Abstract
While both murine and human homologues of the LSP1 gene (lymphocyte-specific gene 1) and its protein products have been identified, studies on human LSP1 have been limited. The present report describes a detailed immunocytochemical study of the distribution and localization of human LSP1 in both normal and neoplastic cells and tissues. The specificity of the monoclonal anti-LSP1 reagent was confirmed by expression cloning and transfection studies. The intracellular 60 000 MW LSP1 protein was found to be present in peripheral blood B cells, monocytes and granulocytes but absent in a subpopulation of circulating T cells (10–15% of CD3-positive T cells). The presence of LSP1 protein in medullary thymocytes, but only in scattered cortical thymocytes, provided additional evidence for heterogeneity of expression in T cells. Novel observations also included the presence of LSP1 in plasma cells, dendritic cells and Langerhans’ cells. The leucocyte-restricted distribution of LSP1 protein means that it may play an important role in haematopathology. LSP1 protein was detected in a wide range of leukaemias and lymphomas, particularly of B-cell origin, and in tumour cells in classical Hodgkin’s disease. Of interest was the indication of a reciprocal relationship in the expression of LSP1 and ALK (anaplastic lymphoma kinase) proteins in patients with anaplastic large cell lymphoma. As the anti-LSP1 reagent used in the present study recognizes a formalin-resistant epitope it should be of considerable value in the diagnosis of routinely fixed material.
INTRODUCTION
The murine LSP1 (lymphocyte-specific gene 1) gene identified in 1988 encodes a 300-amino acid phosphoprotein, present in pre-B cells, B cells, concanavalin A (Con A)-stimulated thymocytes, macrophages and granulocytes,1–5 which appears to be under the control of different combinations of transcription factors in T and B cells.6 The presence of two Ca2+-binding sites, together with the ability of LSP1 to bind to actin and to co-cap with immunoglobulin M (IgM),1–4 suggests a potential role for this protein in signal transduction within the white cells. There has also been a report that its expression is reduced in murine T cells following neoplastic transformation, raising the possibility of a role for LSP1 protein in oncogenesis.7
The human homologue of murine LSP1 (also known as WP34, pp52 or leufactin) was identified in 1990 as a 339-amino acid protein whose predicted sequence shows 67% identity with mouse LSP1.3,8 In contrast to murine LSP1, human LSP1 protein contains only one possible Ca2+-binding site.3 Human LSP1 is encoded by a gene present on chromosome 11 at p15·59 and it has been implicated in a human immunodeficiency syndrome associated with neutrophil actin dysfunction.10,11
Despite the presence of a considerable body of information concerning murine LSP1 protein, studies on the cell and tissue distribution of human LSP1 protein have been limited to Western blotting, immunoprecipitation and mRNA analyses of peripheral blood cells, a few cell lines and tissue lysates such as fetal liver and thymus, using both monoclonal and polyclonal anti-LSP1 antibodies. While these studies have demonstrated that human LSP1 protein is found in lymphoid cells and granulocytes,8,10,12 and appears to play a role in cell motility10,13 little else is known about its distribution.
In the present study we have used a monoclonal antibody (mAb) specific for a denaturation-resistant epitope on the LSP1 protein to document its distribution in a wide range of human cell lines and both normal and neoplastic tissues and to estimate its potential value as a diagnostic tool.
MATERIALS AND METHODS
Cell and tissue samples
Normal and reactive tissues and lymphoma samples were obtained from the Histopathology Department of the John Radcliffe Hospital, Oxford. Blood was obtained from normal volunteers. Human cell lines were obtained and maintained in culture as previously described.14,15 Tumour samples were classified according to the REAL classification.16 U937 cells were incubated with 1·6×10−8 m 12-0-tetradecanoylphorbol 13-acetate (TPA, Sigma Chemical Co., Poole, Dorset, UK) for 3 days to promote their differentiation towards macrophages. Cytocentrifuge preparations were made of all cell suspensions. Sections and cytocentrifuge preparations were fixed and stored as previously described.17
Antibodies
Monoclonal anti-LSP1 antibody (TPD153)
After immunization with tonsil cells, a fusion of the murine spleen cells with NS1 myeloma cells was carried out as described previously.17 One hybridoma supernatant reacted selectively with leucocytes when tested on cryostat sections of human tonsil and kidney, and the cells producing this supernatant were cloned to yield the mAb TPD153 (IgG1 isotype). Concentrated tissue culture supernatant was directly conjugated to fluorescein isothiocyanate (FITC) using the FluoroTag FITC Conjugation Kit (Sigma Chemical Co.).
Other monoclonal reagents
Antibodies to CD3, CD19, CD20, CD68, major histocompatibility complex (MHC) class II, vimentin, actin and cytokeratin, and phycoerythrin (PE)-conjugated anti-CD3 and anti-CD19, were obtained from DAKO A/S (Glostrup, Denmark). Antibodies to anaplastic lymphoma kinase (ALK),14,18 CD15, rabbit immunoglobulin (MR12) and CD45 (2B11 and PD7/26) were prepared in the authors’ laboratory.
Polyclonal antibodies
Rabbit anti-CD3, rabbit anti-S100 protein, horseradish peroxidase (HRP)-conjugated goat antimouse immunoglobulin and HRP goat anti-rabbit immunoglobulin were obtained from DAKO A/S. Polyclonal anti-CD8 was raised in the authors’ own laboratory using a synthetic peptide.19 FITC-conjugated rabbit anti-mouse F(ab′)2 immunoglobulin was obtained from DAKO A/S. Goat isotype-specific anti-mouse immunoglobulin and goat anti-rabbit immunoglobulin conjugated to either FITC or Texas Red™ were purchased from Eurogenetics (Middlesex, UK).
Immunological cDNA library screening
Screening of an oligo dT and random-primed lambda ZAP Express cDNA library prepared from circulating blood (Stratagene, Cambridge, UK) was carried out as described previously.18 Protein expression was detected using antibody TPD153 and HRP-conjugated goat antimouse antibody (DAKO A/S). The antigen–antibody complexes were visualized using diaminobenzidine/H2O2 with metal ion enhancement. A second screen was performed to isolate individual positive clones.
DNA sequencing
The cDNAs encoding the TPD153 antigen in plasmid pBK-CMV were excised in vivo from the lambda ZAP Express vector (Stratagene) according to the manufacturer’s instructions. Individual plasmid DNAs (pAB278–280) were purified using Qiagen columns (Qiagen, Crawley, Surrey, UK) and the DNA was sequenced with M13 Universal and Reverse primers, using a Cy5 Autoread sequencing kit and an ALF Express DNA sequencer (Pharmacia, St Albans, Herts, UK).
Bacterial expression and detection
Escherichia coli strain XLOLR (Stratagene), containing either plasmid pAB280 or the empty vector pBK-CMV, was grown overnight, at 37°, in Luria–Bertoni (LB) medium containing ampicillin.20 Sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS–PAGE) and Western blotting were carried out as previously described.20,21 Protein was detected using antibody TPD153, HRP-conjugated goat antimouse (DAKO A/S) and ECL substrate (Amersham International Ltd, Amersham, Bucks, UK).
Expression of TPD153 antigen in transfected cells
Proteins encoded by the pAB279 and pAB280 plasmids were transiently expressed in the COS cell line using the diethylaminoethyl (DEAE) transfection method.22 The cells were harvested after 24, 48, 72 and 144 hr of incubation and cytocentrifuge preparations were made.
Immunostaining
Immunocytochemical labelling
Paraffin sections were subjected to microwave pretreatment.14 Immunocytochemical staining of cytocentrifuge preparations, cryostat and routinely processed sections was performed using either a two-stage immunoperoxidase technique or the alkaline phosphatase: anti-alkaline phosphatase (APAAP) technique.17,23
Double immunofluorescence staining
Cytocentrifuge preparations and tissue sections were incubated for 30 min with anti-LSP1 supernatant together with either a polyclonal antibody (anti-CD3, CD8 or S100) or one of the following mAb: CD15 (IgM isotype), CD20 (IgG2a isotype) or CD68 (IgG3 isotype). After washing in phosphate-buffered saline (PBS), the slides were incubated for a further 30 min in the dark with goat anti-rabbit immunoglobulin or goat anti-mouse IgG-specific subtype antibodies conjugated to either FITC or Texas Red™. After washing, the slides were counterstained with 0·002% 4′,6-Diamine-2′-phenylindole dihydrochloride (DAPI, Boehringer Mannheim, Lewes, Sussex, UK) for 10 min. The slides were washed in water and mounted in an antifade mountant (DAKO A/S). The slides were then visualized on a Zeiss Axioskop microscope, and pictures were taken using a cooled CCD camera (Hamamatsu, Hamamatsu City, Japan) using Improvision Openlab software (Improvision, Warwick, Warwickshire, UK) running on an Apple Macintosh computer.
Flow cytometric analysis
Suspensions of normal peripheral blood mononuclear cells and tonsil cells were permeabilized using buffered formol acetone (BFA).24 Labelling of surface and intracellular antigens was performed as previously described.25 Double labelling was performed by incubating the cells with FITC–TPD153 and with either PE–CD3 or PE–CD19. Following a final wash, the cells were fixed and the results analysed in a FACScan using CellQuest software (Becton-Dickinson, Oxford, UK). In some experiments the cells were not permeabilized before labelling so that only surface antigens could be studied.
Western blotting
Cell lysates from cell lines and tonsil cell suspensions were prepared as previously described.14 Cytoplasmic extract (10 μg) was resolved in a 7·5% polyacrylamide gel by SDS–PAGE. Proteins were transferred to immobilon polyvinylidene difluoride (PVDF) membrane by semidry electroblotting. The antigens recognized by antibody TPD153 were then detected using the ECL chemiluminescent substrate (see above).
RESULTS
Identification of LSP1 antigen as the target for antibody TPD153
Gene cloning
Seven clones encoding the TPD153 antigen were isolated. The cDNA inserts in the plasmid clones were sequenced in both directions and a search against the GenEMBL databank revealed that they all encoded human LSP1 protein (accession number: M33552). Two of the plasmids encoded the full-length LSP1 (pAB278 and pAB279) while the smallest clone (pAB280) only contained nucleotides 328–1495 of the published 1631-bp LSP1 cDNA.
Antibody TPD153 recognized a 58 000 MW β-galactosidase (β-Gal) fusion protein expressed by plasmid pAB280 in E. coli. (Fig. 1a) but was unreactive with the β-Gal protein expressed by the empty pBK-CMV vector (Fig. 1b). These data confirm that antibody TPD153 recognizes the LSP1 protein and show that its epitope is encoded by nucleotides 328–1495.
Figure 1.
Western blotting studies on lymphocyte-specific protein 1 (LSP1).(a) Recombinant protein. Antibody TPD153 recognizes LSP1 protein expressed by plasmid pAB280 but is unreactive with the products of the ‘empty’ pBK-CMV vector. The smaller bands are assumed to be LSP1 degradation products. The similar amounts of the LSP1 protein present before and 3 hr after the addition of IPTG indicate that the protein expression is not tightly regulated by the IPTG-inducible promoter in the pBK-CMV vector. The positions of molecular weight markers are shown. (b) Lymphoid tissue and haematopoietic cell lines. LSP1 protein is detected as a 60 000 MW band in lysates of normal tonsil cells and of several leucocyte-derived cell lines. The small amount detectable in the U937 myeloid cell line is increased by 12-0-tetradecanoylphorbol 13-acetate (TPA) stimulation. No LSP1 protein is detected in the Nalm-1 pre-B cell, the CEM T-ALL cells, the erythroid cell line (K562) or the megakaryocytic HEL cell line. An isotype-matched antibody, MR12, mouse antirabbit immunoglobulin, was used as a negative control.
Immunocytochemical staining of cytospin preparations of COS cells transfected with the pAB279 and pAB280 plasmids further confirmed that TPD153 antibody recognized both the full-length and truncated LSP1 protein expressed in eukaryotic cells (Fig. 2a)No labelling by antibody TPD153 was observed of cells transfected with ‘empty’ pBK-CMV vector alone.
Figure 2.
Immunolabelling of cells and normal tissues for human lymphocyte-specific protein 1(LSP1).The anti-LSP1 antibody,TPD153,labels cytocentrifuged preparations of(a)COS cells transfected with plasmid pAB279 encoding full-length LSP1 protein and the truncated LSP1 cDNA encoded by plasmid pAB280(containing nucleotides 328–1495).No labelling was seen of those cells that had been transfected with the empty pBK-CMV vector(inset).(b)LSP1 protein is extensively present in germinal centres(GC),mantle zones(MZ)and interfollicular areas of normal tonsil.Tingible body macrophages in the germinal centres were LSP1 negative(arrowed).Higher magnification shows LSP1 to be present in (c) Langerhans’ cells and (d) plasma cells (arrowed).Note the absence of LSP1 protein in epithelium.(e)The Hut 78 T-cell line shows polarized localization of LSP1 protein (arrowed) in the cytoplasm.(f)Double-immunofluoresence labelling of Hut 78 cells for LSP1 (green)and actin(red)shows areas of co-localization(yellow) of these two proteins in some cells (arrowed). Double-immunofluorescence labelling of peripheral blood cells showing (g) an LSP1-positive monocyte(green) and a CD3-positive but LSP1-negative(red)T cell(arrowed).(h)Double labelling with anti-CD20(red)and anti-LSP1 (green) shows two CD20-positive B cells that also express LSP1 protein(arrowed).In(i)LSP1 protein(green)is detected both in lymphocytes and neutrophils, which are identified in(j)by labelling for CD15(red).(k)Double-immunofluoresence labelling of thymus for LSP1(green)andCD3(red).LSP1-positive CD3-positive T cells(yellow)are present in the medulla, while the majority of cortical thymocytes express only CD3 (red). Higher magnification of the cortico-medullary junction in (l) shows the scattered LSP1-positive CD3-positive T cells (yellow) in the cortex in more detail (arrowed). In (m), a section of thymus is double labelled with anti-CD3 (red) and anti-CD45RA (green) to demonstrate that the CD45RA antigen, as found for LSP1, is confined mainly to the thymic medulla. In another high magnification of the thymus(n),double-immunofluorescence labelling shows the presence of a large LSP1-positive(green, arrowed) cell amongst CD68-positive (red) macrophages. (o) LSP1 is expressed in a lymphoid area in the spleen but not in the sinusoidal lining cells or in the majority of red pulp macrophages (inset). The LSP1-positive leucocytes are clearly seen within the sinusoid (arrowed). A comparison of labelling of lung (p) and liver (q) for the presence of LSP1 and CD68 antigens shows that only a minority of alveolar macrophages and no Kupffer cells are LSP1 positive. Labelling was performed by immunoperoxidase (a, e and p), immuno-alkaline phosphatase (b, c, d, o and q) or immunofluorescence (f–n) techniques.
Biochemical and flow cytometric analysis
Western blotting studies using antibody TPD153 on tonsil cell lysates revealed a protein of 60 000 MW (Fig. 1b), in agreement with previously published data on LSP1 protein.3,26 Flow cytometry studies of tonsil cell suspensions showed that LSP1 protein could be detected in 95% of the cells following permeabilization but not in untreated cells (Fig. 3a, b), in agreement with its known intracellular location.3,8
Figure 3.
Flow cytometric studies of the lymphocyte-specific protein 1 (LSP1) protein on tonsil cells and peripheral blood lymphocytes. LSP1 protein is undetectable in fresh tonsil cells (a) but can be detected in more than 95% of these cells after permeabilization (b). Double-labelling studies of permeabilized tonsil cells demonstrate that all of the CD19-positive B cells (c) and CD3-positive T cells (d) express LSP1. In contrast, labelling of peripheral blood lymphocytes demonstrates that while all of the CD19-positive B cells are LSP1-positive (e), a subpopulation of the CD3-positive T lymphocytes do not express LSP1 protein (f, arrowed). For single labelling, cells were incubated for 30 min on ice with anti-LSP1, MR12 (isotype-matched negative control) or antivimentin (to confirm that more than 95% of the cells had been permeabilized).
Distribution of LSP1 protein in cell lines and normal cells
B cells
LSP1 protein was detected as a 60 000 MW band in lysates of the Daudi B cell line and the Thiel plasma cell line (Fig. 1b). In contrast, no LSP1 protein could be detected in the pre-B cell line, Nalm-1 (Fig. 1b). The same results were obtained by immunostaining studies (Table 1).
Table 1.
Lymphocyte-specific protein 1 (LSP1) in human cell lines
The symbols − to + + + refer to the intensity of immunoperoxidase labelling of cytocentrifuge cell preparations.
LSP1 protein was expressed in B lymphocytes in the blood and normal tissue (Table 2). In sections of tonsil and spleen, LSP1 protein was present in B cells in germinal centres, mantle zones and marginal zones (Fig. 2b, o). The strongest labelling was seen in plasma cells, where the cytoplasm stained homogeneously with the anti-LSP1 reagent (Fig. 2d). The presence of LSP1 in all tonsillar B cells was confirmed using flow cytometry (Fig. 3c).
Table 2.
The distribution of lymphocyte-specific protein 1 (LSP1) in normal human cells and tissues
*Immunocytochemical labelling was carried out on formalin-fixed paraffin-embedded tissue samples.
Flow cytometry studies and immunofluorescent double labelling of cytocentrifuge preparations of peripheral blood mononuclear cells demonstrated that all CD19-positive circulating B cells expressed LSP1 protein (Figs. 2h, 3e).
T cells
LSP1 protein was detected by Western blotting (Fig. 1b) and immunocytochemical labelling in only five of the eight T-ALL cell lines studied (Table 1). Diffuse cytoplasmic labelling was observed in the Du-528 and Jurkat cell lines. In contrast, LSP1 protein tended to be localized at one pole of the cell in the Hut 78 cell line (Fig. 2e). In the Molt 4 and RPMI 8402 T-cell lines, only a proportion of the cells (50% and 10%, respectively) were weakly LSP1-positive. LSP1 protein was not detected in the Molt 13, Peer or CCRF-CEM cell lines.
Immunocytochemical labelling of tissue sections (summarized in Table 2) demonstrated that T cells in the tonsil and spleen were LSP1-positive. Flow cytometry results confirmed that all tonsillar CD3-positive T cells expressed LSP1 (Fig. 3d). In contrast, while thymic medullary T cells expressed LSP1 protein, only a few scattered CD3-positive cortical thymocytes were LSP1 positive (Fig. 2k, l). This pattern of labelling was similar to the distribution of the CD45RA antigen (Fig. 2m).
Double immunofluorescence labelling of peripheral blood T-cell populations confirmed the heterogeneity of LSP1 protein expression (see Table 2). LSP1-negative CD3-positive T cells were observed in peripheral blood smears (Fig. 2g) and flow cytometry studies confirmed that a subpopulation (10–15%) of CD3-positive peripheral blood T cells did not express LSP1 (Fig. 3f).
Double immunofluorescence studies were performed on the Hut 78 cell line, using anti-LSP1 and anti-actin. Areas of co-localization of LSP1 protein and actin could be observed in the cells (Fig. 2f), although both proteins were also found distributed separately throughout the cytoplasm.
Monocytes, macrophages and myeloid cells
Western blotting and immunocytochemical studies demonstrated the presence of the 60 000 MW LSP1 protein in HL60 and U937 cell lines (Fig. 1b). TPA stimulation of the U937 cells increased the amount of LSP1 protein (Fig. 1b) and the percentage of LSP1-positive cells (Table 1).
LSP1 protein was strongly expressed in circulating blood monocytes and neutrophils (Fig. 2i, j) being evenly distributed within the cytoplasm in these cells. By contrast, tissue macrophages (as defined by CD68 expression) were either LSP1 negative (tingible body macrophages in germinal centres, Fig. 2b) or only a subpopulation of cells expressed LSP1 (in the thymic cortex, splenic red pulp, lung and liver; Fig. 2o, p, q).
LSP1 protein was also expressed strongly in Langerhans’ cells (Fig. 2c) and in the interdigitating reticulum cells (dendritic cells) in the T-cell areas of the tonsil and in a case of dermatopathic lymphadenopathy (Fig. 4). Large branching LSP1-positive cells were also scattered throughout the thymic cortex (Fig. 2n) and were identified as probable dendritic cells because they also expressed high levels of MHC class II and S100 protein but lacked CD68 and cytokeratin.
Figure 4.
Immunofluorescence labelling to show the distribution of human lymphocyte-specific protein 1 (LSP1) in a case of dermatopathic lymphadenopathy. Double labelling for LSP1 (green) and S100 (red) proteins shows that the large cluster of S100-positive interdigitating reticulum cells in the interfollicular area of the lymph node co-expresses LSP1 protein (yellow). A higher magnification shows the double labelling of these interdigitating reticulum cells (yellow, arrowed) in more detail. Lymphocytes expressing only LSP1 protein (green) are also observed (arrowhead).
Erythroid, megakaryocytic and non-haematopoietic cells
LSP1 protein was not detectable in the K562 and HEL cell lines by using either Western blotting or immunocytochemical methods (Fig. 1b, Table 1). Megakaryocytes and erythroid precursors in fetal liver and cord blood were LSP1 negative (Table 2).
Non-haematopoietic cells
In agreement with the data from the study of cell lines (Table 1), LSP1 protein was not expressed in any of the non-haematopoietic tissues investigated (Table 2).
Distribution of LSP1 protein in leukaemias and lymphomas
The results obtained are summarized in Table 3. LSP1 protein was widely expressed in tumours of B-cell origin, such as hairy cell leukaemia (HCL) (Fig. 5a). Two of four myeloma patients tested were positive for LSP1 (Fig. 5b). LSP1 protein was detected in only six (all precursor B-cell leukaemias) of the 15 patients with acute lymphoblastic leukaemia.
Table 3.
The distribution of lymphocyte-specific protein 1 (LSP1) in leukaemias and lymphomas
*Immunocytochemical labelling was carried out on formalin-fixed paraffin-embedded tissue samples.
Figure 5.
Immunoalkaline phosphatase labelling of routinely processed tissue samples to show lymphocyte-specific protein 1 (LSP1) expression in lymphoid neoplasia. (a) A case of hairy cell leukaemia (HCL) in the spleen showing the presence of a large intrasinusoidal cluster of LSP1-positive neoplastic cells (arrowed). In (b) the myeloma cells show cell-to-cell variation in the intensity of LSP1 expression. The long, elongated, strongly positive (arrowed) cells are macrophages (previously identified as CD68-positive cells). (c) Reed–Sternberg cells are very strongly labelled (arrowed) in a patient with Hodgkin’s disease. The intensity of their labelling exceeds that of the surrounding normal cells (seen in greater detail in the inset). The case of ALCL shown in (d) is LSP1 negative. The LSP1-negative tumour cells (arrowed) can be easily seen amongst the strongly LSP1-positive normal haematopoietic cells. A higher magnification showing double-immunofluorescence labelling for LSP1 (green) and ALK (red) allows the reciprocal distribution of the LSP1 and ALK proteins to be seen in the tumour cells.
Most patients with classical Hodgkin’s disease were LSP1 positive. The Reed–Sternberg cells were particularly strongly stained (Fig. 5c). By contrast, LSP1 protein was detected in the tumour cells of only three of the seven patients with lymphocyte-predominance Hodgkin’s disease.
LSP1 protein was present in only four of the nine patients with anaplastic large cell lymphoma (ALCL). The neoplastic cells in three of these patients were very strongly labelled, indicating the presence of large amounts of LSP1 protein in the cytoplasm. All of these neoplastic cells were shown to be ALK negative (as determined by the antibody ALK1).14 The tumour cells in the fourth patient expressed variable amounts of LSP1 protein, with some neoplastic cells expressing high levels of LSP1 protein while other cells were LSP1 negative. This patient, together with three LSP1-negative patients with ALCL, were shown to contain ALK protein (Fig. 5d). Double-immunolabelling studies of the LSP1-negative lymphomas confirmed the absence of LSP1 protein in the ALK-positive tumour cells (Fig. 5d). The two remaining patients with ALCL were negative for both LSP1 and ALK proteins (Table 3)
DISCUSSION
This study provides the first detailed description of the distribution of LSP1 in a number of different human tissues. The study also resolves discrepancies in published descriptions of the distribution of human LSP1 where contradictory results have been obtained.10,12
The mAb, TPD153, was identified as being specific for human LSP1 protein by expression cloning (and by its reactivity with cells transfected with the LSP1 gene). Flow cytometric studies confirmed that the LSP1 protein was an intracellular protein, in agreement with the lack of a transmembrane region in the predicted sequence of LSP1 protein.4,10 The value of 60 000 obtained in the present study for the molecular weight of the LSP1 protein is at variance with the predicted weight of 36 000 for the LSP1 protein,4 but in keeping with previous biochemical observations of molecular weights, which vary between 47 000 and 60 000.2,3,8,10,26,27
In agreement with Li et al.12 LSP1 protein was expressed in all normal B cells, and in Daudi and Namalwa cell lines, but absent from pre-B cells. An important novel observation, however, was the presence of LSP1 protein in normal plasma cells (and in a plasma-cell line). LSP1 has been identified as the major cellular substrate for protein kinase C, an enzyme that plays a key role in signal transduction in B cells.12,26 This fact suggests that LSP1 protein may be of functional importance in B-cell signalling and development.
Previous reports have suggested that while LSP1 mRNA and/or LSP1 protein is expressed in normal T cells, it is absent or present in only small amounts in transformed T lymphocytes and T-cell lines.2,8,12 Data from the present study are in general agreement with these results in that expression of LSP1 protein differed between the T-cell lines tested. The one exception was the high level of LSP1 protein present in the T-ALL-derived Hut 78 cell line. The polarization of LSP1 protein in this cell line was striking and its co-localization with actin is of interest. Reports that LSP1 protein is partially associated with the cytoskeleton2,4,10 and plays a role in cell movement10,11,28 may be relevant to this observation.
We also report heterogeneity of LSP1 protein expression in normal T cells. This was partly related to cell maturation, in that very few cortical thymocytes expressed LSP1 whereas it was strongly expressed in the medulla (Fig. 2k, 2l). The absence of LSP1 from the thymic cortex in humans is similar to the expression pattern of the CD45RA antigen (Fig. 2m),25 the acquisition of which appears to be necessary for T cells to migrate from the thymus.29 It is possible that human LSP1 protein plays a role in T-cell development in concert with CD45RA, and it may not be a coincidence that the former molecule is a major phosphorylation substrate and the latter acts as an intracellular phosphatase. We also noted that ≈10% of circulating CD3-positive T cells were LSP1 negative. Kadiyala et al.8 reported the presence of LSP1 protein in activated T cells, but the number of LSP1-positive T cells we detected was too great for all of them to represent activated cells. Matsumoto et al.27,30 have reported that LSP1 protein is also a substrate for protein kinase C in T cells, and it will be of interest to determine whether this is relevant to the heterogeneity of LSP1 expression in normal T cells.
In the present study, LSP1 protein was found in circulating granulocytes and monocytes (and in the U937 and HL60 cell lines), in agreement with reports of LSP1 protein expression in human granulocytes8,10 and monocytes10 and in murine macrophages and some myeloid cell lines.4 LSP1 protein has been identified as a major substrate for mitogen-activated protein kinase-activated protein (MAPKAP) kinase 2, phosphatases 1 and 2A (as well as for protein kinase C) in human granulocytes.31,32 In agreement with this, the amount of LSP1 protein was up-regulated in the present study in the myeloid-cell line, U937, following activation of protein kinase C by TPA. Similar results were previously obtained by Li et al.12 using another myeloid cell line, HL60. These results, together with the phosphorylation of LSP1 protein as a result of chemokine stimulation,28 suggest a role for LSP1 in myeloid-cell activation and inflammation.
Tissue macrophages were much more variable than circulating monocytes in their expression of LSP1. Many were LSP1 negative (Fig. 2o, p, q) and there was no clear pattern to the presence or absence of LSP1. Jongstra-Bilen et al.5 also previously reported heterogeneity in the levels of LSP1 mRNA in macrophages and suggested that this was a result of the differentiation stages of the macrophages. The present results suggest that once monocytes have matured to the macrophage stage, LSP1 protein may no longer play a major role in their physiology.
This study reports for the first time the expression of LSP1 protein by dendritic cells and Langerhans’ cells. These cells were identifiable by their morphology and their expression of high levels of MHC class II and S100 protein.33–35
In agreement with other studies, both in the mouse and in humans, it is evident that LSP1 is absent from human erythroid cells and platelets.3,5,8 We also report (for the first time) the absence of LSP1 protein from megakaryocytic cells (i.e. tissue megakaryocytes and the HEL cell line).
It is also clear from our results that LSP1 protein is not expressed in cells and tissues of non-haematopoietic origin. This is in agreement with the previous report by Kadiyala et al.8 who found no evidence of LSP1 mRNA in human liver and brain. We therefore interpret the LSP1 mRNA found in murine tissue homogenates by Jongstra-Bilen et al.5 as being the result of contaminating leucocytes.
The present study is the first detailed report of LSP1 protein distribution in human lymphoid neoplasia. LSP1 protein was detected in a wide range of leukaemias and lymphomas (mainly of B-cell type). Miyoshi et al. have reported the presence of LSP1 in HCL.36 The presence of LSP1 protein in chronic lymphocytic leukaemia in the present study is of interest. Chromosomal translocations involving chromosome 11 at p14-15, and therefore possibly involving the LSP1 gene, have been previously described in B-cell chronic lymphocytic leukaemia.37 In addition, Carballo et al.26 reported LSP1 protein to be the prominent protein kinase C substrate present in B-cell chronic lymphocytic leukaemia and suggested that LSP1 protein was overexpressed in the neoplastic cells.
The variable presence of LSP1 protein in the acute lymphoblastic leukaemias is in agreement with its distribution in pre-B and T-ALL cell lines.3,7,8 The absence of LSP1 protein in the small number of T-cell leukaemias studied here may be indicative of a possible link between the loss of LSP1 expression and the increased survival of neoplastic T cells, as suggested by Brennan et al.7
The Reed–Sternberg and Hodgkin’s cells expressed LSP1 protein in 94% (17/18) of the patients with classical Hodgkin’s disease. In contrast, only 43% (three of seven) of patients with lymphocyte-predominance Hodgkin’s disease contained LSP1 protein. These findings provide further evidence that the lymphocyte predominance form of Hodgkin’s disease is a unique lymphoproliferative disorder rather than a variant of Hodgkin’s disease.16,37,38 As the characteristic cells of this disorder usually express several B-cell antigens,39 this finding was unexpected in view of the distribution of LSP1 in other B-cell neoplasms.
Heterogeneity in the expression of LSP1 protein was observed in ALCL. The tumour cells in three patients were strongly labelled for LSP1 and did not contain ALK protein, a protein associated with ALCL.14,40–42 Conversely, three of the LSP1-negative patients with ALCL were shown to express high levels of ALK protein. These results are in agreement with the absence of LSP1 protein found in the t(2;5)-positive SU-DHL-1 cell line and suggest the possibility of a reciprocal relationship between ALK and LSP1 proteins. A study is currently being undertaken to further investigate the relationship between the expression of LSP1 protein and ALK proteins in ALCL.
The combination of broad reactivity within haematopoietic cells but absence from other cell lineages is strikingly reminiscent of the CD45 antigen, and prompts an investigation of possible interaction between the LSP1 and CD45 antigens. Furthermore, CD45 has been widely used by pathologists as a means of identifying white cell-derived neoplasms (e.g. lymphoma) and differentiating them from tumours, such as carcinomas. However, there are well-recognized examples of CD45-negative leucocyte-derived tumours (e.g. plasmacytomas), and because the anti-LSP1 antibody used in the present study recognizes a formalin-resistant epitope it may be of value in the diagnosis of such problem cases.
In conclusion, the present study confirms and expands existing knowledge concerning the LSP1 protein. The name lymphocyte-specific protein 1 is potentially misleading because LSP1 protein is clearly expressed in myeloid cells, an observation that led Li et al.12 to suggest the alternative name of leufactin for LSP1 protein. However, the title ‘leucocyte-specific protein’ seems to be more appropriate because, fortuitously, it can be indicated by the same initials.
Acknowledgments
We are grateful to Dr K. Micklem for his help in producing the illustrations. This work was supported by the Leukaemia Research Fund.
Glossary
Abbreviations
- ALCL
anaplastic large cell lymphoma
- ALK
anaplastic lymphoma kinase
- APAAP
alkaline phosphatase:antialkaline phosphatase
- BFA
buffered formol acetone
- DAPI
4′,6-Diamine-2′-phenylindole dihydrochloride
- HCL
hairy cell leukaemia
- HRP
horseradish peroxidase
- IPTG
isopropyl-β-D-thiogalactopyranoside
- LSP1
lymphocyte-specific protein 1
- PBST
phosphate-buffered saline containing 0·05% Tween-20
- PE
phycoerythrin
- TPA
12-0-tetradecanoylphorbol 13-acetate
References
- 1.Jongstra J, Tidmarsh GF, Jonstra-Bilen J, Davis MM. A new lymphocyte-specific gene which encodes a putative Ca2+-binding protein which is not expressed in transformed T lymphocyte lines. J Immunol. 1988;141:3999. [PubMed] [Google Scholar]
- 2.Klein DP, Galea S, Jongstra J. The lymphocyte-specific protein LSP1 is associated with the cytoskeleton and co-caps with membrane IgM. J Immunol. 1990;145:2967. [PubMed] [Google Scholar]
- 3.Jongstra-Bilen J, Young AJ, Chong R, Jongstra J. Human and mouse LSP1 genes code for highly conserved phosphoproteins. J Immunol. 1990;144:1104. [PubMed] [Google Scholar]
- 4.Jongstra-Bilen J, Janmey PA, Hartwig JH, Galea S, Jongstra J. The lymphocyte-specific protein LSP1 binds to F-actin and to the cytoskeleton through its COOH-terminal basic domain. J Cell Biol. 1992;118:1443. doi: 10.1083/jcb.118.6.1443. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Jonsgtra J, Ittel ME, Iscove NN, Brady G. The LSP1 gene is expressed in cultured normal and transformed mouse macrophages. Mol Immunol. 1994;31:1125. doi: 10.1016/0161-5890(94)90026-4. [DOI] [PubMed] [Google Scholar]
- 6.Omori SA, Smale S, O-Shea-Greenfield A, Wall R. Differential interaction of nuclear factors with the leucocyte-specific pp52 promoter in B and T cells. J Immunol. 1997;159:1800. [PubMed] [Google Scholar]
- 7.Brennan L, Jongstra J. Low expression of the LSP1 gene in early mouse T-lymphomas induced by N-methyl-N-nitrosourea. Carncinogensis. 1996;17:771. doi: 10.1093/carcin/17.4.771. [DOI] [PubMed] [Google Scholar]
- 8.Kadiyala RK, McIntyre BW, Krensky AM. Molecular cloning and characterisation of WP34, a phosphorylated human lymphocyte differentiation and activation antigen. Eur J Immunol. 1990;20:2417. doi: 10.1002/eji.1830201109. [DOI] [PubMed] [Google Scholar]
- 9.May W, Korenberg JR, Chen XN, et al. Human lymphocyte-specific pp 52 gene is a member of a highly conserved dispersed family. Genomics. 1993;15:515. doi: 10.1006/geno.1993.1102. [DOI] [PubMed] [Google Scholar]
- 10.Howard T, Li Y, Torres M, Guerrero A, Coates T. The 47 kD protein increased in neutrophil actin dysfunction with 47- and 89-kd protein abnormalities is Lymphocyte-Specific Protein. Blood. 1994;83:231. [PubMed] [Google Scholar]
- 11.Zhang Q, Li Y, Howard TH. Specific leufactin domains are required for F-actin binding and ‘hair’ formation characteristic of the human mobility defect -NAD 47/89. Blood. 1996;90(Suppl. 1):1941. Abstract. [Google Scholar]
- 12.Li Y, Guerrero A, Howard TH. The actin-binding protein, lymphocyte-specific protein 1, is expressed in human leucocytes and human myeloid and lymphoid cell lines. J Immunol. 1995;155:3563. [PubMed] [Google Scholar]
- 13.Howard T, Hartwig J, Cunningham C. Human LSP1 expression in eukaryotic cells create cell surface projections resembling the projections on neutrophils in NAD. Blood. 1994;84(Suppl. 1):890. Abstract. [Google Scholar]
- 14.Pulford K, Lamant L, Morris SW, et al. Detection of anaplastic lymphoma kinase (ALK) and nucleolar protein nucleophosmin (NPM)–ALK proteins in normal and neoplastic cells with monoclonal antibody ALK1. Blood. 1997;89:1394. [PubMed] [Google Scholar]
- 15.Pulford K, Micklem K, Thomas J, Jones M, Mason DY. A 72-kD B cell-associated surface glycoprotein expressed at high levels in hairy cell leukaemia and plasma cell neoplasms. Clin Exp Immunol. 1991;85:429. doi: 10.1111/j.1365-2249.1991.tb05744.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Harris NL, Jaffe ES, Stein H, et al. A revised European-American classification of lymphoid neoplasms: a proposal from the International Lymphoma Study Group. Blood. 1994;84:1361. [PubMed] [Google Scholar]
- 17.Erber WN, Pinching AJ, Mason DY. Immunocytochemical detection of T and B cell populations in routine blood smears. Lancet. 1984:1042. doi: 10.1016/s0140-6736(84)91451-x. [DOI] [PubMed] [Google Scholar]
- 18.Banham AH, Turley H, Pulford K, Gatter K, Mason DY. The plasma cell associated antigen detectable by antibody VS38 is the p63 rough endoplasmic reticulum protein. J Clin Pathol. 1997;50:485. doi: 10.1136/jcp.50.6.485. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Mason DY, Cordell JL, Gaulard P, Tse AG, Brown MH. Immunohistological detection of human cytotoxic/suppressor T cells using antibodies to a CD8 peptide sequence. J Clin Pathol. 1992;45:1084. doi: 10.1136/jcp.45.12.1084. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Sambrook J, Fritsch EF, Maniatis T. Molecular Cloning. 2. New York: Cold Spring Harbour Laboratory Press; 1989. [Google Scholar]
- 21.Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 1970;227:680. doi: 10.1038/227680a0. [DOI] [PubMed] [Google Scholar]
- 22.Seed B, Aruffo A. Molecular cloning of the CD2 antigen, the T-cell erythrocyte receptor, by a rapid immunoselection procedure. Proc Natl Acad Sci USA. 1987;84:3365. doi: 10.1073/pnas.84.10.3365. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Cordell JL, Falini B, Erber WN, et al. Immunoenzymatic labelling of monoclonal antibodies using immune complexes of alkaline phosphatase and monoclonal anti-alkaline phosphatase (APAAP complexes) J Histochem Cytochem. 1984;32:219. doi: 10.1177/32.2.6198355. [DOI] [PubMed] [Google Scholar]
- 24.Slaper-Cortenbach ICM, Admiraal LG, Kerr JM, van Leeuwen EF, von dem Borne A, Tetteroo PAT. Flow cytometric detection of terminal deoxynucleotidyl transferase and other intracellular antigens in combination with membrane antigens in acute lymphatic leukaemia. Blood. 1988;72:1639. [PubMed] [Google Scholar]
- 25.Pulford K, Micklem KJ, Jones M, et al. A novel internal antigen which distinguishes germinal centre cells from other B cell types, Immunology. 1994;82:154. [PMC free article] [PubMed] [Google Scholar]
- 26.Carballo E, Colomer D, Vives-Corrons JL, Blackshear PJ, Gil J. Characterisation and purification of a protein kinase C substrate in human B cells. Identification as lymphocyte-specific protein 1 (LSP1) J Immunol. 1996;156:1709. [PubMed] [Google Scholar]
- 27.Matsumoto N, Kojima S, Osawa T, Toyoshima S. Protein kinase C phosphorylates p50 LSP1 and induces translocation of p50 LSP1 in T lymphocytes. J Biochem. 1995;117:222. doi: 10.1093/oxfordjournals.jbchem.a124714. [DOI] [PubMed] [Google Scholar]
- 28.Li Y, Zhang Q, Howard TH. Chemokines induce leufactin phophorylation in human neutrophils. Blood. 1997;90(Suppl. 1):52. Abstract. [Google Scholar]
- 29.Fujii Y, Okumura M, Inada K, Nakahara K, Matsuda H. CD45 isoform expression during T cell development in the thymus. Eur J Immunol. 1992;22:1843. doi: 10.1002/eji.1830220725. [DOI] [PubMed] [Google Scholar]
- 30.Matsumoto N, Kita K, Kojima S, et al. Lymphocyte isoforms of mouse p50 LSP1, which are phosphorylated in mitogen-activated T cells, are formed through alternative splicing and phosphorylation. J Biochem. 1995;118:237. doi: 10.1093/oxfordjournals.jbchem.a124885. [DOI] [PubMed] [Google Scholar]
- 31.Li Y, Zhang O, Howard TH. Leufactin is a substrate of protein kinase C and phosphatases 1/2A in human neutrophils. Blood. 1996;88:609. [Google Scholar]
- 32.Huang CK, Zhan L, Ai Y, Jongstra J. LSP1 is the major substrate for mitogen-activated protein kinase-activated protein kinase 2 in human neutrophils. J Biol Chem. 1997;272:17. doi: 10.1074/jbc.272.1.17. [DOI] [PubMed] [Google Scholar]
- 33.Vanstapel M-J, Gatter KC, de Wolfe-Peeters CD, Mason DY, Desmet VD. New sites of human S-100 immunoreactivity detected with monoclonal antibodies. Am J Clin Pathol. 1986;85:160. doi: 10.1093/ajcp/85.2.160. [DOI] [PubMed] [Google Scholar]
- 34.Steinman RM. The dendritic cell system and its role in immunogenicity. Annu Rev Immunol. 1992;9:271. doi: 10.1146/annurev.iy.09.040191.001415. [DOI] [PubMed] [Google Scholar]
- 35.Mosialos G, Birkenbach M, Ayehunie S, et al. Circulating human dendritic cells differentially express high levels of a 55-kD actin-binding protein. Am J Pathol. 1995;148:593. [PMC free article] [PubMed] [Google Scholar]
- 36.Miyoshi EK, Thompson AA, Kincade PW, Wall R. Aberrant expression of leucocyte-specific pp. 52 in hairy cell leukaemia. J Allergy Clin Immunol. 1997;99:865. [Google Scholar]
- 37.Pittman S, Catovsky D. Prognostic significance of chromosomal abnormalities in chronic lymphocytic leukaemia. Br J Haematol. 1984;58:649. doi: 10.1111/j.1365-2141.1984.tb06112.x. [DOI] [PubMed] [Google Scholar]
- 38.Mason DY, Banks PM, Chan J, et al. Nodular lymphocyte predominance Hodgkin’s disease. A distinct clinocopathological entity. Am J Surg Pathol. 1994;18:526. doi: 10.1097/00000478-199405000-00014. [DOI] [PubMed] [Google Scholar]
- 39.Kuzu I, Delsol G, Jones M, Gatter KC, Mason DY. Expression of the Ig-associated heterodimer (mb-1 and B29) in Hodgkin’s disease. Histopathology. 1993;22:141. doi: 10.1111/j.1365-2559.1993.tb00092.x. [DOI] [PubMed] [Google Scholar]
- 40.Morris SW, Kirstein MN, Valentine MB, et al. Fusion of a kinase gene, ALK, to a nucleolar protein gene, NPM, in non-Hodgkin’s lymphoma. Science. 1994;263:1281. doi: 10.1126/science.8122112. [DOI] [PubMed] [Google Scholar]
- 41.Shiota M, Fujimoto J, Takenaga M, et al. Diagnosis of t(2;5)(p23;q35)-Associated ki-1 lymphoma with immunohistochemistry. Blood. 1994;84:3684. [PubMed] [Google Scholar]
- 42.Mason DY, Bastard C, Rimokh R, et al. CD30-positive large cell lymphomas (‘Ki-1 lymphoma’) are associated with a chromosomal translocation involving 5q35. Brit J Haematol. 1990;74:161. doi: 10.1111/j.1365-2141.1990.tb02560.x. [DOI] [PubMed] [Google Scholar]








