Epitope Spreading

Epitope spreading—also known as determinant spreading—is the dynamic immunological process whereby the specificity of a cellular (T lymphocyte) or humoral (B lymphocyte and antibody) adaptive immune response broadens over time from a single initial, immunodominant antigenic epitope to include secondary, non-cross-reactive (cryptic or subdominant) epitopes located either on the same polypeptide chain (intramolecular epitope spreading) or on distinct proteins residing within the same macromolecular complex or anatomical tissue (intermolecular epitope spreading). First discovered and conceptualized in 1992 by Eli E. Sercarz, Paul V. Lehmann, Thomas Forsthuber, Alexander Miller, and colleagues in murine models of experimental autoimmune encephalomyelitis (EAE), epitope spreading explains a central clinical feature of chronic autoimmune diseases: whereas an autoimmune reaction is often initiated by a narrow trigger such as molecular mimicry against a single viral or bacterial peptide, the mature clinical disease is sustained by a self-amplifying cascade of autoreactive lymphocytes targeting dozens of structurally unrelated tissue self-antigens. Conversely, in infectious disease clearance and cancer immunotherapy, epitope spreading serves as a beneficial protective amplifier that broadens cytotoxic T-cell surveillance against heterogeneous viral variants and tumor neoantigens.

Unlike molecular mimicry or TCR degeneracy—in which a single lymphocyte clone recognizes two structurally homologous peptides—epitope spreading involves the de novo priming and recruitment of new, distinct naive T-cell and B-cell clonotypes whose antigen receptors do not cross-react with the original initiating epitope.

1. Intramolecular Epitope Spreading

In intramolecular epitope spreading, the immune response expands from a primary immunodominant peptide sequence (Epitope A) of a multi-domain protein to secondary subdominant or formerly cryptic peptide determinants (Epitopes B, C, and D) situated elsewhere along the primary amino-acid chain of that same polypeptide.

In the landmark demonstration by Paul V. Lehmann, Thomas Forsthuber, Alexander Miller, and Eli Sercarz (1992), immunization of (SJL × B10.PL)F₁ mice with the single amino-terminal nonapeptide Ac1–11 of guinea pig Myelin Basic Protein (MBP) induced an initial wave of acute central nervous system inflammation mediated exclusively by Ac1–11-specific CD4⁺ T cells restricted to the MHC class II molecule I-A^u. However, within 2 to 4 weeks, as primary inflammatory demyelination caused macrophages and residential microglia to phagocytose and proteolytically degrade intact endogenous murine MBP inside the spinal cord, naive T-cell clones specific for the cryptic distal MBP peptides MBP_35–47, MBP_81–100, and MBP_121–140 (restricted to I-E^u or I-A^s) became primed and expanded, driving chronic relapsing-remitting paralysis.

2. Intermolecular Epitope Spreading

In intermolecular epitope spreading, the immune response expands from an epitope on a primary target protein (Protein X) to epitopes located on structurally and genetically distinct proteins (Proteins Y and Z) that either:

• Associate physically with Protein X inside a non-covalent macromolecular ribonucleoprotein or chromatin complex (such as the spliceosome or nucleosome), or
• Co-localize with Protein X inside the same anatomical cellular structure (such as the myelin sheath of oligodendrocytes, the desmosome/hemidesmosome of epidermal keratinocytes, or the secretory granules of pancreatic β-islet cells) and are co-released during tissue lysis.

Stephen D. Miller and colleagues (1995) demonstrated ordered intermolecular spreading in the SJL/J mouse model of relapsing-remitting EAE: when mice are immunized with the dominant peptide of Proteolipid Protein (PLP_139–151), the initial paralytic attack is driven solely by PLP_139–151-specific Th1/Th17 cells; the first clinical relapse is driven by intramolecular spreading to PLP_178–191; and the second and third clinical relapses are driven by intermolecular spreading to completely separate myelin structural proteins, namely Myelin Oligodendrocyte Glycoprotein (MOG_92–106) and Myelin Basic Protein (MBP_84–104). Crucially, inducing antigen-specific peripheral tolerance to the secondary relapse epitopes (PLP_178–191 or MOG_92–106) during remission completely prevents subsequent clinical relapses even after primary PLP_139–151 damage has already occurred.

Two distinct cellular pathways—one mediated by antigen-specific B lymphocytes and the other by inflammatory myeloid antigen-presenting cells—drive the transition from mono-epitopic recognition to poly-epitopic spreading:

1. B-Cell Receptor-Mediated Endocytosis and Linked Recognition (Physical Complex Capture)

In humoral autoimmunity, B lymphocytes act as ultra-efficient antigen-concentrating antigen-presenting cells (APCs) that capture specific proteins at concentrations 1,000- to 10,000-fold lower than pinocytosing dendritic cells:

• Intramolecular B-Cell Presentation: A B cell whose surface immunoglobulin (BCR) binds a single native surface epitope on a self-protein internalizes the entire intact polypeptide via clathrin-mediated endocytosis. Inside endolysosomes, cathepsin proteases cleave the entire protein into dozens of linear peptides that are loaded onto major histocompatibility complex class II (MHC class II) molecules. Furthermore, as shown by Colin Watts, the high-affinity Fab arm of the BCR physically shields its bound footprint from endosomal proteases while altering the conformational unfolding of the remaining polypeptide, unmasking cryptic peptides (determinants that are normally destroyed or under-processed by conventional dendritic cells) and presenting them to naive CD4⁺ T cells.
• Intermolecular Linked Recognition: When a B cell's BCR binds a single surface epitope on Protein X, and Protein X is non-covalently bound inside a native macromolecular particle (such as a nucleosome containing DNA + histones H1, H2A, H2B, H3, and H4, or a U1 small nuclear ribonucleoprotein [U1-snRNP] spliceosomal particle containing U1 RNA + Sm B/B′, Sm D1–D3, U1-70K, U1-A, and U1-C), the B cell internalizes the entire intact multi-protein complex as a single cargo unit. Following endolysosomal degradation of all constituent subunits, the B cell displays peptides derived from Proteins X, Y, and Z on its surface MHC class II molecules. Consequently, an existing helper T cell specific only for Protein X delivers CD40L and IL-21 follicular help to B cells specific for Protein Y or Z, driving germinal-center class switch recombination and somatic hypermutation of autoantibodies against every subunit of the particle.

2. Inflammatory Tissue Destruction, APC Licensing, and Proteasomal Remodeling

In cell-mediated organ-specific autoimmunity, epitope spreading is driven by local cytotoxic tissue damage inside the target organ:

1. Release of Sequestered Intracellular Self-Antigens: Primary CD4⁺ Th1/Th17 or CD8⁺ cytotoxic T-cell infiltration induces apoptosis, pyroptosis, and secondary necrosis of parenchymal cells (such as pancreatic β-cells or CNS oligodendrocytes), liberating high concentrations of previously sequestered intracellular proteins together with damage-associated molecular patterns (DAMPs: HMGB1, extracellular ATP, mitochondrial DNA).
2. APC Recruitment, Costimulatory Upregulation, and Immunoproteasome Induction: Local IFN-γ and TNF-α license resident macrophages, microglia, and infiltrating conventional dendritic cells (cDC1 and cDC2) to upregulate MHC class I, MHC class II, and the costimulatory ligands CD80 (B7-1) and CD86 (B7-2) while replacing constitutive 20S proteasome subunits (β1, β2, β5) with the IFN-γ-inducible immunoproteasome subunits (LMP2/β1i, MECL-1/β2i, and LMP7/β5i). Because the immunoproteasome alters cleavage specificity after hydrophobic and basic residues, and reactive oxygen/nitrogen species induce post-translational modifications (deamidation, citrullination, oxidation), inflamed APCs generate novel self-peptide repertoires and migrate to draining lymph nodes to prime naive T cells against secondary self-antigens.

Longitudinal biobank serum studies (such as the U.S. Department of Defense Serum Repository studies by Judith A. James and John B. Harley) demonstrate that epitope spreading unfolds in a predictable, years-long preclinical cascade before clinical symptom onset:

• Systemic Lupus Erythematosus (SLE): Up to 5 to 9 years before clinical diagnosis, patients develop an initial autoantibody response restricted to a single proline-rich octapeptide (PPPGRRP) of the spliceosomal protein Sm B/B′ or amino acids 169–180 of 60-kDa Ro/SSA (both of which share structural mimicry with Epstein-Barr Virus EBNA1). Over the ensuing years, the humoral response spreads intramolecularly across non-overlapping epitopes of Ro60 and Sm B/B′, and then spreads intermolecularly to La/SSB, Sm D1, U1-70K, U1-A, histones, and finally native double-stranded DNA (dsDNA) immediately preceding clinical lupus nephritis.
• Type 1 Diabetes Mellitus (T1D): Prospective birth cohorts (TEDDY, DAISY) show that islet autoimmunity typically initiates in early childhood against a single primary autoantigen—most commonly insulin (IAA) (espec. insulin B-chain peptide B_9–23) or glutamic acid decarboxylase 65 (GAD65). Progression to overt hyperglycemia and clinical β-cell destruction correlates strongly with intermolecular spreading from 1 autoantibody to ≥ 2 or 3 distinct islet autoantigens, sequentially recruiting insulinoma-associated antigen-2 (IA-2 / PTPRN), the β-cell-specific zinc transporter 8 (ZnT8 / SLC30A8), and chromogranin A / islet-specific glucose-6-phosphatase catalytic subunit-related protein (IGRP).
• Pemphigus Foliage and Paraneoplastic Syndromes: Patients with endemic pemphigus foliaceus (fogo selvagem) initially harbor non-pathogenic IgG1 antibodies against the C-terminal extracellular domain (EC5) of desmoglein-1 (Dsg1); clinical epidermal blistering erupts only when intramolecular epitope spreading generates pathogenic IgG4 autoantibodies against the calcium-dependent adhesive EC1 and EC2 domains of Dsg1. In severe mucocutaneous pemphigus vulgaris, intermolecular spreading progresses from mucosal desmoglein-3 (Dsg3) to cutaneous desmoglein-1 (Dsg1).

While detrimental in autoimmunity, epitope spreading is a critical determinant of durable clinical remission in oncology:

• Adoptive T-Cell Therapy, Oncolytic Viruses, and Cancer Vaccines: Tumors frequently escape monospecific therapies (such as a chimeric antigen receptor, CAR-T, or a single-peptide vaccine targeting a single tumor-associated antigen like gp100, MART-1, or HER2) by downregulating or losing the targeted antigen (antigen-loss variant escape). However, when primary cytotoxic T lymphocytes or oncolytic viruses lyse a fraction of tumor cells in an immunogenic manner, dying cancer cells release hundreds of patient-specific somatic mutated neoantigens accompanied by calreticulin exposure and cGAS–STING activation.
• Cross-Priming by Batf3-Dependent cDC1s and Checkpoint Blockade: Intratumoral CD103⁺ / XCR1⁺ conventional type 1 dendritic cells (cDC1s) phagocytose tumor debris and cross-present these secondary neoantigens on MHC class I to prime endogenous polyclonal CD8⁺ T-cell clones. Clinical administration of anti-CTLA-4 (ipilimumab) and anti-PD-1 (nivolumab / pembrolizumab) amplifies this intermolecular neoantigen spreading across dozens of distinct passenger and driver mutations, converting a narrow therapeutic attack into a broad, self-sustaining polyclonal anti-tumor response that prevents recurrence of antigen-negative tumor subclones.

1. Lehmann, P. V., Forsthuber, T., Miller, A., & Sercarz, E. E. (1992). "Spreading of T-cell autoimmunity to cryptic determinants of an autoantigen." Nature, 358(6382), 155–157. https://doi.org/10.1038/358155a0
2. McRae, B. L., Vanderlugt, C. L., Dal Canto, M. C., & Miller, S. D. (1995). "Functional evidence for epitope spreading in the relapsing pathology of experimental autoimmune encephalomyelitis." Journal of Experimental Medicine, 182(1), 75–85. https://doi.org/10.1084/jem.182.1.75
3. Vanderlugt, C. L., & Miller, S. D. (2002). "Epitope spreading in immune-mediated diseases: implications for immunotherapy." Nature Reviews Immunology, 2(2), 85–95. https://doi.org/10.1038/nri724
4. Arbuckle, M. R., McClain, M. T., Rubertone, M. V., Scofield, R. H., Dennis, G. J., James, J. A., & Harley, J. B. (2003). "Development of autoantibodies before the clinical onset of systemic lupus erythematosus." New England Journal of Medicine, 349(16), 1526–1533. https://doi.org/10.1056/NEJMoa021933
5. Gulley, J. L., Madan, R. A., Pachynski, R., Mulders, P., Sheikh, N. A., Trager, J., & Drake, C. G. (2017). "Role of antigen spread and distinctive characteristics of immunotherapy in cancer treatment." Journal of the National Cancer Institute, 109(4), djw261. https://doi.org/10.1093/jnci/djw261