Immunogenetics (1995) 42:302-303 © Springer-Verlag 1995
PEPTIDE MOTIF REGISTER
Nagendra R. Hegde • Shirley A. Ellis • Ruth M. Gaddum
Clive A. Tregaskes • Gautam Sarath
Subramaniam Srikumaran
Peptide motif of the cattle MHC class I antigen BoLA-A11
Received: 2 May 1995
MHC class I molecules present self-peptides as well as
peptides derived from intracellular pathogens to receptors
on cytotoxic T lymphocytes (CTLs). The bound peptides
are usually 8-10 amino acids long, and generally have one,
two, or more chemically related amino acid residue(s)
recurring at two or more positions. These anchor residues
define the peptide binding characteristics of an allelic
product (for a review see Rammensee et al. 1993). Co-
dominant expression of MHC alleles has made it necessary
either to use allele-specific monoclonal antibodies (mAbs)
or to use single allelic transfectants for the characterization
of allele-specific peptide motifs.
Data from several bovine lymphocyte antigen (BoLA)
workshops have demonstrated the existence of more than
50 serologically defined MHC class I specificities (Davies
et al. 1994). It is apparent that these specificities may
encompass more than one gene product, and it is at present
unclear how many of the three (or more) class I genes are
transcribed (Davies and co-workers, unpublished data). In
this study, we report the peptide motif for a BoLA-All
gene product.
Mouse fibroblasts (L cells) were transfected with
sheared genomic DNA from a heterozygous animal which
typed serologically as All/A14. Transfectants were
screened initially for cattle MHC class I expression using
the mAb IL-A88 (Toye et al. 1990), which recognizes a
monomorphic determinant on cattle MHC class I heavy
chains, and subsequently for the All specificity, by allo-
specific sera. A gene encoding a product recognized as
BoLA-A11 has been recently cloned, sequenced, and trans-
fected (Sawhney et al. 1995). The transfected cell line used
N_ R. Hegde • S. Srikumaran (~)
Dept. of Veterinaryand Biomedical Sciences,University of Nebraska-
Lincoln, Lincoln, NE 68583-0905, USA
G. Sarath
Dept. of Biochemistry and Protein Core Facility, University of
Nebraska-Lincoln, Lincoln, NE 68583-0718, USA
S. A. Ellis • R. M. Gaddum • C. A. Tregaskes
Institute for Animal Health, Compton Laboratory,Compton,
Nr. Newbury, RG20 7NN, UK
in our study is indistinguishable by serology and 1D IEF
analysis from the cell line described by Sawhney and co-
workers (data not shown).
Transfectants were grown (to 1 ×10 ~0) and lysed in the
presence of 0.5% NP-40. Immunoaffinity columns were
made by covalently cross-linking Protein A (Pharmacia,
Piscataway, NJ), using dimethylpimelimidate (Sigma, St.
Louis, MO), with the mAb [L-A88, or an isotype-matched
control mAb (specific for the envelope glycoprotein gp53
of bovine viral diarrhoea virus). The lysate was passed first
over the control mAb column, then over the anti-class I
mAb column. The cattle class I molecules were eluted, and
dissociated from bound peptides by acid teatment (0.1%
trifluoroacetic acid). The peptide pool was separated from
the MHC class I heavy chain and ~2-microglobulin by using
a 3000 Mr cut-off membrane filter (Amicon, Beverly, MA).
The peptide pool was analyzed by sequential Edman
degradation (Millipore 6600 Protein Sequencer; Millipore,
Predford, MA) for 12 cycles according to the manufac-
turer's protocol, and the results were interpreted as de-
scribed in Falk and co-workers (1991). Table 1 shows the
positional characteristics for peptides bound to the BoLA-
All allelic product, obtained from three independent ex-
periments and five independent sequence analyses.
The majority of the peptides that occupied the binding
groove of BoLA-All were nonamers. There was clear
evidence that position 2 was occupied by Pro, although a
small amount of Gln was observed in some experiments.
The carboxy-terminal amino acid anchor was Ile/Val with a
small proportion of Leu and Tyr. Ala and Pro appeared to be
the preferred residues for P3 and P5, respectively, and these
positions may serve as auxiliary anchors. Pronounced
heterogeneity occurred at P3, while a lesser degree of
heterogeneity was seen at other positions of the peptide.
Tyr produced a significant signal at P10, while signals at
Pll and P12 were not obtained in all of the experiments
(see Table 1). The possibility of a few decamer peptides
associated with B oLA-A 11 cannot be ruled out, particularly
since Pro at one or more dominant and/or auxiliary anchor
positions can kink the peptide, so that the C/F pocket,
which has been implicated in accomodating P9 (Saper et al.