Characterization of Potato Virus M epitopes with the use of synthetic peptides

As a result of thermolysin hydrolysis of a coat protein (CP) of Potato Virus M Ukrainian Strain UJ (PVM)i the heptapeptide AADFEGJC was found to be recognised by two PVM specific monoclonal antibodies (MAbs) M6D5 and M9GJ. This heptapeptide represents the C-terminal part of tryptic tetradecapeptide EARPLPTAADFEGK (P14) which was also recognised by the same MAbs. The peptides represented sequences of tryptic (P14)t thermolysinic (P7) fragments and three heptapeptides containing alanine substitutions for Asp and Gluf were synthesised to determine the contribution of dicarbonic amino acids in the antigen-antibody interaction. All synthetic heptapeptides were recognised by both MAbs weakly in indirect ELISA. These peptides were also used as inhibitors of MAb-CP and MAb-P 14 reactions in inhibition ELISA. The results of inhibition ELISA have shown the following: 1) the same concentrations of peptides were more effective to inhibit the interaction of MAbs with P14 than with CP; 2) substitutions of charged amino acids decreased noticeably the ability of peptides to inhibit the antigen-antibody interaction, especially the substitution of Asp; 3) heptapeptides containing alanine substitutions suppressed more effectively the interaction of MAb M6D5 with antigens and were less effective to inhibit the reaction of MAb M9G1 with the same antigens. Thust the difference in Asp and Glu contributions to the antigen-antibody complex formation has been found.

Introduction.Previously the antigenic analysis of two Potato Virus M Ukrainian strains (PVM Ul and U7) was carried out with the use of three monoclonal antibodies (MAbs M6D5, M9G1 and M4C1) [1].The PVM-specific epitopes were established to be located in the N-terminal region of the coat protein (CP) and sequentially overlapped each other [1 ].In that study the tryptic fragments 22 Glu-Lys 35 (PVM Ul) and 22 Gly-Lys 35 (PVM U7) were recognised by two MAbs M6D5 and M9G1.The amino acid substitution Glu-» -*Gly at the position 22 of PVM U7 CP did not affect the antibody binding of the tryptic peptide and PVM U7 CP.It was shown that MAb M9G1 interfered with MAb M4C1 for binding PVM CP sites in the dot immunobinding assay.The synthetic peptide P14, corresponding to the tryptic fragment 2 Glu-Lys 35 , inhibited the MAbs M6D5 and M9G1 interaction with CP and peptide P14 with different level of ELISA inhibition.The modification of the side chain positive charge of Lys 35 to negative one in tryptic peptides 22 Glu-Lys 35 and 22 Gly-Lys 35 using citraconic anhydride, resulted in two-fold increase of the MAb M9G1 reaction and slightly reduced the MAb M6D5 interaction with both fragments.On the basis of these results, it was concluded that: 1) PVM-specific epitopes are located at the N-terminal region of PVM coat protein; 2) MAbs M6D5 and M9G1 recognise sequentially overlapping epitopes and the common part of both epitopes is presented in the 22 GWGly-Lys 35 fragment; 3) MAbs M4C1 and M9G1 recognise either overlapping or conformational approximated epitopes.Also it was suggested that the region at position 34-49 of PVM CP is disposed to form the loop with я-turn [1 ].
In the present study the synthetic heptapeptides containing alanine substitutions of dicarbonic amino acids were used for more detailed analysis of PVM epitopes.
Materials and Methods.Virus purification.PVM Ul strain was purified as described in [2], using 50 mM KH 2 PO 4 -KOH buffer with 5 mM EDTA and 0.2 % 2-mercaptoethanol, pH 7.5, and 0.3 M glycine-KOH buffer, pH 8.5.All purification procedures were performed at 4 °С.The virus concentration was estimated assuming the extinction coefficient at 260 nm of 2.88 cm^mg 1 .The A 2607280 = 1.21,A max7min =1.12 ratios [3] and SDS-electrophoresis [4] were used to analyse purity and quality of virus preparations.The purified virus were stored at -20 0 C in 0.3 M glycine-KOH buffer, pH 8.5, containing Coat protein preparation.PVM Ul coat protein (CP Ul) was prepared according to [5].The purified virus at the concentration of 3-5 mg/ml in the 0.3 M glycine-KOH buffer, pH 8.5, was treated with the mixture of guanidine-HCl and LiCl at the final concentration of 4 M and 2 M, respectively, and followed by freezing at -70 °С for 3 h and thawing at room temperature.The precipitated virus RNA was collected by centrifugation at 5000 g for 30 min.The supernatant containing coat protein was dialysed against ammoniac water, pH 8.0, and lyophilised.The protein concentration was estimated according to Lowry et al. [6 ].
Fractionation of thermolysinic peptides and sequencing.CP Ul thermolysinic mixture was fractionated by High Performance Liquid Chromatography (HPLC) on a C18 LiChrosorb RP-18 column (5 μτη, 4.6 mm χ 250 mm, «LKB», Sweden) using the 2-60 % linear gradient of acetonitrile («Fluka», Switzerland) in 0.1 % trifluoroacetic acid (TFA) at the flow rate of 1.0 ml/min.Optical density was monitored at 206 nm and 280 nm.The collected fractions were dried and peptide material of each fraction was dissolved in an appropriate buffer before using in further experiments.
The thermolysinic fractions found to be recognised by MAbs were repeatedly fractionated using the same column and combination of linear gradient (2-40 %) and isocratic acetonirtile elution in 0.1 % TFA.The amino acid sequence analysis was performed in a gas sequenator, model 816 («Knauer», Germany), according to the manufacturer protocol.
In the inhibition ELISA experiments, MAbs were incubated with two-fold decreasing concentrations of synthetic peptides P14, P7, PI, P2 and P3 for 1 h at room temperature.Then these mixtures were added to the precoated antigens CP Ul -500 ng (14 pM) per well or peptide Pl4 -50 ng (33 pM) per well with following incubation for 1 h at 37 °С.The other steps of inhibition ELISA were performed as described above in ELISA procedure.
Further in the text all concentrations of antigens and inhibitors mean the amount of the material per well.
Monoclonal antibodies.MAbs M6D5, M4C1 and M9G1 were raised against native virus particles of PVM-Russian strain and were characterised earlier [8 ].Professor M. Saarma, Institute of Chemical Physics and Biophysics (Estonia) has kindly provided the antibodies for this study.
Results and Discussion.For the further study of PVM antigenic structure the thermolysin hydrolysis of PVM CP was carried out.The products of proteolytic cleavage were fractionated by HPLC using linear acetonitrile gradient.The fractions collected were tested by indirect ELISA using three MAbs, M6D5, M4C1 and M9G1.MAb M4C1 did not re-cognise any thermolysinic fraction while MAbs M6D5 and M9G1 reacted with fraction 17.After additional HPLC purification using a combination of linear gradient and isocratic acetonitrile elution it has been shown, that fraction 17 consists of three subfractions: 17,, 17 2 and 17 3 .Only subtraction 17 3 was recognised by MAbs M6D5 and M9G1 in ELISA.The primary structure of 17 3 subfraction peptide AADFEGK was established on a gas sequenator.The sequence of this heptapeptide corresponds to the fragment of PVM Russian strain CP [9] at the positions 29-35 and represents the C-terminal part of tryptic tetradecapeptide 22 EARPLPTAADFEGK 35 that was also recognised by MAbs M6D5 and M9G1 [1].
In previous work [1 ] we pointed out that lack of reaction between MAb M9G1 and lengthened peptides 22 Glu/Gly-Arg 44 may be caused by structural changes in the C-terminal region of this fragment after protein cleavage by trypsin.These structural changes did not affect the recognition of 22 GWGly-Arg 44 peptides by MAb M6D5.The similar phenomenon, when the antibody recognised the short peptide and did not recognise the lengthened peptide, have been observed by other investigators [10,11] and had no explanation until present.In our case we suggested that region 34 Gly-Asp 49 formed the loop with л:-turn [12] and some of M9G1 epitope's amino acids were involved in the formation of this structure.The enzymatic destruction of PVM CP resulted in drastic distortions of the я-turn structure and abolished the recognition of M9G1 epitope in peptides 22 GWGly-Arg 44 by corresponding MAb, whereas the main portion of M6D5 epitope located out of the лг-turn structure and both peptides 22 Glu/Gly-Lys 35 and 22 Glu/Gly-Arg 44 were recognised by MAb M6D5.Thus, most likely, the M6D5 epitope is located at the N-terminal part while M9G1 epitope is shifted to the C-terminal part of 22 GWGly-Arg 44 fragment [1 ].
On the basis of ELISA of thermolysinic fragments, the linear heptapeptide AADFEGK (P7) was synthesised.Additionally, the following peptides, containing alanine substitutions of charged amino acid were synthesised: 1) peptide AAAFEGK (Pl) with Asp 31 Ala substitution; 2) peptide AADFAGK (P2) with Glu 33 Ala substitution; 3) peptide AAAFAGK (P3) with Asp 31 Ala and Glu 33 Ala simultaneous substitutions.The alanine substitutions can help to find out the contribution of negatively charged amino acids in the MAb-antigen interaction.Generally accepted, that in most cases one amino acid substitution on the protein molecule surface resulted only in local changes, although the mutations are known that alter the distant regions conformation and thus have an indirect effect on antigenicity [13][14][15][16].
In indirect ELISA MAbs M6D5 and M9G1 reacted with 500 ng (14 pM) of CP equally (E 450 = 2.5), while the reaction of MAb M6D5 with synthetic peptide P14 (50 ng or 33 pM) E 450 = 2.6 was significantly higher then of MAb M9G1 (E 450 = 1.4).The difference in reactions between MAbs and peptide P14 shows that only the part of M9G1 epitope is presented in peptide P14, whereas this peptide contains the most part of M6D5 epitope.Above mentioned concentrations of CP and peptide P14 were used as coating concentrations in inhibition ELISA.
Synthetic peptides (P7, PI, P2 and P3) were analysed in indirect and inhibition ELISA formats.In indirect ELISA MAbs M6D5 and M9G1 detected the peptide P7 at the minimal concentrations of 10 pM (7 ng) and 1 nM (700 ng), respectively, while peptides PI, P2 and P3 were not recognised by the antibodies at the concentration of 100 nM.The results of analysis of synthetic peptides P7, PI, P2 and P3 as coating antigens were highly contradictory and the level of MAbs reaction did not correlate with peptide concentrations.
This effect can be referred to very poor adsorption of short peptides to the plate material that leads to decrease or lack of ELISA reaction.Usually, to avoid such problem, the short peptides are used as inhibitors of the interaction between antibodies and antigens.In our study we used this approach to analyse the abilities of synthetic peptides P7, PI, P2 and P3 to inhibit the reaction of MAbs M6D5 and M9G1 with PVM CP and synthetic peptide P14.Synthetic peptide i55 DASSSVF 1 61 was used as a negative control and displayed any inhibition capability in MAb-CP and MAb-Pl4 reactions.
The higher concentrations of peptide P7 were required to inhibit the MAbs-CP interaction in com-parison with MAbs-P 14. 50 % inhibition of M6D5-CP reaction was achieved in the presence of 2.5 and 2.2 nM of peptides P7 and P14, respectively, and C raax of both peptides were 200 nM (Fig. 2, lines 7, 2).The interaction M9G1-CP was suppressed by 50 % with 136 nM of peptide P14 (Fig. 2, line 4).These results showed that conformation of peptide P7 was more appropriate for MAb M9G1 paratope while the peptide P14 was stronger inhibitor of reactions M6D5-CP/P14.
Simultaneous substitution of Asp 31 and Glu 33 in peptide P3 decreased drastically the ability of this peptide to inhibit the reaction of MAbs with both antigens.At the concentration 200 nM peptide P3 did not suppress completely any reaction, displaying 72 % and 33 % of inhibition in M6D5-P14 (Fig. 3, line 5) and M6D5-CP (Fig. 3, line 6) interactions, respectively.The reactions of MAb M9G1 with CP and P14 were inhibited with 200 nM of peptide P3 by 1 % (Fig. 4, line 6) and 18 % (Fig. 4, line 3), respectively.
Thus, the comparison of inhibition abilities of peptides P7, PI, P2 and P3 showed that peptide P3 (Asp 31 Ala and Glu 33 Ala) was the weakest.Peptide P2 (Glu 33 Ala) maintained the highest residual inhibition activity in comparison with peptides Pl and P3.Peptide Pl (Asp 31 Ala) also significantly affected the reactions of MAbs with both antigens.Peptides PI, P2 and P3 were more effective inhibitors of MAb M6D5-antigen reactions (Fig. 3) than MAb M9G1-antigen reactions (Fig. 4).The same inhibitor (peptides P14, P7, PI, P2, P3) concentrations more effectively suppressed the interactions MAbs-P 14, than MAbs-CP.These data indicate that the native conformation and environment of this region in the protein molecule are very important for the interaction between MAbs and PVM epitopes, even if all critical residues of the epitopes are assumed to be located in the 22 Glu-Lys 35 region (peptide P14).
In present study the contribution of individual residues (Asp 31 and Glu 33 ) in the MAb-antigen complex formation has been evaluated using the alanine substitutions.It has been established that both residues Asp 31 and Glu 33 are involved in the M6D5 and M9G1 epitopes and contribute differently to the formation of the antigen-antibody complexes.Asp 31 is more crucial residue of both epitopes than Glu 33 .Both amino acids Asp 31 and Glu 3 in M9G1 epitope are more important for recognition by corresponding MAb than the same residues in the structure of M6D5 epitope.The simultaneous substitution of Asp 31 and Glu 33 in the M9G1 epitope drastically decreases the recognition of this epitope by the antibody that shows the key role of these residues in the epitope, while M6D5 epitope in the absence of Asp 1 and Glu 33 residues partially reserves the ability to be recognised by the corresponding antibody.It is obvious that neighbour amino acids in the M6D5 epitope considerably contribute to the antigen-antibody complex formation.