CD97 isoform expression in leukocytes
`Wolfram Eichler
`Faculty of Biosciences, Pharmaceutics and Psychology, University of Leipzig, Talstrasse 33, D-04103 Leipzig,
`Germany
`
`Abstract: Different adhesive capacity in interac-
`tions with CD55 has been ascribed to the isoforms
`of
`the leukocyte CD97 antigen, CD97 (EGF
`1,2,5), CD97 (EGF 1,2,3,5), and CD97 (EGF
`1,2,3,4,5). In the study, coexpression of the three
`CD97 isoforms and predominance of CD97 (EGF
`1,2,5) transcripts in leukocytes are demonstrated.
`The contribution of CD97 (EGF 1,2,3,5) and
`CD97 (EGF 1,2,3,4,5) to total CD97 levels varied
`among most cell types only slightly, although rela-
`tively higher mRNA levels of both isoforms were
`detected in U 937 cells and monocytes. In periph-
`eral blood lymphocytes, CD97 isoforms did not
`show clear variation after PMA stimulation and
`were down-regulated equally after CD97 cross-
`linking. Moreover, the CD97 isoform pattern was
`not altered in monocytes after interferon-g stimu-
`lation and in synovial T cells from patients with
`rheumatoid arthritis. CD97 mRNA levels did not
`necessarily correspond to CD97 surface density.
`The findings suggest that adhesive activity of CD97
`toward CD55 is unlikely to be regulated by differ-
`ential CD97 isoform expression. J. Leukoc. Biol.
`68: 561–567; 2000.
`
`Key Words: leukocytes z CD97 isoforms z EGF-like domains z
`CD55
`
`INTRODUCTION
`
`The biological role of the leukocyte surface antigen CD97 is
`still unknown, although the structural properties of this mole-
`cule strongly suggest a functional significance. Thus, expres-
`sion cloning and sequencing of human [1] and murine [2] CD97
`cDNAs indicated that the CD97 antigen is represented by a
`serpentine transmembrane protein with an extended extracel-
`lular region [1]. The extracellular part of CD97 contains a
`variable number of epidermal growth factor (EGF)-like do-
`mains, which give rise to the existence of different CD97
`isoforms [3]. Similar structural architecture was also described
`for EMR1 [4] and the murine macrophage cell-surface marker
`F4/80 [5]. Based on sequence similarities in their membrane-
`spanning regions, all of these molecules form a novel class of
`seven-span (7-TM) proteins (EGF-TM7 family) within the se-
`cretin/vasoactive intestinal peptide hormone receptor (SecR)
`family [6, 7]. Although a conclusive function of these mole-
`cules remains to be elucidated, their chimeric structure indi-
`cates adhesive and/or signal transducing capability. In line
`
`with possible adhesive properties, CD55 (decay-accelerating
`factor, DAF), a membrane regulatory protein of complement
`activation, was identified as a cellular ligand for CD97,
`strongly suggesting that CD97 and CD55 molecules participate
`in adhesive cellular contacts [8].
`The human CD97 molecule is expressed in cells and cell
`lines of different origin with a preference of surface expression
`to leukocytes [9]. However, resting lymphocytes exhibit low
`levels of cell-surface CD97, but activation of these cells leads
`to strong CD97 up-regulation [10]. A variable number of EGF-
`like domains in CD97 result from alternative splicing of the
`CD97 precursor transcript. Thus far, three different CD97
`isoforms have been shown to possess three (EGF 1,2,5), four
`(EGF 1,2,3,5), or five (EGF 1,2,3,4,5) EGF-like domains [3].
`Although the expression of CD97 isoforms was identified in
`activated human T cells [3], their presence in other cell types
`has not been investigated. The observation that larger CD97
`isoforms, CD97 (EGF 1,2,3,5) and CD97 (EGF 1,2,3,4,5), bind
`with a significantly lower affinity to CD55 (DAF) raised the
`question of functional importance of different isoform expres-
`sion [11]. Thus, potentially, leukocytes could regulate the
`strength of interaction with CD551ve cells via CD97 isoform
`expression. Because the expression of these isoforms in differ-
`ent cell types has not been compared with each other, the study
`described in this investigation was undertaken. It encompasses
`investigations of isoform pattern, mRNA levels, and cell-sur-
`face density of CD97 in leukocytes of different lineages.
`
`MATERIALS AND METHODS
`
`Preparation, culture, and in vitro stimulation of
`cells
`
`The human cell lines used in this study (K 562, Daudi, CEM, Jurkat, U 937,
`HL-60) were obtained from American Type Culture Collection (ATCC; Rock-
`ville, MD) and routinely cultured at 2 3 106 cells/ml in complete RPMI 1640
`medium supplemented with 10% fetal calf serum (FCS). Peripheral blood
`mononuclear cells (PBMC) were separated from blood of healthy donors by
`standard density gradient centrifugation (density51.077). Fractions of lym-
`phocytes and monocytes were obtained by counter-flow elutriation. Purity of
`these cells was 90 –95%. Peripheral blood lymphoctyes (PBL) used to study
`activation-dependent CD97 antigen expression were stimulated with 5 ng/ml
`phorbol 12-myristate 13-acetate (PMA; Calbiochem, Bad Soden, FRG). Stim-
`ulation of PBL with cross-linked monoclonal antibody (mAb) CD97 mAb
`
`Correspondence and present address: Dr. W. Eichler, University of Leipzig,
`Interdisciplinary Centre for Clinical Research, Department of Ophthalmology,
`Liebigstrasse 10-14, D-04103 Leipzig, Germany
`Received July 8, 1999; revised January 24, 2000; revised March 28, 2000;
`accepted April 10, 2000.
`
`Journal of Leukocyte Biology Volume 68, October 2000 561
`
`Sanquin EX2001
`Forty Seven v. Stichting Sanquin Bloedvoorziening
`IPR2016-01529
`
`

`
`BL-Ac/F2 was performed in flat-bottom plates (Greiner, Nuertingen, FRG),
`precoated with 50 mg/ml sheep anti-mouse immunoglobulin (Ig; Roche Mo-
`lecular Biochem, Mannheim, FRG). PBL were added to the washed plates and
`incubated at 37°C for 8 h. Monocytes were stimulated with recombinant 250
`U/ml interferon (IFN)-g or 50 ng/ml tumor necrosis factor (TNF)-a (R&D
`Systems, Minneapolis, MN) for 2 days. Synovial fluid was obtained from two
`patients (one male, age 63 years; one female, age 74 years) with rheumatoid
`arthritis, conforming to the American College of Rheumatology criteria. The
`samples were diluted 10-fold in phosphate-buffered saline (PBS) and centri-
`fuged at 400 g for 15 min. Cells were further separated by density gradient
`centrifugation. T cells of peripheral blood samples and synovial fluid were
`purified by immunomagnetic separation on ice using incubation for 20 min
`with CD3-coated Dynabeads (Dynal, Hamburg, FRG).
`
`RNA preparation and polymerase chain reaction
`(PCR) amplification of CD97 mRNA
`
`Total RNA of cells was prepared using a commercially purchased RNA
`isolation kit (InViTek, Berlin, FRG), according to the manufacturer’s specifi-
`cations. The resulting RNA was precipitated using isopropanol, dried, and
`dissolved in H2O. Contaminating genomic DNA was eliminated with 1 u
`DNase I (Life Technologies, Eggenstein, FRG). PCR amplification was per-
`formed using as template single-stranded cDNA, obtained by reverse tran-
`scription of total RNA preparation. The cDNA was synthesized from 1 mg in a
`20-ml reaction using 200 u of Superscript II reverse transcriptase (dT)15 (Life
`Technologies), 500 mM each of nucleotides, and 0.5 mg of oligo(dT)15 (In-
`ViTek). PCR was performed within the exponential amplification range using
`a 20-ml vol with 0.5 u of InViTAQ DNA polymerase (InViTek), 1 ml of
`single-stranded cDNA, 100 mM deoxynucleotide triphosphates (dNTPs), 125
`nM each of the CD97-specific primers (indicated in Table 1) in 50 mM
`Tris-HCl, pH 8.8, 16 mM (NH4)2SO4, 2.5 mM MgCl2, and 0.01% Triton X-100.
`PCR products were separated by electrophoresis on a 1.8% agarose gel.
`
`Quantitation of PCR products
`
`Relative CD97 mRNA levels were analyzed by reverse transcription followed
`by PCR (RT-PCR). The primers used throughout this study are indicated in
`Table 1. To analyze total CD97 mRNA, cDNA samples were adjusted to equal
`G3PDH inputs by PCR in the presence of a competitor (kindly provided by Dr.
`P. Ruschpler, Institute of Pathology, University of Leipzig, Germany). This
`competing DNA was 283 bp longer than the amplified fragments (566 bp)
`derived from G3PDH cDNA samples. An internal standard for a competitive
`CD97 PCR was constructed as described previously [12]. Briefly, a CD97 DNA
`fragment ((cid:130)CD97), which was 84 bp smaller than the amplicon derived from
`CD97 cDNA (exons 7–10, 331 bp), was generated by PCR and cloned into the
`pCR-Script plasmid (Stratagene, Heidelberg, FRG). Known amounts of (cid:130)CD97
`were coamplified with unknown amounts of sample CD97 cDNA, competing by
`the same set of primers. Relative sample CD97 cDNA levels were expressed in
`arbitrary units (AU) of each (cid:130)CD97 amount necessary for adjustment of the
`ratio (cid:130)CD97/CD97cDNA to 1. The lowest amount of (cid:130)CD97, which yielded a
`visible amplification product in agarose gels using ethidiumbromide staining,
`was defined as 1 AU. PCR of
`target-derived and competitive fragments
`occurred with virtually identical efficiency and was performed within the
`exponential amplification range. Ethidiumbromide-stained agarose gels were
`scanned using a CMD camera of a GelPrint 2000i Station from BioPhotonics
`Corp. (Ann Arbor, MI) and analyzed with the Sigmagel Software (Jandel Corp.,
`San Rafael, CA).
`
`Immunofluorescence analysis of CD97 antigen
`expression
`
`The murine mAb BL-Ac/F2 (IgG1), which is directed to CD97 EGF-like
`domain 1 and was characterized further at the VIth International Workshop on
`Leukocyte Differentiation Antigens [13], was used in this study. Cell-surface
`expression of CD97 antigen was analyzed by staining aliquotes of 2 3 105 cells
`with BL-Ac/F2 followed by incubation with FITC-conjugated goat anti-mouse
`IgG (Sigma, Deisenhofen, FRG). Samples were analyzed on a flow cytometer
`(FACScan, Becton Dickinson, San Jose, CA) using LYSYS version 1.1 soft-
`ware. Gates were set to discriminate between different cell populations and to
`exclude nonviable cells, and histograms were recorded to determine percent-
`age and mean fluorescence intensity (MFI) of labeled cells defined by scatter
`gates.
`
`Immunoprecipitation
`
`Surface labeling of cells was performed by a modification of a previously
`described method [14] using D-biotin-N-hydroxysuccinimide ester (Boehringer
`Mannheim, FRG). Cell lysates were prepared on ice by detergent lysis (53107
`cells/ml) in a buffer containing 50 mM Tris-HCl, 0.15 M NaCl, pH 8, 2%
`Nonidet P-40 (NP-40), 1 mM phenylmethylsulfonyl fluoride (PMSF), 1 mM
`ethylene diaminetetraacetate (EDTA), and precleared twice using goat anti-
`mouse IgG-coated protein A Sepharose and mouse IgG-coupled Sepharose
`beads. For immunospecific isolation [15] of the CD97 antigen, cell extracts
`were incubated overnight at 4°C with mAb BL-Ac/F2. The absorbed antigens
`were eluted and subjected to analysis by sodium dodecyl sulfate-polyacrylam-
`ide gel electrophoresis (SDS-PAGE) [16]. Following electrophoresis, proteins
`were transferred to nitrocellulose, which was blocked afterward for 2 h at37°C
`in 5% powdered nonfat dried milk and probed with streptavidin/alkaline
`phosphatase (Boehringer Mannheim), followed by 0.5 mg/ml nitrobluetetrazo-
`lium and 0.25 mg/ml 5-bromo-4-chlor-3-indolyl phosphate (Sigma). Enzymatic
`digestion of the isolated CD97 antigen was performed by incubation with 0.5
`U endoglycosidase-F (Boehringer Mannheim) in 20 mM phosphate buffer, 50
`mM EDTA, pH 6.1, 1% NP-40, 1 mM PMSF, 1% 2-mercaptoethanol.
`
`RESULTS
`
`CD97 isoform pattern in leukocytes
`
`To analyze CD97 isoform expression in leukocytes, CD97
`mRNA from PBL, peripheral blood monocytes, and cell lines
`representative of erythroid, lymphoid, and myeloid cell lin-
`eages were analyzed. RT-PCR experiments were performed
`with primers that flank the exons encoding for the EGF-like
`domains (Fig. 1). The CD97 (EGF 1,2,3,4,5) form (CD97
`mRNA with exons 5 and 6) generated a 857-bp PCR fragment;
`the CD97 (EGF 1,2,3,5) form (mRNA without exon 6) yielded
`a 710-bp band, and the CD97 (EGF 1,2,5) isoform (exon 52ve,
`exon 62ve) gave rise to a 578-bp band. The results indicated
`that the isoforms CD97 (EGF 1,2,5), CD97 (EGF 1,2,3,5), and
`CD97 (EGF 1,2,3,4,5) are expressed in all cells investigated. It
`was also observed that the presence of exons 5 and/or 6
`
`Primer binding sites
`
`CD97, exons 2 and 9
`
`CD97, exons 7 and 10
`
`G3PDH, exons 6 and 8
`
`TABLE 1. Human CD97 and G3PDH Oligonucleotides
`
`Primer sequences
`forward 59 ACTCTGCCGGGAGCTGAAAC 39
`reverse 59 TGGATGGTGACCTCGGCTGA 39
`forward 59 CAGCATCAGTGTGACAGCTC 39
`reverse 59 CTATGAGGTGCCGGACAGGT 39
`forward 59 GCAGGGGGGAGCCAAAAGGG 39
`reverse 59 TGCCAGCCCCAGCGTCAAAG 39
`
`PCR product
`
`578, 710, 857 bp
`
`331 bp
`
`566 bp
`
`562 Journal of Leukocyte Biology Volume 68, October 2000
`
`http://www.jleukbio.org
`
`Sanquin EX2001
`Forty Seven v. Stichting Sanquin Bloedvoorziening
`IPR2016-01529
`
`

`
`and the other cell types. First, higher expression of CD97 (EGF
`1,2,3,5) and CD97 (EGF 1,2,3,4,5) was characteristic for U
`937 cells. Notably, U 937 cells expressed larger isoforms at a
`level that exceeds that of CD97 (EGF 1,2,5) (Fig. 2B). Second,
`quantitation of CD97 (EGF 1,2,3,4,5) revealed that U 937 cells
`and monocytes express this isoform stronger, at about 75% of
`the CD97 (EGF 1,2,3,5) level (Fig. 2C).
`
`CD97 isoform expression during cellular
`activation and inflammatory reactions
`
`To determine whether cellular activation leads to variations in
`CD97 isoform expression, PBL were activated with PMA, and
`peripheral blood monocytes were stimulated with IFN-g. The
`phorbol-ester PMA stimulates T lymphocytes by activating
`protein kinase C, which is a key event associated with the
`T-cell antigen receptor/CD3 stimulation. IFN-g was selected,
`because this cytokine is able to regulate the expression of
`cell-surface proteins of
`the monocytic lineage [17]. When
`CD97 isoform expression in PMA-stimulated PBL was detected
`by RT-PCR, three distinct amplification products indicated the
`expression of the CD97 isoforms, CD97 (EGF 1,2,5), CD97
`(EGF 1,2,3,5), and CD97 (EGF 1,2,3,4,5). Expression of single
`isoforms was apparently not influenced during two days of PMA
`treatment (Fig. 3A). Similar results were obtained by using
`phytohemagglutinin-activated PBL (unpublished results). The
`findings were confirmed by immunoprecipitation of the CD97
`protein from lysates of PMA-stimulated and cell-surface-la-
`beled PBL. Following Endo-F digestion of the isolated 80 –95
`kDa CD97 antigen, three bands of Mr 58, 64, and 71 kDa were
`visible (Fig. 3B). These bands, which are consistent with the
`
`Fig. 1. Schematic representation of the CD97 structure and position of
`primers used for amplification of CD97 isoforms. Only the 59 terminal part of
`mature CD97 mRNA is shown. This part contains the exons that encode for the
`signal peptide (SP) and EGF-like sequence repeats. The scheme neglects that
`the start of EGF-like domain 1, as defined in ref. 1, does not match exactly the
`beginning of exon 3. The EGF-like domains 3 and 4, which are lacking in the
`CD97 (EGF1,2,3,5) and CD97 (EGF1,2,5) isoforms, respectively, are indicated
`by shaded boxes. The position of primers (F5forward; R5reverse) is indicated
`by solid circles.
`
`encoding for EGF-like domains 3 and 4, respectively, is linked
`in most cells to decreasing levels of the corresponding CD97
`isoforms (Fig. 2A).
`All cells displayed the strongest expression values consis-
`tently for the CD97 (EGF 1,2,5) isoform, which was responsible
`for 40 – 65% of total CD97 mRNA dependent on the cell type
`investigated. Thus, transcripts lacking sequences from CD97
`exons 5 or 6 appear to be generated preferentially in leuko-
`cytes. Furthermore, the relative proportions of the two larger
`isoforms were relatively constant among the cells. Evaluating
`their contribution to the total CD97 fraction and mutual levels
`revealed that relative expression of CD97 (EGF 1,2,3,5) and
`CD97 (EGF 1,2,3,4,5) varied most between the monocytic cells
`
`Fig. 2. Expression of CD97 isoforms in various human hematopoietic cells and
`cell lines. (A) Agarose gel showing separation of the CD97 isoforms. (B) Pro-
`portions of mRNA encoding for CD97 (EGF1,2,3,5) and CD97 (EGF1,2,3,4,5)
`within total CD97 mRNA of the indicated cells and cell lines. (C) Relative
`expression of isoforms CD97 (EGF1,2,3,4,5) and CD97 (EGF1,2,3,5). Pattern of
`CD97 isoforms was analyzed by RT-PCR, which was performed on 1 mg total
`RNA with the primers binding to exons 2 and 9 of the human CD97 gene as
`indicated in Table 1 and Figure 1. Contrary to B and C, the results shown in A
`were obtained after RT-PCR analysis outside the exponential amplification range
`(40 cycles) to ensure clear presentation of the three isoforms in all samples.
`Agarose gels were ethidiumbromide-stained, scanned using a CMD camera, and
`analyzed using commercially available software. Concentrations of CD97 mRNA
`were expressed as proportions of integration values of bands corrected by factors
`that took into account dependence on different DNA lengths of ethidiumbromide
`fluorescence. Data are representative of three or more replicate experiments and
`are given as mean 6 standard deviation.
`
`Eichler
`
`Isoforms of CD97 563
`
`Sanquin EX2001
`Forty Seven v. Stichting Sanquin Bloedvoorziening
`IPR2016-01529
`
`

`
`IFN-g (Fig. 3C), a slight increase of the CD97 (EGF 1,2,3,4,5)
`isoform was observed. The CD97 (EGF 1,2,3,4,5)/(EGF
`1,2,3,5) ratio (0.8360.08) in these cells was slightly higher
`than in cells cultivated with medium alone (0.7560.08) or
`TNF-a (0.7460.05), but
`this effect was not significant
`(p50.07). This result suggests that the CD97 isoform pattern in
`monocytic cells is relatively constant. However, it cannot be
`excluded that the IFN-g-induced signal transduction pathway
`may influence the CD97 isoform expression under other con-
`ditions of cellular activation. The purpose of further experi-
`ments was to down-regulate CD97 mRNA levels and to exam-
`ine whether this affects the expression of CD97 isoforms.
`Therefore, cell-surface CD97 of PBL was cross-linked by the
`mAb BL-Ac/F2, a treatment that resulted in a decrease of
`CD97 mRNA. All three CD97 isoforms appeared to be affected
`equally by this down-regulation (Fig. 3D). Previously pub-
`lished results [3, 18] indicated a potential role of CD97 ex-
`pression in inflammatory processes. Therefore, CD97 isoforms
`were also analyzed in synovial T lymphocytes from patients
`with rheumatoid arthritis, known to be in an activated state
`[19]. Figure 3E demonstrates that the relative expression of
`isoforms in synovial T cells (lanes 1 and 2) did not show clear
`differences from the CD97 isoform pattern of normal peripheral
`blood T cells (lanes 3–5).
`
`Relationship between CD97 mRNA and surface
`expression in leukocytes
`
`To obtain more insight in regulatory mechanisms that control a
`putative CD97-mediated adhesive activity, it was important to
`know whether mRNA expression and cell-surface density of
`CD97 are correlated. Therefore, a comparison of relative CD97
`mRNA levels with CD97 surface expression was performed
`using PBL, peripheral blood monocytes, and the cell lines, U
`937, HL-60, K 562, CEM, Jurkat, Daudi, and Raji (Table 2).
`The results of semiquantitative RT-PCR analysis are shown in
`Figure 4. High CD97 mRNA levels were detected in myeloid
`(HL-60) or monocytic cells, which were paralleled by strong
`CD97 surface expression. In contrast, in spite of high CD97
`mRNA levels, PBL expressed CD97 only at low density on the
`cell surface. In general, CD97 surface expression on PBL was
`comparable to that of the lymphoid cell lines, CEM, Daudi, and
`Raji. Only a fraction of these cells displayed surface CD97 at
`a low level, which was consistent with their low CD97 mRNA
`levels. Surface CD97 staining of Jurkat cells was weak and of
`even lower magnitude in comparison with the other lymphoid
`cells. Thus, higher CD97 mRNA expression in PBL and Jurkat
`T cells does not indicate equally higher CD97 surface density.
`In these cells, CD97 surface expression is likely to be regu-
`lated also independently of the mRNA level, for example, by
`mechanisms that control the redistribution and/or targeting of
`the CD97 protein. The results suggest that the surface expres-
`sion and mRNA levels of CD97 are not necessarily correlated.
`Moreover,
`transcriptional regulation of
`isoform proportions
`and/or regulation of mRNA stability of CD97 might be mech-
`anisms that do not sufficiently determine a functional CD97
`expression.
`
`Fig. 3. Expression pattern of CD97 isoforms during stimulation of PBMC,
`after down-regulation of CD97 mRNA in PBL, and in T lymphocytes derived
`from synovial fluid of patients with rheumatoid arthritis. CD97 mRNA was
`analyzed by RT-PCR using the CD97-specific primers as in Figure 1. The
`position of molecular standards is shown at the margin of each panel. (A)
`Detection of CD97 isoforms in PBL stimulated with PMA (5 ng/ml) for the
`indicated periods of time. (B) Immunoprecipitation of the CD97 antigen from
`PMA-stimulated PBL. At the indicated time points, immunoprecipitates were
`prepared from NP-40 lysates of surface-labeled cells using the mAb BL-Ac/F2.
`The isolated CD97 antigen was incubated without (2) or with (1) endoglyco-
`sidase F and separated by SDS-PAGE. (C) Detection of CD97 isoforms in
`monocytes, which were cultured for two days in medium alone, with 50 ng/ml
`TNF-a or 250 U/ml IFN-g. [See text for CD97 (EGF 1,2,3,4,5)/(EGF 1,2,3,5)
`ratios (n53).]
`(D) Down-regulation of CD97 mRNA in PBL. Cells were
`cultivated for 8 h with cross-linked irrelevant IgG1 mAb (control) or mAb
`BL-Ac/F2. The cDNA in these samples was adjusted to equal G3PDH inputs
`and amplified by PCR as indicated in Materials and Methods. (E) CD97
`isoforms in synovial T cells obtained from two patients with rheumatoid
`arthritis in comparison with T cells from normal peripheral blood samples. The
`CD97 (EGF 1,2,3,4,5)/(EGF 1,2,3,5) ratios in synovial T cells were 0.45 (lane
`1) and 0.52 (lane 2), and in normal peripheral blood T cells, 0.32 (lane 3), 0.51
`(lane 4), and 0.56 (lane 5).
`
`predicted molecular weights of the processed CD97 isoforms
`and with previously characterized CD97 isoforms from trans-
`fectants [1, 3], suggest that the CD97 (EGF 1,2,5), CD97 (EGF
`1,2,3,5), and CD97 (EGF 1,2,3,4,5) proteins are expressed on
`the lymphocyte cell surface. After activation of monocytes with
`
`564 Journal of Leukocyte Biology Volume 68, October 2000
`
`http://www.jleukbio.org
`
`Sanquin EX2001
`Forty Seven v. Stichting Sanquin Bloedvoorziening
`IPR2016-01529
`
`

`
`TABLE 2. Comparison of CD97 mRNA with Corresponding Cell-Surface Expression in Different Cells and Cell Linesa
`
`Cells
`
`PBL
`Monocytes
`U 937 (monocytic)
`HL-60 (promyelocytic)
`K 562 (erythroleukemia)
`CEM (T-cell lymphoma)
`Jurkat (T-cell lymphoma)
`Daudi (B-cell lymphoma)
`Raji (B-cell lymphoma)
`
`CD97 mRNA
`CD97 cell-surface expression
`expression
`5.70 6 0.9
`31.6 6 17.0%
`33.3 6 3.45 AU
`16.60 6 0.7
`114.4 6 22.41 AU
`100%
`13.55 6 3.43
`69.12 6 26.9 AU
`100%
`14.65 6 2.12
`35.84 AUb
`100%
`3.53 6 0.23
`40.3 6 5.9%
`2.2 6 0.48 AU
`7.45 6 3.83
`49.7 6 24.7%
`6.77 6 3.85 AU
`5.49 6 2.63
`11.1 6 3.1%
`11.62 6 3.72 AU
`5.52 6 1.10
`76.2 6 17.3%
`1.26 AUb
`5.17 6 2.44
`71.3 6 7.0%
`1.54 6 0.65 AU
`a CD97 mRNA levels are expressed in relative arbitrary units (AU, mean 6 standard deviation, n 5 3) as indicated in Materials and Methods. CD97
`cell-surface expression is indicated as percent cells expressing the CD97 antigen, in conjunction with corresponding ratios of the linear mean fluorescence
`intensities (MFI) of CD971ve cells divided by the MFI of CD972ve cells. Data from three or more experiments obtained by flow cytometry are given. Monocytes
`and lymphocytes from two different donors were used for determination of CD97 mRNA expression, which was performed in triplicate. b The mean value of two
`experiments with similar results is represented.
`
`DISCUSSION
`
`In this study, expression characteristics of the three CD97
`isoforms, CD97 (EGF 1,2,5), CD97 (EGF 1,2,3,5), and CD97
`(EGF 1,2,3,4,5), were investigated. Differential splicing of
`transcripts leading to the generation of isoforms with a variable
`number of EGF-like domains is a common attribute of the
`members of the EGF-TM7 family, CD97, EMR-1, and F4/80
`[3, 5, 20]. EGF-like domains, which are similarly arranged as
`
`Fig. 4. Representative semi-
`quantitative RT-PCR analysis
`of CD97 mRNA expression in
`various cells and cell
`lines.
`Serial dilutions of an internal
`CD97 competitor DNA (indi-
`cated above the panels in fg)
`were
`added
`to
`unknown
`amounts of sample cDNA. The
`331-bp (sample cDNA) and
`247-bp
`(CD97
`competitor)
`PCR products were analyzed
`by agarose gel electrophoresis
`and by measuring the intensity
`of ethidiumbromide fluores-
`cence as indicated in Materi-
`als and Methods.
`
`in EGF-TM7 members, can also be found in thrombomodulin/
`CD141, the low-density lipoprotein receptor, and the extracel-
`lular matrix proteins, fibrillin, fibulin, and nidogen/entactin.
`These molecules are characterized by tandemly arranged EGF-
`like sequences, the interdomain linkage of which is stabilized
`by hydrophobic interactions between conserved amino acid
`residues [21]. Additional presence of EGF-like domains may
`influence the overall rod-like conformation of one given Ca21-
`binding EGF domain pair or modulate affinity for Ca21 of one
`Ca21-binding domain [22]. Therefore, variable expression of
`EGF-like domains in EGF-TM7 proteins might have conse-
`quences for interactions with putative ligands. Indeed, the
`additional expression of EGF-like domains 3 and/or 4 in CD97
`was linked to reduced binding capacity of CD97-transfected
`COS-7 cells to CD55 (DAF), which has been identified as a
`cellular ligand for CD97 [8]. Thus the questions raised are (1)
`whether certain leukocytes express some isoform(s) preferen-
`tially, which could give rise to particular CD97-mediated ad-
`hesive properties, and (2) whether lineage-specific or activa-
`tion-dependent differences exist in terms of isoform expression
`patterns.
`The single CD97 isoforms are apparently not expressed in a
`lineage-specific manner; the triplet of bands after PCR ampli-
`fication indicates rather a coexpression of the CD97 isoforms.
`The CD97 (EGF 1,2,5) isoform, which is characterized by a
`strong adhesive capacity toward CD55 [11], was expressed
`predominantly in different cells and cell
`lines. No splice
`variants with less than three EGF-like domains were observed,
`confirming earlier findings that indicated that the structural
`and functional integrity of CD97 requires a minimal sequence
`of three EGF-like domains [11]. The contribution of CD97
`(EGF 1,2,3,5) and CD97 (EGF 1,2,3,4,5) to total CD97 levels
`varied among nearly all cells and cell lines and during cellular
`activation only slightly. Thus, activation of PBL by PMA was
`apparently not accompanied by preferential expression of one
`or both larger isoform(s), at least during the time interval (2
`days) investigated. The study of synovial T cells derived from
`patients with rheumatoid arthritis, which may express early
`(CD69), late [major histocompatibility complex (MHC) Class II,
`VLA-4], and very late (VLA-1) activation markers [19], sug-
`gests that inflammatory processes are not linked with changes
`
`Eichler
`
`Isoforms of CD97 565
`
`Sanquin EX2001
`Forty Seven v. Stichting Sanquin Bloedvoorziening
`IPR2016-01529
`
`

`
`of the CD97 isoform pattern. Additionally, down-regulation of
`CD97 mRNA in PBL was initiated by ligation of the CD97
`protein and found to be related to all isoforms. Thus, all
`available data indicate that proportions of CD97 isoform tran-
`scripts are relatively constant in lymphocytes. This is not in
`accordance with a possible interrelationship between the reg-
`ulation of the expression of single isoforms and CD97-mediated
`cellular processes.
`In comparison with monocytes cultured in medium alone
`or in the presence of TNF-a, exposition of the cells to IFN-g
`resulted in only a slightly altered CD97 (EGF 1,2,3,4,5)/
`(EGF 1,2,3,5) ratio. Although this effect was not significant,
`it is possible that cytokines influence the balance of CD97
`isoforms in monocytes under conditions that are still to be
`defined. The function of CD97 in monocytic cells and par-
`ticipation of CD97 in adhesive interactions have not been
`elucidated. However, the lack of a marked shift of the CD97
`(EGF 1,2,3,4,5)/(EGF 1,2,3,5) ratio after cytokine exposi-
`tion is rather consistent with more or less stable CD97
`isoform proportions.
`Regulatory mechanisms leading to differential CD97 iso-
`form expression are expected to act at the mRNA level.
`Stronger mRNA expression of distinct isoforms could con-
`tribute to an increased CD97 surface density, and thus, it is
`relevant
`to possible adhesive interactions. Additionally,
`generation of high CD97 (EGF 1,2,5) mRNA levels could
`indicate adequate expression of this isoform on the cell
`surface and preference for CD97-mediated adhesive con-
`tacts. The comparison of the cell-surface expression levels
`of different cell types suggested that CD97 is functionally
`significant in monocytic and myeloid cells. However, U 937
`cells express CD97 strongly in association with a higher
`level of the CD97 (EGF 1,2,3,5) and CD97 (EGF 1,2,3,4,5)
`isoforms. Compared with other cell types, U 937 cells and
`monocytes exhibit an enhanced expression of CD97 (EGF
`1,2,3,4,5) in proportion to CD97 (EGF 1,2,3,5) also. Be-
`cause the larger isoform(s) bind with a significantly lower
`activity to CD55 (DAF) [11], their overrepresentation in
`monocytic cells might not favor adhesive contacts. Further
`results indicated that even the relatively high levels of
`CD97 mRNA in PBL and Jurkat cells did not determine
`stronger CD97 cell-surface expression. With regard to ad-
`hesive contacts of lymphoid cells, even the predominant
`CD97 (EGF 1,2,5) isoform may be insufficiently expressed
`on the cell surface; e.g., the preferential expression of CD97
`(EGF 1,2,5)
`is insignificant
`to CD97 (EGF 1,2,5)-based
`interactions with CD55 (DAF). A possible explanation for
`regulation of CD97 cell-surface expression in PBL provided
`previous experiments that demonstrated that
`these cells
`possess intracellular CD97 protein. Its redistribution onto
`the cell surface contributes to rapid CD97 up-regulation
`following cellular activation [9]. In the context stated above,
`it must be noted that U 937, HL-60, and lymphoid cells
`failed to adhere to CD551ve erythrocytes (unpublished re-
`sults); this adhesion is a characteristic property of CD97
`(EGF 1,2,5)1ve transfectants [8]. Thus, regardless of the
`prominent CD97 (EGF 1,2,5) isoform expression in all leu-
`kocytes and the strong CD97 cell-surface expression of
`monocytic cells, there is no evidence that different propor-
`
`tions of single CD97 isoforms at the RNA level indicate a
`different adhesive capacity of CD97 on the cell surface.
`Rather than differential regulation of mRNA levels of CD97
`isoforms, additional mechanisms, which determine CD97
`protein expression and distribution and/or accessibility of
`single isoforms to the ligand, are likely to be important. The
`apparent dependence on such processes in conjunction with
`the relative invariability of the CD97 isoform ratio in leu-
`kocytes suggests that the adhesive activity of CD97 is not
`predetermined by the proportion of isoforms.
`
`ACKNOWLEDGMENTS
`
`The technical assistance of S. Petter is greatly appreciated. The
`author is grateful to K. Droessler for his support, to M. Pfister
`for helpful discussion, and to G. Baumbach and S. Hauschildt
`for supplying elutriated cells.
`
`Note added in proof:
`
`In a recent publication [Lin et al. (2000) Genomics 67,
`188 –200] the existence of a CD97-like molecule, EMR2, is
`described. EMR2 shares with CD97 the almost
`identical
`EGF-like domains and is also recognized by the CD97 mAb
`BL-Ac/F2.
`
`REFERENCES
`
`1. Hamann, J., Eichler, W., Hamann, D., Kerstens, H., Poddighe, P. J.,
`Hoovers, J., Hartmann, E., Strauss, M., van Lier, R. A. W.
`(1995)
`Expression cloning and chromosomal mapping of the leukocyte activation
`antigen CD97, a new seven-span transmembrane molecule of the secretin-
`receptor superfamily with an unusual extracellular domain. J. Immunol.
`155, 1942–1950.
`2. Qian, Y-M., Haino, M., Kelly, K., Song, W-C. (1999) Structural charac-
`terization of mouse CD97 and study of its specific interaction with the
`murine decay-accelerating factor (DAF, CD55). Immunology 98, 303–311.
`3. Gray, J. X., Makot, H., Roth, M. J., Maguire, J. E., Jensen, P. N., Yarme,
`A., Stevenson-Stetler, M. A., Siebenlist, U., Kelly, K. (1996) CD97 is a
`processed, seven-transmembrane, heterodimeric receptor associated with
`inflammation. J. Immunol. 157, 5438 –5447.
`4. Baud, V., Chissoe, S. L., Viegas-Pequignot, E., Diriong, S., N’Guyen,
`V. C., Roe, B. A., Lipinski, M. (1995) EMR1, an unusual member in the
`family of hormone receptors with seven transmembrane segments. Genom-
`ics 26, 334 –344.
`5. McKnight, A. J., Macfarlane, A. J., Dri, P., Turley, L., Willis, A. C.,
`Gordon, S. (1996) Molecular cloning of F4/80, a murine macrophage-
`restricted cell surface glycoprotein wit