Human lymphocyte antigen molecular typing: how to identify the 1250+ alleles out there. (Advances in the Science of Pathology).
Application of HLA typing was focused primarily in the transplantation arena. (6) The more closely the donor and recipient alleles matched, the better the outcome of the graft. As the transplantation community moved forward and expanded from the solid organ arena to bone marrow transplantation, it became apparent that the serologic and cellular methods used were insufficient. Grafts were being rejected and graft-versus-host disease was claiming lives when the donor recipient pairs appeared to be well matched with the tools at hand. By this time, the human genome project and the immunology community as a whole had produced tools that could allow a finer discrimination of the alleles of an individual. In fact, the subtypes of HLA-B44 and -B45 had given way to further subdivision of alleles. (7) It was clear that this was a polymorphic system beyond previous thought.
A comprehensive listing of alleles for each loci within the HLA system is not realistic for publication. New members of allele families are added continuously, and a snapshot at best would be provided. For the most current, accepted alleles for any given loci, the following Web sites are provided: http://www.anthonynolan.org.uk/HIG/ data.html or http://www.ebi.ac.uk/imgt/hla/. These resources are updated regularly and are user friendly. Within these sites, the various alleles at a locus are available both as protein sequence and as nucleic acid sequence. The sites provide documentation for the entries listed.
LEVELS OF DISCRIMINATION
Testing vernacular has incorporated terms that need definition. HLA typing can be performed at several levels of discrimination between alleles. The standard terms used are low, medium, and high resolution. (8) Low resolution is basically the equivalent of serologic typing and incorporates the idea of defining broad families of alleles that may have many members. The members of these broad families detected with low-resolution typing can be recognized as distinct using high-resolution testing. High-resolution typing is an effort to definitively state what allele resides at a loci for an individual. Medium resolution is exactly what would be expected, in between high and low, and helps determine that the individual has one of the many alleles within a low-resolution group and eliminates others from possible choices. Instead of stating that an individual has any one of 20 subtypes of a low-resolution group, the option is narrowed to fewer than 20. Solid organ transplant programs operate well with low-resolution typing; bone marrow programs need the support of high-resolution typing to promote long-term graft survival and minimize graft-versus-host disease.
RESTRICTION FRAGMENT LENGTH POLYMORPHISM
With an increasing need for highly discriminatory typing, the HLA community quickly embraced molecular techniques. In some cases molecular testing has replaced traditional methods, but many laboratories still use both traditional and molecular methods to reach a meaningful result.
Early molecular typing employed restriction fragment length polymorphism (RFLP) paradigms. (9,10) Genomic DNA was cut with a restriction endonuclease, and the resultant fragments were size-fractionated in an agarose gel and transferred to a nylon membrane for probing with locus-specific probes. While the testing provides discrimination of alleles at a locus, not all alleles are discernable. Restriction fragment length polymorphism methods are macroscopic at best, and they could not provide the discrimination that practice indicated was necessary. In addition, there were problems with cross-hybridization between loci of a given class. Restriction fragment length polymorphism was useful, but higher resolution or discrimination of alleles was needed.
With the advent of the polymerase chain reaction (PCR), HLA molecular typing expanded almost exponentially. (11) Most current, state-of-the-art, HLA typing laboratories employ amplification protocols (most often by PCR) in their typing scheme. There are numerous approaches and each will be discussed, but not in order of their importance, popularity, or discovery.
PCR-Sequence-Specific Oligonucleotide Probes
Amplification of DNA followed by probing with sequence-specific oligonucleotide probes (SSOP) is performed in several models. (12) The principle of this approach is that different alleles at a locus can be detected by probes recognizing the allele differences and used to detect the complementary region of an amplified target. With that said, it is customary to have one of the DNA strands in the reaction immobilized. HLA typing scientists are creative, and both immobilization of the target and the probe motifs have been developed. If the amplified product is immobilized, then the probe is labeled and applied. This is a traditional dot blot, named such because the target is spotted as dots on a filter or membrane. (13) Not to be constrained by geometry, the material can also be applied as a line, and the process is then referred to as a slot blot. If the probe is immobilized and the amplified target is labeled, the assay is then referred to as a reverse dot blot or a "blot dot." (14) These methods are well suited to high-throughput needs, as many samples can be spotted and probed in a single reaction in the dot-blot scenario, and a sample can be reacted with many probes in one reaction in a blot-dot paradigm. To use many probes at once, the temperature of melting for the probes needs to be optimized.
Amplification of alleles at a locus using allele-specific primers is also used quite extensively. (15) This method is often referred to as sequence-specific primer (SSP) or amplification refractory mutation system (ARMS) typing. (16,17) While SSP and ARMS have a slightly different context, the principle is very much the same. In these systems, the amplification process is performed using one primer that is conserved across many alleles and a second primer specific for a 1-base pair (bp) difference. That is not to say that the primer is 1 bp in length, but that the 3' end of the primer is specific for a complementary base in the target. The 3' end of a primer must be firmly bound to the target for the polymerase to extend the strand. The detection employed in SSP or ARMS methods is usually an agarose gel fractionation of products. Most kits available have multiple primer sets in separate reactions to type for many alleles at a locus or loci. Temperature of melting optimization of the primers is necessary to allow thermal cycling of all reactions in a single instrument. (18) Testing using SSP and ARMS can yield low-, medium-, or high-resolution results, depending on the primer sets used.
PCR-Restriction Fragment Length Polymorphism
Digestion of amplified allele products with a restriction endonuclease can also provide very discrete typing results. (19,20) Kits modeled on RFLP use restriction endonucleases specific for sites that confer the variation between alleles. This type of testing can be used to resolve ambiguous typing results for other assays, but it is not usually employed for high-throughput needs.
PCR-Single-Strand Conformation Polymorphism
Given the proper physical environment, DNA will form secondary structures. The placement of the "loops" and "hairpins" that form are sequence dependent. If 2 alleles have different sequences, the types and placement of secondary structures should differ. This is the basis of double-stranded sequence conformation polymorphism testing. (21) The concept is that alleles allowed to form secondary structures can be discriminated among themselves by the migration of these secondary structures in a gel matrix environment. This type of format works for most types, but there can still be some ambiguities. Resolution of some of these concerns has been addressed by utilizing a reference strand in the duplex formations. (22) This serves as an internal reference across all allele combinations, because only those duplexes with the reference strand are visualized. This can be a very robust system.
The holy grail of HLA typing is knowing exactly what allele resides on the chromosome being tested. This finding is definitive, or is thought to be definitive. There are many ways for HLA laboratories to perform sequence-based typing (SBT). (23,24) Class I and class II typing are performed by slightly different methods. The polymorphic exon for class II is exon 2, and usually just a few hundred bases are sequenced to obtain enough information to get an answer. Ambiguous types can often be resolved by using primers at codon 86 of exon 2 as anchors and sequencing back in the 5' direction. (25) Class I typing is more complex and requires information from several exons for allele assignment. (26) Often several nested amplifications are necessary to get discrete fragments to analyze. Ambiguous results are still possible even with this method. This is considered a high-resolution method. With the development of automated sequencers and robotic stations, SBT testing can be efficacious and utilized for high-throughput needs.
Progress made by the genomic community has led to what is called microarray technology. Microarrays are literally what the word implies; they are an array of discretely placed probes in columns and rows in a very microscopic area. It is possible to label amplified DNA and hybridize it to these probes and detect binding patterns that can be translated to typing patterns. There are reports that future HLA typing may be performed on this platform, but as yet there are no solid prototypes for HLA on the commercial market. (27)
Given the wide selection of molecular methods available for HLA typing, the question becomes which method to select. Many laboratories have stopped performing cellular and serologic testing for HLA class II typing needs and have adopted strictly molecular methods. There are strong indications that there are laboratories considering this same practice for class I HLA typing. Considering the facts that DNA can be isolated from samples long after the viable cells needed for cellular and serologic testing are gone and that a single DNA isolation from a sample can be used for both, the adoption of a single platform is attractive. Molecular-based HLA typing, while efficient and progressive, can still produce ambiguous results. There are several ways this ambiguity may come about and also several options for resolution.
Ambiguity in testing may manifest as the inability to distinguish between 2 or more subtypes of a single allele family or the inability to distinguish between members of 2 separate allele families. Many of the polymorphisms seen in HLA alleles are shared among subtypes of an allele and can be present in other allele families. Since the polymorphic regions are the sites targeted by primers and probes, they are an obvious source for cross-hybridization of primers and probes and the generation of extra bands in amplification reactions. One way to resolve this situation is to use a different set of primers or probes to distinguish the differences in polymorphic regions, and often this clarification is attempted by using a different manufacturer's kit. This method does not often resolve the problem, because many manufacturers target the same sites and have almost identical primer and probe sets. Often by knowing what the allele is and by eliminating possibilities based on data from one platform, a second platform can be used to help identify the allele. A simplistic example is utilizing 2 kits, one that can discriminate most alleles at a locus very broadly and when used on an individual sample, it would yield a result indicating the presence of more than 1 allele (eg, an HLA-A24 and/or HLA-A30 present). If a second kit, at the same resolution, indicates the absence of an HLA-A30 allele member, the conclusion is that the individual tested is either an HLA-A24/-A24 or an HLA-A24 and a blank. A blank can be a yet-undetected or unidentified allele. If this example were further complicated by evidence from the first test yielding the possibility of an HLA-A24, -A30, or an A34, then the second assay could further direct the decision. With the exception of polyploid individuals, any given individual can only have 2 alleles for a specific loci. (28) The detection of 3 antigens by SSP typing in particular is not improbable, based on the high degree of polymorphic sites between alleles. Going to a second kit does not always resolve this problem, and a molecular method augmented by a serologic method can be helpful.
Another area of ambiguity results from the detection of 2 alleles by molecular methods, while the serologic method detects the presence of only 1 molecule on the surface of the cell. This situation is often due to the second allele being a pseudogene with a sequence that has an early stop codon or a gene with a faulty promoter. (29,30) Either way, the decision of what to report must be made. The molecular type can be misleading, in that the second allele is not present and will not contribute to any immune response, nor will it be the target of any immune response.
CONVERSION TO SEROLOGY
The decision of how to report molecular typing results is also contentious, particularly those produced by low-resolution methods. These methods can lead to a very long list of possible allele subtypes, which can be not only confusing but also daunting to those not familiar with the methodology. Many transplant programs prefer to have the information distilled and reported as the equivalent serologic assignment of alleles. Many times that can be accomplished, but some allele families have molecularly defined subtypes that can have no serologic equivalent. The HLA-B14 is an example of this situation. In serologic terms, HLA-B14 can be split to HLA-B64 or -B65, and some of the molecularly detectable subtypes of HLA-B14 can be converted. Other subtypes of HLA-B14 do not have serologic equivalents and are therefore reported as a molecular assignment. When reviewing such a case, it is not clear whether the HLA-B14 that cannot be assigned a serologic type is the equivalent of a molecularly assigned HLA-B64. It is true that both are an HLA-B14, but they may not be the equivalent. The HLA typing community must come to a consensus on how handle this dilemma because the problem will only grow with the addition of new subtypes in all the allele groups.
Molecular typing is far superior to serologic or cellular methods in terms of detecting allele group subtypes, and many reports indicate that there is a higher degree of accuracy with molecular methods. Why not completely drop serologic methods? There are compelling factors to consider before doing this, among them the argument stated earlier supporting the use of serology to validate molecular results in cases of ambiguity and to determine the presence of antigens on the cell surface.
Molecular typing methods present challenges in terms of quality assurance and updates for newly identified alleles. Commercially available kits need to have quality assurance testing performed on receipt. This testing involves demonstrating that the reagents react when indicated and fail to react when appropriate. This validation can be accomplished by running full kits with various DNA samples or using the kit in a patchwork sort of arrangement with various DNA samples, the latter (the patchwork design) being the more economical approach. The other issue is algorithmic software supplied by the manufacturer, which is used to help aid in coming to a conclusive result. With each update of the HLA Nomenclature Report from the World Health Organization, the libraries of each software must be updated, and the possible outcomes of the various new alleles have to be accounted for with the kit's reagents (primers and probes). This is a significant task and should be a variable used in selection of manufacturers. How often do they update alleles, and how do they supply the update? These are 2 salient questions.
Molecular typing is clearly adaptable and suited for HLA typing needs. Many of the questions that need to be answered in assignment of an HLA type for an individual can only be answered by molecular typing methods. There is evidence that supports the maintenance of more than one molecular and/or serologic method to arrive at the answers necessary to support a transplant program.
(1.) Nomenclature for factors of the HLA system: update October 2000. Tissue Antigens. 2001;57:93-94.
(2.) McCloskey DJ, Brown J, Navarrete C. Serological typing of HLA-A, -B and -C antigens. In: Hui KM, Bidwell JL, eds. Handbook of HLA Typing Techniques. Boca Raton, Fla: CRC Press Inc; 1993:175-247.
(3.) Brown J, McCloskey DJ, Navarrete C. HLA-DR and -DQ serotyping. In: Hui KM, Bidwell JL, eds. Handbook of HLA Typing Techniques. Boca Raton, Fla: CRC Press Inc; 1993:249-307.
(4.) Navarrete C, Brown J, McCloskey DJ, Jaraquemada D. Definition of HLA-Dw determinants using homozygous typing cells and the mixed lymphocyte culture. In: Hui KM, Bidwell JL, eds. Handbook of HLA Typing Techniques. Boca Raton, Fla: CRC Press Inc; 1993:309-349.
(5.) Navarette C, McCloskey DJ, Brown J. Definition of HLA-Dw and HLA-DPw determinants by the primed lymphocyte test. In: Hui KM, Bidwell JL, eds. Handbook of HLA Typing Techniques. Boca Raton, Fla: CRC Press Inc; 1993:351-371.
(6.) Bradley BA. Introduction. In: Hui KM, Bidwell JL, eds. Handbook of HLA Typing Techniques. Boca Raton, Fla: CRC Press Inc; 1993:1-8.
(7.) Keever CA, Leong N, Cunningham I, et al. HLA-B44-directed cytotoxic T cells associated with acute graft-versus-host disease following unrelated bone marrow transplantation. Bone Marrow Transplant. 1994;14:137-145.
(8.) Zachary AA, Vogelsang GB, Steinberg AG, Leffell MS. Strategies for determining HLA compatibility in related donor bone marrow transplantation. Transplantation. 1997;64:828-835.
(9.) Duquesnoy RJ, Trucco M. Genetic basis of cell surface polymorphisms encoded by the major histocompatibility complex in humans. Crit Rev Immunol. 1988;8:103-145.
(10.) Bidwell JL, Bidwell EA, Wood DAP, et al. HLA-DR and -DQ typing by DNA-RFLP analysis. In: Hui KM, Bidwell JL, eds. Handbook of HLA Typing Techniques. Boca Raton, Fla: CRC Press Inc; 1993:71-98.
(11.) Begovich AB, Erlich HA. HLA typing for bone marrow transplantation: new polymerase chain reaction-based methods. JAMA. 1995;273:586-591.
(12.) Wordsworth P. Techniques used to define human MHC antigens: polymerase chain reaction and oligonucleotide probes. Immunol Lett. 1991;29:3739.
(13.) Tiercy JM, Grundschober C, Jeannet M, Mach B. A comprehensive HLA-DRB, -DQB and -DPB oligotyping procedure by hybridization with sequence-specific oligonucleotide probes. In: Hui KM, Bidwell JL, eds. Handbook of HLA Typing Techniques. Boca Raton, Fla: CRC Press Inc; 1993:99-116.
(14.) Suberbielle-Boissel C, Chapuis E, Charron D, Raffoux C. Comparative study of two methods of HLA-DR typing: serology and PCR/dot blot reverse. Transplant Proc. 1997;29:2335-2336.
(15.) Olerup O, Setterquist H. HLA-DR typing by polymerase chain reaction amplification with sequence-specific primers (PCR-SSP). In: Hui KM, Bidwell JL, eds. Handbook of HLA Typing Techniques. Boca Raton, Fla: CRC Press Inc; 1993: 149-174.
(16.) Dupont B. "Phototyping" for HLA: the beginning of the end of HLA typing as we know it. Tissue Antigens. 1995;46:353-354.
(17.) Tonks S, Marsh SG, Bunce M, Bodmer JG. Molecular typing for HLA class I using ARMS-PCR: further developments following the 12th International Histocompatibility Workshop. Tissue Antigens. 1999;53:175-183.
(18.) Bunce M, Barnardo MC, Welsh KI. The PCR-SSP Manager computer program: a tool for maintaining sequence alignments and automatically updating the specificities of PCR-SSP primers and primer mixes. Tissue Antigens. 1998;52:158-174.
(19.) Inoko H, Ota M. PCR-RFLP. In: Hui KM, Bidwell JL, eds. Handbook of HLA Typing Techniques. Boca Raton, Fla: CRC Press Inc; 1993:9-70.
(20.) Trejaut J, Hobart D, Kennedy A, Greville WD, Taverniti A, Dunckley H. New DRB1* alleles (HLA-DRB1*1135, DRB1*1430 and DRB1*1433) and a confirmatory sequence (DRB1*1133). Tissue Antigens. 2000;55:89-91.
(21.) Teutsch SM, Bennetts BH, Castle M, Hibbins M, Heard RN, Stewart GJ. HLA-DQA1 and -DQB1 genotyping by PCR-RFLP, heteroduplex and homoduplex analysis. Eur J Immunogenet. 1996;23:107-120.
(22.) Turner DM, Poles A, Brown J, Arguello JR, Madrigal JA, Navarrete CV. HLAA typing by reference strand-mediated conformation analysis (RSCA) using a capillary-based semi-automated genetic analyser. Tissue Antigens. 1999;54:400-404.
(23.) Hurley CK. Acquisition and use of DNA-based HLA typing data in bone marrow registries. Tissue Antigens. 1997;49:323-328.
(24.) Dinauer DM, Luhm RA, Uzgiris AJ, Eckels DD, Hessner MJ. Sequence-based typing of HLA class II DQB1. Tissue Antigens. 2000;55:364-368.
(25.) Lanchbury JS, Hall MA, Welsh KI, Panayi GS. Sequence analysis of HLA-DR4B1 subtypes: additional first domain variability is detected by oligonucleotide hybridization and nucleotide sequencing. Hum Immunol. 1990;27:136-144.
(26.) Kurz B, Steiert I, Heuchert G, Muller CA. New high resolution typing strategy for HLA-A locus alleles based on dye terminator sequencing of haplotypic group-specific PCR-amplicons of exon 2 and exon 3. Tissue Antigens. 1999;53: 81-96.
(27.) Fortina P, Delgrosso K, Sakazume T, et al. Simple two-color array-based approach for mutation detection. Eur J Hum Genet. 2000;8:884-894.
(28.) Pearson G, Mann JD, Bensen J, Bull RW. Inversion duplication of chromosome 6 with trisomic codominant expression of HLA antigens. Am J Hum Genet. 1979;31:29-34.
(29.) Salazar M, Granja CB, Selvakumar A. New HLA-DR haplotypes containing the DRB6 pseudogene. Tissue Antigens. 1996;48:575-579.
(30.) Laforet M, Froelich N, Parissiadis A. A null HLA-A*68 allele in a bone marrow donor. Tissue Antigens. 1999;53:573-575.
Accepted for publication September 28, 2001.
From Medical Technology Program and the Department of Human Pathology, Michigan State University, East Lansing, Mich.
Presented at the 10th Annual William Beaumont Hospital Seminar on Molecular Pathology, DNA Technology in the Clinical Laboratory, Royal Oak, Mich, March 8-10, 2001.
Reprints: John A. Gerlach, PhD, dip ABHI, Medical Technology Program and the Department of Human Pathology, B228 Life Science, Michigan State University, East Lansing, MI 48824-1317 (e-mail: email@example.com).
|Printer friendly Cite/link Email Feedback|
|Author:||Gerlach, John A.|
|Publication:||Archives of Pathology & Laboratory Medicine|
|Date:||Mar 1, 2002|
|Previous Article:||Mitochondrial DNA-related mitochondrial dysfunction in neurodegenerative disease. (Advances in the Science of Pathology).|
|Next Article:||Molecular testing in the diagnosis and management of hepatitis C virus infection. (Advances in the Science of Pathology).|