Immunohistochemistry: From Antigen Retrieval to H-score

Fixation and retrieval pH · DAB chromogenic detection · H-score and Allred · Multiplex IHC/TSA · Controls and validation

Published 2026 · Last updated 2026-09-27 · Hailu Zhou

Immunohistochemistry is the only routine method that places a protein in a tissue context at cellular resolution, and it is the method most often done badly. Fixation, retrieval, and detection each introduce variability, and the temptation to compare staining across blocks, antibodies, or laboratories without quantifying that variability is the field's chronic weakness. The steps below are ordered the way the protocol runs, because an error made in fixation cannot be rescued by a clever retrieval step.

Fixation sets the ceiling on everything downstream

Formalin (10% neutral buffered formaldehyde) crosslinks proteins through methylene bridges, primarily between lysine residues. Overfixation masks epitopes by burying them under a dense crosslinked matrix; underfixation leaves the tissue soft and the epitopes labile. For most tissues, 10% NBF for 6–24 hours is the working range, but the rate of penetration is slow — roughly 1 mm per hour — so large specimens need to be sliced thin before fixation or the centre will autolyse while the surface overfixes.

Alternatives matter for specific antigens. Zinc-based fixatives preserve some phosphoepitopes and membrane antigens better than formalin. Alcohol fixation (methanol, ethanol) precipitates proteins and works well for small molecules and some nuclear antigens, but damages tissue architecture. For phospho-specific antibodies, the choice of fixative often determines whether the signal exists at all, so it should be optimised per target rather than inherited as a lab default.

Antigen retrieval: pH 6 versus pH 9

Antigen retrieval reverses part of the crosslinking and re-exposes hidden epitopes. Two methods dominate: heat-induced epitope retrieval (HIER) in buffer, and enzymatic retrieval with proteinase K, trypsin, or pepsin. HIER is standard for most nuclear and cytoplasmic antigens.

The buffer choice is empirical but not random:

Retrieval can over-shoot: excessive heating strips antigen, produces tissue detachment, or creates false nuclear signal through unmasking of cross-reactive epitopes. The correct retrieval condition is the mildest one that gives maximal specific signal with minimal background. Antibody datasheets give a starting point, not a validation.

DAB detection and why signal amplification is a double-edged sword

The avidin–biotin complex and the more modern polymer-based systems both localise horseradish peroxidase to the antibody. DAB (3,3′-diaminobenzidine) is oxidised by HRP in the presence of hydrogen peroxide into a brown, alcohol-insoluble precipitate. DAB is sensitive and permanent, and its brown colour is well separated from the blue haematoxylin counterstain.

But DAB is also irreversible and amplifies everything, including nonspecific binding. If the primary antibody is dirty, the polymer system will produce beautiful, convincing, artefactual staining. Two safeguards: always include a negative control slide with the primary antibody omitted or substituted with isotype-matched immunoglobulin, and always titrate each new antibody lot. A constant antibody concentration across a dilution series is the only way to find the concentration at which signal is specific and background is absent.

For weak targets, tyramide signal amplification (TSA) deposits activated tyramide near the HRP, greatly increasing sensitivity. The trade-off is that TSA is prone to amplifying endogenous peroxidase and biotin, so peroxidase and biotin blocking are mandatory, and TSA should not be used where a properly titrated standard detection works.

Scoring: H-score, Allred, and their limitations

Pathologist scoring converts colour into a number, and the choice of scheme changes the biology you can see.

Two caveats dominate. First, scoring is only as reproducible as the staining that generated it; a H-score from an over-retrieved slide is precise nonsense. Second, most scoring systems treat intensity as linear, which it is not — the relationship between DAB concentration and optical density saturates. Where quantitative claims are needed, calibrate against a known standard rather than assuming intensity is proportional to protein amount.

Multiplex IHC: TSA and the sequential-staining problem

Sequential TSA uses the same principle repeatedly: stain with one antibody, deposit a tyramide-fluorophore, then elute the antibody before the next cycle. Because tyramide covalently binds the tissue, the deposited signal survives elution, allowing four to seven markers on a single slide. This is the basis of the Opal and similar multiplex platforms.

The discipline required is high. Elution conditions must be strong enough to remove the previous antibody and weak enough to preserve the antigen and the deposited fluorophore. Each new antibody in the panel must be validated individually before being added to the multiplex, and cross-reactivity is the most common failure — an antibody that works singly can produce off-target deposition when combined with a previous cycle's residual HRP. Spectral unmixing and accurate bleed-through correction are required for acquisition, and the unmixing matrix should be derived from single-stain controls.

Controls and validation

A credible IHC figure shows at minimum: a positive control tissue known to express the antigen, a negative control from the same tissue with omitted primary antibody, an isotype control, and — where the antibody is new — a knockdown or knockout tissue as the definitive specificity control. Antibody validation ladders (genetic, orthogonal, independent-antibody) exist because too many commercial antibodies are not specific for their stated target.

Report the antibody clone and lot, the retrieval method and buffer pH, the detection system, and the scoring method with the number of observers and the agreement statistic if scored manually. These details are what allow another lab to reproduce your staining pattern rather than a different one.

References

  1. Harlow E, Lane D. Using Antibodies: A Laboratory Manual. Cold Spring Harbor: Cold Spring Harbor Laboratory Press; 1999.
  2. Taylor CR, Rudbeck L, eds. Immunohistochemical Staining Methods. 6th ed. Carpinteria: Dako/Agilent; 2013.
  3. Shi SR, Key ME, Kalra KL. Antigen retrieval in formalin-fixed, paraffin-embedded tissues: an enhancement method for immunohistochemical staining based on microwave oven heating. J Histochem Cytochem. 1991;39(6):741–748.
  4. Shi SR, Imam SA, Young L, et al. Antigen retrieval immunohistochemistry under the influence of pH using monoclonal antibodies. J Histochem Cytochem. 1995;43(2):193–201.
  5. McCarty KS Jr, Miller LS, Cox EB, et al. Estrogen receptor analyses: correlation of biochemical and immunohistochemical methods using monoclonal antireceptor antibodies. Arch Pathol Lab Med. 1985;109(8):716–721.
  6. Allred DC, Harvey JM, Berardo M, Clark GM. Prognostic and predictive factors in breast cancer by immunohistochemical analysis. Mod Pathol. 1998;11(2):155–168.
  7. Budwit-Novotny DA, McCarty KS, Cox EB, et al. Immunohistochemical analyses of estrogen receptor in endometrial adenocarcinoma using a monoclonal antibody. Cancer Res. 1986;46(10):5419–5425.
  8. Stack EC, Wang C, Roman KA, Hoyt CC. Multiplexed immunohistochemistry, imaging, and quantitation: a review, with an assessment of Tyramide signal amplification, multispectral imaging and multiplex analysis. Methods. 2014;70(1):46–58.
  9. Bordeaux J, Welsh A, Agarwal S, et al. Antibody validation. Biotechniques. 2010;48(3):197–209.
  10. Hewitt SM, Baskin DG, Frevert CW, et al. Controls for immunohistochemistry: the Histochemical Society's standards of practice for validation of immunohistochemical assays. J Histochem Cytochem. 2014;62(10):693–697.