Clin Pathol 1996;49:539-544 539 Papers J Clin Pathol: first published as 10.1136/jcp.49.7.539 on 1 July 1996. Downloaded from Differential expression of T cell antigens in normal peripheral blood lymphocytes: a quantitative analysis by flow cytometry L Ginaldi, N Farahat, E Matutes, M De Martinis, R Morilla, D Catovsky Abstract diagnosis by comparison with findings in Aims-To obtain reference values of the normal counterparts. level of expression of T cell antigens on ( Clin Pathol 1996;49:539-544) normal lymphocyte subsets in order to disclose differences which could reflect Keywords: flow cytometry, T cell antigens, peripheral their function or maturation stages, or blood lymphocytes. both. Methods-Peripheral blood from 15 healthy donors was processed by flow The most common use of flow cytometry is to cytometry with triple colour analysis. For determine the percent of "positive" or "nega- each sample phycoerythrin (PE) conju- tive" cells for each antigen in different cell gated CD2, CD4, CD5, CD8, and CD56 populations. However, valuable information is monoclonal antibodies were combined lost when the relative intensities of the positive with Cy5-R-phycoerythrin (TC) conju- cells, reflecting antigenic densities, are not considered. A cell population with a well gated CD3 and fluorescein isothiocyanate http://jcp.bmj.com/ (FITC) conjugated CD7; CD2- and defined phenotype, positive for one antigen, CD7-PE were also combined with might be heterogeneous with regard to its level CD3-TC and CD4-FITC. Standard mi- of expression. This may reflect different crobeads with different capacities to bind functional or maturational states, or both, or mouse immunoglobulins were used to identify subpopulations on the basis of the dif- convert the mean fluorescence intensity ferent numbers of molecules of antigen per cell. These intensity differences may be charac- (MFI) values of the lymphocyte subsets on September 27, 2021 by guest. Protected copyright. identified by multiparametric flow cytom- teristic of a specific subpopulation-for exam- etry into the number of antigen molecules ple, CD8 "dim" and CD8 "bright" lym- per cell, measured as antibody binding phocytes correspond to different cellular capacity (ABC). subsets with different functional properties.' Recent studies have suggested that an Results-CD4+ (helper/inducer) T cells aberrant antigen density in leukaemic cells can exhibit a higher CD3 antigen expression be used as a parameter with diagnostic and compared with CD8+ (suppressor/ prognostic significance.2 3 Quantitative flow cytotoxic) T lymphocytes. Within the cytometry can be used to distinguish between CD4+ T cells, the CD4+CD7- subset leukaemic and normal lymphoid cells.4 These expressed a lower level of CD3 compared methods are becoming widely used in the study with CD4+CD7+ and CD8+CD7+ cells, of pathological conditions.5 6 Normal lym- and higher CD2 and CD5 expression than phocyte subsets can exhibit different levels of Academic Departmnent the main CD3+CD7+ subset. Major dif- antigen expression and this has to be taken into of Haematology and ferences in antigen expression were also account in order to define "aberrant" density Cytogenetics, detected between CD3+ T cells and The Royal Marsden expression in pathological states and for the Hospital and Institute CD3-CD56+ natural killer (NK) cells: NK detection of minimal residual disease. of Cancer Research, cells exhibited higher levels of CD7 and To address this problem we studied the level London CD56 and lower levels of CD2 and CD5 of expression of a series of T lymphocytic than T cells. Significantly lower CDS Correspondence to: markers (CD2, CD3, CD4, CD5, CD7, CD8) Professor D Catovsky, expression was also detected in the small in different T cell subsets from 15 healthy Academic Department of B Haematology and CD5+ lymphocyte subset compared donors, by means of triple labelling, flow Cytogenetics, with T cells. cytometry quantitative methods. This tech- The Royal Marsden Hospital, Fulham Road, Conclusions-Quantitative flow cytom- nique permits the assessment of the number of London SW3 6JJ. etry with triple colour analysis may be antigen molecules per cell in well defined lym- Accepted for publication used to detect antigen modulations in dis- phoid subsets using standard beads to convert 21 March 1996 ease states and to increase the accuracy of fluorescence intensity into antibody binding 540 Ginaldi, Farahat, Matutes, De Martinis, Morilla, Catovsky Table 1 Monoclonal antibodies used in this study cells were vortexed and incubated for 15 min- utes at room temperature, washed three times Reagent Fluorochrome Volume (,ul) * Source with phosphate buffered saline (PBS) azide, CD2 (Leu-5b) PE 20 Becton Dickinson, San Jose, CA, USA resuspended in 0.5 ml Isoton (Coulter) and J Clin Pathol: first published as 10.1136/jcp.49.7.539 on 1 July 1996. Downloaded from CD3 (Tri-Color) RPE/Cy5 5 Caltage, Burlingame, CA, USA analysed on a flow cytometer. CD4 PE 10 Sera-Lab, Crawley Down, UK CD4 FITC 10 Sera-Lab CD5 (Leu-1) PE 20 Becton Dickinson FLOW CYTOMETRY ANALYSIS CD7 FITC 10 Sera-Lab Flow cytometry was performed using a FACS- CD8 (Leu-2a) PE 20 Becton Dickinson CD56 (Leu-19) PE 20 Becton Dickinson can (Becton Dickinson) with linear amplifica- tion (1024 channels) of the forward- and side- PE = phycoerythrin; RPE/Cy5 = Cy5 conjugated R-PE; FITC = fluorescein isothiocyanate. scatter signals (FS and SS) and logarithmic *Arnount of monoclonal antibody recommended by the manufacturer. amplification (4 logdecades) of the FL1, FL2 capacity (ABC).7 8 Our results show that CD2, and FL3 signals. Compensation parameters for CD3, CD4, CD5, CD7, and CD8 are multiparametric studies were adjusted at the expressed with different intensities on the vari- start of analysis. Correlated data of 10 000 ous positive lymphocyte cell subsets. This may events per sample were acquired with a live reflect different functions or different stages in gate applied on lymphocytes and stored in list the differentiation pathway. mode using the Lysys II software (Becton Dickinson). Analysis was performed using Methods Lysys II and Paint-a-GatePus software (Becton Dickinson). The latter reports the mean BLOOD SAMPLES Peripheral blood from 15 healthy donors fluorescence intensity (MFI) of the investi- (seven men and eight women, mean age 33 gated antigens for each of the seven subsets of years) was analysed in order to evaluate the cells resulting from the three colour analysis. differential expression of the most common T DETERMINATION OF ANTIBODY BINDING lymphocyte antigens, which define the various CAPACITY (QUANTIFICATION) cell subsets, and to obtain reference values for The MFI values of the positive cells were con- the quantitative, three colour, flow cytometry verted into ABC or number of molecules per tests. The blood samples were collected in cell by using Quantum Simply Cellular (QSC) heparin and processed immediately. Microbeads Kit (Sigma, St Louis, Missouri, USA). This is a mixture of four microbead IMMUNOSTAINING populations of uniform size, coated with goat The mononuclear cells were separated by den- anti-mouse antibodies, which have differing sity gradient centrifugation with Lymphoprep capacities to bind mouse monoclonal antibod- (Nycomed, Oslo, Norway). After washing three ies. Included in the mixture is a blank popula- times with RPMI 1640 at room temperature, tion of microbeads having no specific binding http://jcp.bmj.com/ 2 x 106 cells were used per test. Monoclonal capacity for mouse immunoglobulins. The antibodies directly conjugated with fluorescein accompanying software regresses the binding isothiocyanate (FITC), phycoerythrin (PE) capacities of the microbeads, stained with each and CyS-RPE (TC) were appropriately com- fluorochrome conjugated monoclonal anti- bined in each tube. The monoclonal antibodies body, against their corresponding peak chan- used in this study are shown in table 1. The nels. The regression curve permits quantitative following three colour staining protocols were estimates of the monoclonal antibody mol- on September 27, 2021 by guest. Protected copyright. applied: for each sample, PE conjugated CD2, ecules bound to the target cells. Each mono- CD4, CD5, CD8, and CD56 were combined clonal antibody was added in the appropriate with CD3-TC and CD7-FITC. A CD4-FITC amount to 50 ,ul QSC beads. After one hour of monoclonal antibody was also used combined incubation, the mixtures were washed and ana- with CD3-TC and CD2- or CD5-PE ABCs to lysed according to the method used for measure the expression of CD2 and CD5 on T lymphocytes and using the same instrument helper cells as well as to assess the influence of settings. The MFI of the respective antigen different fluorochrome conjugations on ABC analysed on the positive lymphocyte popula- measurements. To each tube 50 gl of 2% AB tions was converted into molecules of antibody serum was added followed by the appropriate bound to the cell membrane (ABC) or number amount of each monoclonal antibody. Ten of antigen molecules per cell. microlitres of double labelled isotypic mouse immunoglobulins (MsIgG-FITC/RDl) (Coul- STATISTICAL ANALYSIS ter) and 5 gl of MsIgG-TC (Caltag) were used The analysis ofvariance (ANOVA) was used to as negative controls in all experiments. The compare the ABC values of the different Table 2 Expression ofCD3, CD7 and CD56 antigens (measured as ABC) in different lymphocyte subsets (CD3-TCICD7-FITC/CD56-PE immunostaining protocol). Results are expressed as mean (SD) ABC x 10' Lymphocyte subset Percentage CD3 CD7 CD56 CD3+CD7+CD56- 62.25 (9.86) p 37.63 (8.41) 26.56 (8.49) p < 0.001 lI CD3+CD7-CD56- 4.35 (2.57) 21.69 (7.19) - p < 0.001 - p < 0.005 CD3+CD7-CD56+ 4.12 (2.82) 36.04 (14.75) 34.04 (19.67) 8.23 (4.50) p < 0.001 6 ( p < 0.05 1 CD3+CD7 CD56+ 9.95 (4.16) - 65.46 (28.01) )12.39 (3.99) Differential expression ofT cell antigens in normal peripheral blood lymphocytes 541 FRINT-R-FjRTE (T Mode * File: 4 3:BENO8l Dots = 5RR t: a:- 984 328 i :Nr WO - mnde; (> s1e gate.