WO2009117785A1 - Method of hla tissue typing - Google Patents

Abstract

Provided herewith is a method of characterising Human Leukocyte Antigen (HLA) expression of a subject, comprising a) enriching a biological sample from the subject for an HLA polypeptide from the subject, b) obtaining an at least partial amino acid sequence of the HLA polypeptide from the subject, and c) comparing the at least partial amino acid sequence of the HLA polypeptide from the subject to a library of known HLA polypeptide sequences and assigning the at least partial amino acid sequence of the HLA polypeptide from the subject to one or more known HLA polypeptides.

Description

Method of HLA tissue typing Related Application

This application claims benefit from Australian provisional patent application No. 2008901498 entitled "Method of HLA tissue typing", which was filed 28 March 2008. The entire contents of this provisional application are incorporated herein by reference.

Technical Field

The present invention relates to methods for characterising the expression of HLA polypeptides from the cells, tissues or organs of an individual by obtaining an at least partial HLA polypeptide sequence from a biological sample from the individual.

Background of the Invention

Human Leukocyte Antigens (HLA) are encoded by a group of genes, the majority of which are found in the major histocompatibility complex (MHC) on human chromosome 6. These genes encode polypeptides which are assembled into the different HLA which are expressed on the surface of many cells. HLA polypeptides fall into two principal classes, the HLA class I and the HLA class II polypeptides. HLA class I polypeptides are expressed on virtually all nucleated cells of the body, whilst HLA class II polypeptides are only expressed by cells of the immune system that can act as antigen presenting cells.

Each HLA class I protein is made up of two polypeptide chains: a heavy chain transmembrane polypeptide usually encoded by either the HLA-A, HLA-B or HLA-C genes, and a light chain beta-2 microglobulin (which is encoded on chromosome 15). The major HLA class I proteins are co-dominantly expressed, so that an individual who is heterozygous at all three loci will express six different HLA class I complexes on the surface of most nucleated cells. Each of the three HLA class I polypeptides are involved in immune responses, which allow for the recognition of "self and "non-self. HLA class I molecules are involved in the presentation of endogenous antigens to cytotoxic T- cells, and provide a mechanism for T-cell mediated killing of viral-infected or tumour antigen-expressing cells. HLA class I molecules are also involved in protecting cells from natural killer (NK) cell mediated cytotoxicity by interacting with HLA-specific inhibitory receptors present on NK cells. The engagement of these receptors by autologous HLA class I molecules blocks the ability of NK cells to attack cells which would otherwise be recognized as foreign due to "missing self recognition.

Each HLA class II protein is a heterodimer made up of two transmembrane polypeptides: an alpha chain and a beta chain, which are encoded by each of the HLA- DR, HLA-DQ and HLA-DP genes. The alpha subunit is encoded by A or Al locus and the beta subunit is encoded by one B locus (Bl for DP and DQ, or Bl, B3, B4 or B5 locus for DR). All individuals possess a DRBl gene. For example, an HLA-DP complex may be encoded by the HLA-DPAl and HLA-DPBl loci, and an HLA-DQ complex may be encoded by the HLA-DQAl and HLA-DQBl loci. Usually the HLA-DR complex is encoded by the HLA-DRAl locus and by one of the HLA-DRBl, HLA-DRB3, HLA- DRB4 or HLA-DRB5 loci, hi humans, each chromosome 6 codes for four different HLA class II molecules (two different HLA-DR, one HLA-DQ and one HLA-DP molecule) such that a heterozygous individual will express eight different HLA class II molecules. In contrast to HLA class I molecules, HLA class II proteins are only expressed on the surface of specialized antigen-presenting cells, such as B-lymphocytes, macrophages and dendritic cells.

HLA class II antigens are involved in the presentation of exogenous antigens, such as bacterial or viral antigens to T-helper cells, which in turn activate the production of specific B-cells and the production of antibodies which allow the neutralization of the antigen.

The HLA loci are amongst the most polymorphic coding sequences in the human genome. More than ten allele groups at each locus have been identified for the majority of the HLA loci discussed above, with approximately 1900 class I alleles and approximately 800 class II alleles being identified. Currently over 600 human HLA B alleles and over 390 alleles for the HLA DRBl loci alone have been identified. At least 1500 class I and 718 class II proteins and 110 null alleles (genetically encoded HLA alleles that are not expressed as protein) have been identified.

Histocompatibility, that is the property of having the same, or mostly the same, alleles of the major histocompatibility complex is of vital importance in determining the probability of whether an individual will acutely reject a transplanted tissue, or whether a transplanted haematopoietic tissue will cause a graft-versus-host disease (GvHD).

In order to minimise the possibility of acute rejection or GvHD, the HLA class I and II expressed by the potential donor and host are matched as closely as possible. Furthermore, certain HLA alleles are associated with an increased risk of autoimmune disorders and other disease states, and so HLA typing is of use in rapidly distinguishing individuals at risk or not at risk of developing certain diseases. It is noteworthy that only a small part of the amino acid sequence of each HLA protein is involved in self- recognition. For the purposes of transplantation, genetic polymorphisms which produce changes to the HLA protein primary structure (amino acid sequence), especially within the αl and α2 domains of the class I HLA and both α and β chains of class II HLA, are involved in self recognition, affect self-recognition and antigen presentation. Genetic variation, such as single nucleotide polymorphisms which do not lead to amino acid substitutions, may have no effect on the clinical outcome of a transplant pairing. HLA matching between donors and patients has the greatest impact on clinical outcome with respect to hematopoietic cell transplantation (HCT). Based on the results of several large retrospective studies, typing is performed on three loci (HLA-A, -B and — DRBl), with best results obtained when at least 5 out of these six alleles are matched. The inclusion of HLA-C typing is recommended, but not mandatory. The degree to which HLA matching affects clinical outcome for different indications remains a matter of contention. Current data has demonstrated a significant correlation between matching HLA-A, -B and -DRBl antigens and improved clinical outcomes in terms of prolonged survival and decreased incidence of Graft versus Host Disease. The benefit of matching HLA-C is less apparent. There is no clear data to suggest that matching other HLA loci (HLA-DRB3/4/5, -DQ and DP) improves clinical outcome.

Traditionally, HLA matching requires testing of white blood cells from the host and the potential donors. Differing testing regimes vary in their ability to detect differences in the expressed HLA. Serological HLA typing allows the identification of the major antigens which make up a subject's HLA. While this form of testing is relatively rapid, it is of limited use for detecting small differences in HLA, such as single amino acid polymorphisms, which may nevertheless have large influences in clinical outcomes. The use of serological typing has led to one system of nomenclature for HLA classification, in which antigens are assigned letters and numbers, in the form "HLA- XY", where X is the HLA class letter and Y is a numerical identifier for the antigen.

Higher resolution testing is also available through the use of DNA molecular typing techniques, which rely on the use of specific polynucleotide probes and primers to allow the amplification of the DNA which expresses the HLA, generally followed by restriction enzyme digestion and fragment analysis to identify the specific alleles which are expressed. The DNA molecular typing techniques allow a more refined description of alleles, leading to an alternative nomenclature where each allele is assigned an identifier in the form "HLA-X*Y", where X is the HLA class letter and Y is an identifier of at least two and more usually four numerals for the allele. Marsh et al. (2005) provides an overview of the nomenclature of serologically and genetically defined HLA antigens.

DNA molecular typing techniques, however, are relatively slow to provide results due to the requirement for repeated polynucleotide amplification reactions, and these techniques expend a large plurality of HLA class-, type- and/or allele-specific oligonucleotide primers and probes for each screening test. Analysis in DNA molecular typing techniques is complex, because heterozygous individuals may express two different bases at multiple positions along an HLA gene, requiring complex analysis to predict the two separate sequences.

Accordingly, there is a desire for a technique for characterising HLA expression which has the potential to provide higher resolution data than serological tests while avoiding or alleviating the extended times and large number of specific reagents required for HLA typing based on DNA sequencing techniques.

Summary of the Invention

Accordingly, in a first aspect there is provided a method of characterising Human Leukocyte Antigen (HLA) expression of a subject, comprising enriching a biological sample from the subject for an HLA polypeptide from the subject, obtaining an at least partial amino acid sequence of the HLA polypeptide from the subject, and comparing the at least partial amino acid sequence of the HLA polypeptide from the subject to a library of known amino acid sequences of HLA polypeptides and assigning the at least partial amino acid sequence of the HLA polypeptide from the subject to one or more known HLA polypeptides.

In certain embodiments the HLA polypeptide which is enriched in the biological sample from the subject is an HLA class I polypeptide. In certain embodiments the HLA polypeptide which is enriched in the biological sample from the subject is an HLA class II polypeptide. In particular embodiments the HLA polypeptide is one or more of HLA-A,

HLA-B and HLA-Cw heavy chain polypeptides.

In certain embodiments the HLA polypeptide which is enriched in the biological sample from the subject is a soluble HLA polypeptide. In particular embodiments the step of obtaining an at least partial amino acid sequence of the HLA polypeptide comprises subjecting the enriched biological sample to at least one mass spectrometry analysis.

In certain embodiments the biological sample from the subject is a blood sample. In particular embodiments the step of enriching a biological sample from the subject for an HLA polypeptide comprises an immuno-purification step. In certain embodiments, the immuno-purification step comprises the use of an HLA class I specific antibody and/or the use of an HLA class II specific antibody.

In certain embodiments the mass spectrometry analysis comprises liquid chromatography -tandem-mass spectrometry or matrix-assisted laser desorption/ ionization.

In certain embodiments the subject is a potential donor of cells, tissue or an organ, hi other embodiments the subject is a potential recipient of donated cells, tissue or an organ. In certain embodiments, the subject is suspected of, or at risk of, or with a predisposition towards a disease or condition associated with the expression of one or more specific HLA alleles. The disease or condition associated with the expression of one or more specific HLA alleles may be an autoimmune disease. The disease or condition associated with the expression of one or more specific HLA alleles may be selected from rheumatoid arthritis, lupus like disease, type I diabetes, Behcet's disease, HLA-B27-Associated Cardiac Disease, ankylosing spondylitis, Systemic Lupus Erythematosus, multiple sclerosis, Sjogren's syndrome, myasthenia gravis, viral infection such as HIV infection or hepatitis infection, and pemphigus vulgaris.

In another aspect there is provided a kit for enriching a biological sample from a subject for an Human Leukocyte Antigen (HLA) when used in a method of characterising

Human Leukocyte Antigen (HLA) expression of the subject, the kit comprising at least one antibody specific to a Class I HLA polypeptide and/or at least one antibody specific to a Class II HLA polypeptide.

Brief Description of the Figures

A preferred embodiment of the present invention will now be described, by way of an example only, with reference to the accompanying drawings wherein:

Figure 1 is a photograph of an SDS PAGE gel illustrating the progressive purification of the monoclonal antibody w6/32 from a hybridoma supernatant. Figure 2 provides a graphical representation of the elution profile of soluble HLA Class I from an immunoaffinity column, with UV absorbance on the Y axis and eluted volume on the X axis.

Figure 3 is a photograph of a Western blot analysis of irnmunoaffmity purified soluble HLA Class I antigens from human serum. This figure follows the immunoaffmity purification of HLA class I antigens from the serum of two individuals (identified as AC and LO). Lanes 1 and 9 contained protein molecular weight standards. Lanes 3 and 6 contained serum from subjects AC and LO respectively. Lanes 4 and 7 contained the immunoaffinity column flow through material from subjects AC and LO respectively. Lanes 5 and 8 contained the eluted HLA-enriched fraction from subjects AC and LO respectively. This figure illustrates that prior to immunoaffinity purification and subsequent concentration, the soluble HLA present in the serum samples was below the detection limits of the assay. No soluble HLA was detected in the flow through material (material which did not bind to the immunoaffinity column). The absence of detectable levels of HLA in this sample indicates that the HLA remained bound to the immunoaffinity agarose and that the column capacity (total number of HLA binding sites) had not been exceeded.

Figure 4 provides the amino acid sequence of the HLA class I, A-I alpha chain precursor (allele HLA-A*0101), overlayed on which in bold and underlined font are the amino acid sequences, which were identified by mass spectroscopy analysis of the immunoaffinity purified soluble HLA from subject HM.

Figure 5 provides the amino acid sequence of the HLA class I, A-11 alpha chain precursor (allele HLA-A* 1101), overlayed on which in bold and underlined font are the amino acid sequences which were identified by mass spectroscopy analysis of the immunoaffinity purified soluble HLA from subject HM.

Figure 6 provides the amino acid sequence of the HLA class I, B-8 alpha chain precursor (allele HLA-B*0801), overlayed on which in bold and underlined font are the amino acid sequences, which were identified by mass spectroscopy analysis of the immunoaffinity purified soluble HLA from subject HM. Figure 7 provides the amino acid sequence of the HLA class I, B-49 alpha chain precursor (allele HLA-B *4901), overlayed on which in bold and underlined font are the amino acid sequences, which were identified by mass spectroscopy analysis of the immunoaffinity purified soluble HLA from subject HM. Figure 8 provides the amino acid sequence of the HLA class I, Cw-7 alpha chain precursor (allele HLA-Cw*0701), overlayed on which in bold and underlined font are the amino acid sequences, which were identified by mass spectroscopy analysis of the immunoaffinity purified soluble HLA from subject HM. Figure 9 provides the amino acid seqiience of the HLA class I, Cw-7 alpha chain precursor (allele HLA-Cw*0706), overlayed on which in bold and underlined font are the amino acid sequences, which were identified by mass spectroscopy analysis of the immunoaffinity purified soluble HLA from subject HM.

Figure 10 provides the amino acid sequence of the HLA class I, Cw-7 alpha chain precursor (allele HLA-Cw*0718), overlayed on which in bold and underlined font are the amino acid sequences, which were identified by mass spectroscopy analysis of the immunoaffinity purified soluble HLA from subject HM.

Figure 11 provides the amino acid sequence of the HLA class I, A-2 alpha chain precursor (allele HLA-A* 0201), overlayed on which in bold and underlined font are the amino acid sequences, which were identified by mass spectroscopy analysis of the immunoaffinity purified soluble HLA from subject LO.

Figure 12 provides the amino acid sequence of the HLA class I, A-2 alpha chain precursor (allele HLA-A*0205), overlayed on which in bold and underlined font are the amino acid sequences, which were identified by mass spectroscopy analysis of the immunoaffinity purified soluble HLA from subject LO.

Figure 13 provides the amino acid sequence of the HLA class I, B-50 alpha chain precursor (allele HLA-B*5001), overlayed on which in bold and underlined font are the amino acid sequences, which were identified by mass spectroscopy analysis of the immunoaffinity purified soluble HLA from subject LO. Figure 14 provides the amino acid sequence of the HLA class I, B-60 alpha chain precursor (allele HLA-B *4001), overlayed on which in bold and underlined font are the amino acid sequences, which were identified by mass spectroscopy analysis of the immunoaffinity purified soluble HLA from subject LO.

Figure 15 provides the amino acid sequence of the HLA class I, A-2 alpha chain precursor (allele HLA-A* 0201), overlayed on which in bold and underlined font are the amino acid sequences, which were identified by mass spectroscopy analysis of the immunoaffinity purified soluble HLA from subject MC.

Figure 16 provides the amino acid sequence of the HLA class I, B-7 alpha chain precursor (allele HLA-B*0702), overlayed on which in bold and underlined font are the amino acid sequences, which were identified by mass spectroscopy analysis of the immunoaffinity purified soluble HLA from subject MC.

Figure 17 provides the amino acid sequence of the HLA class I, B-7 alpha chain precursor (allele HLA-B*0761), overlayed on which in bold and underlined font are the amino acid sequences, which were identified by mass spectroscopy analysis of the immunoaffinity purified soluble HLA from subject MC.

Figure 18 provides the amino acid sequence of the HLA class I, A-I alpha chain precursor (allele HLA-A*0101), overlayed on which in bold and underlined font are the amino acid sequences, which were identified by mass spectroscopy analysis of the immunoaffinity purified soluble HLA from subject AC.

Figure 19 provides the amino acid sequence of the HLA class I, A-2 alpha chain precursor (allele HLA-A*0203), overlayed on which in bold and underlined font are the amino acid sequences, which were identified by mass spectroscopy analysis of the immunoaffinity purified soluble HLA from subject AC. Figure 20 provides the amino acid sequence of the HLA class I, B-8 alpha chain precursor (allele HLA-B*0801), overlayed on which in bold and underlined font are the amino acid sequences, which were identified by mass spectroscopy analysis of the immunoaffinity purified soluble HLA from subject AC.

Figure 21 provides the amino acid sequence of the HLA class I, B-35 alpha chain precursor (allele HLA-B*3501), overlayed on which in bold and underlined font are the amino acid sequences, which were identified by mass spectroscopy analysis of the immunoaffinity purified soluble HLA from subject MC.

Figure 22 provides the known predicted amino acid sequence alignment between

HLA-B8 alpha chain precursor (allele HLA-B*0801) and HLA B-49 alpha chain precursor (allele HLA-B *4901), demonstrating the expected regions of homology between the expressed proteins at the HLA-B locus in individual HM. "Consensus" - shows amino acids which are the same for alleles, and amino acids which differ between these two proteins are listed in the rows HLA-B*0801 and HLA-BM901. This figure demonstrates that the amino acid sequence of the expressed HLA-B 8 and HLA-B49 proteins should be identical in the region spanning amino acid positions 207 to 298. For simplicity, this region has been chosen to outline several ambiguities which remain following full molecular DNA sequence based typing.

Figure 23 provides the forward and reverse DNA sequence which was obtained during bi-directional sequencing of the HLA-B locus over exon 4 for subject HM. "Consensus" - shows nucleotides which are the same within both the forward and reverse DNA sequencing data. "b4f" - indicates DNA sequence obtained from the HLA-B locus, exon 4 forward sequencing primer. "b4r.l" - indicates DNA sequence obtained from the HLA-B locus, exon 4 reverse sequencing primer. "..." - indicates regions of homology between the DNA sequence obtained in the forward and reverse sequencing reactions. The sequence beyond nucleotide 257 indicates that the forward sequencing reaction proceeded beyond the limits of the reverse reaction in this region. The DNA codes for ambiguous positions are listed as:

R = G or A; Y = T or C; K = G or T; M = A or C; S = G or C; W = A or T; B = G or T or C; D = G or A or T; H = A or C or T; V = G or C or A; and N = A or G or C or T

Abbreviations

CHAPS 3-[(3-Cholamidopropyl)dimethylammonio]-l-propanesulfonate

DTT dithiothreitol ECACC European Collection of Cell Cultures

FPLC fast protein liquid chromatography

GvHD Graft versus Host Disease

HLA Human Leukocyte Antigen

HRP horseradish peroxidase mAb monoclonal antibody

PAGE polyacrylamide gel electrophoresis

PBS phosphate buffered saline

SBT sequence-based typing

SCX strong cation exchange TCA trichloroacetic acid

Detailed Description of the Preferred Embodiments

Preferred embodiments of the invention will now be described in more detail, including, by way of illustration only, with respect to the examples that follow. The present inventors have now described a method of characterising the Human

Leukocyte Antigen (HLA) expression of a subject. In a particular embodiment the method comprises enriching a biological sample from the subject for an HLA polypeptide, obtaining an at least partial amino acid sequence of the HLA polypeptide by subjecting the biological sample which is enriched for an HLA polypeptide to a mass spectrometry analysis, and comparing the at least partial amino acid sequence of the HLA polypeptide to a library of amino acid sequences of known HLA polypeptides and assigning the at least partial amino acid sequence of the HLA polypeptide to one or more known HLA polypeptides. The direct sequencing of expressed HLA polypeptides from a subject negates the identification of null alleles that is crucial in DNA-based typing methods, and allows transcriptional (such as alternate mRNA splice variants) and translational (glycosylation) variations to be identified, which cannot be resolved using conventional genotyping techniques. The term "Human Leukocyte Antigen" is intended to encompass HLA class I and HLA class II molecules. Within each of these classes of HLA molecules, there are "types" of HLA. It will be recognised that HLA class I molecules comprise the types HLA-A, -B, -C, -E, -F, -G, -H, -J, -K, -L, -N, -S, -X and -Z, with the type based on the identity of the heavy chain (the chain which is responsible for anchoring the HLA to the cell membrane) HLA class I polypeptide which is expressed. The class of HLA class II molecules comprises the types HLA-DRA, HLA-DRBl to HLA-DRB9, HLA-DQAl, HLA-DQBl, HLA-DQA2, HLA-DQB2, HLA-DQB3, HLA-DOA, HLA_DOB, HLA_DMA, HLA-DMB, HLA-DPAl, HLA-DP A2, HLA-DP A3, HLA-DPBl, and HLA- DPB2, with the type based on the identity of the HLA class II alpha or beta chain polypeptide which is expressed.

The term "Human Leukocyte Antigen-like molecule" is intended to include any one or more of TAPl and 2, PSMB8 and 9, MICA, MICB, MICC, MICD, and MICE. (Marsh et ah, 2005). These proteins are all involved in immune responsiveness, although in most cases the details of the mechanism(s) involving these molecules are yet to be elucidated.

Contained within each of these types of Human Leukocyte Antigens and Human Leukocyte Antigen-like molecules are numerous alternative alleles. The International Immunogenetics Project IMGT/HLA Sequence Database at

<http://www.ebi.ac.uk/imgt/hla/> provides a repository for all the HLA sequences which have been named by World Health Organization Nomenclature Committee for Factors of the HLA System (Robinson et al, 2003). The database includes both thenucleotide sequence and the predicted amino acid sequence of all alleles recognised by this committee. New releases of the database are made every three months. A list of all of the recognised alleles and their accession numbers as of 2004 is provided in Marsh et al, 2005 (supra), the entire contents of which is incorporated herein by reference.

It will be understood that the nature of this invention relates to methods for characterizing HLA expression, which in certain embodiments comprises identifying or distinguishing between different HLA classes, types and alleles. Variations to the sequences of members of different HLA classes, types or alleles present in the database, or the addition or deletion of members of different HLA classes, types or alleles present in the database may occur as new alleles are discovered or duplications or errors of records in the sequence database are identified. In some embodiments, the methods of characterizing HLA expression described herein relate to the identification of an HLA class, type or allele of one or more HLA polypeptide(s) expressed by a subject. The methods comprise the identification of at least a partial sequence of the expressed HLA polypeptides, and in particular embodiments the identification of at least a partial sequence of the expressed HLA polypeptides using Mass spectrometry analysis of these polypeptides. Although techniques such as Edman degradation are also able to provide an at least partial amino acid sequence of a soluble HLA, unlike mass spectrometry methods, Edman techniques require highly pure initial samples and are only capable of sequencing peptides of up to 50-60 amino acid residues, and so protease digestion and peptide separation would be necessary. Other protein identification techniques do not give conclusive amino acid sequence data and therefore would not be useful for distinguishing amino acid sequence polymorphisms.

The step of obtaining an at least partial amino acid sequence of the HLA polypeptide is intended to encompass the identification of sufficient amino sequence information so that a sequence comparison is able to attribute the sequence to a particular HLA class, or a particular HLA type, or a particular HLA allele, or to a group consisting of two or more HLA classes, types or alleles.

A partial amino acid sequence of an HLA polypeptide will comprise one or more series of contiguous amino acid sequences from the complete polypeptide. In one embodiment, the contiguous amino acid sequences will comprise at least 6 contiguous amino acid residues from the complete HLA polypeptide. Shorter amino acid sequences may be used, but these have a greater probability of resulting in false matches in a probability based search. For protein identification, a sequence stretch consisting of three to four amino acid residues may provide enough search specificity when combined with the intact mass of the peptide and the masses of corresponding fragment ions in a peptide sequence tag (Shevchenko, Chernushevic et al. 2002).

In other embodiments, the contiguous amino acid sequences will comprise at least 7, at least 8, at least 9 or at least 10 contiguous amino acid residues from the complete HLA polypeptide.

The number of series of contiguous amino acid sequences which may be used to attribute the sequence to a particular HLA class, or a particular HLA type, or a particular HLA allele, or to a group consisting of two or more HLA classes, types or alleles may vary depending on the size of the contiguous amino acid sequence(s) and the position(s) the sequences lie on the sequence of the complete HLA polypeptide. A single contiguous amino acid sequence may be sufficient to characterize the HLA polypeptide where the contiguous amino acid sequence passes through a unique sequence which is characteristic of the particular HLA class, or the particular HLA type, or the particular HLA allele, or to a group consisting of two or more HLA classes, types or alleles. The number of contiguous amino acid sequences may be at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10 or more.

What constitutes sufficient amino acid sequence to attribute the sequence to a particular HLA class, or a particular HLA type, or a particular HLA allele, or to a group consisting of two or more HLA classes, types or alleles can be determined empirically by comparison of a partial sequence to a database of known HLA polypeptides. Partial sequence information may be sufficient to precisely attribute a particular HLA polypeptide to a particular HLA class, or a particular HLA type, or a particular HLA allele with that partial sequence. Currently the most informative HLA typing data is achieved via sequence-based typing (SBT). Using this technique, the most complete HLA typing approach currently employed involves bi-directional DNA sequencing of exons 2, 3 and 4 of the HLA-encoding gene sequence. This is achieved using either genomic DNA as a sequencing template, or by amplification of exons 1 through 5 via PCR to generate template for all subsequent sequencing reactions. For the most commonly expressed HLA allele, HLA-A*0201 (NCBI accession no. NM_002116), exons 2, 3 and 4 combined represent 822 nucleotides of DNA sequence, or 274 amino acids corresponding to positions 25 to 298 of the expressed HLA protein. Most SBT strategies currently employed use only exons 2 and 3 for HLA class I typing. This would correspond (in the case of HLA- A2) to 546 nucleotides encoding 182 amino acids at positions 25 to 206 of the expressed HLA protein. Amino acid sequence information covering amino acids 25 to 298 (or 25 to 206 for the more commonly employed techniques) of the HLA class I alpha chain would represent the protein-level equivalent of the most informative HLA characterisation methodology currently employed. It should be noted that SBT over exons 2 and 3 is often insufficient for assigning a full "HLA type" due to the thousands of ambiguous allele combinations that remain unresolved in heterozygous individuals.

The current commonly used oligonucleotide hybridisation typing method only physically identifies the part of the sequence that is complementary to the probe, and HLA type is assigned based on the assumption that the remainder of the sequence would align to the same allele if the DNA were entirely sequenced, which represents a low to medium resolution characterisation of the HLA of the subject.

Elevated serum concentrations of soluble HLA antigens have been shown to correlate with rejection of heart, liver, kidney, kidney/pancreas and bone marrow allografts and has been suggested as a practical means of monitoring post-transplant alloreactivity. More specifically, the detection of donor derived soluble HLA in the serum of transplant recipients led to the observation that high and stable concentrations of donor soluble HLA may induce immune tolerance and prolonged, stable graft function.

These observations have necessitated a means of distinguishing between donor- derived and recipient soluble HLA antigens in serum, since DNA molecular typing techniques are unable to distinguish between soluble HLA allotypes. As such, the methods described herein provides a practicable method for screening donor soluble HLA following transplantation.

Serological typing identifies a broad range of poorly defined epitopes present on HLA proteins. In some cases, monoclonal antibodies which bind to defined epitopes on the HLA molecule are employed. However, in many cases, polyclonal antisera which bind to a number of undefined protein targets of "known HLA type" are used to assign a

HLA classification, representing a low resolution of HLA of the subject.

Using the methods described herein, it is possible to identify a peptide sequence within a polymorphic region of the HLA protein with great accuracy. If the identified sequence aligns to only one HLA allele, this may be sufficient to assign a low resolution HLA typing result. Other factors such as the linkage disequilibrium between HLA alleles or the alleleic frequency within ethnicity groups may be applied when assigning a HLA typing result.

The characterizing of HLA expression of a subject may be used in screening a subject, or cells, tissue or organs of a subject for suitability in cell, tissue or organ donation. The characterizing may be used for screening a subject for the expression of specific HLA alleles associated with risk of, or with a predisposition towards a disease or condition. The disease or condition associated with the expression of one or more specific HLA alleles may be an autoimmune disease. The disease or condition associated with the expression of one or more specific HLA alleles may be selected from rheumatoid arthritis, lupus like disease, type I diabetes, Behcet's disease, HLA-B27-Associated Cardiac Disease, ankylosing spondylitis, Systemic Lupus Erythematosus, multiple sclerosis, Sjogren's syndrome, myasthenia gravis, susceptibility to viral infection such as HIV, or hepatitis virus and pemphigus vulgaris. For example, individuals who are HLA- B27 or HLA-B57 positive demonstrate "long-term non progression" in that HIV disease progression is much slower in these individuals.

The step of comparing the at least partial amino acid sequence of the HLA polypeptide to a library of known HLA polypeptide sequences is most conveniently carried out using bioinformatics software packages such as Mascot version 2.2.04, Matrix Science (David N. Perkins 1999). A searchable database of known HLA polypeptide sequences may be found at <ftp://ftp.ebi.ac.uk/pub/databases/imgt/mhc/hla/> (Robinson, Waller et al 2003).

In one embodiment, the HLA polypeptides are soluble HLA polypeptides. Soluble forms of HLA include class I heavy chain polypeptides and class II alpha and beta chains of all the types discussed above. The isolation of soluble HLA polypeptides have been described from the serum of healthy individuals, and in individuals with infections, tumours, following transplantation or blood transfusions (Aultman, Adamashvili et al 1999; Ghio, Contini et al. 1999; Munoz-Fernandez, Martin et al. 2001; Toussirot, Saas et al. 2006; Novikov, Egorova et al. 2007). Biological samples of serum, plasma, cerebrospinal fluid, synovial fluid, sweat, saliva or tears may also be used, as soluble HLA-II is present in these fluids. Methods of identification and/or isolation of soluble HLA from such samples are described in the above references, which are incorporated herein by reference. The concentration of soluble HLA class I present in the serum of normal individuals has been estimated at 1470 +/- 95 ng/mL (Shimura, Tsutsumi et al. 2001) and 518 +/- 340 ng/mL (Melchiorri, Martini et al. 2002) as determined by ELISA. An extensive study (n=319) into the level of sHLA class I in serum identified that all normal people have circulating HLA (mean = 357 ng/mL) and that the population could be divided into one group of low secretors (mean = 162.4 +/- 65.2 ng/mL) and another group of high secretors (mean = 541 +/- 186 ng/mL) (McDonald, Gelder et al. 1992). The concentration of sHLA correlates to the presence of certain HLA allotypes (higher levels for HLA- A23, -24, -29 and -33, and lower for HLA- A2) and the concentration may change due to a number of disease states including cancer (McDonald and Adamashvili 1998).

The analysis of soluble HLA polypeptides from blood serum may offer advantages in the relative ease of obtaining a biological sample which comprises the HLA polypeptides, and the relative ease by which the HLA polypeptides may be enriched and preferably substantially isolated from other expressed polypeptides.

In another embodiment, the HLA polypeptide is a membrane-bound HLA polypeptide, such as an HLA polypeptide which is expressed on the surface of white blood cells or on cells present in a tissue biopsy or biological sample. In this embodiment, the method may further comprise the separation of the HLA polypeptide from the membrane. For example separation of the HLA polypeptide from the membrane may take place before or during the step of enriching the biological sample for the HLA polypeptide. The membrane bound HLA polypeptide may be separated from the membrane by cleavage of the extracellular portion, for instance by using a proteolytic enzyme, or by solubilization of the polypeptide with one or more non-ionic and/or zwitterionic detergents, such as Triton X-100 and/or CHAPS, using techniques which are generally available in the art (for example Roe, 2001).

The characterising of the HLA expression of a subject may comprise the identification of one HLA type expressed by the subject. It may comprise the identification of multiple HLA types expressed by the subject. It may comprise the identification of all the HLA types expressed by the subject. The characterising of the HLA expression of a subject may comprise the identification of one or more HLA alleles expressed by the subject. The characterising of the HLA expression of a subject may comprise the identification that one or more alleles are not expressed by the subject. The characterising of the HLA expression of a subject may comprise the identification that the subject expresses any one of a group of alleles. The characterising of the HLA expression by a subject may comprise the characterising of the expression of an HLA class I polypeptide. The characterising of the HLA expression of a subject may comprise the identification of the expression of an HLA class II polypeptide. The characterising of HLA expression of a subject may comprise the identification of the expression of an HLA-like polypeptide as defined above.

In one embodiment the "subject" is a non-human animal or non-human animal tissues, cells or organs expressing the animal equivalent of HLA or a homologous MHC antigen, such as an animal of commercial, economic or research importance, or tissue or organs donated by or obtained from the subject. Where the subject is a non-human animal the term "HLA" is intended to encompass the animal equivalent of HLA or a homologous MHC antigen. A database of a broad range of non-human species whose MHC has been partially characterised is available at The Immuno Polymorphism MHC Database <http://www.ebi.ac.uk/ipd/mhc/>. This database is described in Ellis et al. 2006 and Robinson et al., 2005, and includes information about the leukocyte antigens from canines, cattle, chickens, felines, fish, horses, non-human primates, pro-simian primates, rats, sheep and swine. The biological sample may be a blood sample, including whole blood, a cellular fraction of whole blood, such as a complete white blood cell fraction or a specific white cell fraction, or, if the animal equivalent of the HLA polypeptide is a soluble polypeptide, the biological sample may be, but is not limited to, a serum, plasma, cerebrospinal fluid, synovial fluid, sweat, saliva or tear sample or another biological fluid in which soluble animal-equivalent of HLA may be found. The biological sample may be a lymph sample. The biological sample may be urine, cerebrospinal fluid, sweat, saliva, tears, or a biological fluid sample which comprises soluble or membrane bound animal equivalent of HLA protein.

In another embodiment the subject is a human subject, or cells, tissue or organs of a human. Thus the subject may be an individual in need of a blood transfusion or needing to receive transplanted cells, tissue or organ. The subject may be an individual donating blood, cells, tissue or an organ. The "subject" may be cells, tissues, or organs which have been donated by an individual but which have not yet been provided to a recipient. The biological sample may be a sample comprising cells which express HLA or where soluble HLA is used, the biological sample may be, but is not limited to, a serum, plasma, cerebrospinal fluid, synovial fluid, sweat, saliva or tear sample or another biological fluid in which soluble HLA may be found. In particular embodiments, the method of characterising the HLA expression of a subject described herein comprises enriching a biological sample from the subject for an HLA polypeptide. The biological sample may be a blood sample, including whole blood, a cellular fraction of whole blood, such as a complete white blood cell fraction or a specific white cell fraction, or, if the HLA polypeptide is a soluble HLA, a blood serum or plasma sample. The biological sample may be a lymph sample. The biological sample may be urine, cerebrospinal fluid, sweat, saliva, tears, or a biological fluid sample which comprises soluble or membrane bound HLA protein.

The biological sample may be a sample of tissue, such as a tissue biopsy. Since detectable levels of HLA class I are expressed on virtually all nucleated cells of the body and in most bodily fluids, there is little restriction on what tissues may be used to provide an HLA class I sample, such as a muscle, skin, or fat sample. For HLA class II however, samples containing or which in vivo are in contact with antigen presenting cells, including samples of blood, blood cells, plasma or serum, spleen or lymph node tissue and tissues containing mononuclear cell infiltrates, such as skin biopsies which contain specialised antigen presenting cells such as B cells, dendritic cells or Langerhan's cells which express HLA-class II may be used. In addition, cell lines such as B-lymphoblastoid cell lines, transformed cell lines such as melanoma cell lines derived from primary human cell cultures also express soluble and membrane bound forms of HLA class I and/or II antigens that are the same as those from the host from which the cell line was derived.

Methods of Enriching for an HLA polypeptide

In certain embodiments, the identification of sufficient sequence data to infer a HLA type may require an enrichment step prior to analysis. The purpose of enriching for HLA polypeptide is to ensure that the HLA is detectable and that the amino acid sequence information from the HLA may be generated. As used herein, the term "enriching" may comprise increasing the concentration of the HLA polypeptide relative to the concentration of at least one non-HLA polypeptide present in the sample prior to the enrichment, for example the enrichment will reduce the proportion of at least one non- HLA polypeptide which may be present in amounts in the biological sample prior to enrichment which may otherwise mask or confound the presence of HLA sequences. In some embodiments enriching may comprise increasing the concentration of the HLA polypeptide relative to the concentration of the HLA polypeptide in the biological sample prior to enrichment. In certain embodiments, enriching may comprise both increasing the concentration of the HLA polypeptide relative to the concentration of at least one non- HLA polypeptide present in the sample and increasing the concentration of the HLA polypeptide relative to the concentration of the HLA polypeptide in the biological sample prior to enrichment. The sensitivity of a mass spectrometry detection system is <1 fmol (attomole range) of a single peptide (e.g. 2.5 pg of a 2,500 Da peptide). For a typical mass spectrometry analysis, 100 fmol of protein would be trypsin digested and applied to the HPLC column for analysis. The purity of the HLA protein (proportion of HLA-derived peptides to total peptide within the sample) will affect the amount of HLA sequence that is identified in a single analysis. As exemplified herein, less than 5 pmol of total peptide was loaded for mass spectrometry analysis and the proportion of the HLA-derived peptides to total peptides within the sample was approximately 10% w/w. Variables such as the mass range of the instrument (50 - 4000 Da), the efficiency of the peptide separation, the ionization efficiency, and the collision induced dissociation efficiency may affect the quality of the data obtained.

A wide variety of means of enriching for an HLA may be used. Illustrative of such a method for enriching for an HLA polypeptide is a method which comprises an immuno- purification step. Although the genes encoding HLA class I and class II polypeptides are amongst the most polymorphic coding sequences in the human genome, there are relatively constant or invariable regions for each of the HLA class I heavy chains and HLA class II alpha and beta chains which can be targeted by antibodies to selectively capture any HLA class I heavy chain or HLA class II alpha or beta chain. For example, HLA-C polypetides are typically expressed by individuals at lower levels than HLA-A and HLA-B. Accordingly, in order to enhance the detection of HLA-C it may be advantageous to provide a specific immunopurification of HLA-C using an HLA-C- sepecific antibody, in addition to other purification methods.

For the purpose of enriching for HLA polypeptides, anti HLA class II antibodies may recognise conserved epitopes presented on either the α or β chain. Since the α and β chains are normally associated with each other in vivo, immuno-purification of the α- chain of an intact soluble HLA may co-precipitate the β-chain and vice versa. The monoclonal antibodies IVA12 (ATCC number HB-145) and LGII-612.14 are examples of monoclonal antibodies recognising conserved regions on HLA class II molecules. These antibodies were chosen for the purpose of simplifying the purification, since a single mAb can be used to capture all of the known polymorphic HLA polypeptides within the sample. Nevertheless, numerous examples of monoclonal or polyclonal antibodies which bind specifically to individual class II chains (e.g. HLA-DR only) are also commercially available. It will be understood that in certain embodiments, antibodies specific to individual HLA classes may be used, or combinations of two or more antibodies may be used to enrich HLA polypeptides from a biological sample. For example, for use in enriching Class I HLA a combination of an antibody specific to HLA-A with any one or more of antibodies specific to HLA-B, HLA-C, HLA-E, HLA-F, HLA-G, HLA-H, HLA- J, HLA-K, HLA-L, HLA-N, HLA-S, HLA-X and HLA-Z may be used. A combination of an antibody specific to HLA-B with any one or more of antibodies specific to HLA-A, HLA-C, HLA-E, HLA-F, HLA-G, HLA-H, HLA-J, HLA-K, HLA-L, HLA-N, HLA-S, HLA-X and HLA-Z may be used. A combination of an antibody specific to HLA-C with any one or more of antibodies specific to HLA-A, HLA-B, HLA-E, HLA-F, HLA-G, HLA-H, HLA-J, HLA-K, HLA-L, HLA-N, HLA-S, HLA-X and HLA-Z may be used. A combination of an antibody specific to HLA-E with any one or more of antibodies specific to HLA-A, HLA-B, HLA-C, HLA-F, HLA-G, HLA-H, HLA-J, HLA-K, HLA-L₅ HLA-N, HLA-S, HLA-X and HLA-Z may be used. Similarly combinations of any one of HLA-F, HLA-G, HLA-H, HLA-J, HLA-K, HLA-L, HLA-N, HLA-S, HLA-X and HLA- Z with any one or more of HLA-A, HLA-B, HLA-C, HLA-E, HLA-F, HLA-G, HLA-H, HLA-J, HLA-K, HLA-L, HLA-N, HLA-S, HLA-X and HLA-Z may be used. In certain embodiments, the combination is of two antibodies specific to two different HLAs. In certain embodiments, the combination is of three antibodies specific to three different HLAs. In particular embodiments the combination is of an HLA-A specific antibody, an HLA-B specific antibody and an HLA-C specific antibody.

Similarly, when enriching for HLA Class II molecules, a combination of two or more antibodies specific to HLA-DRA, HLA-DRBl to HLA-DRB9, HLA-DQAl, HLA- DQBl, HLA-DQ A2, HLA-DQB2, HLA-DQB3, HLA-DOA, HLA_DOB, HLA_DMA, HLA-DMB, HLA-DPAl, HLA-DP A2, HLA-DP A3, HLA-DPBl, and HLA-DPB2 may be used. In certain embodiments, the combination is of two antibodies specific to two different HLAs. In certain embodiments, the combination is of three antibodies specific to three different HLAs.

In contrast, in DNA based typing methods a multitude of oligonucleotide primers are applied in polymerase chain reaction methods to identify the presence of a particular allele. The binding of primers to their complementary sequence and subsequent amplification products (SSP) or sequencing products (SBT) are evidence of the presence of a particular allele. The necessity to run hundreds of PCRs to identify the alleles present is time consuming and therefore expensive.

The term "antibody" as used herein includes IgG (including IgGl, IgG2, IgG3, and IgG4), IgA (including IgAl and IgA2), IgD, IgE, or IgM, and IgY, and is meant to include whole antibodies, including single-chain whole antibodies, and antigen-binding fragments thereof. Antigen-binding antibody fragments include, but are not limited to, Fab, Fab' and F(ab')₂, Fd, single-chain Fvs (scFv), single-chain antibodies, disulfide- linked Fvs (sdFv) and fragments comprising either a VL or VH domain. The antibodies may be from any animal origin. Antigen-binding antibody fragments, including single- chain antibodies, may comprise the variable region(s) alone or in combination with the entire or partial of the following: hinge region, C_HI, CH2, and C_H3 domains. Also included are any combinations of variable region(s) and hinge region, C_HI, CH2, and C_H3 domains. Antibodies may be monoclonal, polyclonal, chimeric, humanized, and human monoclonal and polyclonal antibodies which specifically bind the HLA polypeptide. A person of skill in the art will recognise that a variety of immunoaffinity techniques are suitable to enrich soluble proteins, such as soluble HLA polypeptides or membrane bound HLA polypeptides which have been proteolytically cleaved from the membrane. These include techniques in which one or more antibodies capable of specifically binding to the soluble protein are immobilised to a fixed or mobile substrate, such as plastic wells or resin, latex or paramagnetic beads, and a solution containing the soluble protein from a biological sample is passed over the antibody coated substrate, allowing the soluble protein to bind to the antibodies. The substrate with the antibody and bound soluble protein is separated from the solution and the optionally the antibody and soluble protein are disassociated, for example by varying the pH and/or the ionic strength and/or ionic composition of the solution bathing the antibodies. Alternatively, immunoprecipitation techniques may be used in which the antibody and soluble protein are combined and allowed to form macromolecular aggregates which are separated from the solution by size exclusion techniques or by centrifugation.

In one embodiment, the antibody is the anti-HLA class I antibody w6/32 which is produced by the hybridoma cell line w6/32 (ECACC Accession No. 84112003). The w6/32 mAb binds to a structural epitope which most likely includes amino acid residue 121 in the α2 domain and residues 3, 44 and 89 in the β₂-microglobulin light chain (Wilfred A. Jefferies 1987; Ladasky, Shum et al. 1999). In one embodiment the antibody is the anti-HLA class II antibody LGII-612.14 which recognises a monomorphic determinant expressed on the beta chain of HLA-DR, -DQ and -DP which is likely to be formed by residues 19-25 of these HLA molecules (Temponi et al 1993). The antibody 9.3F10 (ATCC cat. no. HB-180) binds to monomorphic determinants of HLA-DR and - DP molecules (Laura Cuomo 1990). The antibody L243 (ATCC cat. no. HB-55) binds specifically to HLA-DR molecules only (Topalian, Gonzales et al. 2002). There are also numerous anti-HLA monoclonal antibodies which bind to variable regions of the class I and numerous anti-HLA monoclonal antibodies which bind to variable regions of the class II molecules. Although the present examples have focused on those antibodies that bind to all of the classical class I (w6/32) or all class II (IVAl 2 or LGII-612.14) proteins, it will be understood that the use of a plurality of antibodies is also contemplated. HLA class I polypeptides have been immunoprecipitated with mAb w6/32 (Parham 1979; Schneider, Newman et al. 1982), with mAb HC-10 (Tran, Satumtira et al. 2004) which binds to an epitope in the αl domain of unfolded HLA-B and -C heavy chains (Perosa, Luccarelli et al. 2003), and with mAb TP25.99 (Puppo, Bignardi et al. 1999) which binds to a conformational epitope formed by amino acid residues 194-198 of all HLA class I heavy chains as well as a linear epitope mapped to residues 239-242, 245 and 246 in the α3 domain of some HLA-A and -B allospecificities (Desai, Wang et al. 2000).

HLA-class II proteins have been immunoprecipitated with the monoclonal antibody Q5/13 (Quaranta, Walker et al. 1980) specific for the HLA-DRβ chain; and with the monoclonal antibody L243 (ATCC cat. no. HB-55) and LB3.1 (ATCC cat. no. HB- 298) (Gorga, Horejsi et al. 1987), specific for the HLA-DRβ chain and HLA-DRα chain respectively. The use of multiple antibodies specific to different HLA proteins is contemplated by the methods described herein.

HLA polypeptides may also be enriched from a biological sample by chromatography techniques such as HPLC, see for example Malik and Strominger 2000, the entire contents of which are incorporated herein by reference.

The depletion of high abundance serum proteins may be used to enrich for an HLA polypeptide. Methods for removing abundant serum proteins include dye ligands (for albumin), protein A and G (for γ-globulins) or specific antibodies which bind with high affinity and selectively deplete these species from the sample (Govorukhina, Reijmers et al. 2006). Such strategies would increase the number of HLA-derived peptide sequences identified in a single mass spectrometry analysis.

The degree of enrichment desirable to optimise the resolution of particular HLA sequences from a biological sample will depend on the initial concentration of the HLA sequence in the biological sample, and the concentration and nature of other non-HLA proteins in the sample. Based on densitometry analysis of western blot data provided in the examples described herein, it may be estimated that the immunoaffmity purified HLA which was subsequently analysed by mass spectrometry was enriched at least about 1, 000-fold relative to the initial concentration of HLA present in serum.

To enrich HLA protein within a biological sample, classical protein purification techniques may be used alone or in combination with protein affinity methods. Classical protein separation (purification) techniques are based on; size differences (ultrafiltration, gel filtration, or size exclusion chromatography); charge differences (pi) (anion/cation exchange chromatography, or hydrophobic interaction chromatography); and combinations of size and charge differences (ID or 2D electrophoresis). Immunoaffmity purification options include the use of monoclonal or polyclonal antibodies that specifically bind HLA proteins. Other protein affinity purification options involve the use of proteins that are known to bind HLA, these include; CD8, which binds to the α.3 domain of all HLA class I proteins; CD4 which binds to all HLA class II proteins; autologous T-cell receptors; and antigenic peptides which bind HLA with high affinity (computer modelling algorithms can be used to predict peptide/HLA binding characteristics). Any of these high HLA affinity protein options can be immobilized onto an insoluble solid support to prepare an affinity matrix which may be used to capture the HLA from a liquid biological sample. Appropriate elution conditions will result in the concentration and purification (isolation) of the sample's HLA content.

Mass spectrometry Analysis

In a particular embodiment, the step of obtaining an at least partial amino acid sequence of an HLA polypeptide comprises obtaining an at least partial amino acid sequence of the HLA polypeptide by Mass Spectrometry analysis.

Techniques for generating amino acid sequences of polypeptides by Mass Spectrometry are reviewed in Steen and Mann 2004, the entire contents of which are incorporated herein by reference. In brief, samples comprising or suspected of comprising one or more HLA polypeptides are subjected to enzymatic digestion using one or more of trypsin or other sequence-specific endoproteinases to generate a series of oligopeptides or short polypeptides of at least 6, at least 7, at least 8, at least 9 or at least 10 amino acid residues in length. Mass Spectrometry of peptides is most efficient at obtaining sequence information where the peptides are up to 20 amino acid residues in length.

For electrospray ionization techniques, the peptides are then introduced in a graded manner into the Mass Spectrometer by pre-separating the peptide components into different peaks using an inline HPLC or Multidimensional protein identification technology (MudPIT). It will be understood that other ionization techniques, such as techniques employed in MALDI Mass Spectrometry are also applicable in the present methods.

Once inside the Mass Spectrometer, an analysis of mass to charge ratios of each of the peptides is carried out, and the primary structure of the peptides calculated by fragmenting the peptides and measuring the resultant mass to charge ratios of the fragments.

Mass Spectrometry techniques particularly suited for proteomic analysis and contemplated by the present invention include quadrupole Mass Spectrometry, Time of Flight Mass Spectrometry, Fourier transform ion cyclotron resonance (FTICR-MS) or quadrupole ion trap Mass Spectrometry, or combinations thereof.

It will be understood that any single mass Spectrometry analysis may not provide complete sequence information for any given HLA polypeptide. If it is desirable to increase the proportion of the HLA sequence which is identified, for example if the sequence information which is generated is insufficient to characterise the HLA polypeptide to the level desired, the skilled addressee will recognise that further Mass Spectrometry analyses of the same sample may identify different sequence fragments in different runs, and so multiple Mass Spectrometry analyses may be used to increase the amount of sequence information for any given polypeptide and hence form optional steps of the method of the invention. In addition, the use of alternative endoproteinase enzymes to generate different peptide fragments is expected to allow the elucidation of different regions of sequence of the HLA polypeptide, and hence form optional steps of the methods of the invention. Endoproteinases Lys-C, Asp-N, Arg-C, GIu-C, or chymotrypsin either alone or in combination with a trypsin digestion and/or under harsh solubilizing conditions may allow the formation of alternative fragment peptides. Furthermore, if the sequence being sought after is known, such as in the case of attempting to discriminate between two isoforms or two genetically defined alleles, the enzymes necessary to produce peptides which are useful for specifically distinguishing between the isoforms in the sequencing analysis may be selected based on their respective cleavage patterns.

Results obtained from the Mass Spectometry analysis are interpreted as a series of one or more amino acid sequences of short peptides. These may be matched to one or more peptide sequence databases (with allowances for unrecognised mutations) to obtain a probability score for attribution of the sequences to a specific polypeptide.

It should be noted that, as used in the present specification, the singular forms "a", "an" and "the" include plural aspects unless the context clearly dictates otherwise. Thus, for example, reference to "an HLA polypeptide" includes polypeptides of a single HLA type, as well as polypeptides of two or more HLA types; and so forth.

Kits

Also provided are kits suitable for use in the methods described herein. The kits comprise at least one antibody for use in enriching an HLA polypeptide from a biological sample, to enhance the generation of HLA sequence data. The kits may comprise an antibody which binds a common epitope to all Class I HLA. The kits may comprise an antibody which binds a common epitope to all Class II HLA.

The kits may comprise a plurality of different antibodies, either in isolation or admixed, to allow the enrichment of multiple HLA classes or types, using immuno- enrichment techniques which are available in the art. In certain embodiments the kits comprise at least one antibody capable of being used to enrich each of HLA-A and HLA-

B, and at least one antibody capable of being used to enrich HLA-DRBl .

A variety of HLA specific antibodies are described herein, and are contemplated for use in the kits. In addition, other HLA-specific antibodies not expressly identified herein are contemplated.

A variety of techniques suitable for the immuno-enrichment of HLA are described herein, and the kits of this aspect of the invention are contemplated as being suitable for at least one of these methods.

The kits may also comprise reagents, buffers, and salts suitable for use in methods for the immuno-enrichment of HLA using the antibodies provided.

The kits may also comprise substrates suitable for separating HLA bound antibodies from the biological sample.

The kits may also comprise an endoproteinase suitable for use in sample preparation for mass spectrometry. In the present specification, except where the context requires otherwise due to express language or necessary implication, the word "comprise" or variations such as

"comprises" or "comprising" is used in an inclusive sense, i.e. to specify the presence of

5 the stated features but not to preclude the presence or addition of further features in various embodiments of the invention.

It will be clearly understood that, although a number of prior art publications are referred to herein, this reference does not constitute an admission that any of these documents forms part of the common general knowledge in the art in any country. 0

Examples

Example 1. Preparation of HLA class I specific antibody and immunoaffinity column

Hybridoma cell culture was carried out in two stages. The first stage, aimed ats maximising biomass yield, was performed in commercial medium supplemented with 10% fetal bovine serum. The second stage, aimed at maximising antigen-specific monoclonal antibody production, was performed under serum-free conditions to circumvent the possibility of co-purification of any bovine immunoglobulin present in the serum supplement. o The w6/32 monoclonal antibody (isotype IgG2a) binds to a monomorphic conformational determinant formed between the heavy and light chains of all β-2- microglobulin-associated HLA-A, -B, -C, -E and -G molecules. In order to produce this antibody, hybridoma cell line w6/32 (European Collection of Cell Cultures (ECACC) cat no 84112003) was cultured in commercial medium (Hybridoma- SFM, Invitrogen cat no₅ 12045-076) supplemented with 10% fetal bovine serum (Invitrogen cat no 16000-044) at

37°C in a humidified atmosphere of 5% CO₂ in air. The cultures were seeded at an initial cell density of 10⁵ cells/mL and subcultured when the cell density reached 10⁶ cells/mL.

Cells were harvested in mid-logarithmic growth phase, washed in an equal volume of serum-free medium, resuspended at 4 x 10⁵ cells/mL in Hybridoma-SFM and0 incubated at 37⁰C for 6 days. Cells were removed via centrifugation (15 minutes at 500 x g) and the culture supernatants pooled. Sodium azide (0.02% v/v) was added and the pooled supernatants were passed through a 0.45 μm filter (Millipore cat no HAWP 047 00) to remove culture debris. Aliquots (500 mL) were stored at -20°C until further use. To purify the anti-HLA class 1 monoclonal antibody (w6/32), affinity chromatography was performed using a Trichorn 10/20 column (GE Healthcare cat no 18-1163-13) packed with 1.5 mL Protein- A Sepharose (GE Healthcare cat no 17-1279- 02) attached to a Pharmacia FPLC system. All steps were carried out at a constant flow s rate of 0.5 mL/min. The column was pre-eluted with 15 mL 0.1 M glycine buffer (pH 3.0) and equilibrated with 15 mL PBS/0.02% sodium azide (pH 7.4). Hybridoma culture supernatant (500 mL), buffered by the addition of 100 mL PBS/azide (pH 7.4), was applied and the column was washed with 30 mL (20 column volumes) PBS/azide prior to elution. o The bound monoclonal antibody was eluted with 15 mL 0.1 M glycine buffer

(pH 3.0) and 1.0 mL fractions were collected into tubes containing 200 μL 1.0 M Tris buffer (pH 9.0) to immediately neutralise the pH.

Protein concentration of the eluted mAb was determined spectrophotometrically (Nanodrop ND-1000), with the absorbance of a 1% mAb solution assumed to be 13.7 ats 280 nm (ε = 205,500 M⁴Cm^'1). The purification procedure was assessed via SDS-PAGE (NuPAGE 4-12% Bis-Tris gel, Invitrogen cat no NPO322BOX). Aliquots of sample (13 μL), with added dithiothreitol (Invitrogen cat no NP0004) and lithium dodecyl sulphate buffer (Invitrogen cat no NP0007) were heated to 98°C for 5 minutes, loaded onto gels and run at 100 volts constant for 1 hour in MOPS running buffer (Invitrogen cat0 no NPOOOl). The results of this purification procedure analysed by SDS-PAGE are illustrated in Figure 1. w6/32 hybridoma supernatant (lane 2) was applied to the protein A column and the flow through (lane 3) was discarded. Fractions containing purified monoclonal antibody at 9.9 mg/mL (lane 4), 7.7 mg/mL (lane 5), 5.8 mg/mL (lane 6), 6.5 mg/mL₅ (lane 8), 5.7 mg/mL (lane 9) and 1.7 mg/mL (lane 12) were collected. Dilutions (1 :10) of the w6/32 supernatant (lane 10) and column wash (lane 11) were also included for analysis. Lanes 1 and 7 are molecular weight markers (Invitrogen, cat no LC5925).

Bands corresponding to the murine IgG2a light chain (25 kDa), heavy chain (50 kDa), un-denatured light/heavy chain heterodimers (75 kDa) and un-denatured0 heavy/heavy chain homodimers (100 kDa) were evident in the purified fractions (Figure 1).

Pooled mAb fractions were added to a 10 kDa dialysis membrane (Pierce cat no 68100), and dialysed against 100 volumes of 0.1 M sodium carbonate buffer (pH 8.3) overnight at 4⁰C. Sodium azide (0.05% v/v) was added as preservative and aliquots were stored at 4°C.

The purified anti-HLA class I monoclonal antibody was immobilised on a sepharose immunoaffmity column as follows. 0.6 g cyanogen-bromide activated Sepharose 4B (GE Healthcare cat no 17-0430-01) was resuspended in 10 mL of 1 mM HCl, poured onto a sintered glass filter and washed with 250 mL 1 mM HCl. The drained medium was transferred to a 12 x 75 mm test tube and 2.5 mL w6/32 mAb (6.15 mg/mL) in 0.1 M sodium carbonate buffer (pH 8.3) was added. The mixture was rotated end- over-end for 24 hours at 4⁰C to couple the monoclonal antibody ligand to the Sepharose gel.

From this mixture, 2.0 mL of gel was transferred to an FPLC column (GE Healthcare cat no 18-1163-10) and un-coupled ligand was washed from the column with 10 mL 0.1 M sodium carbonate buffer (pH 8.3) containing 0.5 M sodium chloride. Any remaining active groups were blocked by passing 10 mL 0.1 M Tris-HCl buffer (pH 8.0) through the column, followed by incubation overnight at 4°C.

To remove any weakly bound protein ligand, the column was washed with three cycles of alternating pH. Briefly, 10 mL 0.1 M sodium acetate buffer (pH 4.0) containing 0.5 M NaCl was passed through the column, followed by 10 mL Tris-HCl (pH 8.0) containing 0.5 M NaCl and this cycle was repeated three times. Finally the column was washed with 10 mL PBS containing 0.05% sodium azide and stored at 4°C.

Example 2. Immunoaffinity enrichment of HLA from serum

Sixty ml of peripheral venous blood was collected in silicone coated glass collection tubes without additives (Becton Dickinson cat no 366430) and allowed to clot at 37°C for 1 hour. Samples were inverted several times to release the clot from the sides of the tube, incubated for 4 hours at 4⁰C, the serum (approximately 20 ml) decanted and centrifuged at 500 x g for 15 minutes to remove any remaining cells. The resultant serum was passed through a 0.22 μm syringe filter (Millipore cat no SLGP 033 RS) and stored at -20⁰C until HLA enrichment. It is anticipated that similar results may be obtained using a sample of 10 mL or less of blood.

The column wash and elution buffers used in this example were essentially as described in (Parham 1979), with slight modifications. The immunoaffinity column as prepared in Example 1 was pre-eluted with 10 mL 0.05 M diethylamine (pH 11.5), washed with 10 mL 0.1 M Tris-HCl (pH 7.8) and the serum sample (diluted 1 :5 in Tris- HCl pH 7.8) was applied slowly (0.25 niL/min). The antigen-loaded column was sequentially washed with 10 mL each of 0.1 M Tris-HCl (pH 7.8), 0.5 M NaCl, 1.0 M Tris-HCl (pH 7.8) and 0.1 M Tris-HCl (pH 7.8) to remove any non-specifically bound proteins. The bound HLA proteins were eluted with 10 mL 0.05 M diethylamine (pH

11.5) and 0.9 mL fractions were collected into tubes containing 200 μL neutralisation buffer (1 M Tris-HCl, pH 7.8) to adjust to pH 8.0. The elution profile of the bound soluble HLA proteins is illustrated in Figure 2. The A_280nm of the immunoaffmity column eluate was monitored as a function of the volume of 0.05 M diethylamine applied to the column (flow rate = 0.5 mL/min).

The column was washed with 10 mL PBS containing 0.05% azide and stored at 4⁰C. Protein concentration of the eluted HLA antigen was estimated spectrophotometrically (Nanodrop ND-1000), with the absorbance of a 1% soluble HLA solution assumed to be 10.0 at 280 nm (ε = 56,000 M^"1 cm^"1). Samples eluted from the w6/32 immunoaffmity column were analysed via SDS-

PAGE as described above, transferred electrophoretically to nitrocellulose, and stained with monoclonal antibody HC-10 which primarily binds to β2-microglobulin-free HLA class 1 heavy chains. The mAb HC-10 (IgG2a, a gift of Prof. Soldano Ferrone, Roswell Park Cancer Institute, Buffalo New York) recognizes a determinant expressed on all β-2- microglobulin-free HLA-B and -C heavy chains and on β-2-micro globulin-free HLA- A3, AlO, A28, A29, A30, A31, A32, and A33 heavy chains (Stam, Spits et al. 1986) and was visualised using an HRP-conjugated goat-anti mouse Ig and the peroxidase- tetramethylbenzidine reaction to produce a coloured label.

An image of a Western blot analysis of immunoaffmity enrichment of HLA from two individuals (identified as "AC" and "LO") is provided in Figure 3. Bands corresponding to intact (44 kDa) and truncated (40 kDa and 35-37kDa) soluble HLA class 1 heavy chains were visible (see lanes 5 and 8). High molecular weight bands (70 kDa) present in several early gels (results not shown) may represent previously described HLA homodimers formed by incomplete denaturation and disassociation of the HLA prior to loading the SDS-PAGE gel. The 70 kDa bands were not visible in subsequent western blots in which the protein samples were heated for 15 minutes at 98°C prior to loading the SDS-PAGE gels.

Lanes 1 and 9 are molecular weight markers (10 μL pre-stained standard proteins, Invitrogen, cat no LC5925); Lane 2 was blank; Lane 3 was 13 μL 4% v/v of serum from individual "AC"; Lane 4 was 13 μL of immunoaffmity column flowthrough from a sample from individual "AC"; Lane 5 was 13 μL of soluble HLA concentrate from individual "AC"; Lane 6 was 13 μL of 4% v/v of serum from individual "LO"; Lane 7 was 13 μL of immunoaffinity column flowthrough from a sample from individual "LO"; and Lane 8 was 13 μL of soluble HLA concentrate from individual "LO".

Fractions present in lanes 5 and 8 were expected to contain soluble HLA, based on the UV absorbance profile provided in Figure 2, and this was confirmed by the Western blot analysis.

Example 3. Liquid Chromatography Tandem Mass Spectrometry of HLA Class I

Aliquots corresponding to the samples applied to lanes 5 or 8 of the western blot were each concentrated over a 10 kDa ultrafiltration membrane (Sartorius cat no VS0201) to give 18-20 μL per sample with 0.46 μg/μL and 0.57 μg/μL protein respectively in 0.1 M Tris-HCl (pH 8.0). Sequencing grade trypsin (final concentration 16 ng/μL; Promega cat no V5113) was added to the concentrated HLA samples and incubated overnight at 37°C. Trypsin digested peptides were dried in a heated vacuum centrifuge and resuspended in 20 μL 0.1% v/v heptafluorobutyric acid.

Tryptic digest peptides were separated by nano-LC using an Ultimate 3000 HPLC and autosampler system (Dionex, Amsterdam, Netherlands). Samples (5 μl) were concentrated and desalted onto a micro Cl 8 precolumn (500 μm x 2 mm, Michrom Bioresources, Auburn, CA) with H₂O:CH₃CN (98:2, 0.05 % HFBA) at 20 μl/min. After a 4 min wash the pre-column was switched (Valco 10 port valve, Dionex) into line with a fritless nano column (75μ x ~10cm) containing Cl 8 media (5μ, 200 A Magic, Michrom) manufactured according to Gatlin (Gatlin, Kleemann et al. 1998). Peptides were eluted using a linear gradient of H₂O:CH₃CN (98:2, 0.1 % formic acid) to H₂O:CH₃CN (55:45, 0.1 % formic acid) at 350 nl/min over 30 min. High voltage (1800 V) was applied to low dead volume tee (Upchurch Scientific) and the column tip positioned ~ 0.5 cm from the heated capillary (T=200°C) of a LTQ FT Ultra (Thermo Electron, Bremen, Germany) mass spectrometer. Positive ions were generated by electrospray and the LTQ FT Ultra operated in data dependent acquisition mode (DDA).

A survey scan m/z 350-1750 was acquired in the FT ICR cell (Resolution = 100,000 at m/z 400, with an initial accumulation target value of 1,000,000 ions in the linear ion trap). Up to the 7 most abundant ions (>2500 counts) with charge states of +2 or +3 were sequentially isolated and fragmented within the linear ion trap using collisionally induced dissociation with an activation q = 0.25 and activation time of 30 ms at a target value of 30,000 ions, m/z ratios selected for MS/ MS were dynamically excluded for 30 seconds.

Peak lists were generated using Mascot Daemon/extract_msn (Matrix Science, London, England, Thermo) using the default parameters, and submitted to the database search program Mascot version 2.2.04, Matrix Science (David N. Perkins 1999). Search parameters were: taxonomy Homo sapiens; peptide mass tolerance 8 ppm and fragment ion mass tolerance ± 0.6 Da; Cys-acrylamide, Met-oxidation and Cys- carboxyamidomethylation were specified as variable modifications; enzyme specificity was trypsin with allowance for up to 1 missed cleavage site per peptide and the Swiss-

Prot 9_12_08 database searched.

The combined results of three runs of mass spectrometry and sequence searching analysis are illustrated in Figures 4to 10 and in Tables 1 to 3.

Based on partial peptide sequences identified in the trypsin digest by tandem mass spectrometry, the presence of numerous polypeptides was inferred from the sample enriched for soluble HLA. These polypeptides are listed in Table 1.

Table 1 - Polypeptides consistent with the amino acid sequences in soluble HLA samples identified in 3 separate runs by mass spectrometry from individual "HM"

Protein Hit Protein Description Swiss-Prot ID Protein Mass (D

Score Matched

Run l

1 Complement C3 CO3_HUMAN 1724 187030 60 25.1

2 HLA class I histocompatibility antigen, A-1 alpha chain 1A01_HUMAN 736 40820 24 51.2

3 Complement C4-A CO4A_HUMAN 702 192650 17 10.6

4 HLA class I histocompatibility antigen, A-11 alpha chain 1A11_HUMAN 669 40911 20 47.7

5 Clusterin CLUS_HUMAN 666 52461 25 45.2

6 HL-A class I histocompatibility antigen, B-49 alpha chain 1B49_HUMAN 548 40556 21 40.6

7 Apolipoprotein E APOE_HUMAN 546 36132 19 53

8 HLA class I histocompatibility antigen, B-15 alpha chain 1B15JHUMAN 539 40363 20 37 OJ

9 HLA class I histocompatibility antigen, B-53 alpha chain 1B53JHUMAN 537 40470 18 40.1

10 IgGFc-binding protein FCGBP_HUMAN 514 571690 16 2.9

11 HLA class I histocompatibility antigen, B-41 alpha chain 1B41_HUMAN 449 40514 18 34.3

12 HLA class I histocompatibility antigen, B-39 alpha chain 1B39JHUMAN 448 40303 20 35.9

13 HLA class I histocompatibility antigen, A-68 alpha chain 1A68_HUMAN 408 40883 13 32.9

14 CD5 antigen-like CD5LJHUMAN 299 38063 11 24.8

15 Ig gamma-2 chain C region IGHG2_HUMAN 289 35862 9 31.9

16 Ig gamma-4 chain C region IGHG4_HUMAN 193 35918 8 18.3

17 Prothrombin THRB_HUMAN 129 69992 6 7.7

18 Ig lambda chain C regions LAC_HUMAN 122 11230 3 23.8

19 Ig gamma-3 chain C region IGHG3_HUMAN 118 41260 6 14.6

20 Ig heavy chain V-III region BRO HV305_HUMAN 102 13218 1 15.8

21 Ig mu chain C region IGHM_HUMAN 97 49276 2 5.8

Protein Hit Protein Description Swiss-Prot ID Protein Mass (Da) Queries Coverage {%)

Score Matched

22 Keratin, type I cytoskeletal 9 K1C9_HUMAN 87 62092 1 3.7 23 Ig gamma-1 chain C region IGHG1_HUMAN 86 36083 4 10.3 24 Serum albumin ALBILHUMAN 77 69321 1 2.1 25 Keratin, type I cytoskeletal 10 K1C10JHUMAN 62 59475 2 2.7 26 Plasma serine protease inhibitor IPSP_HUMAN 60 45673 1 2.7 27 Haptoglobin HPT_HUMAN 54 45177 2 4.9 28 Haptoglobin-related protein HPTFLHUMAN 45 38983 2 4.3 29 Complement component C6 CO6_HUMAN 44 104718 1 1.1 30 Ficolin-3 FCN3_HUMAN 42 32882 1 3.3 31 Complement C1q subcomponent subunit B C1QB_HUMAN 41 26442 1 2.8 32 Ig alpha-1 chain C region IGHA1JHUMAN 41 37631 1 2.5 33 Keratin, type Il cytoskeletal 4 K2C4_HUMAN 40 57250 1 1.7 34 Complement component C9 CO9_HUMAN 37 63133 1 1.6 OJ to 35 Complement component C7 CO7JHUMAN 36 93457 1 1.3 36 Apolipoprotein A-I APOA1_HUMAN 34 30759 1 4.1 37 Inner centromere protein INCEJHUMAN 33 105365 1 0.8 38 Disintegrin and metalloproteinase domain-containing protein 5 ADAM5JHUMAN 33 47150 1 1.9 39 Apolipoprotein A-IV APOA4_HUMAN 28 45371 1 2.8 40 Keratin, type Il cytoskeletal 1 K2C1_HUMAN 27 65978 1 1.4 41 Complement C1q subcomponent subunit A C1QA_HUMAN 26 26000 1 4.1 42 Ig lambda chain V-I region HA LV102_HUMAN 24 11889 1 7.1 43 Vitronectin VTNC_HUMAN 24 54271 1 2.1

Run 2

1 Complement C3 CO3_HUMAN 830 187030 39 12.4 2 HLA class I histocompatibility antigen, A-11 alpha chain 1A11_HUMAN 419 40911 13 15.9 3 HLA class I histocompatibility antigen, A-1 alpha chain 1A01 HUMAN 281 40820 8 12.9

Protein Hit Protein Description Swiss-Prot ID Protein Mass (Oa) Queries Coverage (%)

Score Matched

4 Apolipoprotein E APOE_HUMAN 279 36132 10 20.8

5 Clusterin CLUS_HUMAN 278 52461 14 19.6

6 HLA class I histocompatibility antigen, B-8 alpha chain 1B08_HUMAN 243 40306 10 19.3

7 HLA class I histocompatibility antigen, B-40 alpha chain 1B40_HUMAN 214 40480 10 18.8

8 Complement C5 CO5_HUMAN 151 188186 8 3.6

9 Complement C4-A CO4A_HUMAN 121 192650 2 1.4

10 Haptoglobin HPT_HUMAN 98 45177 5 6.9

11 Haptoglobin-related protein HPTFLHUMAN 97 38983 3 8.9

12 HLA class I histocompatibility antigen, B-49 alpha chain 1B49_HUMAN 94 40556 7 12.2

13 Keratin, type Il cytoskeletal 1 K2C1_HUMAN 85 65978 2 1.9

14 Serum albumin ALBLLHUMAN 84 69321 3 3

15 Ig kappa chain C region IGKC_HUMAN 79 11602 1 18.9

16 IgGFc-binding protein FCGBP_HUMAN 78 571690 4 0.6 OJ

17 Apolipoprotein L1 APOL1_HUMAN 70 43947 1 3.8

18 Vitronectin VTNCJHUMAN 60 54271 1 2.5

19 Ficolin-3 FCN3JHUMAN 57 32882 1 3

20 Keratin, type I cytoskeletal 10 K1C1 OJHUMAN 56 59475 1 2

21 Ig gamma-1 chain C region IGHG1_HUMAN 48 36083 3 2.1

22 Ig alpha-1 chain C region IGHA1_HUMAN 39 37631 1 2.5

23 Ig mu chain C region IGHM_HUMAN 39 49276 1 2.7

24 Actin, cytoplasmic 1 ACTB_HUMAN 35 41710 1 2.7

25 EMIUN-2 EMIL2JHUMAN 30 115544 2 0.7

26 Plectin-1 PLEC1_HUMAN 22 531466 1 0.2

Run 3

1 Apolipoprotein A-I APOA1_HUMAN 859 30759 30 56.2 2 HLA class I histocompatibility antigen, A-11 alpha chain 1A11 HUMAN 727 40911 19 31

Protein Hit Protein Description Swiss-Prot ID Protein Mass (Da) Queries Coverage (%)

Score Matched

3 HLA class I histocompatibility antigen, A-1 alpha chain 1A01JHUMAN 691 40820 17 29

4 Complement C3 CO3_HUMAN 676 187030 21 12.3

5 Serum albumin ALBILHUMAN 576 69321 21 19

6 Ig gamma-1 chain C region IGHG1_HUMAN 334 36083 13 30

7 HLA class I histocompatibility antigen, B-15 alpha chain 1B15_HUMAN 310 40363 9 25.4

8 Haptoglobin HPT_HUMAN 273 45177 11 17

9 Ig alpha-1 chain C region IGHA1_HUMAN 257 37631 7 19

10 HLA class I histocompatibility antigen, Cw-4 alpha chain 1C04_HUMAN 255 40969 6 12.8

11 HLA class I histocompatibility antigen, B-45 alpha chain 1B45_HUMAN 252 40389 9 24.3

12 HLA class I histocompatibility antigen, B-49 alpha chain 1B49_HUMAN 252 40556 9 24.3

13 Ig kappa chain C region IGKC_HUMAN 216 11602 4 35.8

14 Ig gamma-2 chain C region IGHG2_HUMAN 202 35862 9 22.4

15 Ficolin-2 FCN2_HUMAN 198 33980 4 14.1 u>

-P-

16 Ig gamma-3 chain C region IGHG3JHUMAN 165 41260 7 12.2

17 Ig mu chain C region IGHM_HUMAN 165 49276 4 9.7

18 Ig lambda chain C regions LAC_HUMAN 150 11230 4 32.4

19 Kininogen-1 KNG1_HUMAN 147 71912 6 7.3

20 Apolipoprotein A-Il APOA2JHUMAN 123 11168 6 21

21 Platelet basic protein CXCL7_HUMAN 120 13885 4 25.8

22 Ig lambda chain V-IV region HiI LV403_HUMAN 119 11510 2 17.8

23 Ig heavy chain V-III region BRO HV305_HUMAN 118 13218 1 15.8

24 Ig kappa chain V-Il region Cum KV201_HUMAN 101 12668 2 11.3

25 Antithrombin-lll ANT3_HUMAN 90 52569 4 3.7

26 Beta-2-microglobulin B2MG_HUMAN 86 13706 3 16.8

27 Clusterin CLUS_HUMAN 79 52461 4 6.2

28 Histidine-rich glycoprotein HRG_HUMAN 77 59541 2 1.7

Protein Hit Protein Description Swiss-Prot ID Protein Mass (Da) Queries Coverage (%)

Score Matched

29 Complement factor H-related protein 1 FHR1JHUMAN 75 37637 2 3.3

30 Complement C1s subcomponent C1S_HUMAN 73 76635 2 1.2

31 Complement C1q subcomponent subunit B C1QB_HUMAN 72 26442 1 6

32 Alpha-1 -antitrypsin A1 ATJ-IUMAN 67 46707 5 5.5

33 Pigment epithelium-derived factor PEDF_HUMAN 53 46313 1 2.4

34 Ig lambda chain V-I region HA LV102_HUMAN 48 11889 2 7.1

35 Apolipoprotein C-I APOC1JHUMAN 41 9326 1 8.4

36 Complement C4-A CO4AJHUMAN 40 192650 3 1

37 Apolipoprotein E APOE_HUMAN 37 36132 2 2.8

38 Arachidonate 12-lipoxygenase, 12R type LX12B_HUMAN 36 80304 2 1.3

39 Ig kappa chain V-I region Lay KV113JHUMAN 36 11827 1 8.3

40 Ficolin-3 FCN3_HUMAN 34 32882 1 3

41 Ig lambda chain V-III region LOI LV302JHUMAN 34 11928 1 7.2 ω

42 Complement factor B CFAB_HUMAN 33 85479 1 1.2

43 28S ribosomal protein S26, mitochondrial RT26_HUMAN 29 24197 1 4.9

44 Protein Shroom3 SHRM3_HUMAN 29 216528 1 0.5

45 CD5 antigen-like CD5L_HUMAN 26 38063 1 1.7

46 Prothrombin THRB_HUMAN 24 69992 3 2.7

47 Centaurin-gamma-3 CENG3JHUMAN 22 94985 1 0.7

48 TRIO and F-actin-binding protein TARA_HUMAN 20 261217 1 0.4

Table 1 shows the proteins identified in three separate MS/MS ions searches using the Mascot version 2.2.04 program, with the significance threshold set to p<0.05 and the number of protein hits set to "auto". The proteins are ranked in order of protein score, as displayed in the Mascot peptide summary report. Amongst the proteins which were identified in Run 1 were eight HLA Class I molecules:

HLA class I histocompatibility antigen, A-I alpha chain

HLA class I histocompatibility antigen, A-I l alpha chain

HLA class I histocompatibility antigen, B-49 alpha chain HLA class I histocompatibility antigen, B-15 alpha chain

HLA class I histocompatibility antigen, B-53 alpha chain

HLA class I histocompatibility antigen, B-41 alpha chain

HLA class I histocompatibility antigen, B-39 alpha chain

HLA class I histocompatibility antigen, A-68 alpha chain Amongst the proteins which were identified in run 2 were five HLA class I molecules:

HLA class I histocompatibility antigen, A-I l alpha chain

HLA class I histocompatibility antigen, A-I alpha chain

HLA class I histocompatibility antigen, B-8 alpha chain HLA class I histocompatibility antigen, B-40 alpha chain

HLA class I histocompatibility antigen, B-49 alpha chain

Amongst the proteins which were identified in run 3 were six HLA class I molecules:

HLA class I histocompatibility antigen, A-I l alpha chain HLA class I histocompatibility antigen, A-I alpha chain

HLA class I histocompatibility antigen, B-15 alpha chain

HLA class I histocompatibility antigen, Cw-4 alpha chain

HLA class I histocompatibility antigen, B-45 alpha chain

HLA class I histocompatibility antigen, B-49 alpha chain Based on the Mascot MS/MS ion search algorithm using the parameters outlined above, these proteins represent the most likely HLA isoform matches against the Swiss-

Prot database which were identified in each run.

Since the samples analysed originated from a single individual "HM", there could only have been a maximum of two HLA-A proteins, two HLA-B proteins and two HLA-Cw proteins present within the sample. Additional HLA isoforms which were attributed were simply the next most likely predictions of HLA-associated matches to the Swiss-Prot database based on the single data set.

A large proportion of the peptides identified within each run would be shared among many different HLA isoforms, and therefore did not represent a complete enough data set to definitively assign HLA types at each of the three loci analysed.

Subsequently, the HLA-associated peptides identified in the three mass spectrometry and database search runs were combined. Examples of the combined HLA-

Al and HLA-AIl identified peptides (which were consistently the two highest matching/most likely HLA isoforms) present in the sample from individual "HM" are shown in tables 2 and 3.

In addition the sequences which weer identied were overlaid on the complete predicted amino acid sequence for HLA-Al and HLA-Al 1 as set out in Figures 4 and 5.

Table 2 HLA A-I Matched peptides identified by mass spectrometry over 3 runs (individual HM)

Start- End Observe Mr (expt) Mr (CaIc) Delta Miss Amino Acid Sequence Ions d Score

31 - 41 658.8399 1315.6652 1315.6673 -0.0021 0 R.YFFTSVSRPGR.G 29

46 - 59 815.416 1628.8174 1628.8199 -0.0024 0 R.FIAVGYVDDTQFVR.F 86

46 - 59 815.42 1628.8254 1628.8199 0.0056 0 R.FIAVGYVDDTQFVR.F 83

46 - 59 815.4216 1628.8286 1628.8199 0.0088 0 R.FIAVGYVDDTQFVR.F 75

60 - 68 484.7216 967.4286 967.4247 0.004 0 R.FDSDAASQK.M 48

60 - 72 749.3361 1496.6576 1496.6565 0.0011 1 R.FDSDAASQKMEPR.A 48

60 - 72 499.8928 1496.6566 1496.6565 0 1 R.FDSDAASQKMEPR.A 39 OJ oo

73 - 89 1067.481 2132.9474 2132.9439 0.0035 0 R.APWIEQEGPEYWDQETR.N 86

73 - 89 1067.482 2132.9494 2132.9439 0.0055 0 R.APWTEQEGPEYWDQETR.N 74

73 - 89 1067.476 2132.9374 2132.9439 -0.0065 0 R.APWIEQEGPEYWDQETR.N 67

93 - 106 770.4033 1538.792 1538.7913 0.0007 1 K.AHSQTDRANLGTLR.G 32

100 - 106 372.7216 743.4286 743.429 -0.0003 0 R.ANLGTLR.G 40

139 - 145 398.6786 795.3426 795.3399 0.0028 0 R.QDAYDGK.D 24

139 - 145 398.6788 795.343 795.3399 0.0032 0 R.QDAYDGK.D 23

146 - 155 611.3103 1220.606 1220.6037 0.0024 0 K.DYIALNEDLR.S 52

156 - 168 697.3438 1392.673 1392.6707 0.0023 0 R.SWTAADMAAQITK.R 109

156 - 168 697.3428 1392.671 1392.6707 0.0003 0 R.SWTAADMAAQITK.R 104

Start- End Observe Mr (expt) Mr (CaIc) Delta Miss Amino Acid Sequence Ions d Score

156-168 697.3444 1392.6742 1392.6707 0.0035 0 R.SWTAADMAAQITK.R 103

156-168 705.3398 1408.665 1408.6656 -0.0006 0 R.SWTAADMAAQITK.R 92

156-168 697.3434 1392.6722 1392.6707 0.0015 0 R.SWTAADMAAQITK.R 91

156-169 783.3912 1564.7678 1564.7668 0.0011 1 R.SWTAADMAAQITKR.K 57

170-180 662.842 1323.6694 1323.6683 0.0011 1 R.KWEAVHAAEQR.R 77

170-180 662.8413 1323.668 1323.6683 -0.0003 1 R.KWEAVHAAEQR.R 61

170-180 442.231 1323.6712 1323.6683 0.0028 1 R.KWEAVHAAEQR.R 53

170-180 442.2299 1323.6679 1323.6683 -0.0005 1 R.KWEAVΗAAEQR.R 44

170-180 442.2302 1323.6688 1323.6683 0.0004 1 R.KWEAVHAAEQR.R 46 w

171-180 598.7955 1195.5764 1195.5734 0.0031 0 K.WEAVHAAEQR.R 62

171-180 598.793 1195.5714 1195.5734 -0.0019 0 K.WEAVHAAEQR.R 39

181-187 446.754 891.4934 891.4926 0.0008 1 R.RVYLEGR.C 21

182-187 368.7034 735.3922 735.3915 0.0007 0 R.VYLEGR.C 24

194-200 440.2387 878.4628 878.461 0.0019 1 R.RYLENGK.E 24

194-200 440.2392 878.4638 878.461 0.0029 1 R.RYLENGK.E 26

195 - 200 362.1877 722.3608 722.3599 0.001 0 R.YLENGK.E 24

195 - 205 675.8537 1349.6928 1349.6939 -0.001 1 R.YLENGKETLQR.T 58

195 - 205 450.9048 1349.6926 1349.6939 -0.0013 1 R.YLENGKETLQR.T 38

206 - 226 1219.086 2436.1574 2436.1604 -0.003 1 R.TDPPKTHMTHHPISDHEATLR.C 23

Start- End Observe Mr (expt) Mr (CaIc) Delta Miss Amino Acid Sequence Ions d Score

211 - 226 941.9511 1881.8876 1881.8904 -0.0028 0 K.THMTHHPISDHEATLR.C 72

211 - 226 949.9484 1897.8822 1897.8853 -0.0031 0 K.THMTHHPISDHEATLR.C 62

211 - 226 628.303 1881.8872 1881.8904 -0.0032 0 K.THMTHHPISDHEATLR.C 53

211 - 226 628.3025 1881.8857 1881.8904 -0.0047 0 K.THMTHHPISDHEATLR.C 49

211 - 226 633.6354 1897.8844 1897.8853 -0.001 0 K.THMTHHPISDHEATLR.C 33

244 - 267 879.7345 2636.1817 2636.1838 -0.0022 0 R.DGEDQTQDTELVETRP AGDGTF 57

QK.W

244 - 267 1319.097 2636.1794 2636.1838 -0.0044 0 R.DGEDQTQDTELVETRP AGDGTF 50

QK.W 4^ O

268 - 280 714.3661 1426.7176 1426.7205 -0.0028 0 K.WAAVWPSGEEQR.Y 59

268 - 280 714.367 1426.7194 1426.7205 -0.001 0 K.WAAVWPSGEEQR.Y 53

268 - 280 714.3687 1426.7228 1426.7205 0.0024 0 K.WAAVWPSGEEQR.Y 47

281 - 297 1025.031 2048.0474 2048.0626 -0.0151 0 R.YTCHVQHEGLPKPLTLR.W 58

281 - 297 683.6942 2048.0608 2048.0626 -0.0018 0 R.YTCHVQHEGLPKPLTLR.W 39

281 - 297 513.0224 2048.0605 2048.0626 -0.0021 0 R.YTCHVQHEGLPKPLTLR.W 36

For Tables 2 and 3, the amino acid sequence which is recognised by the Mass Spectometry analysis lies between the periods. The periods represent the position of the cleavage by the typsin. The single amino acid residue on each side of the recognised sequence is part of the greater sequence which is required for typsin digestion.

Table 3 HLA A-Il Matched peptides identified by mass spectrometry over 3 runs (individual HM)

Start- Observed Mr (expt) Mr (CaIc) Delta Miss Amino Acid Sequence Ions

End Score

31 - 41 666.8375 1331.6604 1331.6622 -0.0018 0 R.YFYTSVSRPGR.G 26

46 - 59 815.416 1628.8174 1628.8199 -0.0024 0 R.FIAVGYVDDTQFVR.F 86

46 - 59 815.42 1628.8254 1628.8199 0.0056 0 R.FIAVGYVDDTQFVR.F 83

46 - 59 815.4216 1628.8286 1628.8199 0.0088 0 R.FIAVGYVDDTQFVR.F 75

60 - 68 498.7239 995.4332 995.4308 0.0024 0 R.FDSDAASQR.M 65

60 - 68 498.7241 995.4336 995.4308 0.0028 0 R.FDSDAASQR.M 65

60 - 68 498.7232 995.4318 995.4308 0.001 0 R.FDSDAASQR.M 59

73 - 89 1067.481 2132.9474 2132.9439 0.0035 0 R.APWIEQEGPEYWDQETR.N 86

73 - 89 1067.482 2132.9494 2132.9439 0.0055 0 R.APWEEQEGPEYWDQETR.N 74

73 - 89 1067.476 2132.9374 2132.9439 -0.0065 0 R.APWIEQEGPEYWDQETR.N 67

93 - 106 780.4101 1558.8056 1558.8063 -0.0007 1 K.AQSQTDRVDLGTLR.G 35

100 - 106 387.23 772.4454 772.4443 0.0011 0 R.VDLGTLR.G 34

100 - 106 387.2301 772.4456 772.4443 0.0013 0 R.VDLGTLR.G 31

100 - 106 387.23 772.4454 772.4443 0.0011 0 R.VDLGTLR.G 31

100 - 106 387.2301 772.4456 772.4443 0.0013 0 R.VDLGTLR.G 27

139 - 145 398.6786 795.3426 795.3399 0.0028 0 R.QDAYDGK.D 24

139 - 145 398.6788 795.343 795.3399 0.0032 0 R.QDAYDGK.D 23

Start- Observed Mr (expt) Mr (CaIc) Delta Miss Amino Acid Sequence Ions

End Score

146-155 611.3103 1220.606 1220.6037 0.0024 0 K.DYIALNEDLR.S 52

156-168 697.3438 1392.673 1392.6707 0.0023 0 R.SWTAADMAAQITK.R 109

156 - 168 697.3428 1392.671 1392.6707 0.0003 0 R.SWTAADMAAQITK.R 104

156-168 697.3444 1392.6742 1392.6707 0.0035 0 R.SWTAADMAAQITK.R 103

156-168 705.3398 1408.665 1408.6656 -0.0006 0 R.SWTAADMAAQITK.R 92

156-168 697.3434 1392.6722 1392.6707 0.0015 0 R.SWTAADMAAQITK.R 91

156 - 169 783.3912 1564.7678 1564.7668 0.0011 1 R.SWTAADMAAQITKR.K 57

170-181 712.8544 1423.6942 1423.6956 -0.0014 1 R.KWEAAHAAEQQR.A 74

170-181 712.8566 1423.6986 1423.6956 0.003 1 R.KWEAAHAAEQQR.A 62

170-181 712.8568 1423.699 1423.6956 0.0034 1 R.KWEAAHAAEQQR.A 58

170-181 475.572 1423.6942 1423.6956 -0.0014 1 R.KWEAAHAAEQQR.A 55

170-181 475.5732 1423.6978 1423.6956 0.0022 1 R.KWEAAHAAEQQR.A 50

170-181 475.5731 1423.6975 1423.6956 0.0019 1 R.KWEAAHAAEQQR.A 30

171-181 648.8071 1295.5996 1295.6007 -0.001 0 K.WEAAHAAEQQR.A 61

171-181 648.8098 1295.605 1295.6007 0.0044 0 K.WEAAHAAEQQR.A 56

182-187 354.6872 707.3598 707.3602 -0.0004 0 R.AYLEGR.C 27

194-200 440.2392 878.4638 878.461 0.0029 1 R.RYLENGK.E 26

194-200 440.2387 878.4628 878.461 0.0019 1 R.RYLENGK.E 24

195-200 362.1877 722.3608 722.3599 0.001 0 R.YLENGK.E 24

Start- Observed Mr(expt) Mr(CaIc) Delta Miss Amino Acid Sequence Ions End Score

195-205 675.8537 1349.6928 1349.6939 -0.001 1 R.YLENGKETLQR.T 58

195-205 450.9048 1349.6926 1349.6939 -0.0013 1 R.YLENGKETLQR.T 38

206-226 1219.086 2436.1574 2436.1604 -0.003 1 R.TDPPKTHMTHHPISDHEATLR.C 23

211-226 941.9511 1881.8876 1881.8904 -0.0028 0 K.THMTHHPISDHEATLR.C 72

211-226 949.9484 1897.8822 1897.8853 -0.0031 0 K.THMTHHPISDHEATLR.C 62

211-226 628.303 1881.8872 1881.8904 -0.0032 0 K.THMTHHPISDHEATLR.C 53

211-226 628.3025 1881.8857 1881.8904 -0.0047 0 K.THMTHHPISDHEATLR.C 49

211-226 633.6354 1897.8844 1897.8853 -0.001 0 K.THMTHHPISDHEATLR.C 33

244-267 879.7345 2636.1817 2636.1838 -0.0022 0 R.DGEDQTQDTELVETRP AGDGTF 57 t ιΛ

QK.W

244-267 1319.097 2636.1794 2636.1838 -0.0044 0 R.DGEDQTQDTELVETRPAGDGTF 50

QK.W

268-280 714.3661 1426.7176 1426.7205 -0.0028 0 K.WAAWVPSGEEQR.Y 59

268-280 714.367 1426.7194 1426.7205 -0.001 0 K.WAAVWPSGEEQR.Y 53

268-280 714.3687 1426.7228 1426.7205 0.0024 0 K.WAAVWPSGEEQR.Y 47

281-297 1025.031 2048.0474 2048.0626 -0.0151 0 R.YTCHVQHEGLPKPLTLR.W 58

281-297 513.0224 2048.0605 2048.0626 -0.0021 0 R.YTCHVQHEGLPKPLTLR.W 36

281-297 683.6942 2048.0608 2048.0626 -0.0018 0 R.YTCHVQHEGLPKPLTLR.W 39

The peptide sequences identified for the HLA-C species did not clearly identify the exact allele present. The three sequences identified by Mass Spectrometry occur in a number of HLA-C alleles, so the assignment based on the data currently available was not able to resolve which of the HLA-C loci were expressed by the individual.

The second stage fragmentation performed during the tandem mass spectrometry analysis selects the most abundant peptides from the first stage of mass spectrometry and then fragments the peptide to produce fragment ions. Identification of the mass/charge ratios of these ions and data collation allows a peptide sequence to be obtained. The present analysis is based on probability scoring, and so increasing the enrichment of the sample or the individual isolation of HLA allospecific sequences may improve the amount of sequence data generated. For example, Western blotting to identify the location of HLA heavy chains separated in one-dimensional SDS polyacrylamide gel electrophoresis will allow the appropriate (HLA containing) bands to be excised from identically-electrophoresed gels and these bands to be analysed together or individually via tandem mass spectrometry. Similarly, the generation of different specific enzymic fragments of enriched HLA through the use of one or more other endoproteinases will allow different peptide fragments of HLA to be analysed and identified, and the combination of results from different enzyme digestions will produce greater regions of sequence fragment overlap and thus greater resolution of the HLA sequence.

The direct sequencing of expressed HLA proteins negates the identification of null alleles that is crucial in DNA-based typing methods. Transcriptional (such as alternate mRNA splice variants) and translational (glycosylation) variations can be identified using the techniques described herein, which are not resolved using conventional (genotyping) techniques. Heterozygosity causes problems for SBT, since sequencing requires a DNA template with only one sequence.

Because the database search that was performed searched for amino acid sequences (and not for genotypes), the results are displayed in terms of serologically defined antigens. If 100% sequence coverage is obtained, for example by employing multiple MS/MS runs or by using different enzyme digests, this would be the protein- level equivalent of genotyping patients to four digits. That is, differences in the fifth and following digits do not affect the primary amino acid sequence. However, the current naming convention requires that protein-level typing be limited to 2 digit nomenclature (since serology is the only option currently available for protein-based tissue typing). It is envisaged that this technology may result in a new naming conventions based on amino acid sequence.

The limited coverage of the HLA C sequence by the amino acid sequences identified by the mass spectrometry analysis meant that it was not possible with the data produced to determine which one or two of the three HLA C alleles identified was expressed by the individual. Repeated analysis of the same sample is expected to produce additional sequence coverage and should assist in resolving the identification of which of these HLA is actually expressed. In addition, an alternative digest of the purified HLA sample by other proteases, such as one or more of the endoproteinases Lys-C, Asp-N, GIu-C or chymotrypsin, either alone or in combination with a trypsin digestion and/or under harsh solubilizing conditions should allow the formation of alternative fragment peptides which may be sequenced using the techniques described above, and this is expected to provide additional sequence coverage of the HLA polypeptides. Removal of abundant serum proteins via affinity chromatography using dye ligands, protein A or G, and monoclonal or polyclonal antibodies directed towards these abundant proteins should provide additional HLA sequence, since the mass spectrometry analysis provides sequence data for the most abundant peptides within the sample. The removal of these high concentration serum proteins may be achieved by procurement of the specific reagents themselves, or by applying the sample to any one of a number of commercial kits available for this purpose; such as Sigma cat no PROT20-1KT, Calbiochem cat no 122642, Pierce cat no 89875 and 89876, Agilent Technologies cat no 5188-6560, or Bio- Rad cat no 163-3000.

Example 4. Preparation of HLA class II specific antibody, purification of soluble HLA class II, and mass spectrometry analysis The hybridoma cell line IVAl 2 (ATCC HB-145) is cultured in commercial medium (Hybridoma-SFM, Invitrogen cat no 12045-076) supplemented with fetal bovine serum to produce monoclonal antibody that specifically binds to all HLA-DP, -DQ and DR allotypes (Shaw, Ziegler et al. 1985). Monoclonal antibody is separated from the 0.22 μm filtered culture supernatant by high affinity binding to protein A or protein G immobilised onto a stationary phase such as agarose at neutral pH. The bound mAb is eluted at low pH and neutralized immediately to prevent denaturation and loss of HLA- binding activity. Sodium azide is added as a bacteriostatic agent to preserve the mAb during storage.

The total protein concentration is estimated spectrophotometrically (A₂₈o_nm) or in a biuret reaction (Pierce BCA protein assay reagent cat no 23225), and the purity of the mAb is determined electrophoretically by denaturing SDS-PAGE under reducing conditions. Concentrated, purified IVAl 2 mAb is covalently bound to cyanogen bromide activated agarose (GE Healthcare cat no 17-0430-01 or Sigma cat. nos. C9210, C9142, C9267) under mildly alkaline conditions. Unbound mAb is washed from the gel, and unreacted groups are blocked by the addition of a primary amine buffer such as Tris, ethanolamine or glycine. Cycles of alternating pH (approximately pH4-pH8.5, depending on the expected usage range of the affinity gel) are used to remove any weakly bound mAb.

The gel is transferred to a glass column (dimensions dependant on the required resolution of the separation, e.g. GE Healthcare cat no 18-1163-10) and serum samples applied at mildly alkaline pH (7.5-8.5) to capture HLA class II polypeptides. Non- specifically bound proteins are washed from the column with mildly alkaline buffers of increasing ionic strength to disrupt hydrophobic bonds which occur commonly in concentrated protein mixtures. HLA polypeptides are eluted at high pH (~pH 11.5) to maintain their biological activity. Elution at low pH or with chaotropic agents tends to disrupt HLA tertiary structure and may be unsuitable for certain downstream applications (such as western blotting if necessary). The protein concentration of the eluate is generally monitored by passing through a UV absorbance detector prior to collection. The purified HLA fraction is assayed for the presence of HLA polypeptides via SDS-PAGE and western blotting using an alternative HLA class II specific monoclonal antibody, preferably a mAb capable of binding denatured HLA polypeptides (or native PAGE may be necessary).

The purified HLA polypeptides are digested with trypsin and separated by strong cation exchange and reversed-phase HPLC. Tryptic peptides bound to the Cl 8 column are sequentially eluted using a gradient concentration of solvent in water (typically acetonitrile is used as the solvent). A voltage is applied to the flow and positive ions are generated via electrospray and analysed in the first stage of mass spectrometry. The most abundant ions are sequentially isolated and fragmented in a linear ion trap via collisionally induced dissociation. The mass-to-charge ratios of the tryptic peptides and the mass-to-charge ratios of the fragment ions produced by CID are used to determine the peptide sequence. These peptide sequences are collectively compared and aligned against a database of known protein sequences (most commonly Swiss-Prot or NCBI) to identify the proteins most likely to be within the sample.

Example 5. Comparison of Amino Acid Sequence data with Sequence Based Typing

Four individuals (identified as "HM", "LO", "MC" and "AC") were typed at the HLA-A, -B and -Cw loci using bi-directional sequence based typing (SBT) across exons 2, 3 and 4 at each locus. The assigned typing results are shown in the following table.

Table 4 HLA loci ex ressed b four individuals as determined by SBT

Alleles identified at the same locus are separated by a comma. Alleles separated by a "/" indicated an ambiguous typing result at the corresponding locus. For example, HLA-Cw 0701/06/18 indicated that the SBT results identified the individual as possessing either HLA-Cw*0701, HLA-Cw*0706 or HLA-Cw*0718 allele at this locus.

The HLA expression of the same four individuals was also determined using the methods described in Examples 2 and 3, and the results compared. The results of this analysis are set out in Figures 4 to 21. The sequence based typing (SBT) approach represents the current "gold standard" for HLA typing and is employed only in cases where a very stringent matching between donor-recipient pairs is required (such as for unrelated-donor hematopoietic stem cell transplantation). Therefore, the alleles identified in the SBT analyses were extracted from the IMGT/HLA database < ftp://ftp.ebi.ac.uk/pub/databases/imgt/mhc/hla/> and used as template sequences to align the mass spectrometry peptide sequences against. The consensus sequences used in the alignments were chosen based on bi-directional DNA sequence based typing across exons 2, 3 and 4 of the HLA-A, -B and -C loci. Figures 4 to 21 represent the manual alignment of all of the HLA-derived peptides identified in several mass spectrometry analyses to the predicted amino acid sequences for alleles identified by the SBT analyses.

For each consensus sequence, several polymorphisms are known to exist. For example, the following alleles of HLA-Al are known to exist: A*01010101, A*01010102N, A*010102, A*010103, A*010104, A*010105, A*0102, A*0103, A*0104N, A*0106, A*0107, A*0108, A*0109, A*0110, A*0111N, A*0112, A*0113, A*0114. A*0115N, A*0116N, A*0117, A*0118N, A*0119, A*0120, A*0121, A*0122N, A*0123, A*0124, A*0125, A*0126, A*0127N, A*0128, A*0129, A*0130, A*0131N, A*0132 and A*O133. The sequence shown in the figure is that of A*0101.

In run 1 for individual "HM", the top three HLA proteins identified (HLA-Al, - All and -B49) were in agreement with the SBT results. Neither HLA-B8 nor either of the HLA-Cw proteins were identified as being among the most likely HLA proteins within the sample analysed based on this single data set.

In run 2 for individual "HM", four HLA proteins (HLA-AI l, -Al, -B8 and - B49) identified were in agreement with the SBT results. Neither of the HLA-Cw proteins were identified as being among the most likely HLA proteins within the sample analysed based on this single data set. The single remaining HLA protein identified in this run, HLA-B40, achieved higher protein score than HLA-B49. That is, based on the peptides identified within this sample, the likelihood of HLA-B40 being present within the sample was considered greater than the probability of HLA-B49 being present. In run 3 for individual "HM", the two highest scoring HLA proteins (HLA-Al 1 and -Al) identified were in agreement with the SBT results.

An example of the ability of an amino acid analysis to discriminate between potential ambiguities associated with heterozygosity at HLA class I loci is demonstrated by a comparison between figures 11 and 12. These two closely related alleles differ only at positions 33 and 67 in the amino acid sequence. Spanning the difference at position 33, the technique described herein identified the tryptic peptide R. YFFTS VSRPGR. G unique to HLA-A*0201 and the tryptic peptide R.YFYTSVSRPGR.G unique to HLA-A*0205. Furthermore, spanning the difference at position 67 the techniques described herein identified a single tryptic peptide R.FDSDAASQR.M unique to the HLA-A*0201 sequence.

In order to increase the resolution of the techniques described, multiple endoproteinases which produce different cleavage patterns may be employed to produce regions of overlapping sequence within the HLA molecules. These overlapping sequences may be assembled into multiple alignments and the entire amino acid sequences of the polymorphic HLA proteins can be accurately reconstructed by de novo interpretation. As multiple sequences spanning regions of previously un-characterised polymorphisms may be obtained, sufficient confidence in the integrity of the sequence data would permit the formation of new database entries. This method of data collation is similar to the "shotgun sequencing" approach employed to sequence the human genome. Single endoproteinase digests produce peptide sequences which are mostly non- overlapping rely heavily upon previously identified sequence polymorphisms.

The software which was used in the above examples to attribute amino acid sequence fragments to particular HLA types was not optimised for HLA identification. The peptide summary which is produced by the Mass Spectometry software "assigns" peptides to the highest matched proteins so that the data collapses into a small set of protein hits. This type of report may be misleading, since identified peptides may have been shared among 2 or more of the 6 possible HLA proteins present within the sample. Proteins that matched the same set or a sub-set of peptides were grouped by the software into a single "hit", with the intention of providing a concise report of the search results. A proportion of the tryptic peptides identified in the above examples were common to several different HLA proteins. Optimised sequence searching will allow better recognition of sequence ambiguities, by factoring in information which is not provided in the Mass Spectometry sequence libraries on sequence overlap and sequence uniqueness between HLA types.

Many of the tryptic peptides present within the samples analysed align to conserved regions of the HLA alpha chain, so this scoring strategy relies heavily upon the identification of peptides spanning regions of sequence unique to each HLA protein within the sample. Consequently, particular combinations of tryptic peptides may cover a greater percentage of a closely related protein and hence be scored more highly based on the Mascot search algorithm. The positioning of lysine and arginine residues within the sequence of particular HLA proteins may be in favour of producing greater or lesser proportions of tryptic peptides of the appropriate length for mass spectrometric identification. Greater coverage of any single sequence may be achieved, however, through the use of combining data from a variety of endoproteinase digestions, and/or from the use of digestions using two endoproteinases.

SBT is typically performed only across exons 2 and 3. Exon 4 is usually sequenced in subjects only where additional sequence is required to resolve any remaining ambiguities arising from an exon 2/3 analysis.

Ambiguities which may arise from this form of screening are illustrated in Figures 22 and 23. Figure 22, for example provides the known sequence alignment of HLA-B 8 and HLA-B49, and demonstrates that the amino acid sequence of the expressed HLA-B8 and HLA-B49 proteins should be identical in the region spanning positions 207 to 298.

Figure 23 presents the results of SBT sequencing of an individual expressing HLA-B8 and HLA-B49. This Figure shows that the results of the SBT DNA sequencing reactions were ambiguous at numerous positions, some of which would have resulted in alterations within the predicted expressed protein sequence. For example, the Y (T or C) at position 18 of the DNA sequence which was detected in both the forward and reverse sequencing reactions would not result in an amino acid change (codons CAC and CAT both encode for the amino acid histidine), demonstrating ambiguity at the DNA level. This may represent heterogeneity in the DNA sequences of the alleles resulting in a synonymous amino acid substitution. Sequencing ambiguity which results in a synonymous amino acid substitution such as this would not alter the expressed protein sequence and hence would be unlikely to have any clinical impact.

Where there is no consensus between the DNA sequences obtained from the forward and reverse sequencing reactions at a particular position in the DNA sequence, there is the potential for the identification of an incorrect allele as a consequence of this ambiguity. An example of this is at position 55 of the DNA sequence where both C and M were identified within the sequence data. The M (A or C) or C would result in the triplet codons ATG or CTG which encode the amino acids methionine and leucine respectively. For both HLA~B*0801 and HLA-B*4901 the amino acid at this position in the sequence should have been a leucine (as shown in Figure 22). That is, the correct nucleotide at position 55 in the exon 4 sequence should have been identified as C in both the forward and reverse sequencing reactions. Sequencing ambiguity which results in a non-synonymous amino acid substitution would alter the expressed protein sequence and hence, depending on the position within the expressed protein, could dramatically affect the clinical outcome of a donor-recipient pairing.

The ambiguity inherent in the SBT analysis may contribute to differences which were recognised between the amino acid sequence identified by the present methods and the predicted sequence identified by SBT analysis.

References

Aultman, D., I. Adamashvili, et al. (1999). "Soluble HLA in human body fluids." Human Immunology 60(3): 239-44.

David N. Perkins, D. J. C. P. D. M. C. J. S. C. (1999). "Probability-based protein identification by searching sequence databases using mass spectrometry data." Electrophoresis 20(18): 3551-3567.

Desai, S. A., X. Wang, et al. (2000). "Structural Relatedness of Distinct Determinants Recognized by Monoclonal Antibody TP25.99 on β2-Microglobulin-Associated and β2- Microglobulin-Free HLA Class I Heavy Chains." J Immunol 165(6): 3275-3283.

Ellis SA, Bontrop RE, Antczak DF, Ballingall K, Davies CJ, Kaufman J, Kennedy LJ, Robinson J, Smith DM, Stear MJ, Stet RJM, Waller MJ, Walter L and Marsh SGE. (2006) ISAG/IUIS-VIC Comparative MHC Nomenclature Committee report 2005. Immunogenetics, 57: 953-958

Gatlin, C. L., G. R. Kleemann, et al. (1998). "Protein Identification at the Low Femtomole Level from Silver-Stained Gels Using a New Fritless Electrospray Interface for Liquid Chromatography-Microspray and Nanospray Mass Spectrometry." Analytical Biochemistry 263(1): 93-101.

Ghio, M., P. Contini, et al (1999). "Soluble HLA Class I, HLA Class II, and Fas Ligand in Blood Components: A Possible Key to Explain the Immunomodulatory Effects of Allogeneic Blood Transfusions." Blood 93(5): 1770-1777

Gorga, J. C, V. Horejsi, et al. (1987). "Purification and characterization of class II histocompatibility antigens from a homozygous human B cell line." Journal of Biological Chemistry 262(33): 16087-94.

Govorukhina, N. I., T. H. Reijmers, et al. (2006). "Analysis of human serum by liquid chromatography-mass spectrometry: Improved sample preparation and data analysis." Journal of Chromatography A 1120(1-2): 142-150. Ladasky, J. J., B. P. Shum, et al. (1999). "Residue 3 of β2-microglobulin affects binding of class I MHC molecules by the W6/32 antibody." Immunogenetics 49(4): 312-320.

5 Laura Cuomo, P. T. F. W. G. W. G. K. M. G. M. (1990). "Expression of the Epstein-Barr virus (EB V)-encoded membrane antigen (LMP) increases the stimulatory capacity of EBV-negative B lymphoma lines in allogeneic mixed lymphocyte cultures." European Journal of Immunology 20(10): 2293-2299. o Malik, P. and J. L. Strominger (2000). "Perfusion chromatography for very rapid purification of class I and II MHC proteins." Journal of Immunological Methods 234(1- 2): 83-8.

Marsh SG, ED Albert, WF Bodmer, RE Bontrop, B Dupont, HA Erlich, DE Geraghty, JAs Hansen, CK Hurley, B Mach, WR Mayr, P Parham, EW Petersdorf, T Sasazuki, GM Schreuder, JL Strominger, A Svejgaard, PI Terasaki, and J Trowsdale, (2005) "Nomenclature for factors of the HLA system, 2004". Tissue Antigens; 65(4): 301-69

McDonald, J. C. and I. Adamashvili (1998). "Soluble HLA:: A review of the literature."0 Human Immunology 59(7): 387-403.

McDonald, J. C, F. B. Gelder, et al. (1992). "HLA in human serum—quantitation of class I by enzyme immunoassay. [see comment]." Transplantation 53(2): 445-9. S Melchiorri, L., A. Martini, et al. (2002). "Human leukocyte antigen-A, -B, -C and -DR alleles and soluble human leukocyte antigen class I serum level in Meniere's disease." Acta Oto-Laryngologica Supplement(548): 26-9.

Munoz-Fernandez, S., J. Martin, et al. (2001). "Soluble HLA class I antigens in serum0 and synovial fluid from patients with rheumatoid arthritis and other arthropathies." Rheumatology 40(12): 1365-1369. Novikov, V. V., N. I. Egorova, et al. (2007). "Serum levels of soluble HLA and IL-2R molecules in patients with urogenital chlamydia infection." Adv Exp Med Biol 601 : 285- 9.

Pappin, D. J., P. Hojrup, et al. (1993). "Rapid identification of proteins by peptide-mass fingerprinting." Curr Biol 3(6): 327-32.

Parham, P. (1979). "Purification of immunologically active HLA-A and -B antigens by a series of monoclonal antibody columns." J. Biol. Chem. 254(18): 8709-8712.

Perosa, F., G. Luccarelli, et al. (2003). "(Perosa, Luccarelli et al.) "Beta2-Microglobulin- Free HLA Class I Heavy Chain Epitope Mimicry by Monoclonal Antibody HC-10- Specific Peptide." J Immunol 171(4): 1918-1926.

Puppo, F., D. Bignardi, et al. (1999). "Beta2-micro-free HLA class I heavy chain levels in sera of healthy individuals. Lack of association with beta2 -micro-associated HLA class I heavy chain levels and HLA phenotype." Tissue Antigens 53(3): 253-62.

Quaranta, V., L. E. Walker, et al. (1980). "Purification of immunologically functional subsets of human la-like antigens on a monoclonal antibody (Q5/13) immunoadsorbent." J Immunol 125(4): 1421-1425.

Perosa, F., G. Luccarelli, et al. (2003). "Beta-2-Microglobulin-Free HLA Class I Heavy Chain Epitope Mimicry by Monoclonal Antibody HC-10- Specific Peptide." J Immunol 171(4): 1918-1926.

Robinson, J., M. J. Waller, et al. (2003). "IMGT/HLA and IMGT/MHC: sequence databases for the study of the major histocompatibility complex." Nucl. Acids Res. 31(1): 311-314.

Robinson J, Waller MJ, Stoehr P, Marsh SGE IPD - the Immuno Polymorphism Database. Nucleic Acids Research (2005), 331:D523-526 Roe, S (editor). "Protein Purification Applications: A Practical Approach" 2nd edition, Oxford University Press

Schneider, C, R. A. Newman, et al. (1982). "A one-step purification of membrane proteins using a high efficiency immunomatrix." J. Biol. Chem. 257(18): 10766-10769.

Shaw, S., A. Ziegler, et al. (1985). "Specificity of monoclonal antibodies directed against human and murine class II histocompatibility antigens as analyzed by binding to HLA- deletion mutant cell lines." Hum Immunol 12(4): 191-211.

Shevchenko, A., I. Chernushevic, et al. (2002). ""De novo" sequencing of peptides recovered from in- gel digested proteins by nanoelectrospray tandem mass spectrometry." Molecular Biotechnology 20(1): 107-18.

Shimura, T., S. Tsutsumi, et al. (2001). "Clinical significance of soluble form of HLA class I molecule in Japanese patients with pancreatic cancer." Human Immunology 62(6): 615-619.

Stam, N. J., H. Spits, et al. (1986). "Monoclonal antibodies raised against denatured HLA-B locus heavy chains permit biochemical characterization of certain HLA-C locus products." J Immunol 137(7): 2299-2306.

Steen H and Mann M (2004) "The ABC'S (and XYZ' S) of peptide sequencing". Nature Reviews Molecular Cell Biology 5:699-711.

Temponi M, Kekish U, Hamby CV, Nielsen H, Marboe CC₅ Ferrone S J, (1993) "Characterization of anti-HLA class II monoclonal antibody LGII-612.14 reacting with formalin fixed tissues." Immunol Methods. 161:239-56.

Topalian, S. L., M. I. Gonzales, et al. (2002). "Revelation of a Cryptic Major Histocompatibility Complex Class Il-restricted Tumor Epitope in a Novel RNA- processing Enzyme." Cancer Res 62(19): 5505-5509. Tran, T. M., N. Satumtira, et al. (2004). "HLA-B27 in Transgenic Rats Forms Disulfϊde- Linked Heavy Chain Oligomers and Multimers That Bind to the Chaperone BiP." J Immunol 172(8): 5110-5119.

Toussirot, E., P. Saas, et al. (2006). "Decreased levels of serum soluble HLA class I antigens in HLA-B27 positive spondyloarthropathies." Ann Rheum Dis 65(2): 279-280.

Wilfred A. Jefferies, G. G. M. (1987). "Expression of the W6/32 HLA epitope by cells of rat, mouse, human and other species: critical dependence on the interaction of specific MHC heavy chains with human or bovine β₂-microglobulin." European Journal of Immunology 17(9): 1257-1263.

Claims

Claims

1. A method of characterising Human Leukocyte Antigen (HLA) expression of a subject, comprising a) enriching a biological sample from the subject for an HLA polypeptide from the subject, b) obtaining an at least partial amino acid sequence of the HLA polypeptide from the subject, and c) comparing the at least partial amino acid sequence of the HLA polypeptide from the subject to a library of known amino acid sequences of HLA polypeptides and assigning the at least partial amino acid sequence of the HLA polypeptide from the subject to one or more known HLA polypeptides.

2. The method according to claim 1, wherein the HLA polypeptide which is enriched in the biological sample from the subject comprises an HLA class I polypeptide.

3. The method according to claim 1, wherein the HLA polypeptide which is enriched in the biological sample from the subject comprises an HLA class II polypeptide.

4. The method according to any one of claim 1 or claim 2, wherein the HLA polypeptide selected from HLA-A, HLA-B or HLA-Cw heavy chain polypeptides.

5. The method according to any one of claims 1 to 4, wherein the HLA polypeptide which is enriched in the biological sample from the subject is a soluble HLA polypeptide.

6. The method of any one of claims 1 to 5, wherein the step of obtaining an at least partial amino acid sequence of the HLA polypeptide comprises subjecting the enriched biological sample to at least one mass spectrometry analysis.

7. The method according to any one of claims 1 to 6, wherein the biological sample from the subject is a blood sample.

8. The method according to any one of claims 1 to 7, wherein the step of enriching a biological sample from the subject for an HLA polypeptide comprises immuno- enrichment of the HLA polypeptide.

9. The method according to claim 8, wherein the immuno-enrichment comprises the use of an HLA class I specific antibody and/or an HLA class II specific antibody.

10. The method according to claim 6, wherein the mass spectrometry analysis comprises liquid chromatography-tandem-mass-spectrometry or matrix-assisted laser desorption/ionization.

11. The method of claim 1 wherein the subject is donating cells, tissue or an organ.

12. The method of claim 1 wherein the subject receiving donated cells, tissue or an organ.

13. A kit for enriching a biological sample from a subject for a Human Leukocyte Antigen (HLA) when used in the method of claim 1, the kit comprising at least one antibody specific to a Class I HLA polypeptide and/or at least one antibody specific to a Class II HLA polypeptide.