1 INTRODUCTION

The ability to discriminate between self and nonself lies at the centre of the immune defence. Major histocompatibility complex (MHC) molecules are key players in this process. By displaying small fragments of the proteome on the plasma membrane of cells, MHC molecules are crucial for T-cell-mediated immune recognition of both extracellular and intracellular pathogens, as well as other cellular abnormalities. While expression of MHC class II molecules is predominantly restricted to professional antigen-presenting cells and recognized by CD4+ T cells, MHC class I molecules are present on all nucleated cells in the body, recognized by CD8+ T cells, and monitored by a number of other cells types including natural killer (NK) cells. However, only a small fraction of the peptides that are capable of associating with MHC make it to the plasma membrane for immune recognition (Yewdell, 2006; Yewdell et al., 2003). There are a variety of reasons for this, including the affinity of a peptide for a particular MHC molecule. Furthermore, we now appreciate that as an MHC molecule navigates its way through the complexities of the antigen processing and presentation (APP) pathway, peptide exchange also occurs on MHC, which ultimately controls the cargo that gets presented for immune recognition. Tapasin and HLA-DM, peptide editors for MHC class I and II molecules respectively, which have been extensively studied for over 20 years, are known to strongly influence the immunopeptidome (Chen & Bouvier, 2007; Denzin & Cresswell, 1995; Fleischmann et al., 2015; Garbi et al., 2000; Pos et al., 2012; Sadasivan et al., 1996; Wearsch & Cresswell, 2007; Williams et al., 2002). However, in 2015, a second peptide editor for MHC class I was discovered, the tapasin-related molecule TAPBPR (Boyle et al., 2013; Hermann et al., 2015; Morozov et al., 2016), thus adding further complexity regarding the way the immunopeptidome presented on MHC class I is shaped. A recent study exploring the ability of TAPBPR to bind to 97 distinct MHC-I allotypes, spanning a broad range of human leukocyte antigens (HLA) -A, -B and -C molecules, revealed that polymorphisms in MHC-I strongly influence whether a particular allotype is susceptible to peptide editing by TAPBPR (Ilca et al., 2019). This work emphasizes that polymorphisms in MHC-I not only influence peptide presentation by altering the properties of pockets in the peptide-binding groove, but also by influencing how an MHC-I molecule navigates its way to the plasma membrane. Here, we review what is currently known regarding how variation in HLA class I molecules influences their interaction with key components of the APP pathway.

2 MHC-I ANTIGEN PROCESSING AND PRESENTATION PATHWAY

Before delving into details regarding specific components of the APP pathway, a brief overview of the pathway is needed (for an in-depth review, see Blum et al., 2013) (Figure 1). Firstly, the MHC-I heavy chain (also known as the α chain) is cotranslationally inserted into the endoplasmic reticulum (ER), where it folds and associates with the β-2 microglobulin (β2m) chain to form a heterodimer. This process is aided by the chaperones calnexin and calreticulin (Jackson et al., 1994; Nossner & Parham, 1995). Peptides that are presented by MHC-I generally originate from proteasomal degradation of cytosolic proteins (Kloetzel, 2001). In order for peptides to meet the peptide-receptive MHC-I, they need to be transported into the ER. This is mediated by the transporter associated with antigen processing (TAP), which is a heterodimer of TAP1 and TAP2 proteins in humans (Abele & Tampe, 2018; van Endert et al., 1994; Kelly et al., 1992; Meyer et al., 1994; Neefjes et al., 1993). Once in the ER, peptides can be further trimmed by ER-associated aminopeptidases, ERAP1 and ERAP2, to generate shorter peptides that fit into the MHC-I peptide-binding groove (Hammer et al., 2006; Saric et al., 2002; Saveanu et al., 2005; Serwold et al., 2001, 2002; York et al., 2002). The peptide-receptive MHC-I/β2m heterodimer is recruited to a multiprotein complex called the peptide loading complex (PLC), which consists of the TAP transporter, the chaperone calreticulin, the disulphide isomerase ERp57 and tapasin (Blees et al., 2017). This is the main site of peptide loading onto MHC-I. The PLC stabilizes the peptide-receptive MHC-I and provides a peptide-rich environment to enable peptide loading. Once loaded with an optimal peptide, peptide–MHC-I (pMHC) complexes dissociate from the PLC and were thought to subsequently traffic to the cell surface via the Golgi. However, we now appreciate that pMHC complexes undergo further quality control in the ER and cis-Golgi. One such checkpoint is controlled by UDP-glucose:glycoprotein glucosyltransferase 1 (UGT1) (Ito et al., 2015). This can reglycosylate the N-linked glycan on MHC-I loaded with suboptimal peptides, which subsequently recycles MHC-I back to the PLC for another attempt of peptide loading (Wearsch et al., 2011; Zhang et al., 2011). In addition, MHC-I can undergo further peptide editing by TAPBPR, a tapasin-related protein that can either directly edit peptides (Hermann et al., 2015; Morozov et al., 2016) and interacts with UDP-glucose:glycoprotein glucosyltransferase 1 (UGT1) to recycle peptide-receptive MHC-I complexes back to the PLC (Neerincx et al., 2017; Zhang et al., 2011). TAPBPR functions outside of the PLC, and while it does not contain an ER-retention motif, is mainly localized in the ER and Golgi (Boyle et al., 2013).

Overview of the MHC class I antigen processing and presentation pathway. (1) Proteasomal degradation of cytosolic proteins results in the generation of peptides. (2) Peptides are transported into the ER via the transporter associated with antigen processing (TAP) complex. (3) Peptides can be further trimmed by ER-associated aminopeptidases (ERAP). (4) MHC-I folding and heterodimerization with β2m is aided by chaperones calnexin and calreticulin (CRT). (5) Peptide loading on MHC-I occurs in the peptide loading complex (PLC), a multiprotein complex that includes the peptide editor tapasin. (6) UDP-glucose:glycoprotein glucosyltransferase 1 (UGT1) assists in MHC-I quality control. (7) Additional peptide editing by TAPBPR further shapes the immunopeptidome presented on MHC-I. (8) MHC-I molecules loaded with high-affinity peptides are subsequently released to the plasma membrane for immune monitoring. This figure was created with BioRender.com

3 THE IMPACT OF DEFECTS IN THE ANTIGEN PROCESSING AND PRESENTATION PATHWAY

The critical role of a protein is often best revealed by the phenotype observed upon its loss. The loss of specific components of the APP pathway can have drastic consequences on MHC-I surface expression levels and stability, as well as the size and quality of the presented peptide repertoire. As a result, components of the MHC-I pathway are often directly targeted by viruses to avoid immune recognition (Schuren et al., 2016). Furthermore, defects in the components of the MHC-I APP pathway often arise in tumours, offering a selective advantage in the context of an anti-tumour T-cell response (Dersh et al., 2021; Friedrich et al., 2019; Jhunjhunwala et al., 2021; Leone et al., 2013). Such defects are frequently associated with increased metastasis and reduced patient survival.

Loss of either TAP1 or TAP2 results in dramatically reduced levels of peptide-loaded MHC-I complexes at the cell surface (Chen et al., 1996; Kelly et al., 1992) and is the cause of the inherited immune deficiency type I bare lymphocyte syndrome (Gadola et al., 2000; Raghavan, 1999). Additionally, TAP1 is commonly downregulated by herpesviruses (Hansen & Bouvier, 2009) and in many cancers (Chang et al., 2003). Loss of tapasin has similar effects on MHC-I expression both in human cell lines (Ortmann et al., 1997) and in mice, where tapasin loss is also associated with an altered peptide repertoire and impaired immune responses (Garbi et al., 2000). Similarly, calreticulin-deficient murine cell lines exhibit reduced surface MHC-I expression and impaired peptide loading (Gao et al., 2002). Additionally, some myeloproliferative diseases are associated with mutant forms of calreticulin, which do not localize to the PLC (Arshad & Cresswell, 2018). When these mutants are expressed in HEK293T cells that lack wild-type calreticulin, it leads to reduced surface MHC-I levels due to impaired peptide loading (Arshad & Cresswell, 2018). Components of the APP that are not part of the PLC have more subtle effects on the peptide repertoire. Loss of ERAP in mice leads to an altered peptide repertoire characterized by the presence of longer, so-called ‘unedited’ peptides and the loss of some trimmed peptides (Blanchard et al., 2010; Nagarajan et al., 2016). While there are no comparable studies in human cells, human ERAP1 has a similar trimming activity to mouse ERAP and has been shown to preferentially trim 9-16mer peptides to 8–9 residues (Chang et al., 2005; Evnouchidou & van Endert, 2019; Gandhi et al., 2011; Kochan et al., 2011; Nguyen et al., 2011; Saric et al., 2002; York et al., 2002). Finally, while loss of TAPBPR in human cell lines only significantly affects surface MHC-I levels if tapasin is also absent, it does lead to a larger peptide repertoire with a lower proportion of peptides containing canonical anchor residues, which may indicate a role for TAPBPR in optimizing the peptide repertoire by editing low-affinity peptides (Hermann et al., 2015).

4 VARIATION IN MHC-I AND DISEASE

Given their role in immune recognition and clearance of infected cells and tumours, MHC-I genes have evolved as the most polymorphic in the human genome. To date, 6,425 HLA-A, 7,754 HLA-B and 6,329 HLA-C alleles have been described. After accounting for synonymous mutations, these alleles encode 3,929 HLA-A, 4,885 HLA-B and 3,719 HLA-C proteins (allotypes) (http://hla.alleles.org/nomenclature/stats.html). The vast majority of variation is found in or near the peptide-binding groove and thus affects the presented peptidome.

The high degree of variation in MHC represents a major barrier to transplantation. Furthermore, associations between inheritance of particular MHC-I alleles and susceptibility to a variety of conditions have been, and continue to be, the subject of intense study (Matzaraki et al., 2017; Trowsdale & Knight, 2013). This includes susceptibility to autoimmune conditions, hypersensitivities, cancer and infectious diseases. For example, HLA-B*27 is associated with the spondyloarthropathies (Bowness, 2015), HLA-B*51 is linked to Behçet's disease (McGonagle et al., 2015), and HLA-C*06:02 is associated with psoriasis (Okada et al., 2014). HLA-B*57:01 is linked to drug hypersensitivities (Illing et al., 2012), and multiple associations exist between polymorphisms in MHC peptide-binding regions and allergic rhinitis (Waage et al., 2018).

In the context of cancer, a genome-wide association study followed by fine mapping of the MHC locus has identified polymorphisms associated with susceptibility to lung cancer (Ferreiro-Iglesias et al., 2018). Furthermore, cancer patients exhibiting a diverse HLA class I genotype often respond better to immune checkpoint inhibition (Chowell et al., 2018, 2019) and the HLA-B44 supertype has been specifically linked to survival outcome in both melanoma and non-small-cell lung cancer (Chowell et al., 2018; Cummings et al., 2020).

For infectious diseases, risk variants in the MHC-I region have been found for many common infections including herpesvirus infections, hepatitis A and B, bacterial meningitis and yeast infections (Tian et al., 2017). A classic example is the hierarchal relationship of inheritance of particular MHC-I being linked to control of HIV and progression time to AIDS (Naranbhai & Carrington, 2017). Finally, while inheritance of particular MHC-I alleles has been linked to susceptibility to SARS-CoV (Lin et al., 2003; Ng et al., 2004; Wang et al., 2011), a recent study suggests MHC-I epitope load is inversely correlated with population mortality from SARS-CoV-2, with MHC-I molecules presenting more unique SARS-CoV-2 epitopes potentially being associated with lower mortality (Wilson et al., 2021).

Although the underlying causes linking specific MHC-I to most diseases are still unclear, many of the polymorphisms associated with disease risk are found in the peptide-binding region of MHC molecules, suggesting a role for specific peptide motifs and interaction with T cells.

5 WHAT IMPACT DOES VARIATION IN MHC-I HAVE ON ITS INTERACTION WITH APP COMPONENTS?

While it is perhaps generally assumed that all MHC-I molecules obtain peptide in a homogenous fashion, the incredible diversity in MHC-I means there is potentially substantial variation governing the processes through which they gain peptide. Thus, it is worth considering whether our understanding of the interactions between MHC-I and components of the APP, and the dependency of MHC-I on the pathway, holds true across this huge diversity of MHC-I allotypes. Here, we summarize what is currently known regarding the impact of polymorphisms in human MHC-I on their interactions with key players in the MHC-I APP pathway: TAP/tapasin, ERAP and TAPBPR.

5.1 TAP and tapasin

There is a significant variability in how dependent MHC-I allotypes are on the presence of TAP and tapasin for peptide acquisition and surface expression (Bashirova et al., 2020; Geng et al., 2018; Rizvi et al., 2014). The level of MHC-I dependence on TAP and tapasin is correlated, which may be explained by the fact that both are integral to the PLC (Geng, Zaitouna, et al., 2018). One notable example of a TAP- and tapasin-independent allotype is HLA-B*44:05, which does not show reduced surface expression on TAP-deficient cells and has a low tapasin dependence score based on similar surface expression studies (Bashirova et al., 2020; Rizvi et al., 2014; Williams et al., 2002). Interestingly, the closely related allotype HLA-B*44:02 is highly dependent on both TAP and tapasin. These two proteins only differ by one residue, namely residue 116, which is a tyrosine in HLA-B*44:05 and an aspartic acid in HLA-B*44:02 (Williams et al., 2002). Residue 116 is located close to the F-pocket of the peptide-binding groove and affects the binding motif of the groove, with HLA-B*44:05 preferring phenylalanine at the C-terminus of the bound peptide, whereas HLA-B*44:02 can bind phenylalanine or tyrosine (Zernich et al., 2004).

There are multiple models that explain how MHC-I proteins can be independent of either TAP or tapasin. While some TAP-independent allotypes have been shown to traffic to the cell surface in a stable, peptide-receptive form (Geng, Zaitouna, et al., 2018) and can be recognized by CD8+ T cells (Geng et al., 2018), cytotoxic T cells (CTLs) targeting specific antigens presented by TAP-deficient cells can be found in healthy donors for a range of allotypes (Lampen et al., 2010). Independence from TAP can be linked to the thermal stability for certain allotypes, as well as their ability to present peptides derived from the transmembrane domains and signal sequences of cellular proteins that do not require TAP to enter the ER (Geng, Zaitouna, et al., 2018). Similarly, a comparison of HLA-B proteins showed that tapasin-independent allotypes exhibit reduced aggregation and bind peptide more readily when refolded in vitro, compared with more tapasin-dependent allotypes (Rizvi et al., 2014). Among HLA-B allotypes, the presence of the Bw4 epitope shows a positive correlation with tapasin dependence, whereas the Bw6 epitope is associated with tapasin independence, but this epitope alone does not account for all of the variation in dependence across HLA-Bs (Rizvi et al., 2014). A recent study has shown that across all MHC-I allotypes, there is a wide range of tapasin dependence and that more tapasin-independent allotypes present a broader peptide repertoire, possibly indicating an ability to acquire peptides outside the PLC in the absence of tapasin (Bashirova et al., 2020). This study also linked the overall tapasin dependence of an HLA complement to progression to AIDS in HIV-positive individuals (Bashirova et al., 2020).

5.2 ERAP

There are two models that describe how ERAPs trim peptides (for an in-depth review, see Evnouchidou & van Endert, 2019). The first model is known as the molecular ruler. In this model, ERAP1 trims free peptides in solution, with a preference for peptides 9–16 residues in length and with a hydrophobic C-terminal side chain. The C-terminus of the peptide binds a hydrophobic binding site, and the N-terminus fits into the active site, which then trims the peptide until it is 8–9 residues long, at which point the active site no longer has access to the N-terminus of the peptide and the trimmed product dissociates from the enzyme (Chang et al., 2005). While this model assumes a preference for peptides with a hydrophobic C-terminus, the activity of ERAP1 is independent of other peptide motifs that would influence binding to MHC-I, and thus presumably should not lead to some MHC-I alleles depending on the availability of ERAP1-trimmed peptides more than others. The second model describes how ERAP1/ERAP2 heterodimers can trim peptide precursors already bound to MHC-I (Chen et al., 2016). While this model is based on studies performed on HLA-B*08:01 and has not been tested on a broader range of HLA-I allotypes, it could indicate a novel role for an ERAP complex as a peptide editor on MHC-I, in which case the ability of the ERAP complex to bind to MHC-I might be influenced by polymorphisms in MHC-I and thus lead to a greater variation of ERAP dependence among MHC-I allotypes.

There are few examples of specific allotypes that are highly ERAP-dependent or ERAP-independent. One such example is HLA-B*27:05, which exhibits reduced surface expression levels in ERAP1-deficient cells or cells expressing certain alleles of ERAP1. These ERAP1 alleles are also associated with the development of ankylosing spondylitis when co-expressed with HLA-B*27:05 (Pepelyayeva & Amalfitano, 2019). HLA-B*27 has also shown to be ERAP-dependent in a humanized mouse model of influenza infection, where loss of ERAP affected surface expression of HLA-B*27 in naïve and flu-infected mice, as well as CTL responses restricted to the nucleoprotein 383–391 epitope on HLA-B*27. In contrast, loss of ERAP had no effect on HLA-B*07 expression or CTL responses to the HLA-B*07-restricted epitope from nucleoprotein 418–426 (Akram et al., 2014).

5.3 TAPBPR

TAPBPR is the most recently identified component of the MHC class I antigen processing and presentation pathway (Boyle et al., 2013). While TAPBPR shares tapasin's ability to function as a peptide editor (Hermann et al., 2015; Morozov et al., 2016), the influence of TAPBPR on MHC-I surface expression and peptide presentation appears to be comparatively subtle (Hermann et al., 2015). Thus, perhaps the true role of TAPBPR in the APP can only be understood by studying its activity across a range of MHC-I allotypes. However, the fact that the loss of TAPBPR does not result in a radical surface MHC-I phenotype makes large studies on functional MHC-I dependence on TAPBPR such as the ones performed for TAP and tapasin more difficult. The discovery that recombinant TAPBPR can be used to edit peptides on MHC-I at the plasma membrane, combined with the fact that fluorescent MHC-I-binding peptides can be used to quantify peptide editing, made studying the functional interaction of TAPBPR with MHC-I in a cellular model possible (Ilca et al., 2018; Ilca & Boyle, 2020). Subsequent studies using recombinant TAPBPR, TAPBPRKO and overexpression models, as well as mutant forms of TAPBPR, revealed that TAPBPR exhibits a preference for some MHC-I allotypes over others. In particular, the ability of TAPBPR to edit peptides is strongest on HLA-A molecules, especially members of the A2 and A24 superfamilies, compared with HLA-B and HLA-C allotypes (Ilca et al., 2019). Furthermore, a loop of TAPBPR comprised of residues K22-D35 has been shown to play a role in peptide editing by competing with the C-terminus of the bound peptide and either promoting peptide dissociation (Ilca, Neerincx, Hermann, et al., 2018) or preventing peptide rebinding (Sagert et al., 2020). While one structure of the TAPBPR-MHC-I complex suggested this loop was located close to the F-pocket of the peptide-binding groove (Thomas & Tampe, 2017), another structure, published simultaneously, could not assign the loop to a specific position in the complex due to the less-defined crystallographic data in this region (Jiang et al., 2017). MHC-I allotypes that accommodate hydrophobic residues in their F-pocket are particularly dependent on the activity of this loop and generally interact more strongly with TAPBPR (Ilca, Neerincx, Hermann, et al., 2018). In particular, residues 114 and 116 of MHC-I play a role in determining the interaction of MHC-I with TAPBPR, as shown in mutagenesis studies involving pairs of allotypes that only differ in these residues, such as HLA-A*68:01/A*68:02, B*27:05/B*27:09 and B*35:01/B*35:03, and by mutating residue 114 of B*44:05 from aspartic acid to histidine. Additionally, HLA-B*44:05 interacts more strongly with TAPBPR than with HLA-B*44:02, as shown by co-immunoprecipitation and in vitro peptide editing experiments, which stands in contrast to the independence of HLA-B*44:05 surface levels from TAP or tapasin loss (Ilca et al., 2019). Furthermore, HLA-A allotypes that exhibited increased binding to TAPBPR, and were consequently more susceptible to TAPBPR-mediated peptide exchange (Ilca et al., 2019), were intriguingly found by Bashirova et al. (2020) to be tapasin-independent allotypes. This inverse relationship may indicate a complementarity between TAPBPR and the PLC in the generation of optimal peptide–MHC-I complexes.

6 CONCLUSIONS

In summary, while the respective functions of different components of the APP pathway have been studied in detail, these functions may differ depending on the MHC-I allotypes present, and our overall understanding of the pathway cannot be complete without taking this into account. As discussed, allotype-specific interactions between MHC-I and its chaperones can have consequences for disease susceptibility, and thus, it is important to bear in mind that the vast variation in the MHC locus not only affects the presented peptidome, but also affects how that peptidome is acquired and refined. Finally, while there is no comprehensive understanding of the mechanisms that account for the dependence of MHC-I on different components of the APP, some factors, such as overall stability of the MHC-I-β2m heterodimer, the promiscuity of the peptide-binding groove and the configuration of the F-pocket, seem to play important roles.

ACKNOWLEDGEMENTS

This work was supported by the Wellcome [Grant nos.: 220012/Z/19/Z and 219479/Z/19/Z].

AUTHOR CONTRIBUTIONS

AA conceptualized the work, drafted the original manuscript, and reviewed and edited the manuscript. LHB conceptualized the work, drafted the original manuscript, reviewed and edited the manuscript, and performed supervision and funding acquisition.

Data sharing is not applicable—no new data are generated.

REFERENCES

Abele, R., & Tampe, R. (2018). Moving the cellular peptidome by transporters. Frontiers in Cell and Developmental Biology, 6, 43. https://doi.org/10.3389/fcell.2018.00043 Akram, A., Lin, A., Gracey, E., Streutker, C. J., & Inman, R. D. (2014). HLA-B27, but not HLA-B7, immunodominance to influenza is ERAP dependent. The Journal of Immunology, 192(12), 5520– 5528. https://doi.org/10.4049/jimmunol.1400343 Arshad, N., & Cresswell, P. (2018). Tumor-associated calreticulin variants functionally compromise the peptide loading complex and impair its recruitment of MHC-I. Journal of Biological Chemistry, 293(25), 9555– 9569. https://doi.org/10.1074/jbc.RA118.002836 Bashirova, A. A., Viard, M., Naranbhai, V., Grifoni, A., Garcia-Beltran, W., Akdag, M., Yuki, Y., Gao, X., O'hUigin, C., Raghavan, M., Wolinsky, S., Bream, J. H., Duggal, P., Martinson, J., Michael, N. L., Kirk, G. D., Buchbinder, S. P., Haas, D., Goedert, J. J., … Carrington, M. (2020). HLA tapasin independence: Broader peptide repertoire and HIV control. Proceedings of the National Academy of Sciences USA, 117(45), 28232– 28238. https://doi.org/10.1073/pnas.2013554117 Blanchard, N., Kanaseki, T., Escobar, H., Delebecque, F., Nagarajan, N. A., Reyes-Vargas, E., Crockett, D. K., Raulet, D. H., Delgado, J. C., & Shastri, N. (2010). Endoplasmic reticulum aminopeptidase associated with antigen processing defines the composition and structure of MHC class I peptide repertoire in normal and virus-infected cells. The Journal of Immunology, 184(6), 3033– 3042. https://doi.org/10.4049/jimmunol.0903712 Blees, A., Januliene, D., Hofmann, T., Koller, N., Schmidt, C., Trowitzsch, S., Moeller, A., & Tampé, R. (2017). Structure of the human MHC-I peptide-loading complex. Nature, 551(7681), 525– 528. https://doi.org/10.1038/nature24627 Blum, J. S., Wearsch, P. A., & Cresswell, P. (2013). Pathways of antigen processing. Annual Review of Immunology, 31, 443– 473. https://doi.org/10.1146/annurev-immunol-032712-095910 Bowness, P. (2015). Hla-B27. Annual Review of Immunology, 33, 29– 48. https://doi.org/10.1146/annurev-immunol-032414-112110 Boyle, L. H., Hermann, C., Boname, J. M., Porter, K. M., Patel, P. A., Burr, M. L., Duncan, L. M., Harbour, M. E., Rhodes, D. A., Skjødt, K., Lehner, P. J., & Trowsdale, J. (2013). Tapasin-related protein TAPBPR is an additional component of the MHC class I presentation pathway. Proceedings of the National Academy of Sciences USA, 110(9), 3465– 3470. https://doi.org/10.1073/pnas.1222342110 Chang, C. C., Campoli, M., & Ferrone, S. (2003). HLA class I defects in malignant lesions: What have we learned? The Keio Journal of Medicine, 52(4), 220– 229. https://doi.org/10.2302/kjm.52.220 Chang, S. C., Momburg, F., Bhutani, N., & Goldberg, A. L. (2005). The ER aminopeptidase, ERAP1, trims precursors to lengths of MHC class I peptides by a "molecular ruler" mechanism. Proceedings of the National Academy of Sciences USA, 102(47), 17107– 17112. https://doi.org/10.1073/pnas.0500721102 Chen, H. L., Gabrilovich, D., Tampe, R., Girgis, K. R., Nadaf, S., & Carbone, D. P. (1996). A functionally defective allele of TAP1 results in loss of MHC class I antigen presentation in a human lung cancer. Nature Genetics, 13(2), 210– 213. https://doi.org/10.1038/ng0696-210 Chen, H., Li, L., Weimershaus, M., Evnouchidou, I., van Endert, P., & Bouvier, M. (2016). ERAP1-ERAP2 dimers trim MHC I-bound precursor peptides; implications for understanding peptide editing. Scientific Reports, 6, 28902. https://doi.org/10.1038/srep28902 Chen, M., & Bouvier, M. (2007). Analysis of interactions in a tapasin/class I complex provides a mechanism for peptide selection. The EMBO Journal, 26(6), 1681– 1690. https://doi.org/10.1038/sj.emboj.7601624 Chowell, D., Krishna, C., Pierini, F., Makarov, V., Rizvi, N. A., Kuo, F., Morris, L. G. T., Riaz, N., Lenz, T. L., & Chan, T. A. (2019). Evolutionary divergence of HLA class I genotype impacts efficacy of cancer immunotherapy. Nature Medicine, 25(11), 1715– 1720. https://doi.org/10.1038/s41591-019-0639-4 Chowell, D., Morris, L. G. T., Grigg, C. M., Weber, J. K., Samstein, R. M., Makarov, V., Kuo, F., Kendall, S. M., Requena, D., Riaz, N., Greenbaum, B., Carroll, J., Garon, E., Hyman, D. M., Zehir, A., Solit, D., Berger, M., Zhou, R., Rizvi, N. A., & Chan, T. A. (2018). Patient HLA class I genotype influences cancer response to checkpoint blockade immunotherapy. Science, 359(6375), 582– 587. https://doi.org/10.1126/science.aao4572 Cummings, A. L., Gukasyan, J., Lu, H. Y., Grogan, T., Sunga, G., Fares, C. M., Hornstein, N., Zaretsky, J., Carroll, J., Bachrach, B., Akingbemi, W. O., Li, D., Noor, Z., Lisberg, A., Goldman, J. W., Elashoff, D., Bui, A. A. T., Ribas, A., Dubinett, S. M., … Garon, E. B. (2020). Mutational landscape influences immunotherapy outcomes among patients with non-small-cell lung cancer with human leukocyte antigen supertype B44. Nature Cancer, 1(12), 1167– 1175. https://doi.org/10.1038/s43018-020-00140-1 Denzin, L. K., & Cresswell, P. (1995). HLA-DM induces CLIP dissociation from MHC class II alpha beta dimers and facilitates peptide loading. Cell, 82(1), 155– 165. https://doi.org/10.1016/0092-8674(95)90061-6 Dersh, D., Holly, J., & Yewdell, J. W. (2021). A few good peptides: MHC class I-based cancer immunosurveillance and immunoevasion. Nature Reviews Immunology, 21(2), 116– 128. https://doi.org/10.1038/s41577-020-0390-6 Evnouchidou, I., & van Endert, P. (2019). Peptide trimming by endoplasmic reticulum aminopeptidases: Role of MHC class I binding and ERAP dimerization. Human Immunology, 80(5), 290– 295. https://doi.org/10.1016/j.humimm.2019.01.003 Ferreiro-Iglesias, A., Lesseur, C., McKay, J., Hung, R. J., Han, Y., Zong, X., Christiani, D., Johansson, M., Xiao, X., Li, Y., Qian, D. C., Ji, X., Liu, G., Caporaso, N., Scelo, G., Zaridze, D., Mukeriya, A., Kontic, M., Ognjanovic, S., … Brennan, P. (2018). Fine mapping of MHC region in lung cancer highlights independent susceptibility loci by ethnicity. Nature Communications, 9(1), 3927. https://doi.org/10.1038/s41467-018-05890-2 Fleischmann, G., Fisette, O., Thomas, C., Wieneke, R., Tumulka, F., Schneeweiss, C., Springer, S., Schäfer, L. V., & Tampé, R. (2015). Mechanistic basis for epitope proofreading in the peptide-loading complex. The Journal of Immunology, 195(9), 4503– 4513. https://doi.org/10.4049/jimmunol.1501515 Friedrich, M., Jasinski-Bergner, S., Lazaridou, M.-F., Subbarayan, K., Massa, C., Tretbar, S., Mueller, A., Handke, D., Biehl, K., Bukur, J., Donia, M., Mandelboim, O., & Seliger, B. (2019). Tumor-induced escape mechanisms and their association with resistance to checkpoint inhibitor therapy. Cancer Immunology, Immunotherapy, 68(10), 1689– 1700. https://doi.org/10.1007/s00262-019-02373-1 Gadola, S. D., Moins-Teisserenc, H. T., Trowsdale, J., Gross, W. L., & Cerundolo, V. (2000). TAP deficiency syndrome. Clinical and Experimental Immunology, 121(2), 173– 178. https://doi.org/10.1046/j.1365-2249.2000.01264.x Gandhi, A., Lakshminarasimhan, D., Sun, Y., & Guo, H. C. (2011). Structural insights into the molecular ruler mechanism of the endoplasmic reticulum aminopeptidase ERAP1. Scientific Reports, 1, 186. https://doi.org/10.1038/srep00186 Gao, B., Adhikari, R., Howarth, M., Nakamura, K., Gold, M. C., Hill, A. B.,