T-lymphocyte and B-lymphocyte subpopulations infiltrating human mammary carcinomas.

Both T- and B-lymphocytes were found in human primary mammary carcinomas and were distributed in widely varying amounts, but in most tumors, T-lymphocytes predominated. A small percentage of the T-lymphocytes expressed receptors for the Fc portion of IgG (Fc gamma R), but very few had receptors for C3 (C3+) (comparable to the findings in blood). A prominent subset of lymphocytes had Fc gamma R and were C3+, and most of these were surface immunoglobulin (slg)-bearing cells. The majority of lymphocytes from this subset were Fc+ C3+, and only a small percentage were Fc+ C3- (in contrast to the findings in blood). IgD and IgM were the predominant classes of findings in blood). IgD and IgM were the predominant classes of immunoglobulin found on the B-lymphocytes. The different preparative techniques did not result in a selective loss of lymphocyte subsets, but collagenase digestion did lead to a loss of expression of the C3 receptor on the lymphocyte surface, which was recoverable when lymphocytes were reincubated at 37 degrees C. No evidence was found for blocking of the C3 receptor by immune complexes with activated complement.

Discrimination between innate and cytophilic immunoglobulin on human peripheral blood lymphocytes: analysis by the direct antiglobulin rosette-forming reaction.

Bovine red blood cells linked to polyclonal or monoclonal anti-immunoglobulin antibodies are used in the direct antiglobulin rosetting reaction to detect surface-Ig on human lymphocytes. The sensitivity of this test is markedly increased by pretreating the red cells with trypsin. Enzyme-treated red cells, coupled to anti-human Fab or anti-light chain antibodies, react not only with innate Ig on B lymphocytes but also with smaller amounts of passively adsorbed, cytophilic Ig on up to 25% of freshly prepared peripheral blood (non-B) lymphocytes. In contrast, trypsinized red cells carrying anti-Ig isotype-specific antibodies react exclusively with B cell surface-Ig. Cytophilic Ig is abnormally firmly bound to lymphocytes separated on Ficoll-Hypaque at 20 degrees C or below, and is released very slowly during 3 h or more at 37 degrees C in vitro. Lymphocytes are free of detectable cytophilic Ig when isolated on Ficoll-Hypaque at 37 degrees C, and very little Ig is retained by non-B cells in suspensions purified on Percoll which, unlike Ficoll, does not increase Ig binding affinity. These lymphocyte separation procedures are recommended as a preliminary to B cell assays by sensitive antiglobulin techniques.

Artefactual variations in the B and T subpopulations of rabbit blood lymphocytes depending on method of isolation, and blocking of C3 receptors due to in vitro activation of complement.

Marked differences were found in the proportions of lymphocyte subpopulations in rabbit peripheral blood depending on the techniques used for the purification of lymphocytes. Rosette-forming reactions were used to find the numbers of T lymphocytes, Ig-bearing cells and cells with receptors for C3 or IgG-Fc. Some of the methods used for lymphocyte separation altered the relative numbers of T and B lymphocytes, through a disproportionate loss of T cells. Other changes were due to in vitro activation of complement detectable by the presence of C3 on the lymphocyte cell-membrane and causing partial blocking of C3 receptors. Highest yields of lymphocytes were obtained from defibrinated blood treated with carbonyl iron to remove phagocytes and methyl cellulose to sediment erythrocytes. This procedure was accompanied by in vitro activation of complement, with the consequences mentioned. Complement activation was inhibited by taking the blood into either EDTA or citrate. As EDTA was cytotoxic for rabbit T lymphocytes, citrate was considered best although the resulting lymphocyte suspensions were contaminated with up to 25% granulocytes and monocytes owing to the inhibition of carbonyl iron uptake by the prior exposure to citrate.

The detection of heat-aggregated IgG (as a model for immune complexes) by reverse passive haemagglutination using a human monoclonal rheumatoid factor coupled to erythrocytes.

A human monoclonal IgM rheumatoid factor (RF) produced in vitro by an Epstein-Barr virus (EBV)-immortalized cell line was purified by protein A-Sepharose adsorption and coupled by the chromic chloride method to human erythrocytes. The RF-coupled cells were incorporated in reverse passive haemagglutination (RPH) assays to detect immune complexes (IC) using heat-aggregated human IgG as a model system. The sensitivity of the RPH was comparable to an enzyme-linked immunosorbent assay (ELISA) using sheep C1q for the detection of ICs.

The detection of alloantibodies to subpopulations of human lymphocytes: an adaptation of the indirect anti-immunoglobulin rosetting reaction (IARR).

An indirect anti-immunoglobulin rosetting reaction (IARR) can be successfully used to detect alloantibodies to human lymphocytes. In this paper we describe an adaptation of the IARR which allows detection of alloantibodies to human lymphocyte subpopulations. Fluorescein labelled sheep red cells, as a T cell marker, are incorporated into the rosetting reactions and by looking for rosettes with two indicator cell types, sheep and ox, one can determine if the alloantibodies are reacting with T cells or non-T cells. This type of assay is more sensitive than a standard cytotoxic test and can detect non-cytotoxic antibodies. The results show that this rosetting assay with two types of indicator cell can be useful in the study of pregnancy anisera, particularly those that are thought to contain reactivity directed solely against non-T cells. These antisera probably recognise the human DR antigens which are thought to be equivalent to the rodent Ia antigens.

Detection of ecto-5'-nucleotidase on rat B and T lymphocyte subpopulations using a monoclonal antibody in rosetting reactions.

A mouse monoclonal antibody to rat 5'-nucleotidase (5N 4-2 McAb) was used in the direct anti-determinant rosetting reaction (DARR) to demonstrate the ecto-5'-nucleotidase molecule in preparations of rat lymphocytes. Results indicated that 35.5 +/- 7.5% of peripheral blood lymphocytes (PBL), 37.3 +/- 4.8% of lymph node cells (LN) and 37.0 +/- 8.5% of spleen lymphocytes expressed the 5N 4-2 antigen. Depletion studies and mixed rosetting reactions (MRR) showed that the 5N 4-2 antigen was mainly expressed on rat T lymphocytes rather than on B lymphocytes: In fact 59.6 +/- 3.2% (in PBL), 76.5 +/- 0.6% (in LN) and 67.1 +/- 1.3% (in spleen) of T lymphocytes exhibited the 5N 4-2 antigen compared to only 26.5 +/- 2.6% (in PBL), 34.0 +/- 2.1% (in LN) and 46.1 +/- 12.0% (in spleen) of B lymphocytes. As expected a strong association was found between the expression of 5N 4-2 antigen and 5'-nucleotidase enzyme activity on lymphocytes. Both 5N 4-2 positive cells and enzyme activity were preferentially exhibited in the T lymphocyte subpopulation, and 92% of the enzyme activity was observed in a 5N 4-2 antigen positive subpopulation.

Detection of monoclonal antibodies against cell surface antigens: the use of antiglobulins coupled to red blood cells.

An antiglobulin-coupled red cell assay is described for screening monoclonal antibodies against cell surface antigens. A monoclonal antibody specific for rat immunoglobulin kappa chains was coupled to red blood cells and used to detect binding of rat monoclonal antibodies to cells attached to the wells of microtitre plates. The method was found to be simpler and more rapid than the alternative enzyme-linked binding assay and useful for rapid screening and selection of antibodies for use as differentiation markers of human and mouse haemopoietic cells.

Direct antibody rosette-forming reactions using monoclonal markers of lymphocyte subpopulations. Methodology and applications illustrated by investigations with rat pan-T antibodies of the CAMPATH series.

A sensitive direct antibody rosette assay has been developed for the detection of antigens on the lymphocyte cell membrane. Indicator cells for rosette tests were prepared by chromic chloride coupling of rat or mouse monoclonal IgG or IgM anti-lymphocyte antibodies to untreated or trypsinized bovine red blood cells. The monoclonal antibodies used were reactive with a range of cell surface markers which identify various lymphocyte subpopulations, including T cell antigens, HLA class II (Ia-like antigens), Leu-7 (HNK-1) and VEP 13, a determinant of Fc gamma receptors on large granular lymphocytes. Results obtained by direct rosette formation correlated well with those of parallel tests using indirect immunofluorescent antibodies staining. Several applications of the direct antibody rosetting procedure are described in further investigations with a series of pan-T monoclonal (CAMPATH) antibodies. These include the morphological examination of antibody-binding cells in cytocentrifuge smears, the separation of lymphocyte subsets by density gradient centrifugation, and the use of a rosette inhibition assay to identify monoclonal antibodies binding to the same (or closely associated) epitopes of the lymphocyte cell membrane.

Glutaraldehyde stabilisation of antibody-linked erythrocytes for use in reverse passive and related haemagglutination assays.

A simple method for stabilising antibody-linked red blood cells by the addition of low concentrations of glutaraldehyde is described. Fresh and stabilised reagent-linked cells were shown to compare favourably in reverse passive haemagglutination for the measurement of human immunoglobulin isotypes, G, A and M and for the detection of respiratory syncytial and herpes simplex viruses. Stabilised cells were also used to detect antibodies to bacteria and to a soluble antigen adsorbed to a solid phase by mixed haemagglutination reactions.

Red cell-labelled monoclonal antibodies for assay of human chorionic gonadotropin and luteinising hormone by reverse passive haemagglutination.

The assay of human chorionic gonadotropin and luteinising hormone by reverse passive haemagglutination reaction, using monoclonal antibodies coupled to red cells, is described. Quantitation is achieved by end-point determination for serial dilutions of standard or sample, the haemagglutination reaction being observed after settling under gravity for 90 min. Red cell-labelled antibodies were stabilised with glutaraldehyde without loss of sensitivity and allowing long term storage. Various antibody combinations were assessed, and the best combination under optimum conditions gave a positive haemagglutination reaction down to 0.2 ng/ml with HCG.

Measure of rheumatoid factor in human sera by passive haemagglutination of human erythrocytes carrying immunoglobulin linked by chromic chloride.

Serum rheumatoid factor in sero-positive patients with rheumatoid arthritis may be measured by passive haemagglutination of trypsin-treated human red cells linked with heat-aggregated human IgG by chromic chloride. The results show excellent correlation with those obtained with the classical Rose-Waaler test. The sera may be tested unheated and do not require preliminary absorption with red cells. By this test procedure it should also be possible to analyse the species, allotypic and conformational specificity of different rheumatoid factors.

Isolation of low-frequency class-switch variants from rat hybrid myelomas.

Class-switch variants have been isolated from rat-rat hybrid myelomas by sib selection using a simple assay based on red cell-labelled antiglobulins. The variants detected are consistent with the gene order deduced from molecular cloning. They appear to arise spontaneously at a rate approximately ten-fold lower than for mouse cell lines but the rate of switching back to the parental isotype is substantial in comparison. An IgG2b variant antibody having the same specificity as CAMPATH-1 for human lymphocytes and monocytes is active in antibody-dependent cell-mediated killing (unlike the parental IgG2a) and may prove to be a valuable therapeutic antibody for immunosuppression and treatment of leukaemia and lymphoma.

An indirect anti-immunoglobulin rosetting reaction to detect alloantibodies to human lymphocytes.

A test is reported which detects alloantibody, absorbed onto the surface of human lymphocytes from multiparous antisera, by means of a red cell rosette assay. The red cells, trypsinised ox, are coupled with anti-immunoglobulin using chromic chloride. The antiglobulin used is rabbit anti-human IgG (Fc), chosen to avoid reaction with the surface immunoglobulin naturally present on human B lymphocytes. The reaction is termed the Indirect Anti-immunoglobulin Rosetting Reaction (IARR). The IARR is shown to be specific in the following ways: anti-immunoglobulin coupled ox cells do not react with normal human lymphocytes nor with lymphocytes treated with non-reactive serum. Red cells coupled with normal rabbit IgG do not react with normal or alloantibody coated lymphocytes. Multiparous sera (reactive with other individuals) do not react with cells of the serum donor in the IARR. Finally, the coupled red cells do not usually react with lymphocytes which have absorbed immune complexes onto their Fc receptors. The IARR is shown to be more sensitive than a standard cytotoxic test for detection of alloantibody. Several possible applications of the IARR are discussed.

A simplified procedure for measuring the class of anti-bacterial antibodies by mixed reverse passive antiglobulin haemagglutination (MRPAH).

A modified procedure is described for performing the MRPAH (mixed reverse passive antiglobulin haemagglutination) reaction as a simple micro-method to measure the classes of bacterial antibodies. This 'bacterial dilution procedure' gave results closely correlated with those obtained by the 'serum (sample) dilution procedure' previously reported and with great economy of materials, labour and time. The method was used to investigate human serum antibodies to Br. abortus and S. enteritidis and serum and secretory antibodies to Strep. mutans. The good reproducibility of the MRPAH reaction was demonstrated by re-examining brucellosis sera tested one year previously. MRPAH was sufficiently sensitive to demonstrate the small amounts of IgG and IgM antibodies to Strep. mutans in human colostrum and early milk. A rise of antibody levels in the different immunoglobulin classes G, A and M was readily demonstrated in sera from individuals with salmonellosis.

Detection of protein A-like substances on haemolytic streptococci prior to use in mixed reverse passive antiglobulin haemagglutination (MRPAH).

Before mixed reverse passive antiglobulin haemagglutination tests (MRPAH) can be used to measure the class of bacterial antibodies, the bacteria have to be shown to be free of Protein A or Protein A-like substances on their surfaces. Two basic procedures have been examined: haemagglutination of red cells coated with immunoglobulin by the bacteria, and the MRPAH reaction itself to reveal absorption of purified gamma Fc by the bacterial suspension. The use of a purified gamma Fc component has proved successful in providing a sensitive test for the detection of Protein A-like substances on the surface of bacterial. In addition to both the Cowan and Wood strains of Staph, aureus, strains of haemolytic streptococci of groups A, C and G had Protein A-like substances on their surfaces. In contrast, strains of group B and group D, as well as Strep. milleri, had no detectable Protein A-like activity.

MRSPAH: a simple microtitre plate test for rheumatoid factors of different classes.

A simple and inexpensive microtitre plate test (the mixed reverse (solid-phase) passive antiglobulin haemadherence test, or MRSPAH) has been developed for the measurement of antiglobulins (RFS) of different classes. Results obtained for IgM RF by this test have been compared with results of latex and Rose-Waaler (DAT) tests on rheumatoid arthritis (RA) sera. Levels of RFs of IgM, IgG and IgA classes have been measured by MRSPAH using rabbit IgG as antigen in RAs and normal people. 94% of RA sera tested were above the upper limit of normal for IgM and/or IgA RF. There was a considerable overlap between IgG RF levels in RAs and normals, although the means of the two groups were significantly different.

The demonstration of sIg, MHC and T cell antigens and Fc receptors on the lymphocyte surface by anti-globulin rosetting reactions: some technical considerations.

The previously shown marked difference in sensitivity of antiglobulin rosette formation and immunofluorescence was confirmed in detection of varying amounts of anti-Ig bound to B cells. The direct and indirect rosette method revealed twice the number of sIg+ cells shown by direct immunofluorescence (DIF), and at least an order of magnitude more antibody on the cell surface is necessary to detect the conventional B cells (shown by DIF) by indirect immunofluorescence compared with indirect antiglobulin (IARR) rosette formation. Variants of the sensitive IARR test were developed to reveal general lymphocyte, MHC class I and class II and T cell-specific antigens and Fc receptors using xenogeneic or allogeneic reagents. Sensitive indicator cells were used which form almost no rosettes with unsensitized lymphocytes although coated with IgG which would otherwise bind to the Fc receptor. The suggestion that chromic chloride treatment inactivates the relevant Fc determinants was supported by experiments showing the failure of IgG antibody-sensitized bovine RBC (standard Fc indicator cells) to form rosettes after treatment with chromic chloride.

Cryopreservation of antibody-coupled red cells for use in immunoassays.

Red cells, trypsin-treated to render them more agglutinable and coupled with antiglobulin reagents, may be preserved by droplet freezing in liquid nitrogen. A 2% cell suspension in 0.45% w/v sodium chloride, 5% w/v sucrose and 10% w/v dextran 40 was used. After thawing the frozen pellets in phosphate-buffered saline at 40 degrees C, more than 80% cell recovery was obtained. Sheep and ox red cells preserved in this way were as satisfactory in antiglobulin and in reverse passive haemagglutination tests as unfrozen indicator red cells. Trypsin-treated human red cells coupled with anti-IgE could likewise be frozen and on reconstitution used to assay IgE in human serum. Reconstituted ox red cells were slightly less efficient in rosetting than cells which had not been frozen.

Histological diagnosis of carcinoma of the parathyroid gland.

During the period 1950-81, 678 cases of primary hyperparathyroidism were surgically treated at University College Hospital, London. The causes were a single adenoma in 575, two adenomas in 25, carcinoma in 20 (two of which had coexistent adenomas), chief cell hyperplasia in 56, and water clear cell hyperplasia in two. Histological diagnosis is not difficult except in some cases of carcinoma and in a few in which differentiation between recurrent hyperplasia and recurrent carcinoma is exceptionally difficult. In this paper we review all the cases of primary carcinoma of the parathyroid seen during this period to define those pathological features of value in the diagnosis.

A solid phase assay with radioactively-labelled antibody for the detection of Brucella abortus.

An assay for the detection of Brucella abortus is described. IgG from rabbit antisera against live brucellae or brucella extracts was chemically linked to cellulose to form a solid phase reagent capable of binding brucella antigens present in buffer solutions or serum. After washing away any unbound material the presence of bound antigen was revealed by incubation with radioactively-labelled anti-brucella antibodies. The assay was capable of detecting less than 100 pg brucella antigen in a 20 microliter sample. Experiments in which IgG of non-related specificity was used in place of anti-brucella IgG showed that the test was specific. Normal human serum had only a slight inhibitory effect but anti-brucella antibodies were strongly inhibitory if present in the test sample. The extent of this effect and its relationship to antibody titre was investigated in 12 sera from brucellosis patients.