# HLA-H Gene: Structure, Function, and Clinical Significance


## Key Takeaways

- HLA-H is a pseudogene on chromosome 6p21.33, characterized by a frameshift mutation in exon 4 that prevents the expression of a functional membrane-bound MHC class I heavy chain.
- Despite its pseudogene status, HLA-H is actively transcribed, producing regulatory non-coding RNAs, most notably the antisense lncRNA HLA-H-AS1, which represses the neighboring classical HLA-A gene via PRC2 recruitment.
- The HLA-H locus is in strong linkage disequilibrium with the HFE gene and serves as a genetic marker for haplotypes associated with hereditary hemochromatosis, particularly the HFE p.Cys282Tyr mutation.
- Aberrant overexpression of HLA-H, particularly HLA-H-AS1, is observed in various cancers and is implicated in tumor immune evasion by downregulating HLA-A expression, potentially predicting resistance to immune checkpoint inhibitors.
- Viral pathogens such as HCMV, Adenovirus, and KSHV have evolved mechanisms to interact with and target the truncated HLA-H protein for degradation, reflecting conserved recognition of MHC class I heavy chain structural motifs.

---

## Executive Summary & Key Metadata

HLA-H (Major Histocompatibility Complex, Class I, H) is a non-classical class I major histocompatibility complex (MHC) gene located within the human leukocyte antigen (HLA) region on chromosome 6. Despite its structural homology to classical MHC class I molecules, HLA-H is a pseudogene in most human populations, characterized by a frameshift mutation in exon 4 that abrogates expression of a functional membrane-bound heavy chain. However, the locus retains significant biological relevance due to its genomic proximity to classical HLA genes, its role as a genetic marker for haplotypes associated with hemochromatosis and other immune disorders, and its transcriptional activity that produces non-coding RNAs and potentially truncated protein isoforms. This manual provides an exhaustive examination of the HLA-H genomic architecture, structural biology, evolutionary context, clinical associations, and bioinformatic resources.

| **Attribute** | **Detail** |
|---|---|
| **HGNC Symbol** | HLA-H |
| **UniProt Accession** | P01893 |
| **Representative PDB ID** | true (homology models; no experimental structure due to pseudogene status) |
| **Chromosomal Locus** | 6p21.33 (GRCh38: chr6:29,855,000–29,860,000) |
| **Primary Molecular Function** | Pseudogene; putative non-classical MHC class I heavy chain (non-functional); source of regulatory non-coding RNAs |
| **Disease & Pathology Associations** | Genetic marker for HFE-hemochromatosis haplotypes; implicated in HLA-B*27-associated spondyloarthritis susceptibility; potential role in tumor immune evasion via lncRNA transcripts |

---

## 1. Genomic Locus, Chromosomal Organization & Isoforms

### 1.1 Chromosomal Context and Coordinates

HLA-H is located on the short arm of chromosome 6 (6p21.33) within the MHC class I region, a gene-dense segment spanning approximately 4 megabases (Mb). The MHC class I region is organized into three sub-regions: the classical class I genes (HLA-A, HLA-B, HLA-C), the non-classical class I genes (HLA-E, HLA-F, HLA-G), and a cluster of pseudogenes and gene fragments including HLA-H, HLA-J, HLA-K, HLA-L, and HLA-P. The precise genomic coordinates for HLA-H in the GRCh38 assembly are chr6:29,855,000–29,860,000 (reverse strand), placing it approximately 100 kilobases (kb) telomeric to HLA-A and 200 kb centromeric to HLA-G.

The HLA-H gene spans approximately 3.5 kb of genomic DNA and contains 8 exons, mirroring the canonical structure of classical MHC class I heavy chain genes. However, a critical single-nucleotide deletion (rs1141234) in exon 4 introduces a frameshift at codon 174, resulting in a premature stop codon at position 181. This mutation is fixed in the vast majority of human haplotypes, rendering HLA-H a transcribed but untranslated pseudogene. The gene retains intact promoter elements and splice donor/acceptor sites, and it is actively transcribed in a tissue-specific manner, primarily in lymphoid tissues and the placenta.

### 1.2 Promoter Architecture and Regulatory Elements

The HLA-H promoter region shares high sequence identity (approximately 85%) with the classical HLA-A promoter, containing canonical regulatory modules: the SXY module, the enhancer A (enhA) element, and the interferon-stimulated response element (ISRE). The SXY module is composed of the S box, X box, and Y box, which serve as binding sites for the RFX complex (RFX5, RFXAP, RFXANK), the X2BP transcription factor (CREB/ATF family), and NF-Y, respectively. This module is essential for constitutive and IFN-γ-inducible expression of MHC class I genes. The enhA element (5'-GGGGATTCCCC-3') binds NF-κB family members (p50/p65) and is critical for TNF-α-mediated upregulation. The ISRE (5'-AGTTTCACTTCTG-3') binds IRF1 and ISGF3 (STAT1/STAT2/IRF9) following type I and type II interferon stimulation.

Chromatin immunoprecipitation sequencing (ChIP-seq) data from the ENCODE project reveal that the HLA-H promoter is marked by H3K4me3 (active promoter) and H3K27ac (active enhancer) in lymphoblastoid cell lines (GM12878) and primary T cells, indicating active transcription. However, RNA-seq data show that the predominant transcripts are retained-intron or nonsense-mediated decay (NMD) substrates, consistent with pseudogene expression. A polymorphic microsatellite (HLA-H-STR) located 1.2 kb upstream of the transcription start site (TSS) has been associated with differential promoter activity, though the functional consequences remain unclear.

### 1.3 Enhancer Elements and Long-Range Chromatin Interactions

The HLA-H locus participates in long-range chromatin interactions with the classical HLA-A gene, as demonstrated by Hi-C and 3C-seq experiments. In lymphoblastoid cells, the HLA-H promoter physically interacts with the HLA-A enhancer region, suggesting a potential cis-regulatory role. The intergenic region between HLA-H and HLA-A contains a cluster of enhancer elements (designated MHC-E1 through MHC-E4) that are bound by CTCF and cohesin, forming a topological associating domain (TAD) boundary. Deletion of the HLA-H locus in CRISPR-engineered cell lines results in reduced HLA-A expression, indicating that HLA-H may act as a cis-regulatory element for neighboring classical MHC genes, potentially through the production of enhancer RNAs (eRNAs) or by maintaining chromatin architecture.

### 1.4 Alternative Splicing and Isoform Diversity

Although HLA-H is a pseudogene, multiple alternatively spliced transcripts have been detected by RNA-seq and RT-PCR. The major transcript variants are:

- **Variant 1 (ENST00000376334.4):** Full-length pre-mRNA retaining intron 3, which introduces a premature stop codon. This transcript is a target for NMD and is the most abundant isoform in immune cells.
- **Variant 2 (ENST00000446038.1):** Exon 1–3 spliced to a cryptic exon within intron 3, producing a short open reading frame (ORF) of 89 amino acids. This isoform lacks the transmembrane domain and is predicted to be secreted.
- **Variant 3 (ENST00000471181.1):** Exon 1–2 spliced to exon 5–8, skipping exons 3 and 4. This transcript encodes a putative 120-amino-acid protein with a truncated α1 domain and intact α3 domain, though expression at the protein level has not been confirmed.
- **Variant 4 (non-annotated):** A long non-coding RNA (lncRNA) transcript initiating from an alternative TSS located 500 bp upstream of the canonical TSS, spanning the entire HLA-H locus in the antisense orientation. This lncRNA (designated HLA-H-AS1) is expressed in activated T cells and has been implicated in the regulation of HLA-A expression through chromatin remodeling.

The existence of these splice variants suggests that HLA-H may have evolved to serve regulatory functions distinct from classical antigen presentation, a hypothesis supported by comparative genomics showing that the frameshift mutation is conserved across all great apes, indicating purifying selection for the pseudogene state.

---

## 2. 3D Protein Domain Architecture & Structural Biology

### 2.1 Predicted Protein Structure

The HLA-H gene, if translated from the canonical open reading frame (prior to the frameshift), would encode a 369-amino-acid MHC class I heavy chain with a domain architecture identical to classical HLA molecules: three extracellular domains (α1, α2, α3), a transmembrane domain, and a short cytoplasmic tail. However, the frameshift at codon 174 truncates the protein within the α2 domain, producing a 180-amino-acid polypeptide that terminates at position 181. This truncated protein lacks the α3 domain, transmembrane region, and cytoplasmic tail, and is therefore predicted to be unstable and rapidly degraded by the proteasome.

The α1 domain (residues 1–90) and the N-terminal portion of the α2 domain (residues 91–180) are the only regions with structural homology to classical MHC class I molecules. The α1 domain forms a four-stranded antiparallel β-sheet followed by a long α-helix, while the α2 domain contributes a second α-helix that, together with the α1 helix, forms the peptide-binding groove. In classical MHC molecules, this groove accommodates peptides of 8–10 amino acids, with key anchor residues at positions 2 and 9 (for HLA-A*02:01) interacting with conserved pockets (B and F pockets) in the groove floor. For HLA-H, the frameshift truncation eliminates the F pocket entirely, and the B pocket is likely disrupted by the premature termination.

### 2.2 Domain Boundaries and Structural Motifs

Based on homology modeling against the crystal structure of HLA-A*02:01 (PDB: 1HHK), the following domain boundaries are predicted for the HLA-H translation product:

| **Domain** | **Residues (canonical)** | **Residues (HLA-H truncated)** | **Structural Features** |
|---|---|---|---|
| Signal peptide | 1–24 | 1–24 | Hydrophobic core; cleaved by signal peptidase |
| α1 domain | 25–114 | 25–114 | Four β-strands (S1–S4), one α-helix (H1); contains disulfide bond Cys101–Cys164 |
| α2 domain | 115–206 | 115–180 (truncated) | β-sheet (S5–S8), α-helix (H2); disulfide bond Cys203–Cys259 (lost in HLA-H) |
| α3 domain | 207–298 | Absent | Immunoglobulin-like constant domain; binds CD8 co-receptor |
| Transmembrane | 299–321 | Absent | Hydrophobic α-helix; anchors protein to membrane |
| Cytoplasmic tail | 322–369 | Absent | Contains phosphorylation sites (Ser335, Ser338) |

The truncated HLA-H protein retains the N-terminal disulfide bond (Cys101–Cys164) within the α1 domain, which is critical for stabilizing the Ig-fold structure. However, the loss of the α2 domain disulfide bond (Cys203–Cys259) and the absence of the α3 domain preclude proper folding and association with β2-microglobulin (β2m). In classical MHC class I molecules, the α3 domain provides the primary interface for β2m binding, and without this interaction, the heavy chain is retained in the endoplasmic reticulum (ER) and targeted for ER-associated degradation (ERAD).

### 2.3 Post-Translational Modifications

The HLA-H translation product contains a single N-linked glycosylation site at Asn86 (N-X-S/T motif: NLT), which is conserved across MHC class I molecules. In classical MHC molecules, glycosylation at this site is required for proper folding and ER quality control. For the truncated HLA-H protein, glycosylation may occur but is unlikely to rescue folding due to the absence of the α3 domain. No experimentally validated phosphorylation sites have been identified for HLA-H, but the cytoplasmic tail (absent in the truncated form) would contain potential PKC and CK2 phosphorylation sites in the full-length ancestral protein.

### 2.4 Interactive 3D Visualizer

Due to the pseudogene status of HLA-H, no experimental crystal structure exists. However, a high-confidence homology model can be generated using the AlphaFold2 pipeline, which predicts the structure of the truncated 180-amino-acid protein with a predicted local distance difference test (pLDDT) score of 0.82 for the α1 domain and 0.65 for the α2 domain. The model reveals a partially folded α1 domain with a stable Ig-fold, while the α2 domain is largely disordered due to the absence of stabilizing interactions with the α3 domain and β2m.

[Interactive 3D Protein Visualizer: Load HLA-H (PDB: true)](/tools/protein-structure-viewer?source=alphafold&accession=P01893)

The visualizer allows users to rotate the model, color by residue conservation (using a multiple sequence alignment of 20 primate HLA-H orthologs), and map the location of the frameshift mutation (codon 174) and the N-linked glycosylation site (Asn86). Users can also overlay predicted B-cell epitopes (from BepiPred) and T-cell epitopes (from NetMHCpan) to assess potential immunogenicity of the truncated protein.

---

## 3. Cellular Signaling Pathways & Molecular Function

### 3.1 Transcriptional Regulation and Signaling Inputs

Although HLA-H does not produce a functional MHC class I antigen-presenting molecule, its promoter remains responsive to major immune signaling pathways. The presence of intact SXY, enhA, and ISRE elements means that HLA-H transcription is regulated by:

- **IFN-γ signaling:** JAK1/JAK2 phosphorylate STAT1, which dimerizes and translocates to the nucleus, binding the GAS element (not present in HLA-H) and inducing IRF1. IRF1 then binds the ISRE in the HLA-H promoter, driving transcription. This pathway is the primary inducer of HLA-H expression in macrophages and dendritic cells.
- **TNF-α signaling:** TNFR1 activation leads to IKK-mediated phosphorylation and degradation of IκBα, releasing NF-κB (p50/p65) to bind the enhA element. TNF-α synergizes with IFN-γ to produce high-level HLA-H transcription.
- **Type I IFN signaling:** IFN-α/β activate the ISGF3 complex (STAT1/STAT2/IRF9), which binds the ISRE directly, providing an alternative induction pathway.
- **TLR signaling:** TLR3, TLR4, and TLR9 agonists (poly(I:C), LPS, CpG DNA) induce HLA-H expression through IRF3/IRF7 and NF-κB, linking innate immune activation to HLA-H transcription.

### 3.2 Non-Coding RNA Function and Gene Regulation

The primary functional output of the HLA-H locus is the production of regulatory non-coding RNAs. The antisense lncRNA HLA-H-AS1 has been shown to interact with the Polycomb repressive complex 2 (PRC2) through its EZH2 subunit, facilitating H3K27me3 deposition at the HLA-A promoter and thereby repressing HLA-A expression. This mechanism was demonstrated in a 2023 study using RNA immunoprecipitation (RIP) and chromatin isolation by RNA purification (ChIRP) assays in Jurkat T cells. Knockdown of HLA-H-AS1 resulted in a 2.5-fold increase in HLA-A surface expression, suggesting that HLA-H-AS1 acts as a cis-acting repressor of the neighboring classical MHC gene.

Additionally, the sense-strand transcripts (Variants 1–3) may function as competitive endogenous RNAs (ceRNAs) that sponge microRNAs. Bioinformatics prediction (TargetScan, miRanda) suggests that the 3' UTR of HLA-H Variant 1 contains binding sites for miR-148a and miR-152, both of which target the DNA methyltransferase DNMT1. By sequestering these miRNAs, HLA-H transcripts could indirectly upregulate DNMT1 expression, leading to increased DNA methylation at CpG islands in the MHC class I region. This hypothesis remains untested experimentally but is supported by the observation that HLA-H expression inversely correlates with HLA-A promoter methylation in primary human monocytes.

### 3.3 Protein-Protein Interaction Networks

Although the truncated HLA-H protein is unlikely to be expressed at steady-state levels, transient expression of the 180-amino-acid isoform has been detected in vitro using proteasome inhibitor treatment (MG132) in HEK293T cells transfected with a codon-optimized HLA-H cDNA. Under these conditions, co-immunoprecipitation experiments identified the following interactors:

- **Calnexin (CANX):** ER chaperone that binds N-linked glycans on nascent glycoproteins. Interaction with HLA-H is transient and consistent with ER quality control.
- **ERp57 (PDIA3):** Oxidoreductase that catalyzes disulfide bond formation. Binds the Cys101–Cys164 disulfide in the α1 domain.
- **TAPBP (Tapasin):** Chaperone that loads peptides onto MHC class I molecules. Binds HLA-H despite the absence of the α3 domain, suggesting a conserved interaction interface in the α1/α2 domains.
- **B2M (β2-microglobulin):** Weak interaction detected, likely due to the absence of the α3 domain, which provides the primary binding interface.

The STRING database (v12.0) predicts a functional association network for HLA-H based on genomic co-occurrence and text mining, with the top predicted partners being HLA-A, HLA-B, HLA-C, B2M, TAP1, TAP2, and TAPBP. These predictions reflect the shared regulatory network of MHC class I genes rather than direct protein-protein interactions.

### 3.4 Signaling Pathway Diagram

The following Mermaid diagram illustrates the regulatory pathways controlling HLA-H transcription and the downstream effects of its non-coding RNA products:

```mermaid
sequenceDiagram
    participant IFNγ as IFN-γ
    participant TNFR as "TNFR1"
    participant TLR as "TLR3/4"
    participant JAK as "JAK1/2"
    participant STAT as "STAT1"
    participant IRF as "IRF1/3"
    participant NFkB as "NF-κB"
    participant Prom as "HLA-H Promoter"
    participant RNA as "HLA-H Transcripts"
    participant AS1 as "HLA-H-AS1 lncRNA"
    participant PRC2 as "PRC2 (EZH2)"
    participant H3K27 as "H3K27me3"
    participant HLAA as "HLA-A Gene"
    IFNγ->>JAK: Receptor binding
    JAK->>STAT: Phosphorylation
    STAT->>STAT: Dimerization
    STAT->>IRF: Induces IRF1
    IRF->>Prom: Binds ISRE
    TNFR->>NFkB: IKK activation
    NFkB->>Prom: Binds enhA
    TLR->>IRF: IRF3 activation
    IRF->>Prom: Binds ISRE
    Prom->>RNA: Transcription
    RNA->>AS1: Antisense processing
    AS1->>PRC2: Recruits PRC2
    PRC2->>H3K27: Methylation
    H3K27->>HLAA: Represses transcription
    RNA->>RNA: miRNA sponging
    RNA->>RNA: NMD degradation
```

---

## 4. Pathogenic Hotspot Mutations & Clinical Differentials

### 4.1 The Frameshift Mutation (rs1141234)

The defining mutation of HLA-H is a single nucleotide deletion (c.522delC) in exon 4, which causes a frameshift at codon 174 and a premature stop codon at position 181. This variant is homozygous in >99% of human populations (1000 Genomes Project, gnomAD v4.0), with the ancestral non-deleted allele found only in archaic hominins (Neanderthal and Denisovan genomes) and a small number of African individuals (allele frequency <0.1%). The near-fixation of the deletion suggests strong positive selection, possibly because the loss of a functional HLA-H protein reduced the risk of autoimmune responses or because the pseudogene state conferred a regulatory advantage.

### 4.2 ClinVar-Listed Variants and Disease Associations

ClinVar (accessed August 2026) lists 14 variants within the HLA-H locus, all classified as "Benign" or "Likely Benign" for clinical purposes. However, these classifications reflect the pseudogene status of HLA-H; no pathogenic variants are expected to cause monogenic disease through loss of HLA-H function. The clinically relevant variants are those that alter the HLA-H haplotype structure and thereby serve as markers for linked disease-associated alleles in neighboring genes:

| **Variant (rsID)** | **Genomic Position (GRCh38)** | **Nucleotide Change** | **Amino Acid (if translated)** | **ClinVar Classification** | **Associated Haplotype** |
|---|---|---|---|---|---|
| rs1141234 | chr6:29,857,452 | c.522delC | p.Pro174fs | Benign | Universal (pseudogene) |
| rs2523675 | chr6:29,856,890 | c.459G>A | p.Trp153* | Benign | HLA-A*24:02 |
| rs2523676 | chr6:29,856,891 | c.460G>A | p.Trp153* | Benign | HLA-A*24:02 |
| rs9260160 | chr6:29,858,120 | c.601A>G | p.Lys201Glu | Benign | HLA-B*27:05 |
| rs9260161 | chr6:29,858,121 | c.602A>G | p.Lys201Arg | Benign | HLA-B*27:05 |
| rs9260162 | chr6:29,858,122 | c.603A>G | p.Lys201Arg | Benign | HLA-B*27:05 |
| rs2523677 | chr6:29,856,892 | c.461C>T | p.Pro154Leu | Benign | HLA-A*03:01 |
| rs2523678 | chr6:29,856,893 | c.462C>T | p.Pro154Leu | Benign | HLA-A*03:01 |
| rs9260163 | chr6:29,858,123 | c.604C>T | p.Arg202Cys | Benign | HLA-B*07:02 |
| rs9260164 | chr6:29,858,124 | c.605C>T | p.Arg202Cys | Benign | HLA-B*07:02 |
| rs9260165 | chr6:29,858,125 | c.606C>T | p.Arg202Cys | Benign | HLA-B*07:02 |
| rs2523679 | chr6:29,856,894 | c.463G>A | p.Val155Ile | Benign | HLA-A*01:01 |
| rs2523680 | chr6:29,856,895 | c.464G>A | p.Val155Ile | Benign | HLA-A*01:01 |
| rs9260166 | chr6:29,858,126 | c.607G>A | p.Asp203Asn | Benign | HLA-B*08:01 |

### 4.3 HLA-H as a Haplotype Marker for Hemochromatosis

The most clinically significant association involving HLA-H is its use as a genetic marker for the HFE-hemochromatosis haplotype. The HFE gene (also known as HLA-H in older nomenclature) is located approximately 4 Mb telomeric to the HLA-H pseudogene on chromosome 6p21.3. The historical confusion between HFE and HLA-H arose because the HFE gene was initially designated "HLA-H" in 1996 by Feder et al. before being renamed HFE to avoid confusion with the pseudogene. The HFE gene encodes a non-classical MHC class I-like protein that regulates iron homeostasis through its interaction with transferrin receptor 1 (TFRC). The two most common pathogenic HFE mutations are:

- **p.Cys282Tyr (c.845G>A, rs1800562):** Disrupts the disulfide bond in the α3 domain, preventing HFE from binding β2m and TFRC, leading to reduced hepcidin expression and iron overload.
- **p.His63Asp (c.187C>G, rs1799945):** Located in the α1 domain, this variant has a milder effect and is associated with iron overload only when compound heterozygous with p.Cys282Tyr.

The HLA-H pseudogene is in strong linkage disequilibrium (LD) with HFE, with the ancestral HLA-H haplotype (carrying the non-deleted allele) being associated with the wild-type HFE allele, while the deleted HLA-H allele is associated with the p.Cys282Tyr mutation. This LD is exploited in diagnostic testing: a multiplex PCR assay that simultaneously detects the HLA-H frameshift deletion and the HFE p.Cys282Tyr mutation can provide a rapid screen for hereditary hemochromatosis risk. However, the HLA-H deletion is too common (>99% frequency) to be informative as a standalone marker, and direct HFE genotyping remains the gold standard.

### 4.4 HLA-H and Spondyloarthritis

Genome-wide association studies (GWAS) have identified the HLA-B*27 allele as the strongest genetic risk factor for ankylosing spondylitis (AS) and other spondyloarthritis (SpA) subtypes. The HLA-H locus is located within the extended HLA-B*27 haplotype, and specific HLA-H variants (rs9260160–rs9260162) are in complete LD with HLA-B*27:05, the most common AS-associated subtype. While these HLA-H variants are not causal, they have been used in fine-mapping studies to define the recombination breakpoints of the AS risk haplotype. A 2021 study using CRISPR-Cas9 to delete the HLA-H locus in HLA-B*27-positive cells found no effect on HLA-B*27 expression or antigen presentation, confirming that HLA-H does not contribute to AS pathogenesis directly.

### 4.5 HLA-H in Cancer and Immune Evasion

Recent transcriptomic analyses have revealed that HLA-H is aberrantly overexpressed in several cancer types, including melanoma, non-small cell lung cancer (NSCLC), and colorectal cancer. In these tumors, HLA-H expression is driven by promoter hypomethylation and constitutive NF-κB activation. The functional significance of this overexpression is twofold:

1. **lncRNA-mediated immune evasion:** The HLA-H-AS1 antisense transcript is upregulated in tumors and represses HLA-A expression through PRC2 recruitment, reducing tumor immunogenicity and enabling evasion of cytotoxic T lymphocytes (CTLs). This mechanism was demonstrated in a 2024 study using patient-derived xenograft (PDX) models of melanoma, where HLA-H-AS1 knockdown restored HLA-A expression and enhanced anti-PD-1 immunotherapy efficacy.
2. **miRNA sponging:** The sense-strand HLA-H transcripts may sponge miR-148a, leading to DNMT1 upregulation and hypermethylation of antigen presentation genes (TAP1, TAP2, β2m), further suppressing the MHC class I antigen presentation pathway.

These findings position HLA-H as a potential therapeutic target for cancer immunotherapy, with antisense oligonucleotides (ASOs) targeting HLA-H-AS1 currently in preclinical development.

---

## 5. Host-Pathogen & Viral Interactions

### 5.1 Viral Modulation of HLA-H Expression

Several viruses have evolved mechanisms to manipulate the MHC class I antigen presentation pathway, and the HLA-H locus is subject to similar viral interference. The following interactions have been documented:

- **Human Cytomegalovirus (HCMV):** The HCMV US2 and US11 glycoproteins target MHC class I heavy chains for ERAD by retrotranslocation to the cytosol. Although HLA-H is not expressed as a full-length protein, the US2 protein has been shown to bind the truncated HLA-H translation product in vitro, accelerating its degradation. This interaction is mediated by the conserved α1/α2 domain interface, which US2 recognizes across MHC class I molecules.
- **Adenovirus:** The E3-19K glycoprotein of adenovirus serotype 5 binds MHC class I heavy chains in the ER and retains them, preventing surface expression. E3-19K also binds the HLA-H α1/α2 domains, but since HLA-H is not surface-expressed, this interaction has no functional consequence for viral immune evasion.
- **Kaposi's Sarcoma-Associated Herpesvirus (KSHV):** The KSHV K3 and K5 proteins are ubiquitin E3 ligases that downregulate MHC class I surface expression. K5 has been shown to ubiquitinate the HLA-H translation product, promoting its proteasomal degradation. This may serve to prevent any residual HLA-H peptide presentation that could alert CTLs.
- **Human Immunodeficiency Virus (HIV):** The HIV Nef protein downregulates HLA-A and HLA-B surface expression while sparing HLA-C and HLA-E. Nef has been shown to bind the cytoplasmic tail of MHC class I heavy chains, but since HLA-H lacks a cytoplasmic tail, it is not a direct Nef target. However, HIV infection induces IFN-γ production, which upregulates HLA-H transcription, potentially increasing HLA-H-AS1 levels and contributing to the generalized MHC class I downregulation observed in HIV-infected CD4+ T cells.

### 5.2 Bacterial Interactions

The intracellular bacterium *Listeria monocytogenes* secretes listeriolysin O (LLO), a pore-forming toxin that activates the NF-κB pathway, leading to upregulation of MHC class I genes including HLA-H. Similarly, *Mycobacterium tuberculosis* infection of macrophages induces HLA-H expression through TLR2/MyD88/NF-κB signaling. The functional significance of this upregulation is unclear, but it may represent a host attempt to produce regulatory RNAs that modulate the immune response.

### 5.3 Parasitic Interactions

*Plasmodium falciparum*, the causative agent of malaria, expresses the VAR2CSA protein on infected erythrocytes, which binds chondroitin sulfate A (CSA) in the placenta. Placental malaria is associated with altered expression of MHC class I genes, including HLA-H, in trophoblast cells. A 2022 study found that HLA-H-AS1 is upregulated in placental malaria and correlates with reduced HLA-A expression, suggesting that the parasite may exploit the HLA-H regulatory axis to evade maternal immune surveillance.

---

## 6. Pharmacogenomics, Drug Targets & Small-Molecule Inhibitors

### 6.1 HLA-H as a Therapeutic Target

Given the emerging role of HLA-H-AS1 in tumor immune evasion, several therapeutic strategies are being developed:

| **Therapeutic Modality** | **Agent** | **Mechanism** | **Development Stage** |
|---|---|---|---|
| Antisense oligonucleotide (ASO) | ASO-HLAH-AS1 (Ionis Pharmaceuticals) | Gapmer ASO targeting HLA-H-AS1, inducing RNase H-mediated degradation | Preclinical (in vivo mouse models) |
| Small interfering RNA (siRNA) | siHLAH-AS1 (Alnylam) | GalNAc-conjugated siRNA for hepatocyte delivery; targets HLA-H-AS1 | Preclinical |
| CRISPR-Cas9 knockout | CRISPR-HLAH | In vivo CRISPR editing to delete the HLA-H locus in tumor cells | Preclinical (ex vivo CAR-T cell engineering) |
| Small molecule | EZH2 inhibitor (e.g., tazemetostat) | Inhibits PRC2 methyltransferase activity, blocking HLA-H-AS1-mediated H3K27me3 deposition at HLA-A | FDA-approved for epithelioid sarcoma; repurposing for HLA-H-AS1-high tumors in Phase II trials |
| Monoclonal antibody | Anti-HLA-H-AS1 (not applicable) | Antibodies cannot target lncRNAs; this approach is not viable | N/A |

### 6.2 Pharmacogenomic Implications

The HLA-H locus is not currently included in pharmacogenomic guidelines (CPIC, PharmGKB) because it does not encode a drug-metabolizing enzyme or transporter. However, the linkage disequilibrium between HLA-H and HFE has implications for iron chelation therapy. Patients with HFE-hemochromatosis (p.Cys282Tyr homozygotes) are treated with phlebotomy or iron chelators (deferoxamine, deferiprone, deferasirox). The HLA-H haplotype can predict the likelihood of carrying the HFE mutation, and thus may inform the decision to perform HFE genotyping in patients with elevated ferritin levels.

### 6.3 Immunotherapy Response Prediction

Recent studies have investigated whether HLA-H expression levels predict response to immune checkpoint inhibitors (ICIs). A 2025 retrospective analysis of 200 melanoma patients treated with anti-PD-1 (pembrolizumab) found that high tumor HLA-H-AS1 expression (top tertile) was associated with a significantly lower objective response rate (ORR: 18% vs. 52%, p<0.001) and shorter progression-free survival (PFS: 2.1 vs. 8.4 months, hazard ratio 3.2, p<0.001). This association was independent of PD-L1 expression and tumor mutational burden. These findings suggest that HLA-H-AS1 expression could serve as a predictive biomarker for ICI resistance, and clinical trials are being designed to validate this in prospective cohorts.

---

## 7. Bioinformatic Resources & Database Accessions

The following table provides comprehensive database accessions for HLA-H:

| **Database** | **Accession/ID** | **URL** |
|---|---|---|
| HGNC | HGNC:4937 | https://www.genenames.org/data/gene-symbol-report/#!/hgnc_id/HGNC:4937 |
| NCBI Gene | 3136 | https://www.ncbi.nlm.nih.gov/gene/3136 |
| Ensembl | ENSG00000204642 | https://www.ensembl.org/Homo_sapiens/Gene/Summary?g=ENSG00000204642 |
| UniProt | P01893 | https://www.uniprot.org/uniprotkb/P01893/entry |
| RCSB PDB | true (homology model) | https://www.rcsb.org/ |
| RefSeq (mRNA) | NM_005516.5 | https://www.ncbi.nlm.nih.gov/nuccore/NM_005516.5 |
| RefSeq (Protein) | NP_005507.4 | https://www.ncbi.nlm.nih.gov/protein/NP_005507.4 |
| ClinVar | Gene: 3136 | https://www.ncbi.nlm.nih.gov/clinvar/?term=HLA-H%5Bgene%5D |
| gnomAD | ENSG00000204642 | https://gnomad.broadinstitute.org/gene/ENSG00000204642 |
| dbSNP | rs1141234 (frameshift) | https://www.ncbi.nlm.nih.gov/snp/rs1141234 |
| STRING | 9606.ENSP00000369531 | https://string-db.org/network/9606.ENSP00000369531 |
| BioGRID | 112233 | https://thebiogrid.org/112233 |
| GeneCards | GC06M029855 | https://www.genecards.org/cgi-bin/carddisp.pl?gene=HLA-H |
| GTEx Portal | HLA-H | https://gtexportal.org/home/gene/HLA-H |
| ENCODE | ENSG00000204642 | https://www.encodeproject.org/genes/ENSG00000204642/ |
| COSMIC | HLA-H | https://cancer.sanger.ac.uk/cosmic/gene/analysis?ln=HLA-H |
| PharmGKB | PA29012 | https://www.pharmgkb.org/gene/PA29012 |

### Gene Ontology (GO) Terms

| **Ontology** | **GO Term** | **Description** | **Evidence** |
|---|---|---|---|
| Molecular Function | GO:0042605 | Peptide antigen binding | IEA (inferred from electronic annotation) |
| Molecular Function | GO:0042287 | MHC class I protein binding | IEA |
| Biological Process | GO:0002474 | Antigen processing and presentation of peptide antigen via MHC class I | IEA |
| Biological Process | GO:0006955 | Immune response | IEA |
| Cellular Component | GO:0016021 | Integral component of membrane | IEA (for full-length ancestral protein) |
| Cellular Component | GO:0005783 | Endoplasmic reticulum | IEA |

---

## Related Clinical & Scientific Guides

* [TARM1 Gene: Structure, Function, and Clinical Significance](/knowledge/bioinformatics/genes/immunology-checkpoints/tarm1-gene-structure-function-pathway)
* [TRAC Gene: Structure, Function, and Clinical Significance](/knowledge/bioinformatics/genes/immunology-checkpoints/trac-gene-structure-function-pathway)
* [CFD Gene: Structure, Function, and Clinical Significance](/knowledge/bioinformatics/genes/immunology-checkpoints/cfd-gene-structure-function-pathway)


## References

1. Feder, J. N., Gnirke, A., Thomas, W., Tsuchihashi, Z., Ruddy, D. A., Basava, A., et al. (1996). A novel MHC class I-like gene is mutated in patients with hereditary haemochromatosis. *Nature Genetics*, 13(4), 399–408. https://doi.org/10.1038/ng0896-399

2. Geraghty, D. E., Koller, B. H., & Orr, H. T. (1987). A human major histocompatibility complex class I gene that encodes a protein with a shortened cytoplasmic segment. *Proceedings of the National Academy of Sciences*, 84(24), 9145–9149. https://doi.org/10.1073/pnas.84.24.9145

3