigen (ATA) to select hybridoma clones from fusion products. t Greatest dilation of antibody to give A4o5 greater than 0.20 after overnight incubation of substrate in the antibody-trapped antigen form of indirect ELISA, using PLRV-l-containing P.floridana sap as antigen. In tests with virus-free sap, A,o5 was less than 0.01. secreting clones. These hybridomas were used to produce M A b s in culture supernatant and ascitic fluids; these M A b s were designated SCR-1 to SCR-10. Some of their properties are listed in Table 2. In addition, all the M A b s were used in immunodiffusion tests but none gave a precipitin line whether or not the gel contained P E G , conditions in which the rabbit and mouse polyclonal antisera reacted strongly. Similarly, various combinations of two, five or all 10 M A b s failed to produce a precipitin line when tested against purified virus.
Reactivity of MAbs with British PLR V isolates M A b s SCR-1 to -10 in tissue culture supernatant fluids, or the two C a n a d i a n M A b s in diluted ascitic fluids, were used in indirect E L I S A with antibody-trapped antigen, and r a b b i t antiserum Ig was used in D A S - E L I S A , to investigate serological relations among 30 British P L R V isolates (Table 1) which were cultured in potato cv. Maris Piper and/or P. floridana. Some of these isolates consistently induced severe stunting in potato (PLRV-Y) or P.floridana (PLRV-11), and some induced intermediate symptoms in P.floridana (PLRV-1 and PLRV-F), whereas others caused only mild symptoms in potato (PLRV-Z) or P.floridana (PLRV-30). All the isolates were readily detected in either host, but could not be differentiated using the polyclonal antibody detection system or either of the C a n a d i a n MAbs, both of which gave values that consistently reflected those from the polyclonal antibody tests. Similarly, M A b s SCR-1 to SCR-10 readily detected, but did not differentiate, 28 of the 30 isolates. However, one isolate (PLRV-V), originally obtained from potato cv. Up-to-Date, reacted much more weakly with SCR-8 and SCR-10 in four replicate tests than did the 28 other P L R V isolates. In these tests PLRV-V was not distinguished from the 28 other P L R V isolates using SCR-1 to SCR-7, SCR-9, the C a n a d i a n MAbs, or polyclonal antibody. To investigate further the serological distinctiveness of PLRV-V, its particles and those of three other isolates were purified and tested at known concentrations in indirect E L I S A with SCR-1 to SCR-10, using I g G purified from ascitic fluids. The o p t i m u m concentration of ascitesderived I g G for P L R V detection was first determined in the antibody-trapped antigen form of indirect ELISA. At concentrations of 100 ~tg/ml or greater, SCR-1 to SCR-10 all gave positive readings in wells used for the material purified from virus-free leaves or for buffer alone. A t 1 to 10 ~tg/ml, however, the M A b s all gave strong positive reactions against PLRV-1 and did not react detectably in the wells used for material from virus-free leaves or for buffer. The M A b s were therefore used at 1 ~tg I g G / m l in tests with purified particles of PLRV-V, PLRV-1, P L R V -
MAbs to potato leafroll luteovirus
1817
Table 3. Reactions of purified particles of four strains of potato leafroll virus with 10 MAbs to
PLR V- 1 Reaction in Reaction of MAbs in indirect ELISA~: (A~os) DAS-ELISA]" r Antigen* (A4o5) S C R - 1 SCR-2 SCR-3 SCR-4 SCR-5 SCR-6 SCR-7 SCR-8 SCR-9 SCR-10 PLRV-V 1.53~ 1-38 1.27 1.45 1.13 1.31 1.14 0.60 0-16 1.66 0.14 PLRV-15 0.80 1-45 1-00 1-55 1.42 1-35 0.86 0.40 0-17 1.00 0.18 PLRV-I 1 1.66 1.58 1.08 1.55 1.37 1-36 0.71 1.00 0.92 0.92 0.93 PLRV-1 1.26 1.31 1.04 1.30 1.21 1.20 0.86 0.72 0-60 0.70 0.70 Control 0.06 0-06 0.04 0.06 0.13 0.06 0.07 0-04 0.06 0.04 0.07 * For antigen, purified particles of each PLRV isolate were used, diluted to 500 ng/ml in 0.05 M-phosphate buffer pH 7-0. Control antigen consisted of material purified from virus-free potato plants and diluted to approx. 1 lag/ml in the same buffer. t Rabbit anti-PLRV reagents were used for DAS-ELISA; for indirect ELISA, rabbit anti-PLRV Ig (1 ~tg/ml) was used to coat the plates and the MAbs (1 Ixg/ml) were used as the second-stage antibodies. :~A405values are means for duplicate wells, minus values for buffer controls, and were recorded after incubating for 90 min at 23 °C (DAS-ELISA) or after an additional 16 h at 5 °C (indirect ELISA). 11 and PLRV-15, the three last-named being isolates that could not be distinguished from one another by immunodiffusion tests with polyclonal antibody ( T a m a d a et al., 1984). The results for one concentration of P L R V in one of three similar tests are given in Table 3. In all three tests, SCR-8, SCR-9 and SCR-10 had similar reactivities against purified PLRV-1 and PLRV-11, as indeed they did against the other 26 P L R V isolates tested as infective sap. In contrast, SCR-8 and SCR-10 reacted much more weakly with P L R V - V and PLRV-15 than did SCR-9. SCR-7 seemed to react somewhat less strongly with P L R V - V and PLRV-15 than with PLRV-1 or PLRV-11. The other M A b s and the polyclonal antibody all reacted strongly with all four isolates (Table 3).
Transmission of virus isolates by M. persicae T a m a d a et al. (1984) showed that PLRV-15 could be distinguished from PLRV-1, PLRV-11 and PLRV-30 by its poor transmissibility by M. persicae. To ascertain whether failure to react with SCR-8 and SCR-10 is characteristic of poorly aphid-transmissible isolates, the aphidtransmissibility of the other 26 isolates listed in Table 1 was determined. Twenty-five of these isolates were readily transmitted by M. persicae from the potato cultivar in which they were received to either P.floridana or potato cv. Maris Piper, using one to 10 aphids per test plant. In contrast, P L R V - V was not transmitted by M. persicae from potato cv. U p - t o - D a t e to either P. floridana or potato cv. Marls Piper and had to be graft-inoculated to Maris Piper potato to produce virus source plants for more detailed tests. The results of these tests confirmed that P L R V - V was not transmitted by M. persicae from potato to P.floridana in conditions in which three other isolates were readily transmitted, and they showed that the failure of P L R V - V to be transmitted could not be ascribed to an inadequate virus content of the source plants (Table 4). Recent tests have also substantiated the difficulty in transmitting PLRV-15 by M. persicae. In three experiments purified particles of PLRV-15, stored for 3 to 4 years at - 7 0 °C and used as inoculum at 25 to 100 ~tg/ml, were acquired by aphids feeding through m e m b r a n e s as described by T a m a d a et al. (1984) but none of 20 P.floridana seedlings, each exposed to 5 to 10 such aphids, became infected. In contrast, 12 out of 24 seedlings became infected in a comparable test with PLRV-30, using inoculum purified and stored in the same way and diluted to 25 ~tg/ml (H. Barker, unpublished results). These findings therefore show that the two isolates which are transmitted by M. persicae only poorly, if at all, both lack an antigenic determinant or determinants that occur(s) in all 28 aphidtransmissible isolates tested.
Reactivity of MAbs with other luteoviruses In indirect ELISA, in which homologous polyclonal Ig was used as the coating antibody and SCR-1 to SCR-IO as the second-stage antibody, no reaction was obtained with the British
1818
P. R. MASSALSKI AND B. D. HARRISON Table 4. Transmissibility of four isolates of P L R V by Myzus persicae Virus isolate Expt. 1 (1 aphid/test plant) PLRV-V PLRV-1 PLRV-11 PLRV-F Expt. 2 (3 aphids/test plant) PLRV-V PLRV-1 PLRV-11 PLRV-F
PLRV concentration Transmissionto in source plant* test plantst 825 760 775 420
0/24 9/24 22/24 14/24
740 680 650 390
0/24 36]48 27/48 45]48
* Potato cv. Maris Piper plants were used as the source of each isolate. Figures represent the concentration of purified PLRV particles, expressed as ng virus/g leaf tissue, that would have given the same A,o5 value in DASELISA as the extract from leaves of the virus source plants. t Numerator is number of P. floridana test plants infected, denominator is number inoculated. isolates of BWYV or CRLV, or with the B or F isolates of BYDV, in conditions in which strong reactions were given by each virus in DAS-ELISA using homologous polyclonal reagents. BWYV polyclonal antibody can be used in DAS-ELISA to detect G R A V in rosette-affected groundnut foliage (Casper et al., 1983) although in tests at this institute this was not consistently effective. However, in indirect ELISA, in which polyclonal Ig to BWYV or PLRV was used as the coating antibody, and SCR-1 to SCR-10 were used as the second-stage antibody, SCR-6, SCR-8 and SCR-10 gave strong reactions (A405 values of 0.72 to 1.55) whereas virus-free groundnut sap gave values less than 0-04. The other seven MAbs did not react with G R A V (R. Rajeshwari, personal communication).
Properties of virus epitopes The above results make it possible to distinguish several epitopes on the particle protein of PLRV-1. Two MAbs, SCR-8 and SCR-10, detect epitopes, which might be identical, that are shared with G R A V but missing from PLRV-V and PLRV-15. SCR-6 detects an epitope shared by GRAV and all the PLRV isolates tested, and SCR-7 detects an epitope that is modified in PLRV-V and PLRV-15 in such a way that these isolates react less strongly than the others. Of the other six MAbs, SCR-2 and SCR-9 detect epitopes, which might be identical, that are lacking in some Australian isolates of PLRV (J. E. Thomas, P. R. Massalski & B. D. Harrison, unpublished results). SCR-1, SCR-3, SCR-4 and SCR-5 gave reaction patterns that differ from those of the other MAbs but are indistinguishable from one another. Five epitopes were therefore distinguished: those detected by SCR-1, SCR-2, SCR-6, SCR-7 and SCR-8, respectively. The initial screening tests used to identify antibody-secreting hybridomas indicated that more PLRV-specific MAbs reacted with antibody-trapped virus from sap than with plate-trapped purified virus as antigen. In these two systems the antigen was suspended in different media: extraction buffer (pH 7.0) for antibody-trapped antigen and coating buffer (pH 9-6) for platetrapped antigen. When purified particles of PLRV-1 were incubated overnight at 5 °C, or for 3 h at 37 °C, in either PBS (pH 7.4) or coating buffer, the samples in PBS were found by electron microscopy to contain many intact virus particles whereas no such particles were seen in the samples kept in coating buffer. Moreover, when two comparable samples were sedimented in sucrose density gradients, that in PBS produced an absorption profile with one sharply defined main componefit typical of purified PLRV particles, whereas the sample in coating buffer contained ill-defined components and there was an increased amount of u.v.-absorbing material at the top of the gradient (Fig. 1). These results show that PLRV particles are extensively degraded in coating buffer, with the probable loss of some conformation-specific epitopes on the surface of PLRV particles. Particles of BYDV, another luteovirus, are likewise degraded in coating buffer (Diaco et al., 1986).
MAbs to potato leafroll luteovirus (a)
1819
(b)
2~
Sedimentation
~"
Fig. 1. Sedimentation profile of PLRV-1 after centrifugation at 40000 r.p.m, for 1 h in density gradients of 10 to 40% sucrose in 0.01 M-phosphate buffer pH 7, using a Beckman SW50.1 rotor. (a) Virus particles (40 ~tg)previously incubated overnight at 5 °C in 0.01 M-phosphate-buffered saline, pH 7.4; (b) virus particles (40 p.g) previously incubated overnight at 5 °C in 0.05 M-sodium carbonate buffer, pH 9.6. Table 5. Effect of antigen buffer on reaction of plate-trapped PLRV antigen with MAbs in
indirect ELISA Antigen buffer~f A
Antibody* SCR- 1 SCR-2 SCR-3 SCR-4 SCR-5 SCR-6 SCR-7 SCR-8 SCR-9 SCR-10 Polyclonal Ig
0.01 M-PBS pH 7.4 > 2.00* 1.75 1-80 1.81 1.73 > 2-00 > 2.00 1.54 1.73 1.59 > 2.00
0.05 M-Na2CO3 pH 9.6 > 2.00 > 2.00 > 2.00 > 2.00 > 2.00 0.37 0.07 0.01 > 2-00 0.04 1.77
* MAbs (SCR-1 to SCR-10) were tested in indirect ELISA as tissue culture supernatant fluids; polyclonal Ig was tested in DAS-ELISA. t Purified PLRV-1 particles were diluted to 2 ~tg/ml in sodium carbonate buffer (coating buffer) or PBS and these mixtures incubated overnight at 5 °C in wells of microtitre plates. :~Figures are mean A,os values for duplicate wells, minus the buffer values, and were recorded after incubation with substrate for 2 h at 23 °C. In further tests, the reactivity of the M A b s with plate-trapped antigen was examined in indirect E L I S A using PLRV-1 particles in either coating buffer or PBS as the antigen. The results show that all the M A b s reacted with PLRV-1 antigen which was t r a p p e d from PBS, suggesting that all the epitopes detected are on the surface of virus particles (Table 5). In contrast, only six M A b s reacted strongly with PLRV-1 antigen which was t r a p p e d from coating buffer. SCR-7, SCR-8 and SCR-10 did not react, and SCR-6 reacted only weakly, suggesting that these M A b s probably detect epitopes dependent on quaternary structure. The epitopes associated with aphid-transmissibility in P L R V and those shared by P L R V and G R A V are therefore all of this conformation-sensitive type. DISCUSSION The 10 murine M A b s whose production and testing are described in this p a p e r show some features of special interest. W h e n used at high concentrations (100 ~g/ml or more) in indirect E L I S A they gave unexpectedly large A4o5 values in tests on virus-flee sap or control buffer. This was independent of isotype and did not h a p p e n with l0 M A b s to African cassava mosaic virus that were made by the same methods (Thomas et al., 1986). However, Martin & Stace-Smith (1984) observed a similar phenomenon with the two M A b s to a C a n a d i a n isolate of P L R V and
1820
P . R . MASSALSKI AND B. D. HARRISON
suggested that it was caused by 'stickiness' of the IgG or a tendency of IgG to exchange with other proteins bound to the microtitre plates. Why this should occur with MAbs to some viruses but not those to others is not clear. A second unusual feature is that none of the 10 MAbs to PLRV-1 gave a precipitin line when allowed to react with purified virus particles in immunodiffusion tests. In contrast, five out of 14 MAbs to ilarviruses (Halk et al., 1984) and six out of 10 MAbs to African cassava mosaic virus (Thomas et al., 1986) precipitated homologous virus, although none of three MAbs to BYDV did so (Torrance & Pead, 1986). This suggests that non-precipitating antibodies may be the norm among MAbs to luteoviruses. A third observation, possibly merely coincidental, is that all the MAbs that reacted with disrupted PLRV particles were of the IgG2a isotype whereas those that reacted with the conformationsensitive epitopes were of other isotypes. The tests with the MAbs to PLRV-1 provide fresh information on the antigenic constitution of PLRV particles. The predominant impression is of great antigenic uniformity among the 30 British isolates tested. Our results considerably extend those of previous work, which showed that PLRV isolates from potato in different countries are strongly related, by using tests with polyclonal (Rowhani & Stace-Smith, 1979; Roberts et al., 1980; Kojima, 1981) or monoclonal (Martin & Stace-Smith, 1984; International Potato Center, 1985) antibodies, but which did not include tests designed to detect small antigenic differences between isolates. The relatively great antigenic uniformity among PLRV strains implies that MAbs of the SCR-1 type should be valuable for routine detection and identification of PLRV. Indeed, British PLRV isolates that differ in severity of symptoms induced in potato, P. floridana and/or M. perfoliata were all readily detected by SCR-1, and there was no evidence that virulence differences between PLRV strains were associated with antigenic differences, suggesting that PLRV particle protein does not play a key role in determining symptom type. The considerable antigenic uniformity among PLRV isolates also implies that a strong selection pressure exists for antigenic conservation in PLRV particle protein. An obvious constraint on variation in this protein is that imposed by the aphid vector, M. persicae, because there is increasing evidence that virus particle proteins have a crucial function in transmission of luteoviruses by their vectors (Harrison & Mutant, 1984). The data presented in this paper support this view. Thus both examples of antigenic variation that were found involve isolates characterized by poor aphid transmissibility or aphid non-transmissibility (PLRV-15 and PLRV-V). PLRV-15 was characterized in some detail by Tamada et al. (1984) and was first detected in a vegetatively maintained stock of potato cv. Up-to-Date. Interestingly, PLRV-V, isolated 5 years later, also came from cv. Up-to-Date from the same supplier and there is some reason to suspect that the two isolates are derived from the same virus culture. We may therefore have detected but a single example of antigenic variation among 29 isolates. The greatly decreased reactivity of SCR-8 and SCR-10 with this variant suggests that the epitope(s) recognized by these MAbs, and which occur(s) on the particles of the 28 aphid-transmissible isolates, has (have) undergone a change that results in loss of aphid transmissibility. Presumably, such an isolate would soon be eliminated from the population of isolates that occur naturally in seed potato crops and rely on aphid transmission for survival. PLRV-V and PLRV15 have survived only because of intentional vegetative propagation of infected potato clones. The exact mechanism of passage of a luteovirus through the body of a vector aphid is unknown but the existence of a poorly transmissible variant facilitates some experimental approaches to examining this question. Results of tests with PLRV-15 led to the conclusion that its poor transmission was not caused by an inadequate virus concentration in source plants, or by inadequate virus uptake by, or instability in, M. persicae (Tamada et al., 1984). With BYDV, there is strong evidence that vector transmissibility is determined by the ability of virus particles to pass from the haemolymph of the aphid into cells of its accessory salivary glands (Gildow & Rochow, 1980), and that this ability depends on the specificity of the virus particle protein (Rochow, 1970; Gildow & Rochow, 1980). The results of our work with PLRV isolates differing in aphid transmissibility, and of that of Tamada et al. (1984), are compatible with this view and raise the possibility that the epitopes recognized by SCR-8 and SCR-10 have a key role in the passage of PLRV particles from the haemolymph to salivary gland cells of M. persicae. The
MAbs to potato leafroll luteovirus
1821
a p p a r e n t d e p e n d e n c e o f the epitope(s) d e t e c t e d by these M A b s on q u a t e r n a r y structure would also constitute a m e c h a n i s m for p r e v e n t i n g the passage o f d a m a g e d virus particles t h r o u g h salivary gland cells. In this c o n n e c t i o n , it is w o r t h n o t i n g t h a t b o t h the e p i t o p e s s h a r e d by P L R V and G R A V are o f this c o n f o r m a t i o n - d e p e n d e n t type, suggesting t h a t the m e c h a n i s m o f transmission o f G R A V by its vector, Aphis craccivora, m a y h a v e i m p o r t a n t details in c o m m o n with that o f P L R V by M. persicae. We are indebted to C. Robertson, DAFS, E. Craigs, Edinburgh and to our colleague R. M. Solomon for supplying PLRV-infected potato tubers; to J. E. Duffus, R. R. Martin, A. F. Murant, L. Torrance and D. G. A. Walkey for providing virus antisera and/or virus cultures; to I. M. Roberts for electron microscopy; to R. Rajeshwari and J. E. Thomas for help with some tests and to Ashley Craig for technical assistance. REFERENCES BARKER,H. &HARRISON,B. D. (1985). Restricted multiplication of potato leafroU virus in resistant potato genotypes. Annals of Applied Biology 107, 205-212. CASPER, R., MEYER, S., LESEMANN, D. -E., REDDY, D. V. R., RAJESHWARI, R., MISARI, S. M. & SUBBARAYADU, S. S. (1983). Detection of a luteovirus in groundnut rosette diseased groundnuts (Arachis hypogaea) by enzyme-linked immunosorbent assay and immunoelectron microscopy. Phytopathologische Zeitschrift 108, 12-17. DIACO, R., LISTER, R. M., HILL, J. H. & DURAND, D. P. (1986). Demonstration of serological relationships among isolates of barley yellow dwarf virus by using polyclonal and monoclonal antibodies. Journal of General Virology 67, 353-362. GALFRI~, G. & MILSTEIN,C. (1981). Preparation of monoclonal antibodies: strategies and procedures. Methods in Enzymology 73, 3--46. t:;XLI~OW,V. E. & Rocnow, W. r. (1980). Role of accessory salivary glands in aphid transmission of barley yellow dwarf virus. Virology 104, 97-108. HALK, E. L. & DE BOER, S. H. (1985). Monoclonal antibodies in plant-disease research. Annual Review of Phytopathology 23, 321-350. rlALK, E. L., I-LSU,n. T., AEmG,J. & FRArOZE,J. (1984). Production of monoclonal antibodies against three ilarviruses and alfalfa mosaic virus and their use in serotyping. Phytopathology 74, 367-372. HARRISON,B. D. (1984). Potato leafroll virus. Commonwealth Mycological Institute/Association of Applied Biologists Descriptions of Plant Viruses, no. 291. HARRISON,B. D. & MURANT,A. F. (1984). Involvement of virus-coded proteins in transmission of plant viruses by vectors. In Vectors in Virus Biology, pp. 1-36. Edited by M. A. Mayo & K. A. Harrap. London: Academic Press. INTERNATIONALPOTATOCENTER(1985). Variability of PLRV. Annual Report CIP 1984, 66. KOJIMA,M. (1981). Note on the serological relationship between Japanese and foreign isolates of potato leafroll virus. Bulletin of the Faculty of Agriculture, Niigata University 33, 73-77. MARTIN,R. R. & STACE-SMITH,R. (1984). Production and characterization of monoclonal antibodies specific to potato leaf roll virus. Canadian Journal of Plant Pathology 6, 206-210. REDDY, D. V. R., MURANT, A. F., DUNCAN, G. H., ANSA, O. A., DEMSKI, J. W. & KUHN, C. W. (1985). Viruses associated with chlorotic rosette and green rosette diseases of groundnut in Nigeria. Annals of Applied Biology 107, 57-64. ROBERTS,I. M., TAMADA,T, & HARRISON,B. D. (1980). Relationship of potato leafroll virus to luteoviruses: evidence from electron microscope serological tests. Journal of General Virology 47, 209-213. ROCHOW,w. F. (1970). Barley yellow dwarf virus: phenotypic mixing and vector specificity. Science 167, 875-878. ROWHANI,A. &STACE-SMITn,R. (1979). Purification and characterization of potato leafroll virus. Virology98, 45-54. TAMADA,T. & HARRISON,13. D. (1980). Factors affecting the detection of potato leafroll virus in potato foliage by enzyme-linked immunosorbent assay. Annals of Applied Biology 95, 209-219. TA.V,ADA,T., HARRISON,B. D. &ROBERTS,I. M. (1984). Variation among British isolates of potato leafroll virus. Annals of Applied Biology 104, 107-116. THOMAS, J. E., MASSALSKI, P. R. & HARRISON, B. D. (1986). P r o d u c t i o n o f m o n o c l o n a l a n t i b o d i e s to A f r i c a n c a s s a v a
mosaic virus and differences in their reactivities with other whitefly-transmitted geminiviruses. Journal of General Virology 67, 2739-2748. TORRANCE,L. & PEAD,M. T. (1986). The application of monoclonal antibodies to routine tests for two plant viruses. In Developments and Applications in Virus Testing, pp. 103-118. Edited by R. A. C. Jones & L. Torrance. Wellesbourne: Association of Applied Biologists. TORRANCE, L., PEAD, M. T., LARKINS, A. P. & BUTCHER, G. W. (1986). Characterization of monoclonal antibodies to a U.K. isolate of barley yellow dwarf virus. Journal of General Virology 67, 549-556. WATERHOUSE,V. M. & MURANT,A. F. (1981). Purification of carrot red leaf virus and evidence from four serological tests for its relationship to luteoviruses. Annals of Applied Biology 97, 191-204. WEBB, R. E., LARSON, R. H. & WALKER, I. C. (1951). N a t u r a l l y o c c u r r i n g s t r a i n s o f t h e p o t a t o l e a f roll virus. American
Potato Journal 28, 667-671. (Received 10 February 1987)
