The 1980s: Class II and Molecular Nomenclature
Up until this point, most analysis was performed by serology testing (leukoagglutination, cytotoxicity, and complement fixation) which yielded the identities of Class I specificities. HLA Class I specificities are found on almost all cells in the body.
HLA Class II is a different matter. HLA Class II specificities are found on antigen presenting cells (APCs) such as B lymphocytes.
Testing advancements in the latter part of the 1970s led to cellular testing, focusing specifically on T and B lymphocytes. These cellular tests included mixed leukocyte culture (MLC, which stimulates T cell reactivity) and primed lymphocyte tests (PLT) which paved the way for HLA Class II discovery, beginning with Dw specificities.6
By 1984, the HLA-D region was officially proclaimed, though it’s status as “Class II” was more hesitant. While Dw specificities were not as easy to identify, the DR, DQ, and DPw loci had more success.
In 1987, allele counts exploded thanks to a new testing method: nucleotide sequencing. This test came with extensive molecular data and additional nomenclature categories.
Nucleotide sequencing introduced the concept of alleles as opposed to antigen specificities. A known genetic sequence of an antigen specificity was given an allele name. The correlation between the allele (genetic sequence) and it’s 3D antigen structure (serological specificity) on the cell is known as serological equivalency.
Thanks to nucleotide sequencing, HLA Class I loci were reaffirmed as single letter loci designations: A, B, C, E, etc. The letter D was excluded due to its use in Class II. HLA Class II loci were officially codified as double letter classifications (DN, DO, DP, DQ, DR) with additional alpha (A) and beta (B) chain designations and pseudogene/non-expressed designation numbers.
What is Cellular Typing? How is this different from Serological Typing?
Cellular Typing: This type of test requires pitting two different cell sets against each other (stimulator cells vs responder cells). Stimulator cells are unable to react (irradiated). Their HLA type is known and they are combined with the responder cells to see if the responder cells…well…respond. The immune system is designed to detect anything that doesn’t look like itself. This means that if the responder cells (mainly T cells) don’t react, it’s because they recognize the HLA type. If the responder cells do react, it’s because they don’t recognize the type. This result relies on detecting a cellular response (ie: cellular testing).
Cellular tests: Mixed Leukocyte Culture (MLC), Primed Lymphocyte Testing (PLT)
Serological Typing: This type of test is an antibody-antigen reaction. Typing HLA antigens serologically means using known antibody reagents to identify unknown patient HLA antigens. If a reaction is detected, it means the antibody bonded to the cells, indicating the presence of a specific antigen. This result relies on detecting the presence or lack of antibody-antigen binding.
Serological tests: leukoagglutination, complement fixation, microcytotoxicity testing.
You might notice that these assays tell you when something IS or IS NOT that HLA type. Ultimately, to type a patient, a laboratory needs reagents for all of the known HLA antigen specificities. Is it A1? No. Is it A2? No. Is it A3? No…on and on. That’s a lot of reagents and a lot of testing to mix cells/antibodies or cells/cells.
MLC testing was ultimately looking at the entirety of the responder’s Class II antigens rather than any individual Class II locus. Because of this, the Dw series of specificities discovered by MLC testing were not easily confirmed within the nomenclature. See ‘What is Dw? How do DR, DQ, and DP fit in?‘ for more information on how these assays contributed to HLA nomenclature.
What is Nucleotide Sequencing? How is this different from Serological and Cellular Typing? (High Resolution vs Low Resolution)
How does nucleotide sequencing work?
In the simplest terms, nucleotide sequencing begins with breaking the cell membrane to release the cell’s DNA. It then uses that cell DNA as a template to create additional copies, all with varying lengths, beginning at the first nucleotide. These varying lengths of DNA can then be separated by size.
In this example, one base pair is the shortest sequence and six base pairs is the longest sequence. Organizing them by size and identifying the base at the end of each sequence essentially gives you the order of DNA nucleotides that matches to the original DNA strand.
The specific order of nucleotides (A, T, C, G) matches an HLA allele, which is your typing result. The allele identified then corresponds to a serological specificity (protein product) located on the cell’s surface. To learn more about serological equivalents, read ‘What are serological equivalents?‘.
Nucleotide Sequencing vs Serological And Cellular Typings
Nucleotide sequencing focuses on patient DNA, whereas serological and cellular typings both focus on the HLA cell surface antigens.
The sequence of nucleotide bases in the patient’s DNA identifies the HLA allele that is present. In protein synthesis, this allele is transcribed, translated, and then folded into a protein. This protein is the antigen specificity that ends up on the patient’s cell surface that the serological and cellular testing identifies.
High Resolution and Low Resolution Typings
High Resolution Typing: Identifies differences in HLA typing down to single nucleotide base differences (genetic/DNA differences).
Low Resolution Typing: Identifies what HLA specificity is on the cell surface.
Nucleotide sequencing is a form of molecular typing for HLA specificities. Nucleotide sequencing is also considered high resolution typing because it is a more specific typing, down to single nucleotide bases. It gives us more detail about how the serological specificity is built, which means that when we are matching HLA, high resolution allele typings will be a more exact match. For bone marrow transplants, high resolution typing is required.
Low resolution typing gives us the bare minimum of information for an HLA type. Serological typing and cellular typing are low resolution HLA typing results that are identifying what is on the surface of the patient’s cells without going into detail about their DNA sequence.
You can see ‘What Does Each Part of an Allele’s Molecular Name Mean?‘ to learn more.
What is Dw? How do DR, DQ, and DP fit in?
There two main HLA classes and several classical HLA products that we know today:
HLA Class I loci of A, B, and C
HLA Class II loci of DR, DQ, and DP
But have you heard of ‘D’ specificities? What is just plain old ‘D’ (or Dw, as these specificities were never fully accepted as part of HLA nomenclature)?
After the discovery of the C locus, D was the next letter for use. Cellular testing methods such as the mixed leukocyte culture (MLC) and T cell cloning showed noteworthy reactions that became classified as the Dw specificities.3
The D’s were the beginning of Class II locus discovery but the assays in use to distinguish Dw were not distinct enough to identify specific antigens. The mixed leukocyte culture (MLC) assay was essentially looking at all of the HLA class II antigens (DR, DQ, and DP) as a whole rather than looking at any one locus, so the noteworthy reactivity could’ve possibly been related to multiple loci and specificities instead of one single specificity.5 While the reactions were able to provisionally introduce specificities that were labeled Dw, the reactions were not consistent enough to fully identify and accept the D specificities until after the DR, DQ, and DP loci were established. By this time, some of the Dw reactions could be tied to accepted DR or DQ specificities already in use. This essentially made the Dw locus obsolete.
For example:
Dw1 was associated with the established DR1 reactivity.2
Dw2 and Dw12 were both associated with DR2.2
If Dw1, Dw2, and Dw12 already have accepted names (DR1, DR2) there’s no need to add them to the nomenclature again.
So how did we go from D to DR, DQ, and DP?
As research focused more on isolated B lymphocyte populations, they identified consistent reactivity patterns that often, but not always, matched the Dw’s. They decided to call it DR (for D-Related) and the Nomenclature Committee gave DRs a provisional (‘w’) status as well.7
By 1980 the D’s were still not sufficiently identified to drop provisional status. Some of the DRs, however, achieved full designation in HLA nomenclature. Attempts were made by the Nomenclature Committee to keep the D and DR specificities aligned in name but this just wasn’t working due to increased complexity (ie: Dw9 reactivity did not correlate to DRw9 reactivity).1
The DQ specificities were also designated in 1984 as distinct from DR but still similar to the Dw’s. The DP specificities were introduced by use of primed lymphocyte testing (PLT).2
With the development of nucleotide sequencing assays, researchers were able to establish that multiple D subsets were present and that each of these subsets had multiple alpha and beta chains present, though the details about the number and types of chains were still sketchy. DQ and DP became established in the list of Class II loci due to their position and genetic similarity to DR, giving us DR, DQ, and DP loci in the HLA Class II region.2
Dw, because of the testing’s difficulty identifying singular specificities, became outdated and nucleotide sequencing essentially sealed this with the identification of specific loci tied to the DR, DQ, and DP specificities.
As a side note, the foresight of the Nomenclature Committee accounted for the expectation that, eventually, we would use ‘A’ as a designation for alpha chains and ‘B’ as a designation for beta chains in the Class II loci.2 By 1987, this was effectively in use. (ex: DQA [DQ alpha chains] and DQB [DQ beta chains].3
Why are the DR51, 52, and 53 specificity numbers so high?
Initially, DRw52 and DRw53 were identified as cross-reacting specificities (late 70’s and early 80’s).
In 1984 researchers knew, based on reactivity patterns, that the DRw52 and DRw53 were distinctly tied to the DR subset. Because of their high levels of cross-reactivity they were given high specificity values to distinguish them from the more distinct, narrower specificities that were clearly identifiable (Like DR1,2,3 etc).2 Though I have not confirmed this, I assume using numbers in the 50’s were also an attempt to make sure they had room to continue numbering narrower specificities.
Once nucleotide sequencing came into the picture, these became clear as alternative beta chains and were given beta chain locus numbers in sequence, where:
DRB1 (ie: DR beta chain locus #1) = associated with our currently known narrow DR specificities (DR1, DR2, etc)
DRB2 = pseudogene
DRB3 = DR52
DRB4 = DR53
DRB5 = DR51
The Beginning of Molecular Nomenclature
Work in Progress
What’s with the asterisk in an HLA typing?
In 1987, with the use of molecular typing, the asterisk was implemented to distinguish between molecularly typed HLA alleles and serologically typed specificities. Serological specificities contain no asterisk.
A203 = serological specificity
A*0203 or A*02:03 = molecular typing (aka allele or allelic specificity)
The A203 was formally identified with both serological testing and molecular testing so it has official names using both forms of nomenclature.
Now, use of the asterisk gets complicated by the use of italics. B2701 (in italics) is an acceptable alternative for B*2701 (with asterisk) as the italics identify a gene product and keeps with molecular nomenclature in other fields. This method works for Class I but not for Class II.
Class II has numbers for additional alpha or beta chain genes following the alpha/beta designation. Without the asterisk these numbers run together and confuse the designation. For example, DRB4*0101 without the asterisk would be DRB40101. If someone saw this, would they think “DRB*40101”? This is a confusing interpretation so italics do not work for Class II.
What are serological equivalents?
Some molecular specificities have serological equivalents. The Nomenclature Committee attempted to tie the alleles they were discovering via molecular typing to serological reactivity.
Researchers were using molecular typing to identify so many allelic specificities so fast (and then developing nomenclature to fit them) that the field had trouble keeping up with the development of new serological reagents to match that specific allele.
There were hundreds of alleles, and then thousands of alleles. Creating reagents for serological specificities to match these alleles was not realistic. There were simply too many alleles.
There was debate as to whether they should continue identifying and publishing new serological confirmations given the amount of work required to formally identify serological specificities. They decided to approve new serological specificities only if they matched a corresponding molecular allele (thus removing the need for a provisional ‘w’ for any future serological specificities). This is how alleles such as A*0203 and A*0210 became accepted with serological nomenclature (A203, A210).
Eventually, though, the Committee stopped publishing new serological specificities, focusing instead on matching molecular alleles to serological specificities that already existed. For example, A*0201 is matched to the A2 serological specificity instead of putting in the work to develop reagents specific to A*0201(which would’ve created an A201 serological specificity if they’d gone through with it. Spoiler alert: they did not).
The developing nomenclature did its best to account for matching serological specificities to corresponding alleles though there were many difficulties due to the increasing complexity of the HLA system. This became known as serological equivalency. Some alleles have serological equivalents. Some do not.
Why are some splits named after the broad/parent antigen (B*14:01, B*14:02) while some are named after the subtype (B*51, B*52)?
B14 splits into B64 (B*14:01) and B65 (B*14:02).
B5 splits into B51 (B*51) and B52 (B*52).
Note that the molecular typing for B14 splits follow the parent’s nomenclature (B*14) while the B5 splits follow their own individual molecular nomenclature (B*51, B*52).
Why the naming difference? This is because of nucleotide sequence similarity. Sequences that were extremely similar were given a molecular name under the parent antigen’s specificity (ex: B14). Sequences with more differences were named using their individual serological specificity as the basis for their molecular name (ex: B51,52)
Another example of sequence similarity includes DQ3 splitting into DQB1*03:01 (DQ7), DQB1*03:02 (DQ8), and DQB1*03:03 (DQ9). The sequences were very similar. DQ8 and DQ9 were especially difficult to tell apart in their DNA sequence.4 This meant that molecular nomenclature used the parent/broad antigen for molecular naming of DQ7, 8, and 9.
Another example of sequence differences between splits includes A28’s split into A68 (A*68) and A69(A*69). These were given molecular names based on their split due to their distinct difference in DNA sequence.
This goes into further depth during the 1990’s decade where the following question is answered: “What’s the difference between a two-digit split and a four-digit variance?”
Nomenclature Report Summaries Through the 1980s
Nomenclature History: The 1990s
References
You can find a link to all relevant nomenclature reports at: https://hla.alleles.org/nomenclature/nomenc_reports.html
- WHO Nomenclature Committee (1980). Nomenclature factors of the HLA System, 1980. Bulletin of the World Health Organisation, 58, pp. 945-8.
- Bodmer, W.F., Albert, E., Bodmer, J.G., Dausset, J., Kissmeyer-Nielsen, F., Mayr, W., Payne, R., van Rood, J.J., Trnka, Z., & Walford, R.L. (1985). Nomenclature for factors of the HLA system, 1984. Bulletin of the World Health Organisation, 63, pp. 399-405.
- Albert, E. (1988). Nomenclature for factors of the HLA system, 1987. Tissue Antigens, 32, pp. 177-87.
- Bodmer, J.G., Marsh, S.G.E., Parham, P., Erlich, H.A., Albert, E., Bodmer, W.F., Dupont, B., Mach, B., Mayr, W.R., Sasazuki, T., Schreuder, G.M.Th., Strominger, J.L., Svejgaard, A., & Terasaki, P.I. (1990). Nomenclature for factors of the HLA system, 1989. Tissue Antigens, 35, pp. 1-8.
- Rodey, G.E. (2000). HLA Beyond Tears: Introduction to Human Histocompatibility (2nd Ed.). De Novo, Inc. ISBN: 0-9678268-0-2
- Albert, E., Amos, D.B., Bodmer, W.F., Ceppellini, R., Dausset, J., Kissmeyer-Nielsen, F., Mayr, W., Payne, R., Rood, J.J., Terasaki, P.I., Trnka, Z., & Walford, R.L. (1978). Nomenclature for factors of the HLA System, 1977. Bulletin of the World Health Organisation, 56, pp. 461-5.
One Response