CD3 is a reliable and specific marker for the detection of cells of the T-cell lineage and has been, therefore, routinely used by pathologists for many years. The PS1 clone was a great advancement when it was first released. It was robust and able to work in decalcified bone marrow trephines where only the rabbit polyclonal antibody had been previously effective. After an internal product review at Leica Biosystems the Research and Development team were challenged to produce a superior clone. LN10 was the product of this work, which has exceeded all expectations. The localisation and all round efficacy have provided an advancement in the evaluation of T-cell infiltrates and the identification of T-cell derived neoplasms.
CD3 is a multi-subunit complex containing extracellular, trans-membrane and intracellular domains and is spatially related but separate from the T-cell receptor.
As CD3 is restricted to the T-lineage and is expressed virtually throughout T-cell differentiation, it represents the most specific as well as the most sensitive T-cell lineage marker available for the immunophenotypic analysis of lymphohaematopoietic malignancies. The Novocastra clone PS1 was developed almost 10 years ago and fully characterised by work conducted by Steward et al.1 Since then a review of the product led the team to see if they could produce a superior clone. This resulted in a new clone LN10, which while fully assessed and characterised by Leica Microsystems we describe here our experiences within our laboratory. Therefore, the aim of this study has been the comparison of the immunohistochemical properties of the two different monoclonal antibodies directed against the non-glycosylated epsilon chain of CD3 molecule: the older PS1 clone and the novel LN10.
Materials and methods
As a first step, optimal immunostaining protocols were developed using tonsil tissue specimens. Both clones required an antigen retrieval procedure, which was found to be identical in both instances: it consisted of high pressure-cooking in a citrate buffer of pH 6.0 for 2 minutes. After deparaffinization and antigen retrieval the monoclonal antibodies were applied on the tissue sections with an incubation time of 30 min. Specifically bound antibodies were made visible using the alkaline phosphatase detection system and FastRed as chromogen. After establishing the most optimal dilution (in both instances it has been 1:100) both monoclonal antibodies were applied routinely to stain 260 consecutive cases of the Department of Pathology at Charit_ University Medical School Berlin, Campus Benjamin Franklin, Berlin, Germany.
The majority of the cases stained were lymph node (n = 140 cases) and EDTA-decalcified bone marrow specimens (n = 70 cases) which had been referred to this department for diagnosis or exclusion of a malignant lymphoma manifestation or for the diagnosis of a haematological disease. The collective analyzed also included 10 cases of small intestine biopsies, which had been referred to the Department in order to diagnose or exclude the presence of gluten-sensitive enteropathy.
The samples were evaluated in the first instance independently and then with all experts involved at a multiheaded microscope. The following points were addressed for each pair of immunostains:
- Labelling of the appropriate cell population
- Comparison of the staining intensity between PS1 and LN10
- Comparison of the ability to deliver a crisp signal with a clear labelling of the cell membrane
- Possible influence of the fixation
- Possible influence of EDTA-decalcification in the case of bone marrow biopsy specimens
- In the case of peripheral T-cell lymphomas, which can show an aberrant reduction to deletion of CD3 expression among other, pan T-cell antigens, whether either of these clones was able to identify even a low level of CD3 expression.
Figure 1. Comparison between PS1 and LN10 clones in normal and neoplastic T-cells
Labelling of the appropriate cell population: in all samples harbouring non-neoplastic T-cells, both clones (PS1 and LN10) delivered similar patterns of specifically stained cells, as has been extensively described previously.(2, 3 and references therein) We did not observe any inappropriate labelling of non-lymphoid cells with the novel anti-CD3 clone LN10 in the numerous specimens investigated.
Comparison of staining intensity: Application of the novel anti-CD3 clone LN10 resulted in more intensive immunostains than those obtained with the anti-CD3 clone PS1 (Figures 1 and 2). This was appreciable not only in small specimens (intestinal biopsies) but also in large tissue sections. In 10% of the slides analyzed both antibodies delivered identical immunostains (Figure 3). There was no instance where the anti-CD3 clone PS1 delivered stronger signals than the LN10 clone.
Signal crispiness: In addition to the increase in the staining intensity the novel anti-CD3 antibody led in most cases to a clear-cut and crisp labelling of the T-cell membrane (Figures 1b, 2b). In contrast, the anti-CD3 PS1 clone delivered not only a weaker but a somehow slightly “blurred” label (Figures 1a, 2a).
Influence of fixation time: among the cases investigated there were two large lymph node specimens showing a gradient decrease in CD3 labelling particularly visible with the PS1 clone. This effect is frequently seen to occur in conjunction with heterogeneous fixation of a given specimen where the border has experienced a different fixation time to the centre of the tissue. This staining heterogeneity was not visible with the LN10 clone.
Influence of decalcification procedures: in all bone marrow specimens investigated the anti-CD3 clone LN10 led to a crisp labelling of the normal T-cell population occurring in the bone marrow (Figure 3b).
Identification of low level CD3 expression in peripheral T-cell lymphomas: in one case of the present series with a very weak CD3 expression in the neoplastic cells as detected with the PS1 clone (Figure 1c), the novel clone has demonstrated a considerably stronger labelling of the neoplastic population (Figure 1d).
Figure 2. Comparison between PS1 and LN10 clones in samples from small intestine
The novel anti-CD3 clone LN10 reliably labels T-cells in paraffin embedded tissue specimens. The LN10 clone predominantly produces stronger immunostaining than the PS1 clone. In addition, the labelling is more precise regarding T-cell membrane staining. In this respect LN10 seems to be more efficient in the identification of reactive and neoplastic T-cells than the PS1 clone.
The LN10 clone seems to be less affected by variation in fixation than the PS1 clone. Decalcification procedures – at least involving EDTA – do not interfere with LN10 reactivity making it suitable for investigation of bone trephines. It is therefore arguably an advancement in the detection and assessment of T-cell neoplasms and a welcome addition to the laboratory.
- Steward M, Bishop R, Piggott NH et al. Production and characterization of a new monoclonal antibody effective in recognizing the CD3 T-cell associated antigen in formalin-fixed embedded tissue. Histopathology 1997; 30 (1): 16–22.
- Williamson SLH, Steward M, Milton I et al. New monoclonal antibodies to the T-cell antigens CD4 and CD8: Production and characterization in formalin-fixed, paraffin-embedded tissue. Am J Pathol 1998; 152 (6): 1421–6.
- Leong FJW-M and Leong AS-Y. Essential markers in malignant lymphoma: a diagnostic approach. The Journal of Histotechnology 2002; 25 (4): 215–227.