Antigen presentation, MHC and T-cell responses
MHC class I and II pathways, the three signals for T-cell activation, CD4 helper subsets and CD8 cytotoxicity, and the drugs (calcineurin inhibitors, abatacept, checkpoint inhibitors) and HLA associations Paper 1 builds questions around, with a worked CD4 count and abacavir screening NNT.
Tier 1Medical sciences conceptFree sample
Written with AI assistance and fact-checked against the sources below. For exam study, not clinical decisions. How notes are made
At a glance
- T cells only see peptide bound to MHC (HLA in humans). CD8 T cells read class I (HLA-A, -B, -C) and kill. CD4 T cells read class II (HLA-DR, -DQ, -DP) and help.
- Naive T-cell activation needs three signals: antigen (TCR plus peptide-MHC), co-stimulation (CD28 with CD80/CD86) and cytokine (which sets the helper subset).
- CD4 subsets (Th1, Th2, Th17, Tfh, Treg) each match a type of pathogen and a family of immune diseases and biologics.
- HLA genes are the most polymorphic in the genome, which explains transplant matching, disease associations and pharmacogenetic drug reactions.
- Paper 1 tests mechanisms of immunosuppressants and checkpoint inhibitors, HLA-linked drug reactions, and immunodeficiency patterns. Worked numbers: a CD4 count from a differential, and an NNT for HLA screening.
Two presentation pathways
| MHC class I | MHC class II | |
|---|---|---|
| Expressed on | All nucleated cells (not red cells) | Dendritic cells, macrophages, B cells; induced on others by interferon gamma |
| Peptide source | Cytosol: self, viral, tumour proteins | Endosomes: phagocytosed or endocytosed proteins |
| Processing | Proteasome, then TAP pumps peptides into the ER | Endosomal cathepsins; invariant chain and HLA-DM load the groove |
| Peptide length | 8 to 10 amino acids | 13 to 25 amino acids |
| Read by | CD8 T cells | CD4 T cells |
| Deficiency | TAP deficiency (bare lymphocyte type 1): chronic sinopulmonary infection, necrotising granulomatous skin lesions | Class II deficiency (type 2, CIITA or RFX): very low CD4 count, combined immunodeficiency |
Cross-presentation. Specialised dendritic cells (cDC1) put exogenous antigen onto class I. That lets CD8 T cells respond to viruses that don't infect dendritic cells, and to tumours.
Dendritic cells link innate and adaptive immunity. Immature DCs sample tissue. PAMP signalling through TLRs (see innate immunity) makes them mature, upregulate CD80/CD86 and CCR7, and migrate to the lymph node. Antigen shown without these danger signals produces tolerance.
Pitfall: NK cells kill cells that have lost class I. CD8 T cells need class I to see their target. Viruses that downregulate HLA class I (CMV) escape CD8 cells but become NK targets.
Activating a naive T cell
- Signal 1: TCR binds peptide-MHC. CD4 or CD8 binds the non-polymorphic part of the MHC. CD3 transmits the signal (ZAP-70, then calcium and calcineurin, NFAT, NF-kB, AP-1).
- Signal 2: CD28 binds CD80/CD86. Without it the cell becomes anergic.
- Signal 3: cytokines from the dendritic cell choose the subset.
- IL-2 and CD25 (the high-affinity IL-2 receptor alpha chain) drive clonal expansion. One naive clone can expand 10,000-fold in a week.
Brakes: CTLA-4 is upregulated after activation and binds B7 with higher affinity than CD28. PD-1 binds PD-L1 and PD-L2 on tissue and tumour cells and dampens effector cells in the periphery.
Drugs mapped to the signals
| Step | Drug | Mechanism |
|---|---|---|
| Signal 1 | Anti-thymocyte globulin | Polyclonal depletion of T cells |
| Calcineurin | Tacrolimus, ciclosporin | Block NFAT dephosphorylation, so less IL-2 |
| Signal 2 | Abatacept (CTLA-4-Ig), belatacept | Bind CD80/CD86 and block CD28 |
| IL-2 receptor | Basiliximab | Anti-CD25 induction in kidney transplant |
| Proliferation | Mycophenolate, azathioprine | Purine synthesis inhibition (IMPDH; thiopurine) |
| mTOR | Sirolimus, everolimus | Block IL-2 receptor signalling and cell cycle progression |
| Brakes released | Ipilimumab (CTLA-4), nivolumab, pembrolizumab (PD-1), atezolizumab, durvalumab (PD-L1) | Restore antitumour T-cell activity |
| Leucocyte trafficking | Natalizumab (alpha-4 integrin), fingolimod (S1P receptor) | Keep lymphocytes out of the CNS or in lymph nodes |
Clinical consequences: immunosuppression complications, systemic therapy organ toxicity for immune-related adverse events, and rheumatoid arthritis for abatacept.
Pitfall: checkpoint inhibitor toxicity reflects the brake removed. Ipilimumab causes more colitis and hypophysitis; PD-1 inhibitors cause more thyroiditis and pneumonitis. The combination has the highest grade 3 to 4 rate (over 50% in CheckMate 067).
CD4 helper subsets
| Subset | Driven by | Signature cytokines | Defends against | Disease when excessive | Targeted by |
|---|---|---|---|---|---|
| Th1 | IL-12, IFN-gamma (STAT4, T-bet) | IFN-gamma, TNF | Intracellular bacteria, mycobacteria | Granulomas, type IV hypersensitivity | Anti-TNF, anti-IL-12/23 |
| Th2 | IL-4 (STAT6, GATA3) | IL-4, IL-5, IL-13 | Helminths | Asthma, atopic dermatitis, eosinophilia | Dupilumab, mepolizumab |
| Th17 | IL-6, TGF-beta, IL-23 (STAT3, ROR-gamma-t) | IL-17, IL-22 | Extracellular bacteria, Candida | Psoriasis, spondyloarthritis | Secukinumab, ustekinumab, risankizumab |
| Tfh | IL-6, IL-21 (BCL6) | IL-21, CD40L | Help for B-cell germinal centres | Autoantibodies (SLE) | |
| Treg | TGF-beta, IL-2 (FOXP3) | IL-10, TGF-beta | Self tolerance | Loss: IPEX syndrome (FOXP3) |
Anti-TNF therapy breaks down granulomas, so latent TB reactivates. Screen with an interferon-gamma release assay before starting. Anti-IL-17 drugs raise mucocutaneous candidiasis rates, mirroring STAT3 deficiency (hyper-IgE syndrome) and AIRE deficiency (anti-IL-17 and IL-22 autoantibodies).
CD8 cytotoxicity
CD8 cells kill by perforin and granzymes and through Fas ligand. Perforin deficiency causes familial haemophagocytic lymphohistiocytosis: uncontrolled activation, ferritin often above 10,000 µg/L, cytopenias and hepatosplenomegaly. CD8 T cells mediate SJS/TEN through granulysin release (see SJS/TEN) and drive graft rejection and graft-versus-host disease.
Superantigens
Staphylococcal TSST-1 and enterotoxins, and streptococcal pyrogenic exotoxins, bind class II and the TCR V-beta region outside the peptide groove. Up to 20% of all T cells activate, compared with about 1 in 10^5 for a conventional antigen. The cytokine surge causes fever, shock and a desquamating rash. IVIG may neutralise toxin in streptococcal toxic shock, and clindamycin suppresses toxin synthesis.
Worked example 1: CD4 count from a differential
A man with newly diagnosed HIV has a white cell count of 3.8 × 10^9/L with 22% lymphocytes. Flow cytometry shows CD4 cells at 18% of lymphocytes.
- Lymphocytes: 3.8 × 0.22 = 0.84 × 10^9/L = 840 cells/µL.
- CD4 count: 840 × 0.18 = 151 cells/µL.
That is below 200, so he meets the PJP threshold even though his CD4 percentage (18%) is above the 14% backup threshold. He needs PJP prophylaxis (co-trimoxazole) as well as prompt ART. The percentage is more stable than the absolute count when the total white count changes, for example after splenectomy (raises it) or during acute illness (lowers it). Clinical management: acquired immunodeficiency.
Worked example 2: HLA screening NNT
HLA alleles present drugs or drug-modified peptides to T cells. In PREDICT-1 (Mallal, 2008, 1,956 patients), immunologically confirmed abacavir hypersensitivity occurred in 2.7% of the unscreened arm and 0% of the HLA-B*57:01-screened arm.
- Absolute risk reduction: 2.7% - 0% = 2.7%.
- NNT: 1 / 0.027 = about 37 patients screened to prevent one confirmed reaction.
- About half of HLA-B*57:01 carriers react (positive predictive value 48% in PREDICT-1), and the negative predictive value is close to 100%. A negative test means abacavir can be used.
| Allele | Drug | Reaction | Highest-risk ancestry |
|---|---|---|---|
| HLA-B*57:01 | Abacavir | Hypersensitivity syndrome | European (about 6%) |
| HLA-B*15:02 | Carbamazepine, oxcarbazepine, phenytoin | SJS/TEN | Han Chinese, Thai, Malaysian |
| HLA-A*31:01 | Carbamazepine | DRESS, maculopapular eruption | European, Japanese |
| HLA-B*58:01 | Allopurinol | SJS/TEN, DRESS | Han Chinese, Thai, Korean |
See drug allergy and DRESS. Disease associations work the same way: HLA-B27 with ankylosing spondylitis (see ankylosing spondylitis), HLA-DQ2 and DQ8 with coeliac disease (negative predictive value close to 100%), HLA-DR3 and DR4 with type 1 diabetes, and shared-epitope HLA-DRB1 with seropositive RA.
Thymic selection
T cells mature in the thymus. Positive selection keeps cells that bind self MHC weakly. Negative selection deletes cells that bind self peptide strongly. AIRE lets thymic epithelium express tissue-restricted proteins (insulin, thyroglobulin), so loss of AIRE causes APECED: candidiasis, hypoparathyroidism and adrenal failure. DiGeorge syndrome (22q11.2 deletion) gives a small or absent thymus and low T-cell numbers. See primary defects of cellular immunity and combined immunodeficiency.
How it's tested
- Mechanism of an immunosuppressant or checkpoint inhibitor, and the adverse effect that follows from it.
- HLA pharmacogenetics: which test before which drug, in which ancestry, and why a negative result matters.
- Matching an infection pattern to a T-cell defect: PJP and CMV (CD4 or combined), candidiasis (Th17), TB after anti-TNF.
- Calculating a CD4 count, or an NNT from screening data.
Pitfalls
- Pitfall: HLA-B*57:01 testing prevents immunological hypersensitivity, not all reactions. Clinically diagnosed reactions still occurred in the screened arm of PREDICT-1 (3.4%), mostly false positives.
- Pitfall: CD4 T cells read class II and CD8 T cells read class I. The product rule (4 × 2 = 8 × 1 = 8) is a quick check.
- Pitfall: abatacept blocks co-stimulation by binding CD80/CD86, not CD28.
- Pitfall: tacrolimus and ciclosporin don't kill T cells. They stop IL-2 production, and their toxicity (nephrotoxicity, hyperkalaemia, hypomagnesaemia, tremor) comes from calcineurin inhibition in other tissues.
- Pitfall: the CD4 percentage helps when the total white count is distorted. Don't stop PJP prophylaxis on a single count rising above 200.
Finished reading?
Sources
- Cellular and Molecular Immunology (10th ed.)Elsevier (Abbas, Lichtman and Pillai), 2022
- Janeway's Immunobiology (10th ed.)W. W. Norton (Murphy, Weaver and Berg), 2022
- HLA-B*5701 screening for hypersensitivity to abacavir (PREDICT-1)New England Journal of Medicine (Mallal et al., 2008;358:568-79), 2008
- Australian HIV guidelines (commentary on the DHHS guidelines, opportunistic infection prophylaxis)AU/NZASHM (Australasian Society for HIV, Viral Hepatitis and Sexual Health Medicine)
- Clinical Pharmacogenetics Implementation Consortium guideline for HLA genotype and use of carbamazepine and oxcarbazepine: 2017 updateClinical Pharmacology and Therapeutics (Phillips et al., 2018;103:574-81), 2018
- Last updated
- 14 November 2026
What changed
- 14 November 2026 Turnstile bot check on sign-in and signed-out feedback