MHC Class I, Class II, Antigen Processing, and Presentation

Major histocompatibility complex

• MHC-encoded proteins were first identified in the 1930s during studies of tissue rejection in transplantation experiments.
• Therefore, these proteins were given the name histocompatibility (histo meaning “tissue” and compatibility meaning “getting along”).
• The genes influencing the histocompatibility of tissue transplantation have been mapped to a wide genomic area with several loci, hence the word “complex.”
• Moreover, it was discovered that the proteins generated by these genes had profound effects on histocompatibility. To differentiate these proteins from others (encoded elsewhere in the genome) that had very minimal impacts on histocompatibility, these molecules were dubbed “major” histocompatibility molecules.
• Thus, the genes encoding these proteins were designated as MHC genes (major histocompatibility complex genes). Soon later, it was shown that MHC-controlled rejection of transplanted tissue was caused by the recipient’s immunological reaction to the donor cells.
• Although this observation suggested that MHC gene products were directly engaged in immune responses, it took immunologists several more decades to establish the physiological role of MHC-encoded proteins in presenting antigenic peptides to T cells.
• MHC class I and MHC class II molecules are the MHC-encoded proteins involved in the majority of antigen recognition by T lymphocytes. TCRs of CD8+ T cells identify MHC class I-bound peptides, whereas TCRs of CD4+ T cells recognise MHC class II-bound peptides.
• Additionally, the CD8 coreceptor of CD8+ T cells attaches to MHC class I, whereas the CD4 coreceptor of CD4+ T cells binds to MHC class II. The MHC class I protein is a heterodimer composed of a long transmembrane α chain that is non-covalently connected to β2-microglobulin (β2m).
• The MHC class I α chain, but not β2m, is encoded within the MHC. Class II MHC protein consists of a α chain and a somewhat smaller β chain, both of which are transmembrane proteins encoded by MHC genes.
• With the exception of the peptide-binding groove, the tertiary structures of MHC class I and class II molecules are extremely similar despite this change in composition.
• While virtually all nucleated cells express MHC class I, only a few cell types that serve as APCs (such as DCs, macrophages, and B cells) express MHC class II.
• Consequently, virtually any cell can serve as a target cell and provide antigen to CTLs produced from CD8+ Tc cells, whereas only APCs can activate CD4+ Th cells.

Major histocompatibility complex I (MHC Class I)

• The majority of nucleated cells have an assortment of MHC class I proteins. The peptides bound by these MHC class I molecules are typically endogenous, or derived from the breakdown of intracellularly produced proteins.
• The vast majority of these peptides will be “self” because the majority of proteins normally synthesised within a cell are of host origin at any given moment (as opposed to proteins of non-self origin, such as those generated during a viral infection).
• The MHC class I molecule does not distinguish between “self” and “non-self” peptides; CD8+ T cell TCRs provide this function.
• Self peptide–MHC complexes do not elicit an immunological response because T cells with the relevant specificity are typically absent from the T cell repertoire as a result of the formation of central tolerance.
• In contrast, non-self peptides complexed with MHC class I are identified and activate CD8+ T cells.

MHC Class I Structure

• Peptide antigens are presented by MHC class I molecules to CD8+ T cells. MHC class I molecules are expressed by the majority of nucleated cells, however the amount on the cell surface varies substantially between cell types and inflammatory circumstances.
• MHC class I molecules are single—chain transmembrane proteins that connect with β2-microglobulin for appropriate folding and trafficking to the cell surface.
• MHC class I molecules are characterised by their α1, α2, and α3 domains, which constitute a globular protein with a β-pleated sheet forming the floor of the peptide-binding groove and two helical sections forming the sidewalls of the groove.
• 1
• Since the MHC class I groove is closed on both sides, the length of the peptide that can be tolerated is normally limited to 9 amino acids. This is a significant characteristic of the groove.
• Class I MHC molecules are extremely polymorphic. Although different protein products (allomorphs) generated by MHC class I genes share structural similarities and bind a variety of peptides, the peptides that bind to different allelic forms of MHC class I are diverse.
• Due to structural variations between the peptide-binding grooves, each allomorph of MHC class I binds a separate collection of peptides.
• The majority of MHC class I molecules contain a groove with two or three deeper pockets to accommodate bulky amino acid side chains.
• The form, depth, and charge of the groove and binding pockets dictate the peptide sequences to which a particular MHC class I molecule can bind.
• Because the majority of persons express six distinct MHC class I alleles (two each of HLA-A, HLA-B, and HLA-C), T cells can be presented with a variety of pathogen-derived peptide sequences.
• Elution of peptides from the HLA-A2 MHC class I molecule and mass spectrometry analysis revealed that HLA-A2 can bind thousands of different peptides that share a common motif but are derived from a vast array of biological proteins.
• Despite the fact that binding of peptides and interaction with TCRs are exclusively mediated by the MHC class I protein, sustained surface expression of peptide/MHC class I complexes needs the connection with β2-microglobulin.
• This 12-kDa protein stabilises membrane-bound MHC class I molecules by binding to them. Most MHC class I molecules fold incorrectly and are destroyed before exiting the endoplasmic reticulum in the absence of β2-microglobulin.

Major histocompatibility complex II (MHC Class II)

• As stated previously, MHC class II molecules are nearly exclusively located on APCs. The peptides bound by MHC class II are produced from the breakdown of proteins that have entered the cell from the outside via phagocytosis or receptor-mediated endocytosis.
• Because APCs also catch and digest exhausted host proteins, the vast majority of peptides presented on MHC class II molecules are “self” and do not activate CD4+ T cells because the formation of central tolerance has eliminated these specificities from the Th cell repertoire.
• When an APC presents a non-self peptide coupled to MHC class II, a Th response is produced.

MHC Class II Structure

• MHC class II molecules present peptides to CD4+ T cells. In contrast to MHC class I molecules, the expression of MHC class II molecules is restricted to a small number of cell types, including macrophages, DCs, and B lymphocytes.
• Endothelial cells can express MHC class II molecules upon activation. The class II transactivator protein (CIITA), a transcription factor that promotes expression of MHC class II molecules, the associated invariant chain, and other molecules associated with MHC class II antigen processing, contributes to the regulation of MHC class II expression.
• Because circulating peripheral blood mononuclear cells do not produce surface MHC class II molecules, mutations in the CIITA gene result in bare lymphocyte syndrome, a significant human primary immunodeficiency condition associated with substantial immunosuppression.
• The folded MHC class II molecule is composed of two transmembrane proteins, a α-chain and a β-chain, which together form a peptide-binding groove-containing protein.
• The open-ended nature of the MHC class II groove is responsible for the binding of peptides that are significantly longer than those bound by MHC class I molecules.
• Peptides bound by MHC class II molecules are typically longer than ten amino acids and occasionally exceed twenty.

MHC Class I Antigen-Processing Pathway

• In general, antigens presented by MHC class I and class II molecules originate from distinct cellular compartments.
• The antigen processing route for MHC class I antigens begins in the cytosol with the breakdown of an endogenous self-protein. After microbial infection, intracellular proteins produced from microbes enter the MHC class I processing pathway.
• Cross-presentation is a mechanism through which extracellular host or microbe proteins that have been absorbed into membrane-bound compartments via endocytosis or phagocytosis can enter the MHC class I processing pathway.
• Proteasomes breakdown misfolded or unnecessary proteins through proteolysis in the cytoplasm, thereby regulating the protein composition of the cell. These multicomponent proteases have a barrel shape and are composed of four rings containing seven subunits apiece.
• Proteasomes degrade proteins by a variety of methods. Misfolded proteins and faulty ribosomal products, which both fail to assume their native conformation states, may expose peptide sequences that are identified by proteasomes, resulting in their fast destruction.
• Alternatively, numerous proteins destined for fast proteasomal breakdown are conjugated to ubiquitin by enzymes that identify the phosphorylation of particular amino acids.
• Proteasomal breakdown of microbial antigens is essential for the presentation of microbial peptides by MHC class I proteins during microbial infection.
• Some proteasomal subunits are replaced by others that improve the formation of antigenic peptides, and other components are introduced to the ends of the barrel to alter the effectiveness and specificity of protein degradation in activated cells or after exposure to IFN-γ.
• It is unknown if pathogen-derived polypeptides and proteins are degraded selectively and targeted for presentation by MHC class I molecules during an infection.
• Bacterial proteins that enter the cytosol are rapidly degraded due to the presence of unique amino-terminal amino acids or internal amino-acid sequences that promote fast breakdown.
• In the majority of cases, pathogen-derived antigens are probably degraded nonselectively alongside indigenous proteins, and pathogen-derived peptides compete with more abundant endogenous peptides for a binding groove in MHC class I.
• Proteasomes create peptides between 9 and 12 amino acids in length due to the length of the proteasome channel. Transporter Associated with Antigen Processing binds peptides synthesised by proteasomes (TAP).
• This ATP-dependent, heterodimeric transporter efficiently transports peptides from the cytosol to the lumen of the endoplasmic reticulum.
• Peptides less than 6 amino acids or longer than 14 amino acids are poorly transported by TAP.
• TAP is the primary peptide transporter involved in the formation of peptide/MHC class I complexes; animals with genetic deletions of TAP have significantly reduced levels of surface MHC class I and significantly reduced numbers of CD8+ T lymphocytes.
• Rarely identified in people, TAP deficiency is associated with a substantial reduction in circulating CD8+ T lymphocytes and mild immunodeficiency.
• TAP1 and TAP2 molecules carry peptides from the cytosol to the endoplasmic reticulum lumen.
• The development of the peptide loading complex involves the association of newly synthesised MHC class I molecules with TAP in the endoplasmic reticulum and the recruitment of many additional endoplasmic reticulum resident proteins and chaperones (PLC).
• The PLC consists of the MHC class I/β2-microglobulin complexes coupled to tapasin, which serves as a molecular adapter for TAP, as well as calreticulin and thiol reductase ERp57.
• The primary function of β2-microglobulin and the PLC is to maintain the shape of the MHC class I peptide-binding groove that favours the binding of high-affinity peptides. Due to the fact that TAP delivers numerous peptides into the endoplasmic reticulum lumen that are too long to fit into the MHC class I peptide–binding groove, the endoplasmic reticulum resident proteases ERAP1 and ERAP2 trim peptides prior to their final incorporation into the MHC class I peptide–binding groove.
• If the affinity between peptide and MHC class I is high enough, the PLC releases the MHC class I/β2-microglobulin/peptide complex, allowing it to reach the cell surface via the Golgi complex. If the affinity between peptide and MHC class I is low, the MHC class I heavy chain undergoes re-glycosylation of an N-linked glycan, which redirects MHC class I molecules into the PLC for peptide exchange and blocks their release into the secretory pathway.
• Inflammation controls the antigen processing pathway of MHC class I. IFN-γ in particular is a cytokine with numerous effects on the MHC class I pathway.
• IFN-γ increases the transcription of numerous components of the MHC class I pathway, including MHC class I molecules, TAP, tapasin, and a number of proteasome components. Three proteasome subunits, LMP-2, LMP-7, and MECL, are specifically induced and replace three subunits of the core proteasome complex.
• IFN-γ generates other accessory proteins that affect proteasome efficiency and specificity, with PA28, a six-subunit activator that forms rings that can cap the ends of the proteasome, playing a key role.
• PA28 can enhance the presentation of MHC class I–restricted, virus-derived epitopes to CD8+ T lymphocytes.
• In the presence of an infection, the MHC class I antigen processing pathway is amplified, allowing for greater presentation of pathogen-derived peptides to CD8+ T cells.

Viral Intervention with the MHC Class I Antigen-Processing Pathway

• CD8+ T cells and the MHC class I antigen processing pathway are primarily responsible for viral infection defence. Fascinatingly, viral infections have gone to great lengths to subvert the MHC class I antigen-processing pathway, highlighting the significance of this antiviral defence mechanism.
• It was discovered early on that herpesvirus-infected cells downregulate MHC class I expression. Exploration of the mechanism underlying this discovery found numerous viral proteins that inhibit various MHC class I antigen processing pathway steps.
• Herpes simplex virus encodes ICP47, which disrupts human TAP by blocking the peptide transport channel from the cytosolic side.
• The human cytomegalovirus-encoded protein US6 hinders TAP transport by inhibiting the peptide transporter from the endoplasmic reticulum luminal side using a similar method but a completely different protein.
• Additionally, human cytomegalovirus employs a number of additional ways to prevent MHC class I molecules from reaching the cell surface. US3 binds MHC class I molecules in the endoplasmic reticulum, preventing their transport to the cell surface.
• Adenoviruses, which encode the type I membrane protein E3-19K, adopt a similar method. By expressing an endoplasmic reticulum retention motif on its cytoplasmic tail, this protein binds MHC class I molecules in the endoplasmic reticulum lumen and blocks their egress from the endoplasmic reticulum.
• The displacement of MHC class I molecules residing in the endoplasmic reticulum into the cytoplasm, where they are promptly ubiquitinated and destroyed by proteasomes, is an additional method for downregulating surface MHC class I retention. Two human cytomegalovirus-encoded proteins, US2 and US11, are responsible for mediating this process.
• Remarkably, transport of misfolded or otherwise dysfunctional proteins from the endoplasmic reticulum lumen to the cytosol via the Sec61 translocon is a normal mechanism.
• US2 and US11 appear to expedite this process for MHC class I molecules alone. The binding of US2 to MHC class I molecules has been studied by x-ray crystallography.
• Using a distinct mechanism, the Kaposi sarcoma herpesvirus also inhibits surface MHC class I expression. K3 and K5 expressed by the Kaposi sarcoma herpesvirus are ubiquitin ligases that specifically conjugate ubiquitin to the cytoplasmic tails of MHC class I and B7.2 co-stimulatory components.
• Surface MHC class I molecules are rapidly absorbed and destined for lysosomal destruction upon ubiquitination. Additionally, HIV has evolved methods to inhibit surface expression of MHC class I molecules.
• By connecting with the clathrin adaptor complex, the retrovirally generated Nef protein specifically downregulates the production of HLA-A and HLA-B molecules.
• Importantly, the downregulation of MHC class I renders afflicted cells vulnerable to NK cell-mediated lysis. On contact with MHC class I molecules, NK cells express receptors that suppress NK cell activation.
• To prevent NK cell–mediated lysis of virally infected cells, human CMV generates an MHC class I–like protein, UL18, which works as a decoy for the NK cell inhibitory receptor LIR-1, providing the viral pathogen with an additional layer of camouflage.

MHC Class I Cross-Priming

• The antigen processing route for MHC class I has two primary functions. First, it delivers antigens to CD8+ T lymphocytes that are not yet activated, proliferating, or differentiating in a manner that promotes their activation, proliferation, and differentiation.
• Second, it signals cellular infection to activated CD8+ T cells by presenting antigens. DCs mediate the first function largely, if not solely.
• Any infected cell that expresses MHC class I can execute the second function. Different antigen processing criteria apply in these two instances. As discussed in the preceding sections, the normal MHC class I antigen processing pathway applies to the second function.
• Pathogen-derived antigens are degraded and displayed on the cell surface in the presence of MHC class I molecules when cells are directly infected.
• The presentation of MHC class I antigens by DCs is more complicated and is not limited to indigenous cytosolic proteins.
• Due to the fact that CD8+ T-cells are rarely infected directly, the major route for CD8+ T-cell priming involves uptake of debris from infected cells by DCs and re-presentation of pathogen-derived peptides by an antigen-processing pathway involving endocytosis and TAP-mediated transport of antigen into the endoplasmic reticulum.
• The CD8+ fraction of DCs is especially efficient at taking up and delivering exogenous antigens to the MHC class I antigen processing pathway.
• Antigen-containing phagosomes in DCs fuse with endoplasmic reticulum membranes, which recruits retrotranslocation machinery that shuttles misfolded proteins or antigens from the phagosome lumen into the cytosol, where they are degraded by proteasomes and enter the conventional MHC class I processing pathway.
• This convoluted route for CD8+ T-cell priming guarantees that CD8+ T-cell priming happens in a regulated manner and is likely a critical barrier to prevent excessively powerful CD8+ T-cell responses to systemic viral infections.

MHC Class II Antigen-Processing Pathway

• The antigen processing pathway of MHC class II provides peptides to CD4+ T cells. Although there are similarities to the MHC class I antigen processing system, there are significant differences.
• Firstly, the majority of peptides presented by MHC class II molecules originate from extracellular proteins endocytosed by MHC class II– expressing cells.
• During transport to the cell surface, MHC class II molecules also display peptides from membrane or secretory proteins that have been degraded in endosomal compartments.
• Regarding antimicrobial responses, the MHC class II antigen-processing pathway has been associated mostly with the response to extracellular pathogens and vacuolar-dwelling pathogens, such as S. typhimurium and M. tuberculosis.
• Due to the fact that CD4+ T-cell responses are also necessary for the complete priming, activation, and differentiation of CD8+ T-cell responses, it is difficult to directly attribute immunological vulnerability to CD4+ T-cell deficit.
• The HIV epidemic has made the ramifications and complications of CD4+ T-cell depletion evident.
• The initial stage in the processing of MHC class II antigens is the translocation and assembly of MHC class II α- and β-chains in the endoplasmic reticulum, a mechanism regulated by a chaperone, the invariant chain.
• In the endoplasmic reticulum, MHC class II molecules do not bind antigenic peptides. The α- and β-chains fold by substituting the membrane-bound protein invariant chain for peptide.
• As the complex exits the endoplasmic reticulum, traverses the Golgi complex, and travels to the endosomal compartments, a fragment of the invariant chain fills the MHC class II groove.
• On acidification of the endosomal compartment, proteases such as cathepsin D and cathepsin B are activated and destroy all portions of the invariant chain excluding the section that is protected by the MHC class II groove.
• During the degradation process, MHC class II molecules might interact with endocytosed antigens in the acidified endosome compartment known as MIIC.
• The invariant chain segment is replaced by a proteolytically produced peptide in a complex, topologically demanding sequence of events.
• Similar to MHC class I molecules, MHC class II molecules are selective with regard to peptide binding; however, because the groove is open and can accommodate peptides in different registers, the range of peptides that are bound is expanded and the binding rigour is loosened.
• HLA-DM, an MHC-encoded, MHC-class II-like protein that dwells in MIICs, accelerates the binding mechanism of MHC class II peptides.
• Recent crystallographic investigations demonstrate that HLA-DM stabilises empty MHC class II proteins in a conformation that favours the insertion of a high-affinity peptide. HLA-DM catalyses the extraction of the invariant chain peptide fragment from the MHC class II molecule.
• In addition to proteases, GILT (gamma interferon–inducible lysosomal thioreductase) is involved in the denaturation of certain antigens prior to their breakdown and presentation by MHC class II molecules.
• GILT has also been linked to the activation of the primary secreted virulence factor of the bacterial pathogen L. monocytogenes, listeriolysin-O, offering an exceptional example of microbial exploitation of antigen-processing pathways.
• Upon peptide attachment in the MIIC, MHC class II molecules travel to the cell surface, where CD4+ T lymphocytes can detect the MHC class II/peptide complex. MHC class II molecules undergo re-internalization, and it is possible that these complexes recycle to endosomal compartments and acquire new peptides before returning to the cell surface.
• Uncertainty surrounds the contribution of this pathway to the MHC class II antigen processing pathway during immunological responses to infection.

What is CD1?

• The CD1 family contains antigen-presenting molecules that are structurally similar to MHC class I molecules and are associated with β2-microglobulin.
• Human chromosome 1’s CD1 locus has five different genes that code for the proteins CD1a through CD1e.
• Consistent with their structural resemblance to MHC class I, CD1 molecules are often strongly expressed on antigen-presenting cells, present antigens for TCR recognition, and interact with T lymphocytes. CD1 molecules differ significantly from MHC molecules in the class of displayed antigens.
• In contrast to the peptide antigens presented by the MHC, CD1 proteins present lipid and glycolipid antigens to T cells, a discovery that dramatically increased the number of antigens recognised by T lymphocytes.
• The illustration depicts the structure of the human CD1b molecule displaying ganglioside GM2.
• This section discusses the antigens given by CD1, the lymphocytes that identify these antigens, and the significance of this antigen-presenting system in the defence of the host against specific infectious illnesses.

CD1 Protein Structure

• CD1 proteins are transmembrane proteins with a brief intracellular domain. The extracellular component of CD1 consists of three antigen-binding domains, while particular patterns in the intracellular region regulate CD1 trafficking to intracellular compartments (see later discussion).
• The antigen-binding groove formed by the extracellular domains of CD1 molecules is structurally similar to the peptide-binding groove of the MHC.
• Consistent with CD1’s function of lipid presentation, the antigen-binding groove has several hydrophobic pockets that can accept the aliphatic chains of lipid antigens.
• The three-dimensional structure of mouse CD1266 and human CD1b265 reveals a complex network of hydrophobic channels capable of accommodating various lipids with differing aliphatic chain lengths.
• In contrast to the five CD1 isoforms found in humans, mice lack CD1a, CD1b, and CD1c and possess two copies of the CD1d gene.
• This significant difference between mouse and human CD1 hampers the experimental analysis of CD1 function since genetically modified mice cannot be used to examine the function of CD1a, CD1b, and CD1c.

Antigens Presented by CD1

• T lymphocytes are presented with lipid and glycolipid antigens from various bacteria and fungi by CD1. CD1-presented antigens include diacylglycerols from S. pneumoniae, a fungal glycosphingolipid (i.e., asperamide B), and a marine sponge-derived synthetic glycolipid, -galactosyl ceramide.
• These last three chemicals are delivered to NKT cells in a CD1-restricted way, and -galactosyl ceramide has been utilised extensively to explore the effect of CD1d-mediated NKT-cell activation on the host’s defence against infections and malignancies.
• CD1 is essential for delivering mycobacterial lipids to T lymphocytes, particularly glycosylated and free mycolic acids and lipoarabinomannan, two key lipid and glycolipid components of M. tuberculosis cell envelope.
• Further investigations of the structure of CD1-presented lipids indicated that T-cell recognition of CD1b-presented glycolipids was extraordinarily sensitive to the fine structure of the carbohydrate head group, but rather insensitive to structural changes in the lipid tail.
• Together, these observations and the subsequent characterization of CD1c-presented mycobacterial isoprenoid glycolipids have led to a model of CD1 lipid antigen presentation in which the hydrophilic head group is exposed and available for TCR interactions while the hydrophobic lipid tails of the antigen are bound within the hydrophobic pockets of the CD1 protein.

Cell Biology of CD1 Antigen Processing and Loading

• CD1 isoforms are abundant in diverse intracellular compartments, indicating that each isoform of CD1 has evolved to detect microbial antigens that are present in unique regions of the endosomal lysosomal network.
• The figure provides a summary of CD1 trafficking pathways. Endocytosis results in the internalisation of all CD1 isoforms at the cell surface. C D1a is primarily found in early endosomes, CD1c in late endosomes, and CD1b/d in both late endosomes and lysosomes.
• Specific amino acid residues in the short intracellular tails of CD1 isoforms bind to cytosolic adaptor molecules that mediate organelle trafficking, thereby targeting these isoforms to their respective compartments.
• Although the specific physiological ramifications of this trafficking pattern for the immune response are still being explored, each CD1 isoform may survey a unique intracellular compartment for unique lipid structures from unique pathogens.

References

• Kindt, T., Goldsby, R., Osborne, B., Kuby, J. and Kuby, J. (2007). Kuby immunology. New York: W.H. Freeman.
• Natarajan K, Li H, Mariuzza RA, Margulies DH. MHC class I molecules, structure and function. Rev Immunogenet. 1999;1(1):32-46. PMID: 11256571.
• Li XC, Raghavan M. Structure and function of major histocompatibility complex class I antigens. Curr Opin Organ Transplant. 2010 Aug;15(4):499-504. doi: 10.1097/MOT.0b013e32833bfb33. PMID: 20613521; PMCID: PMC3711407.
• Natarajan, K & Li, Hongmin & Mariuzza, RA & Margulies, David. (1999). MHC class I molecules, structure and function. Reviews in immunogenetics. 1. 32-46. 
• Wieczorek, M., Abualrous, E. T., Sticht, J., Álvaro-Benito, M., Stolzenberg, S., Noé, F., & Freund, C. (2017). Major Histocompatibility Complex (MHC) Class I and MHC Class II Proteins: Conformational Plasticity in Antigen Presentation. Frontiers in Immunology, 8. doi:10.3389/fimmu.2017.00292 
• Janeway CA Jr, Travers P, Walport M, et al. Immunobiology: The Immune System in Health and Disease. 5th edition. New York: Garland Science; 2001. The major histocompatibility complex and its functions. Available from: https://www.ncbi.nlm.nih.gov/books/NBK27156/
• Hohl, T. M. (2015). Cell-Mediated Defense against Infection. Mandell, Douglas, and Bennett’s Principles and Practice of Infectious Diseases, 50–69.e6. doi:10.1016/b978-1-4557-4801-3.00006-0 
• Mak, T. W., & Saunders, M. E. (2006). MHC: The Major Histocompatibility Complex. The Immune Response, 247–277. doi:10.1016/b978-012088451-3.50012-0 
• The Major Histocompatibility Complex. (2014). Primer to the Immune Response, 143–159. doi:10.1016/b978-0-12-385245-8.00006-6 
• Rammensee, H.G. (1993). Structure and Function of MHC Class I Molecules. In: Eibl, M.M., Huber, C., Peter, H.H., Wahn, U. (eds) Symposium in Immunology I and II. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-78087-5_9