The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ling, V.
Right arrow Articles by Collins, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ling, V.
Right arrow Articles by Collins, M.
The Journal of Immunology, 2001, 166: 7300-7308.
Copyright © 2001 by The American Association of Immunologists

Differential Expression of Inducible Costimulator-Ligand Splice Variants: Lymphoid Regulation of Mouse GL50-B and Human GL50 Molecules

Vincent Ling1, Paul W. Wu, Joy S. Miyashiro, Suzana Marusic, Heather F. Finnerty and Mary Collins

Department of Immunology, Genetics Institute, Wyeth Research, Cambridge, MA 02081


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The process of immunological costimulation between APC and T cells is mediated by protein ligand:receptor interactions. To date, costimulatory receptors known to be expressed by T cells include the structurally related proteins CD28 and the inducible costimulator (ICOS). The ligands to human and mouse ICOS, human GL50 (hGL50), and mouse GL50 (mGL50) were recently cloned and demonstrated to have sequence similarity to the CD28 ligands B7-1 and B7-2. Examination of mGL50 cDNA transcripts by 3'RACE revealed an alternatively spliced form, mGL50-B, that encoded a protein product with a divergent 27-aa intracellular domain. Both mGL50- and mGL50-B-transfected cells exhibited binding to human and mouse ICOS-Ig fusion protein, indicating that the alternate cytoplasmic domain of mGL50-B does not interfere with extracellular interactions with ICOS receptor. Flow cytometric and RT-PCR analysis of BALB/c and RAG1-/- mice splenocytes demonstrate that freshly isolated B cells, T cells, macrophages, and dendritic cells express both splice variant forms of ICOS ligand. Comparative analyses with the human ICOS ligand splice variants hGL50 and B7-H2 indicate that differential splicing at the junction of cytoplasmic exon 6 and exon 7 may be a common method by which GL50-ICOS immunological costimulatory processes are regulated in vivo.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
The costimulatory molecules, B7-1 and B7-2 are related members of the Ig superfamily of proteins with functional similarities, reflected by the ability of both these molecules to signal through CD28, an activating surface receptor protein, and through CTLA4, an inhibitory receptor protein (1). Recently, the inducible costimulator (ICOS),2 was described (2) that had amino acid sequence similarity with CD28 (24%) and CTLA4 (17%). Up-regulation and engagement of ICOS on stimulated T cells resulted in the secretion of a panel of cytokines unlike those mediated by CD28 engagement, most notably by the lack of IL-2 and the superinduction of IL-10 (2). We recently reported the cloning of mouse and human ICOS ligand GL50 (mGL50 and hGL50) (3). Mouse GL50 has since been published as B7 h (4), B7RP-1 (5), and mouse LICOS (6), while a variant of hGL50 has been reported as human LICOS (6) and B7-H2 (7). ICOS ligand exhibited ~40% sequence identity across species and ~20% sequence identity to B7-1 and B7-2. Northern blot analysis revealed that both mGL50 and hGL50 were expressed as multiple transcripts, suggesting transcriptional heterogeneity of this gene product. Although the expression of mGL50 was clearly detected in lymphoid tissues, mGL50 transcripts were also detected in many adult tissues and in embryonic hemopoietic tissues, suggesting that developmentally regulated as well as immunologically relevant tissues use this ligand-receptor system. Despite structural similarities to B7, mGL50 was found to bind specifically to mICOS-mIg and not to either CTLA4-Ig or CD28-Ig fusion proteins. FACS analysis demonstrated mICOS-mIg fusion protein binding primarily to freshly isolated naive mouse splenic B cells and at lower levels to CD3+ T cells (3), consistent with the predicted T cell/B cell interactions initially described for ICOS localization in immunohistochemical studies (2). One aspect of ICOS ligand transcriptional regulation was shown by its constitutive expression pattern on lymphoid cells while exhibiting induced protein expression in cytokine-treated embryonic fibroblasts (4). Examination of mice injected with LPS resulted in testes, kidney, and peritoneum samples induced ICOS ligand, and was deduced to be linked to the TNF-{alpha} signaling cascade.

Splice variants of B7-1 have been described in which the exons encoding Ig-V, transmembrane, and cytoplasmic domains were found to be deleted or fused to novel coding sequences (8, 9, 10, 11). More recently, a splice variant of B7-2 encoding a soluble costimulatory protein was reported (12). For costimulatory receptors, splice variants of CD28 and CTLA4 also have been identified (13, 14, 15). Whether transcriptional heterogeneity is present in all B7- and CD28-type proteins of the Ig superfamily remained to be determined. In this paper we describe the characterization of a splice variant of mGL50, further extending alternative splice variation as a method of transcriptional regulation of costimulatory ligand biosynthesis.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Sequence analysis

Geneworks 2.5.1 (Oxford Molecular Group, Campbell, CA) and Vector NTI Suite 6 (Informax, North Bethesda, MD) was used for DNA sequence data entry and manipulation. Lasergene DNAstar Genequest module (Lasergene, Madison, WI) was used for delineating intron exon boundaries of hGL50 against GenBank HS21C098.

3'RACE

3'RACE was performed (3) using nested mGL50 primers VL118 and VL116 and nested anchor primers VL054 and VL055 (Table IGo). Mouse PBLs were enriched for lymphocytes by density centrifugation using Lympholyte M (Cedarlane Laboratories, Hornby, Ontario, Canada) according to the manufacturer’s protocol. Total RNA was extracted from lymphocytes as described below. Reverse transcription was accomplished using primer VL053, 5 µg total RNA, and SuperScript RT (Life Technologies, Gaithersburg, MD) according to the manufacturer’s protocols. cDNA synthesis was performed in 20-µl reactions. 0.5–1.0 µl of reverse transcriptase-synthesized cDNA was used per RACE procedure.


View this table:
[in this window]
[in a new window]
  		Table I. DNA oligonucleotides used

 
RNA isolation and analysis

Total RNA derived from CCE embryonic stem (ES) cells, Swiss-Webster embryos/yolk sacs, and C57BL/6 peripheral blood lymphocytes was extracted using RNAstat 60 (Tel-Test B, Friendswood, TX) in conjunction with Phase-Lock gel barrier (Eppendorf, Westbury, NY). RNA was fractionated using the Northern Max system (Ambion, Austin, TX) and blotted onto Zetaprobe GT (Bio-Rad, Hercules, CA) according to the manufacturer’s protocols. Multiple tissue RNA panels were purchased and used according to the instructions provided (Clontech, Palo Alto, CA). Blots were hybridized to radiolabeled DNA fragments encompassing nt 984-1340 of the mGL50-B clone (357 bp) corresponding to the 3' untranslated region (UTR), whereas fragments corresponding to the coding sequence of mGL50 were used to detect both mGL50 and mGL50-B transcripts. Hybridizations were performed at 65°C with Express Hyb (Clontech, Palo Alto, CA) overnight and subsequently washed with 0.1x SSC and 1% SDS at hybridization temperatures until a suitable signal to noise ratio was reached. Blots were exposed against phosphorimage plates and autoradiographic film for imaging (Fuji Medical Systems, Stamford, CT).

Cell staining

Cell suspensions for cytometric analysis were isolated from BALB/c and RAG1-/- splenocytes (~3 mo old) and washed once with DMEM, 10% (v/v) heat-inactivated FCS (Sigma, St. Louis, MO), 2 mM L-glutamine, 100 U/ml penicillin, 100 µg/ml streptomycin (Irvine Scientific, Santa Ana, CA), 20 µM 2{beta}-ME (Sigma), MEM sodium pyruvate, and MEM nonessential amino acids (Life Technologies). RBC were lysed with ACT lysing buffer and washed once. Splenocytes (~1 x 107 cells/1 ml/well) from BALB/c mice were cultured with 25 µg/ml LPS (Sigma) or with 1 µg/ml Con A. Cells were stained with FITC-labeled Abs (BD PharMingen, San Jose, CA) and mICOS-mIgG2am reagent (3), followed by flow cytometric analysis using FACSCalibur and CellQuest software package (BD PharMingen). Cell separation was performed using anti-FITC microbead magnetic selection (Miltenyi Biotec, Auburn, CA), followed by flow cytometric determination of T cell enrichment.

RT-PCR and PCR-Southern blots

For enriched mouse lymphocyte and cell line RNA, first-strand cDNA synthesis was performed as described above for RACE procedures, followed by duplicate 25-µl amplification reactions (Advantage Taq; Clontech) with the primers RLEE 001 and RLEE005 specific for mGL50, primers RLEE 001 and RLEE003 specific for mGL50-B, and primers VL139 and VL121 common to both mGL50 and mGL50-B (Table IGo). Primers GAPDH-F and GAPDH-R for were used as positive amplification controls. Mouse GL50 and mGL50-B PCR was performed at 95°C for 1 min, 60°C for 1 min, and 72°C for 2 min for 33 cycles, while GAPDH PCR was performed for 30 cycles. For human ICOS ligand RT-PCR, first-strand cDNA was obtained from Human Multiple Tissue cDNA panel 1, panel 2, and Immune System (Clontech K1420-1, K1421-1, and K1426-1). Oligonucleotide primer sets specific for hGL50 (VL141-VL162B) and B7-H2 (VL141-VL274) were used to amplify human ICOS ligand transcripts under the PCR conditions: 95°C for 1 min, 60°C for 1 min, and 68°C for 1 min for 38 cycles. For PCR Southern blots, amplified samples were fractionated on 1% agarose gels and alkaline transfer-blotted onto Zetaprobe membranes (Bio-Rad) followed by hybridization in 0.4x White Rain Shampoo with Conditioner (Gillette, Boston, MA) (16) at 40°C for 1–2 h using mGL50-specific (RLEE004), mGL50-B specific (RLEE002), mGL50 extracellular domain (VL122), or hGL50/B7-H2 (VL138) radio-end-labeled oligonucleotide probes. Membranes were washed with 2x SSC/1% SDS and 0.2x SSC/1% SDS at room temperature until adequate signal to noise levels were attained.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
To determine the extent of mGL50 transcript heterogeneity, 3'RACE was performed to examine splice variants of this molecule. Using specific, nested 5' oligonucleotide primers corresponding to sequences upstream to and including the initiation startsite of mGL50, amplified PCR products were generated fromoligo(dT)-primed cDNAs derived from extracted mouse PBL RNA. Two sets of PCR products, represented by multiple clones with extensive polyadenylation of differing lengths, were found to encode an alternatively spliced form of mGL50. One representative 1759-bp RACE product, termed mGL50-B (Fig. 1GoA), encoded a polypeptide 347 aa in length with a predicted molecular mass of 39 kDa (Fig. 2Go). Alignment of the mGL50 and mGL50-B sequences revealed complete identity from initiation methionine/RACE priming site to position 1027, with the exception of two nucleotides found in multiple RACE products located at positions 531 and 710, leading to an arginine to histidine residue change at position 237 of the derived peptide sequence (Fig. 2Go). These two nucleotide discrepancies may be due to strain differences between the mice used for RNA starting material, because separate PCR products encoded the identical mismatch. Subsequent RT-PCR analysis revealed the presence of both mGL50 and mGL50-B transcripts in BALB/c, Swiss-Webster, Sv129, RAG1-/-, and C57BL/6. Although sequences were divergent downstream of position 1027 of mGL50 and position 961 of mGL50-B, both RACE and cDNA sequences contained a consensus AATAAA polyadenylation signal upstream from the poly(A) tail (13 bp for mGL50-B, 16 bp for mGL50). As a result of the alternative 3' sequences encoding the carboxyl terminus, mGL50-B lacked the final two amino acids of mGL50, but incorporated a novel additional 27 aa in the cytoplasmic domain. The predicted amino acid sequence of mGL50-B indicated the presence of three tyrosine residues, Y325, Y328, and Y333, in addition to the cytoplasmic tyrosine residues Y299 and Y307 shared by both RACE and cDNA molecules. ProfileScan analysis of predicted peptide motifs did not result in matches with known motifs with potential tyrosine signaling function. GenBank database search revealed no cDNA sequences with significant similarity to the divergent 3' domain of the mGL50-B product. Blast search against GenBank revealed the presence of a complex repetitive sequence motif (bases 1349–1554) within a portion of the 3'-UTR of mGL50-B also found in numerous genomic sequences (e.g., accession no. AC005818, AC006508, and AF115517) as well as known mRNAs (mouse desmin Z18892 and mouse servivin AF115517). No such untranslated repetitive sequences were found in the original mGL50 cDNA sequence.



View larger version (17K):
[in this window]
[in a new window]
  		FIGURE 1. A, DNA sequence diagram between the ICOS ligands mGL50 and mGL50-B. Sequence divergence, indicated by the vertical line, occurs at nt 1027 for mGL50 and nt 960 for mGL50-B. {dotsquare}, Repetitive sequence found in the 3'-UTR of mGL50-B encompassing nt 1326–1531. The locations of oligonucleotide PCR primers and probes used to differentially detected mGL50 and mGL50-B are indicated by arrows and lines, respectively. B, Genomic organization of human ICOS ligand splice variants hGL50, B7-H2, and KIAA0653 based on chromosome 21 BAC clone HS21C098.

 


View larger version (46K):
[in this window]
[in a new window]
  		FIGURE 2. Alignment of protein sequences encoded by mGL50-B, mGL50, hGL50, and B7-H2. Exon boundaries derived from available human and mouse genomic data are displayed.

 
RNA blot analysis of mouse tissue samples was performed using probes corresponding to the shared extracellular domain of mGL50/mGL50-B or to a unique nonrepetitive 3'-UTR of mGL50-B (Fig. 1Go) revealed differential hybridization patterns (Fig. 3Go). Whereas the shared extracellular domain probe hybridized to an ~2.7-kb message clearly detectable in heart, brain, spleen, lung, liver, skeletal muscle, kidney, and testis samples (Fig. 3GoB), identical with the pattern previously seen in blots specific for mGL50 3'-UTR (3), the 3'-UTR probe specific to mGL50-B hybridized to a 1.7-kb transcript detected only in heart, spleen, and kidney samples. These results indicate the dominance of the mGL50 transcript across tissues compared with the limited distribution of mGL50-B transcripts (Fig. 3GoA). In poly(A)+ RNA blots, hybridization to mGL50-B 3'-UTR was clearly detected in samples representing undifferentiated ES cells, day 10 embryoid bodies, day 12.5 embryonic yolk sac, and day 15 fetal liver (Fig. 3GoC). In contrast, hybridization against mGL50 cDNA-coding sequence probe clearly revealed transcript in all samples examined (Fig. 3GoD).



View larger version (64K):
[in this window]
[in a new window]
  		FIGURE 3. RNA blot analysis of mGL50-B. Multiple adult tissue panel consisting of heart, brain, spleen, lung, liver, skeletal muscle, kidney, and testis RNA samples plus embryonic tissue panel consisting of undifferentiated ES cells, day 10 embryoid bodies, day 12.5 yolk sac, day 15 fetal liver, and control WEHI 231 cell lines were hybridized against radiolabeled DNA probes corresponding to either the sequence-specific mGL50-B 3'-UTR (A and C) or the conserved extracellular domain shared between mGL50 and mGL50-B (B and D).

 
To determine the genomic exon organization of the known GL50 genes, we performed ICOS ligand sequence alignments to human chromosome 21 (17) and the partially completed, commercially available Celera mouse genomic database. In contrast to the two mGL50 splice variants shown here, at least 3 human ICOS ligand splice variants have been reported in the form of hGL50 (3), KIAA0653 (GenBank accession no. AB014553), and B7-H2 (7) (Fig. 1GoB). For KIAA0653, the splice junction between exons 6 and 7 was not used, resulting in a read-through of 2.9 kb into the putative intron 6. Further, the sequence in this region of the 3' end of KIAA0653 was not in agreement with the human chromosome 21 BAC clone HS21C098. Eight sequence discrepancies comprised of seven mismatches and one 17-bp deletion were revealed, calling into question the existence of the large cytoplasmic domain attributed to the coding sequence of KIAA0653 (18). In contrast, exon sequence alignment of hGL50 and B7-H2 to HS21C098 revealed no sequence dissimilarities up to and including the polyadenylation site. Human GL50, B7-H2, mGL50, and splice variant mGL50-B show some amino acid sequence identity near the splice site for cytoplasmic domain exons 6 (mGL50 residues 316–318, E-L-T; Fig. 2Go). The shared point of splice variation between hGL50/B7-H2/KIAA0653 and between mGL50/mGL50-B suggests the potential of a conserved mechanism that allows or promotes alternative splicing of exon 6 cytoplasmic domain, perhaps to offer alternate signaling through the combinatorial addition of alternate cytoplasmic functional domains.

In a comparative study of human ICOS ligand expression within different tissues, RT-PCR was performed, followed by Southern blotting using oligonucleotide primer and probe sets specific for hGL50 or B7-H2. B7-H2 transcripts were widely distributed in all tissues examined, including brain, heart, kidney, liver, lung, pancreas, placenta, skeletal muscle, bone marrow, colon, ovary, prostate, testis, lymph node, leukocyte, spleen, thymus, and tonsil samples, while hGL50 was amplified only from lymph node, leukocyte, and spleen samples (Fig. 4Go). These differential results suggest a very broad tissue distribution of B7-H2 transcripts compared with hGL50 transcripts, analogous to the degree of tissue specificity difference seen between mGL50 and mGL50-B.



View larger version (27K):
[in this window]
[in a new window]
  		FIGURE 4. RT-PCR analysis of the transcripts representing human ICOS ligands hGL50 and B7-H2. Independent cDNA panels representing multiple human tissues were amplified using primer sets specific to either hGL50 or B7-H2, followed by DNA blotting and hybridization with transcript-specific oligonucleotide probes.

 
To demonstrate that mGL50-B transcripts encode functional surface proteins, mGL50 expression vectors were used to transfect COS cells followed by flow cytometric detection of surface staining. Both mICOS-mIgG2am and hICOS-mIgG2am were found to bind mGL50- and mGL50-B-transfected cells (9–14%), while negligible binding was observed with mCTLA4-mIgG2am (<1%), indicating that the domains encoded by the additional residues in the alternate carboxyl tail found in mGL50-B neither affect surface mobilization of this protein nor alter the specific binding to mICOS (Fig. 5Go). It is also notable that hICOS-mIgG2am binds both mouse and human GL50 molecules, suggesting that the ICOS receptors, like CTLA4 and CD28 receptors, retain ligand binding capacity when assayed against targets across primate/rodent species boundaries. Other mouse cell lines were examined for the presence of surface ICOS ligand. WEHI 231 cells have been previously shown to have surface expression of both B7-1 and B7-2, whereas ES cells have been shown to display only B7-1. To rule out the possibility that B7 proteins contribute to ICOS binding (6), experiments were performed using mCTLA-4 blocking reagents (Fig. 6GoA). Mouse CTLA4-mIgG2am staining of WEHI 231 cells was clearly detectable from 8 ng/ml of reagent, whilemICOS-mIgG2am staining was detectable at levels starting at 1 µg/ml, suggesting that the mGL50 binding affinity to mICOS-mIgG2am is lower than that of B7 proteins to mCTLA4. In the presence of blocking B7-1 and B7-2 Abs, mCTLA4-mIgG2am binding to WEHI 231 was totally abrogated, while no effect of mICOS-mIgG2am binding to cells was observed. These results confirm that neither WEHI 231 B7-1 nor B7-2 potentiates specific binding with mICOS-mIgG2am in this assay system. To corroborate evidence from RNA blot analysis demonstrating the presence of mGL50 in cells representative of the very early embryonic environment, undifferentiated CCE ES cells were analyzed by direct staining with Abs to B7-1 and indirect staining with mICOS-mIgG2am fusion protein. Undifferentiated ES cells stained with anti-B7-1 (Fig. 6GoB) revealed a 1-log fluorescence shift over background, consistent with previous observations (19), and a half-log fluorescence shift over background with mICOS-mIgG2am staining, demonstrating the simultaneous surface display of both B7 and mGL50 type molecules in a system that reflects the undifferentiated inner cell mass of early preimplantation embryos.



View larger version (47K):
[in this window]
[in a new window]
  		FIGURE 5. Flow cytometric analysis of ICOS binding to mouse and human ICOS ligands. COS cells transfected with expression plasmids encoding mGL50, mGL50-B, and hGL50 were incubated with mICOS-mIgG2am, hICOS-mIgG2am, or mCTLA4-mIgG2am, followed by secondary staining with anti-mouse IgG2a biotin and detection with streptavidin-PE.

 


View larger version (38K):
[in this window]
[in a new window]
  		FIGURE 6. ICOS binding to WEHI 231 and undifferentiated ES cells. A, Titrated amounts of mICOS-mIgG2am (A and C) or mCTLA4-mIgG2am (B and D) was used to stain WEHI 231 cells in the presence of blocking anti B7-1 and B7-2 Abs or Ig isotype controls. B, Anti-B7-1 and mICOS-mIgG2am counterstaining of undifferentiated ES cells.

 
Previous phenotypic analysis of the major splenic cell types exhibiting mGL50 surface proteins revealed mICOS-mIg binding to be most readily detectable on phenotypic CD19+ B cells, although it was apparent that other splenic cell types exhibited ICOS-Ig staining. To further identify other freshly isolated cells that display mGL50, wild-type BALB/c splenocytes were compared with RAG1-/- splenocytes lacking mature B and T cells. As expected, high levels of mICOS-mIgG2am binding (Fig. 7Go, A and B) were detected on BALB/c phenotypic B cells (CD19+, B220+, and CD40+; >94%), while lower levels were found on phenotypic T cells and T cell subsets (CD3+, CD4+, and CD8+; <10%), macrophage (CD11b; 26%), dendritic cells (CD11c; 43%), and NK cells (pan-NK; 20%). Mouse ICOS-mIgG2am binding was also detected on the cells expressing the more general lymphoid markers CD24 and class II (94%). By Northern blot analysis we previously demonstrated the presence of mGL50 transcripts in RAG1-/- mice, suggesting that in the absence of mature T or B cells, mGL50 was still expressed on other splenic subpopulations (3). Consistent with these observations, RAG1-/- splenocytes (Fig. 7GoB) were CD3-, CD8-, CD19-, and CD40-, whereas the remaining CD11b+ (35%) and CD11c+ (55%) cells readily counterstained with mICOS-mIgG2am. Mouse ICOS-mIgG2am staining of CD4+ (45%) and CD24+ (28%) cells was also apparent in RAG1-/- splenocytes despite the absence of staining for other T cell markers. CD4+ staining was previously reported on dendritic cells (20), and this was supported by the presence of a CD4+, CD11c+ double-positive cell population in RAG1-/- mice (Fig. 7GoC). The presence of mGL50 transcripts along with mICOS-mIgG staining of macrophage and dendritic cells in RAG1-/- splenocytes verify the presence of ICOS ligand on professional APC.



View larger version (55K):
[in this window]
[in a new window]
  		FIGURE 7. Immunophenotyping of BALB/c and RAG1-/- splenocyte subsets. Two-dimensional FACS plots of 10,000 events are presented with numbers representing the percentage of cells in each quadrant displayed. Samples with 50,000 events are indicated by asterisks. A, Enriched splenocytes from BALB/c or RAG1-/- mice were stained with mICOS-mIgG2am and FITC-conjugated Abs against CD3, CD24, CD45R/B220, pan-NK, MHC class II, or CD40. To further phenotype the CD4+, ICOS ligand+ cells, RAG1-/- cells were stained with PE-labeled anti-CD4 and FITC-labeled anti-CD11c. B, Enriched splenocytes from RAG1-/- and BALB/c mice, after in vitro treatment (untreated, Con A activation, LPS activation), were counterstained with mICOS-mIgG2am and Abs to CD4, CD8, CD19, CD11b, CD11c, or CD69.

 
Because ICOS ligand appeared to exist in at least two splice variants, experiments were performed to semiquantitatively assess the presence of mGL50 and mGL50-B transcripts in splenocyte cell populations. BALB/c splenocytes cultured in the presence of LPS or Con A were found to up-regulate ICOS ligand in all splenocytes examined (Fig. 6GoB). To correlate ICOS ligand surface display to gene expression, RT-PCR was performed using transcript-specific oligonucleotide primers and hybridization probe sets (Fig. 1Go and Table IIGo). As target samples, BALB/c CD4+, CD8+, and CD19+ cell subsets and RAG1-/- CD11b+ and CD11c+ cell subsets were enriched to >90% purity by bead separation (data not shown), followed by RNA extraction. Duplicate RT-PCR analysis of quantity-normalized total RNA samples revealed mGL50 and mGL50-B transcripts to be present in nontreated CD4+ T cells and CD19+ B cells, consistent with results from flow cytometric analysis. However, neither mGL50 nor mGL50-B transcripts were amplified in CD8+ T cells despite surface protein detection by FACS and enrichment of ICOS ligand-positive cells (Fig. 7Go). To ensure the presence of ICOS ligand in CD8+ samples, RT-PCR analysis was performed using primers within the IgV domain common to mGL50 and mGL50-B, resulting in the generation of an amplification product confirmed by hybridization to internal DNA probe (data not shown). These results suggest that CD8+ cells may preferentially express a third form of mGL50 with alternate exon 6/exon 7 configuration. LPS activation led to similar mGL50 and mGL50-B expression profiles compared with control cells, except that low levels of mGL50 were detected in CD8+ samples, suggesting that LPS stimulation of B cells may indirectly up-regulate the expression of this form of ICOS ligand on T cells. Con A stimulation of splenocytes resulted in the amplification of mGL50 transcripts across all samples, with a decrease in product in CD19+ cells. Mouse GL50-B transcripts were induced in CD8+ samples and were not detected in CD19+ samples. The decrease in amplified product of both mGL50 and mGL50-B in CD19+ cells suggests a regulation of B cell transcription upon exposure to Con A. In RAG1-/- splenocytes, mGL50 and mGL50-B were detected in CD11b+ and CD11c+ cells, whereas cultured F5M dendritic cells and WEHI 231 B cells exhibited mGL50 transcripts. Only low levels of mGL50-B were detected in WEHI 231, and LPS activated F5M cells, while no amplified products were detected in uninduced F5M cells. In samples representing embryonic tissues, mGL50 and mGL50-B transcripts were detected in all samples, with abundant levels of both variants present in undifferentiated ES cells. High levels of mGL50 were also detected in day 12.5 embryo and 11.5 yolk sac samples. These results correlate with the degree of transcript hybridization shown by RNA blot analysis (Fig. 3Go).


View this table:
[in this window]
[in a new window]
  		Table II. RT-PCR analysis of mGL50 isoforms1

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Among the members of the extended B7 family of proteins, the genomic structures of B7-1, B7-2, butyrophilin, and the human ICOS ligand variant KIAA0653 have been reported. Although the absolute number of exons that comprise these genes vary from 5 to 12, these genes share structure, in that distinct exons encode the two Ig-like extracellular domains, one exon encodes the transmembrane domain, and multiple exons encode cytoplasmic domains (two exons for hGL50, multiple exons for B7-2 (21, 22), one to two exons for B7-1 (11), three exons for butyrophilin (23)). The exon structure of mGL50 was more similar to that of mB7-2 than to that of mB7-1; whereas mB7-1 is encoded by five exons encoding discrete domains (leader/signal peptide, IgV, IgC, transmembrane, and cytoplasmic domains), in contrast both mature mB7-2 and mGL50 transcripts are encoded by seven exons where the signal sequence domain is split between exon 1 (both mB7-2 and mGL50 = 4 aa) and exon 2 (mB7-2 = 17 aa; mGL50 = 39 aa), and the exons encoding the cytoplasmic domain is split between exon 6 (mB7-2 = 33 aa; mGL50 = 10 aa) and exon 7 (mB7-2 = 5 aa; mGL50 = 3 aa residues). Despite these differences, upon protein alignment of mGL50, mB7-1, and mB7-2 extracellular domains, the predicted relative locations of exon splice junctions located between the IgV, IgC, and transmembrane domains are conserved. It remains to be determined whether additional alternate 5'-UTR/signal sequence exons exist for mGL50 analogous to the genomic structure observed for mB7-2.

Examination of the exons encoding GL50 proteins shows variation in the use of exons encoding the cytoplasmic domains. The splice junction between exons encoding cytoplasmic domains 1 and 2 is not used in KIAA0653, resulting in a read-through of 2.9 kb into the putative intron 6. The variation in splice junction at the point after cytoplasmic domain 1 (exon 6) leads to the differences observed between hGL50/B7-H2/KIAA0653 and between mGL50/mGL50-B, suggesting the potential of a conserved mechanism that allows or promotes alternative splicing of cytoplasmic domain 2, perhaps to offer alternate signaling through the combinatorial addition of alternate functional domains. In this study we show that mGL50 and B7-H2 transcripts are widely distributed and observe that hGL50 and mGL50-B variants are transcribed with certain lymphoid tissue specificity. This suggests that variation in ICOS ligand splice regulation is dependent on physiological locale and cellular activation state, perhaps influencing the cell signaling potential during ligand:receptor interactions. Furthermore, based on the lymphoid-specific distribution of mGL50-B and hGL50, these splice variants may confer a unique signaling function in the lymphoid compartment. Based on transcript abundance, tissue distribution, and exon organization, these results suggest that mGL50 and B7-H2 are orthologous sequences, and mGL50-B and hGL50 are differentially regulated splice variants.

To clearly define the cell subsets that show surface expression of mGL50, comparative phenotyping of RAG1-/- and BALB/c splenocyte subsets was performed. In this study we first show that under these assay conditions mICOS-mIgG2am reagent specifically binds to ICOS ligand and not to B7-1 or B7-2 in nontransfected WEHI 231 cells. These results are in contrast to a previous report that suggested that multivalent ICOS-Ig fusion proteins may bind to B7 proteins (6) and validates the use of the mICOS-mIgG2am reagent in this assay to phenotype freshly isolated splenocytes. Mouse ICOS-mIgG2am staining detected on CD19+CD4+ and CD8+ cells as well as RAG1-/- CD11c+ cells indicated that T cells, B cells, and dendritic cells display surface mGL50 proteins. Con A stimulation of splenocytes resulted in increased levels of mICOS-mIgG2am staining of CD4+ cells (11-fold stimulation) and CD8+ cells (6-fold stimulation). Similarly, LPS treatment of splenocytes resulted in an increase in mICOS-mIgG2am staining by 10-fold in CD4+ cells and 7-fold in CD8+ cells. RT-PCR analysis of bead-enriched cell subpopulations confirmed the phenotypic results, showing that in uninduced CD8+ splenocytes, transcripts for mGL50 and mGL50-B were at levels below assay detection, but were readily detected upon activation with either Con A or LPS. Because of the scarcity of CD11b+ and CD11c+ cells in BALB/c splenocytes, RAG1-/- splenocytes were analyzed for the presence of ICOS ligand on these cell types, revealing that 40% of phenotypic macrophage and dendritic cells display surface ICOS ligand. RT-PCR analysis of enriched cells confirmed the expression of both mGL50 and mGL50-B transcripts in these cell types. In addition, mICOS-mIgG2am staining was detected in both resting and activated F5M dendritic cells (data not shown), correlating to the constitutive gene expression of mGL50 and the LPS induction of mGL50-B in these cells. The presence of both mGL50 and mGL50-B transcription in the same cell population suggests that both splice variants contribute to the surface display of ICOS binding. These results are distinct from those of previous studies where ICOS ligand was reported to be absent in T cell lines (20) and some dendritic cell lines (5).

Phenotypically, both B7-1 and B7-2 have been found on T cells; however, it has been speculated that B7-2 expression on T cells serves an alternative function to signaling based on the hypoglycosylated state of the protein on T cell surfaces (24). Based on the lack of T cell tumor immunity conferred by T cell B7-2, it was further postulated that T cell B7-2 may function only to inhibit T cell responses by preferential binding to CTLA4 over CD28 (25). In a separate study DO11.10 T cells deficient in B7-1 were found to produce more IL-4 compared with nondeficient DO11.10 T cells, suggesting that the presence of T cell B7-1 regulates the differentiation of cells into IL-4 producers (26). It was further demonstrated that while T cell B7-1 regulated IL-4 production, the APC B7-1 promoted IL-4 production, indicating a potential signaling difference for B7-1 molecules depending on the site of expression. In this study we demonstrate the induction of both mGL50 and mGL50-B transcripts and surface protein on CD4+ and CD8+ T cells by Con A or LPS activation, but the relative contribution of each variant to surface expression and whether functional differences exist between the signaling potential of these T cell mGL50 variants and of those found on APC remain to be determined.

In addition to APC, we previously demonstrated that the initial expression of costimulatory ligands occurs early in the ES cell model of embryonic development, with the expression of B7-1 and mGL50 transcripts in undifferentiated cells and in embryoid bodies cultured 10 days in vitro (19). In this study we further show that mGL50-B transcripts are found within these tissues by RNA analysis. By day 9 of embryoid body differentiation, emergent hemopoietic cells phenotypically resemble yolk sac hemopoietic progenitors in vivo, as evidenced by the potential of c-Kit+/PECAM-1+ cells to produce mixed hemopoietic progenitors and CD45+ cells to produce macrophage progenitors in colony assays (27, 28). These CD45+ cells were also found to be B7-1+ and B7-2+, strongly suggesting costimulatory ligand surface expression very early in lymphopoiesis. Correspondingly high levels of mGL50 and mGL50-B RNA expression were found in sites of embryonic hemopoiesis, such as embryonic day yolk sac and fetal liver. It is noteworthy that ICOS ligand is inducible in embryonic fibroblast cultures, a cell type derived from a time point before definitive lymphopoiesis, suggesting that costimulatory molecules may play a distinct role in embryogenesis independent of the adaptive immune response. It has been postulated that metazoans share common evolvable pathways that occur at the phylotypic stage of embryogenesis, and that certain core physiological processes that have special properties relevant to complex development were reflected during this period of embryonic development (29). It remains to be determined whether the extended B7/GL50 family of costimulatory ligands is part of a core process used by both embryonic and adult systems.


   Footnotes
 
1 Address correspondence and reprint requests to Dr. Vincent Ling, Department of Immunology, Genetics Institute, 87 CambridgePark Drive, Cambridge, MA 02081. E-mail address: vling{at}genetics.com Back

2 Abbreviations used in this paper: ICOS, inducible costimulator; mGL50, mouse GL50; hGL50, human GL50; ES, embryonic stem; UTR, untranslated region. Back

Received for publication January 26, 2001. Accepted for publication April 4, 2001.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Greenfield, E. A., K. A. Nguyen, V. K. Kuchroo. 1998. CD28/B7 costimulation: a review. Crit. Rev. Immunol. 18:389.[Medline]
  2. Hutloff, A., A. M. Dittrich, K. C. Beier, B. Eljaschewitsch, R. Kraft, I. Anagnostopoulos, R. A. Kroczek. 1999. ICOS is an inducible T-cell co-stimulator structurally and functionally related to CD28. Nature 397:263.[Medline]
  3. Ling, V., P. W. Wu, H. F. Finnerty, K. M. Bean, V. Spaulding, L. A. Fouser, J. P. Leonard, S. E. Hunter, R. Zollner, J. L. Thomas, et al 2000. Cutting edge: identification of GL50, a novel B7-like protein that functionally binds to ICOS receptor. J. Immunol. 164:1653.[Abstract/Free Full Text]
  4. Swallow, M. M., J. J. Wallin, W. C. Sha. 1999. B7h, a novel costimulatory homolog of B7.1 and B7.2, is induced by TNF{alpha}. Immunity 11:423.[Medline]
  5. Yoshinaga, S. K., J. S. Whoriskey, S. D. Khare, U. Sarmiento, J. Guo, T. Horan, G. Shih, M. Zhang, M. A. Coccia, T. Kohno, et al 1999. T-cell co-stimulation through B7RP-1 and ICOS. Nature 402:827.[Medline]
  6. Brodie, D., A. V. Collins, A. Iaboni, J. A. Fennelly, L. M. Sparks, X. N. Xu, P. A. van der Merwe, S. J. Davis. 2000. LICOS, a primordial costimulatory ligand?. Curr. Biol. 10:333.[Medline]
  7. Wang, S., G. Zhu, A. I. Chapoval, H. Dong, K. Tamada, J. Ni, L. Chen. 2000. Costimulation of T cells by B7–H2, a B7-like molecule that binds ICOS. Blood 96:2808.[Abstract/Free Full Text]
  8. Faas, S. J., M. A. Giannoni, A. P. Mickle, C. L. Kiesecker, D. J. Reed, D. Wu, W. L. Fodor, J. P. Mueller, L. A. Matis, R. P. Rother. 2000. Primary structure and functional characterization of a soluble, alternatively spliced form of B7-1. J. Immunol. 164:6340.[Abstract/Free Full Text]
  9. Zhang, H., D. Haasch, K. B. Idler, G. F. Okasinski. 1996. Isolation and promoter mapping of the gene encoding murine co-stimulatory factor B7-1. Gene 183:1.[Medline]
  10. Inobe, M., P. S. Linsley, J. A. Ledbetter, Y. Nagai, M. Tamakoshi, T. Uede. 1994. Identification of an alternatively spliced form of the murine homologue of B7. Biochim. Biophys. Acta 200:443.
  11. Borriello, F., G. J. Freeman, S. Edelhoff, C. M. Disteche, L. M. Nadler, A. H. Sharpe. 1994. Characterization of the murine B7-1 genomic locus reveals an additional exon encoding an alternative cytoplasmic domain and a chromosomal location of chromosome 16, band B5. J. Immunol. 153:5038.[Abstract]
  12. Jeannin, P., G. Magistrelli, J. P. Aubry, G. Caron, J. F. Gauchat, T. Renno, N. Herbault, L. Goetsch, A. Blaecke, P. Y. Dietrich, et al 2000. Soluble CD86 is a costimulatory molecule for human T lymphocytes. Immunity 13:303.[Medline]
  13. Oaks, M. K., K. M. Hallett, R. T. Penwell, E. C. Stauber, S. J. Warren, A. J. Tector. 2000. A native soluble form of CTLA-4. Cell. Immunol. 201:144.[Medline]
  14. Magistrelli, G., P. Jeannin, N. Herbault, A. Benoit De Coignac, J. F. Gauchat, J. Y. Bonnefoy, Y. Delneste. 1999. A soluble form of CTLA-4 generated by alternative splicing is expressed by nonstimulated human T cells. Eur. J. Immunol. 29:3596.[Medline]
  15. Magistrelli, G., P. Jeannin, G. Elson, J. F. Gauchat, T. N. Nguyen, J. Y. Bonnefoy, Y. Delneste. 1999. Identification of three alternatively spliced variants of human CD28 mRNA. Biochim. Biophys. Acta 259:34.
  16. May, B. P.. 1998. Southern hybridization in shampoo. BioTechniques 25:582.[Medline]
  17. Hattori, M., A. Fujiyama, T. D. Taylor, H. Watanabe, T. Yada, H. S. Park, A. Toyoda, K. Ishii, Y. Totoki, D. K. Choi, et al 2000. The DNA sequence of human chromosome 21: the chromosome 21 mapping and sequencing consortium. Nature 405:311.[Medline]
  18. Ishikawa, K., T. Nagase, M. Suyama, N. Miyajima, A. Tanaka, H. Kotani, N. Nomura, O. Ohara. 1998. Prediction of the coding sequences of unidentified human genes. X. The complete sequences of 100 new cDNA clones from brain which can code for large proteins in vitro. DNA Res. 5:169.[Abstract]
  19. Ling, V., R. C. Munroe, E. A. Murphy, G. S. Gray. 1998. Embryonic stem cells and embryoid bodies express lymphocyte costimulatory molecules. Exp. Cell Res. 241:55.[Medline]
  20. Aicher, A., M. Hayden-Ledbetter, W. A. Brady, A. Pezzutto, G. Richter, D. Magaletti, S. Buckwalter, J. A. Ledbetter, E. A. Clark. 2000. Characterization of human inducible costimulator ligand expression and function. J. Immunol. 164:4689.[Abstract/Free Full Text]
  21. Jellis, C. L., S. S. Wang, P. Rennert, F. Borriello, A. H. Sharpe, N. R. Green, G. S. Gray. 1995. Genomic organization of the gene coding for the costimulatory human B- lymphocyte antigen B7-2 (CD86). Immunogenetics 42:85.[Medline]
  22. Borriello, F., J. Oliveros, G. J. Freeman, L. M. Nadler, A. H. Sharpe. 1995. Differential expression of alternate mB7-2 transcripts. J. Immunol. 155:5490.[Abstract]
  23. Ogg, S. L., M. V. Komaragiri, I. H. Mather. 1996. Structural organization and mammary-specific expression of the butyrophilin gene. Mamm Genome 7:900.[Medline]
  24. Hollsberg, P., C. Scholz, D. E. Anderson, E. A. Greenfield, V. K. Kuchroo, G. J. Freeman, D. A. Hafler. 1997. Expression of a hypoglycosylated form of CD86 (B7-2) on human T cells with altered binding properties to CD28 and CTLA-4. J. Immunol. 159:4799.[Abstract]
  25. Greenfield, E. A., E. Howard, T. Paradis, K. Nguyen, F. Benazzo, P. McLean, P. Hollsberg, G. Davis, D. A. Hafler, A. H. Sharpe, et al 1997. B7.2 expressed by T cells does not induce CD28-mediated costimulatory activity but retains CTLA4 binding: implications for induction of antitumor immunity to T cell tumors. J. Immunol. 158:2025.[Abstract]
  26. Schweitzer, A. N., A. H. Sharpe. 1999. Mutual regulation between B7-1 (CD80) expressed on T cells and IL-4. J. Immunol. 163:4819.[Abstract/Free Full Text]
  27. Ling, V., S. Neben. 1997. In vitro differentiation of embryonic stem cells: immunophenotypic analysis of cultured embryoid bodies. J. Cell Physiol. 171:104.[Medline]
  28. Ling, V., D. Luxenberg, J. Wang, E. Nickbarg, P. J. Leenen, S. Neben, M. Kobayashi. 1997. Structural identification of the hematopoietic progenitor antigen ER-MP12 as the vascular endothelial adhesion molecule PECAM-1 (CD31). Eur. J. Immunol. 27:509.[Medline]
  29. Kirschner, M., J. Gerhart. 1998. Evolvability. Proc. Natl. Acad. Sci. USA 95:8420.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Phil Trans R Soc BHome page
B. R Blazar and W. J Murphy
Bone marrow transplantation and approaches to avoid graft-versus-host disease (GVHD)
Phil Trans R Soc B, September 29, 2005; 360(1461): 1747 - 1767.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Pathol.Home page
M. G. Petroff, E. Kharatyan, D. S. Torry, and L. Holets
The Immunomodulatory Proteins B7-DC, B7-H2, and B7-H3 Are Differentially Expressed across Gestation in the Human Placenta
Am. J. Pathol., August 1, 2005; 167(2): 465 - 473.
[Abstract] [Full Text] [PDF]

Home page
 BloodHome page
W.-I Lee, T. R. Torgerson, M. J. Schumacher, L. Yel, Q. Zhu, and H. D. Ochs
Molecular analysis of a large cohort of patients with the hyper immunoglobulin M (IgM) syndrome
Blood, March 1, 2005; 105(5): 1881 - 1890.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
L. Wassink, P. L. Vieira, H. H. Smits, G. A. Kingsbury, A. J. Coyle, M. L. Kapsenberg, and E. A. Wierenga
ICOS Expression by Activated Human Th Cells Is Enhanced by IL-12 and IL-23: Increased ICOS Expression Enhances the Effector Function of Both Th1 and Th2 Cells
J. Immunol., August 1, 2004; 173(3): 1779 - 1786.
[Abstract] [Full Text] [PDF]

Home page
 BloodHome page
S.-C. Wong, E. Oh, C.-H. Ng, and K.-P. Lam
Impaired germinal center formation and recall T-cell-dependent immune responses in mice lacking the costimulatory ligand B7-H2
Blood, August 15, 2003; 102(4): 1381 - 1388.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
S. Khayyamian, A. Hutloff, K. Buchner, M. Grafe, V. Henn, R. A. Kroczek, and H. W. Mages
ICOS-ligand, expressed on human endothelial cells, costimulates Th1 and Th2 cytokine secretion by memory CD4+ T cells
PNAS, April 30, 2002; 99(9): 6198 - 6203.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ling, V.
Right arrow Articles by Collins, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ling, V.
Right arrow Articles by Collins, M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS