CD27 is a lymphocyte-specific member of a re- cently identified receptor family with at least 10 members that includes the receptors for nerve growth factor and.
0022-1767/92/14912-3937$02.00/0 THEJOURNAL OF IMMUNOLOGY Copyright 0 1992 by The American Association of Immunologists
Vol. 149.
3937-3943. No. 12,December 15. 1992 Prlnted in U.S.A.
GENOMIC ORGANIZATION AND CHROMOSOMALLOCALIZATION OF THE HUMAN CD27 GENE' WILA.
M. LOENEN,'"
LOES A. GRAVESTEIN,"SARA BEUMER,'CORNELIS ANNEHAGEMEIJER,' AND JANNIEBORST"
J. M. MELIEF,+
From the *Divisionof Immunology. The Netherlands Cancer Institute, Antoni van Leeuwenhoek Huis Plesmanlaan 121, 1066 C X Amsterdam; 'Department of Immunohaematology and Bloodbank, Academic Hospital, Leiden; and 'Department of Cell Biology and Genetics. Erasmus University,Rotterdam, The Netherlands
CD27 is a lymphocyte-specific memberof a re- within thememory population, characterizedby the presence of the CD45RO protein isoform(7).In vitro activation cently identified receptor family with at least 10 members thatincludesthereceptors fornerve of peripheral blood cells via the TCR/CD3 complex gives growthfactorand TNF, CD40,and Fas.Several rise to a strong transient increase in CD27 membrane members of this family play a role in cell differen- expression; it reachesa maximum about3 to 4 days after tiation, proliferation, and survival. Within the activation and subsides during the next4 days (1, 3, 4, amino terminal ligand binding domain of these re- 6).Since this activation-related increase occurs relatively ceptors, repeat motifs have been identified. These late, CD27 may function as a receptor for a growth or repeats contain many cysteine residues in a condifferentiation factor. Consistent with this idea, CD27 servedpattern,characteristicof this family. We mAb have been shown to affect the TCR/CD3-induced have isolated and characterized the human CD27 gene to gain insight into the evolutionof this type proliferative T cell response (3,4, 8). On resting T cells, CD27 is a disulfide-linked homodiof receptor domain. The gene was localized on chromer of 50- to 55-kDa polypeptide chains ( 3 , 4, 9). The mosome 12, band 1 2 ~ 1 3Sequence . analysis showed mature protein is generated froma 30-kDa precursor by no correlation between the intron/exon organization and the subdivision ofthe protein into distinct extensive 0-linked glycosylation and addition of one Ndomains. Structural information for the cysteine- linked carbohydrate group (10).A 28- to 32-kDa soluble richdomain is containedwithinthreeexons. In form of this protein(sCD27)is present in the supernatant addition, the splice sites in the CD27 gene are lo- of in vitro-activated T cells, as well as in vivo in human cated in a different position from those in the reserum and urine[ 1 1). sCD27 levels may be increased in lated nerve growth factor receptor gene. However, a patient samples, which is presumably indicative of T cell comparison of the splice sites within the regions activation in vivo (1 2).sCD27 is not generated via alterencoding the respective ligand-binding domains of native RNA splicing, but arises via proteolysis of the the CD27 and nerve growth factor receptor genes membrane receptor (10). This takes place after expresidentifies the archetypal cysteine-rich building sion of CD27 protein at the cell surface and does not blocks, from which the members of this family may seem to require receptor internalization. It remains to be have arisen during the course of evolution. From resolved what physiologic response triggers the release this observation, we propose a new organization of of sCD27. the repeat motifs.
Sequence analysisof the cDNA (13)has identified CD27 a s a member of a novel receptor family (1 4, 15). including CD27 is a membrane glycoprotein that is exclusively the NGFR3 (16, 17). twodifferent TNFR(TNFR-I and B cell molecule CD40 (23). expressed within the lymphocyte lineage. It is found on TNFR-II(l8-21)). Fas (22). the the marker protein for malignantcells in Hodgkin's disthe surface of maturemedullarythymocytesandthe majority of the peripheralblood T cells, distributed evenly ease CD30 (24). the rat T cell Ag- OX40 (25). and the among the CD4+ and CD8+ subsets (1 -4).In addition, it murine T cell protein 4-1BB (26).At least theNGFR (27). is presenton a small populationof B cells(5).Only CD27+ TNFR (28).CD40 (29, 30),and Fas (31, 32) areinvolved cells can stimulateB cells to produce IgM (6).In the CD4+ in the regulationof cell proliferation, differentiation, and T cell subset in peripheral blood, all CD27- cells reside survival. These receptors aretype I transmembrane proteins, with a n amino-terminal Cys domain. This region probably folds into a globular structure, due to extensive Received for publication April 28. 1992. intramolecular disulfide bonding, and is involved in liAccepted for publication September 25, 1992. The costs of publication of this article were defrayed in part by the gand binding(33).In a number of cases (includingCD27). payment of page charges. This article must therefore be hereby marked the Cys domain is followed by a n extracellular, memadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. brane proximal region, the STP domain. The transmem' This work was supported by Grant 900-509-128 from the Nether- brane region is followed by a n intracellular domain of lands Organization for Scientific Research.J.B. was supported by a C . & variable length, ranging from 36 to 220 amino acids. The C. Huygens Fellowship from the Netherlands Organization for Scientific Research, Grant H93-155. Address correspondence andreprint requests to Dr. Wil A. M. Loenen. Division of Immunology. The Netherlands Cancer Institute, Antoni van Leeuwenhoek Huis, Plesmanlaan 121. 1066 CX Amsterdam, The Netherlands.
Abbreviations used in this paper: FISH, fluorescent in situ hybridization; Cys. region rich in cysteine residues; STP, region rich in serine, threonine,and proline residues: NGFR. nerve growth factor receptor; TNFR. TNF receptor; kb. kilobase pair.
3937
3938
ORGANIZATION OF THE HUMAN CD27 GENE
mega). Briefly, the plasmid was linearized at the EcoRI site in the cytoplasmic domains share a consensus sequence for vector and the sticky ends filled in with a-phosphorothioate nucleophosphorylation by protein kinase C but lack obvious tides, which render the DNA ends resistant to exonuclease 111. This homology. All members lack intrinsic kinase activity, and DNA wascut with XhoI, exonuclease I11 was added inlimiting have no homology to any sequences with known signal amounts. and atvarious time points samples were taken. The deletion plasmids were separated from nondigested plasmid on agarose transduction function. gel, recovered via electroelution, treated with S1 nuclease, and liThe common feature of the receptor family is theCys gated. Of each time point, DNA from 25 independent clones was domain. Repeatsof about 40 amino acids long have been isolated using the standard protocol with a n additional LiCl purifirecognized in this Cys region, containing cysteine resi- cation step. The clones were sequenced using the T7 polymerase sequencing kit with standard primers (Pharmacia) (41). Remaining dues in a conserved pattern with variable sequences in gaps were closed using specific oligonucleotide primers. between (14, 15). The NGFR protein has four such reAll sequence analysis was carried out using the Genetics Compeats, each with six cysteine residues. Though reminis- puter Group's software (421, based on sequence comparison proof Pearson and Lipman (43). Themultiple alignment program cent of zinc finger motifs (34). the spacing of the cysteines grams CLUSTAL (44) wasused to align the CD27 and NGFR proteins. Letter is characteristic of this particular receptor family (15). codes for amino acids and nucleotide ambiguities are as proposed by the Nomenclature Committee of the InternationalUnion of BiochemThe other members of the family can be aligned with NGFR and have up to six repeats, though some appear istry to (42).All other DNA manipulations were performed a s described (39).All enzymes were obtained from Boehringer Mannheim. unless be truncated.Theserepeatssuggestgeneduplication otherwise stated. from a single unit, allowing genes with varying domain Mapping of the CD27 gene. Chromosome localization of CD27 lengths. The presence of even numbersof cysteines sug- was determined by ;FISH. The 8.4-kb EcoRI fragment was labeled with Biotin- 1 -dUTp 1 by nick translation anddiluted in hybridization gests disulfide bridges between all residues. Only the buffer (50%formamide, 10%dextran-sulfate in 2X SSCP).After heat chromosomal locus of the NGFR gene has been isolated denaturation, 100 ngprobe was preannealed with 10 pg Cot-1 DNA and sequenced so far (35).The repeats in this case are for 1 to 2 h a t 37°C and hybridized overnight to denatured DNA of metaphase chromosomes from a normal volunteer. After stringent as, e.g., the washing, not contained within discrete separate exons hybridization sites were visualized using avidin-FITC, bioIg domains (36). and the single splice site in the Cys tinylated goat-anti-avidin antibody, and avidin-FITC in successive layers a s reported by Pinkel et al. (45). The DNA was counterstained domain does not occurat the endof a repeat. To investigate the evolutionary origin of the cysteine- with propidium iodide and 4',6-diamino-2-phenylindoleand observed by fluorescent microscopy using the appropriate filters. To rich repeats that form theligand binding domain of this confirm the chromosomal assignment, cohybridization of 100 ng receptor family, we have isolated the chromosomal locus preannealed CD27 probe and 5 ng of pa12H8, a n alphoid repeat encoding CD27 and mapped it to chromosome 12, band repetitive probe for the centromere of chromosome12. was per1 2 ~ 1 3 We . present the sequence of the human CD27 formed. gene and compare its intronlexon organization with that RESULTS of the NGFR gene. Based on this comparison, we have identified the cysteine-rich archetypal unit, from which Isolation and characterization of the human CD27 the ligand-binding domains may have arisen during the gene. A partial library of size-selected EcoRI-generated course of evolution. fragments of human genomic DNA was made in phage-)\ as described in Materials and Methods. Three phages M A T E R I A LA S ND METHODS were isolated from this library that hybridized strongly Isolation of the CD27 gene. Southern blot analysis of human to the CD27 cDNA (13). The 8.4-kb insert of one of these, genomic DNA using theCD27 cDNA (13)as probe identified a single XWL300, was extensively mapped and sequenced. A parEcoRI fragment of about 8 kb. This fragmentlacked restriction sites tial restriction map showing the sites for six restriction for both XbaI and EcoRV. A partial EcoRI-generated library was therefore prepared as follows: Human DNA was isolated from PBL enzymes is presented inFigure 1. and restricted with EcoRI, XbaI. and EcoRV to reduce the number Using the cDNA as a probe, a large intron waslocated of EcoRI fragments in the 8-kb range.After size separation on a n towards the center of the 8.4-kb fragment. All coding agarose gel, DNA ranging from 6.5 to 9.5 kb was isolated by elecleft troelution. This DNA was cloned into the bacteriophage-A EcoRI informationappearedtobecontainedwithinthe insertion vector NM607 (b538 srIA3' irnrn434 srlA4'srIA5' (37)), hand EcoRI-Hind111fragment of 1 kb and theright hand and plated on therecom- HindIII-BamHI fragment of 2.4 kb,respectively. packaged in vitro (Packagene kit: Promega). bination- and restriction-deficient Escherichia coli host DL538 Sequence of the CD27 gene. The 5' EcoRI-Hind111 (hsdR rncrA rncrB recD sbcC supE thyA (38)). ANM607 recombifragment of 1 kb was subcloned and sequenced as denants can easily be distinguished from the vector itself, since the plaque phenotype changes from turbid toclear after uptake of DNA. scribed in Materialsand Methods. Thisregion contains Optimal numbers of recombinant phage (45% clear plaques) were the 5' part of the gene, and the sequence is shown in obtained using a 2:l M ratio of insert to phage a t a final vector concentration of 50 pglml; 50.000 plaques were screened using the Figure 2A. The sequence of the 3' part of the gene was CD27 cDNA as probe. Six positive plaques were identified and ana- assembled from sequence informationof deletion clones lyzed. Three of these hybridized strongly to the cDNA in Southern starting at the unique XhoI site downstream of the gene blots. The 8.4-kb insertof one of the positive phages, XWLBOO, was as described in Materials and Methods. The sequence of used for all additional studies. A detailed restriction map was made after transferof the insert intopUC18 (39).All sites were confirmed the 2.4-kb HindIII-BamHI fragment downstream of the by Southern blot analysis of human genomic DNA to exclude delelarge intron is shown in Figure 2B. tions or rearrangements in thecloned fragment. The CD27 gene consistsof six exons, separatedby five Sequence analysis of the 8.4-kb EcoRlfragrnent. After HindIII introns, the largest of which is nearly 5 kb and lies in digestion of the 8.4-kb EcoRI fragment, two fragments of 1.4 and 2.5 kb did not hybridize to the human cDNA, whereas coding se- between exons I1 and 111 (Fig. 1). The sequenceof the six quences were identified within the two EcoRI- HfndIII fragments of exons is in agreement with the previously determined 1 and 3.5 kb. The 1-kb EcoRI- Hind111 fragment was subcloned in sequence of the CD27 cDNA (13).There are no separate M 13 mpl8 (39) and sequenced using standardM 13 primers,as well as specific internal oligonucleotides. To sequence the rest of the exons in the 5'- and 3"nontranslated region. All splice locus, the EcoRI sites of the 8.4-kb fragmentwere filled in, and the sites follow the consensus splice sequence identified for fragment was cloned into the SrnaI site of plasmid vector pEP30 (40). This construct was used to make unidirectional deletions, start- the various TCR splice sites (46) and always occur being from the unique XhoI site using the Erase-a-Base system (Pro- tween the first and second base of the codons for g1yZ6,
3939
ORGANIZATION OF THE HUMAN CD27 GENE Figure 1 . Partial map of the 8.4-kb EcoRI fragment containing the human CD27 gene. Boxes, Exons: openboxes, nontranslated 5' and 3' sequences. Restrictlon sites are shown for EcoRI (R). HlndlIl (HI. BamHI (B).Sac1 (SI,Scal (Sc).and XhoI (X). In addition, the locus contains multiple Smal and PstI sites [data not shown). but no sites for EcoRV or'XbaI. Bars below the figure. Location of the 5' and 3' sequences, whlch Include all six exons. and are presented in Figure 2, A and B.
2
1 I
1
1
S H
5'
1
4
3
1
I
1
B
H
SS
1
1
5 1
1
1
SH
8
7
h
1
1
I
l
E X
sc
I
I
II
gly50,g l ~ ' ~ pro'6o, ', and asnZoo, respectively, of the mature protein. The sequences around the splice sitesare shown in Table I. The firstexon encodes the 5'-nontranslatedregion, the signal peptide (20 amino acids), and 26 amino acids of the mature receptor. This exon contains the firstfour of the 16 cysteine residues present in the amino-terminal Cys domain, which is the putative ligand-binding domain. The second exon consists of 44 amino acidswith 30 amino acids with the six cysteines. The remaining last six cysteines of the Cys domain lie beyond the largest intron in exon 111, together with the first 30 residues of the STP domain. Like the Cys domain, the STP domain is spread over three exons. Exon IV contains 30 residues and exon V the remaining1 1 amino acidsof this domain. In addition, exon V encodes the transmembrane region and nine amino acids of the cytoplasmic domain, splicing within the consensus sequence for phosphorylation by protein kinase C (serine/asparagine/lysine). The last and largest exon VI (347 bases) contains the last 40 amino acids of the cytoplasmic domain and the untranslated3' region. This exon contains a AATAAA poly(A) attachmentsignal200 basesdownstream of the TGA stop codon. Comparison of the genomic organizationof the CD27 a n d N G F R genes. A s outlined in the above section, the intronlexon organization of the CD27 gene does not coincide with the distinctive protein domains. We compared the splice sites of the CD27 gene with thoseof the N G F R gene, the only other memberof this family that hasbeen characterized in detail (Fig. 3).The Cys domain of CD27 spans 100 amino acids (with 16 cysteines), and that of N G F R , 161 residues (with24 cysteines]: both are followed by a STP domain. The Cys domains of neither CD27 nor N G F R are contained within separate exons, though in the N G F R gene the splice siteat theend of exon 111 allows the precise division of the Cys and STP domains. The first and thirdexon of CD27, as well as thesecond exon of N G F R , contain amino acidsfrom the adjacent protein domains. The other domains are also spread over different exons. Analysis of the 5' region of the CD27 gene. A s judged from sequence analysisof two CD27 cDNA, transcription may start 100 bases upstream of the ATG codon (Fig. 2A), at thesecond G of a string of 5 G (13),though primer extension experiments have so far failed to show a specific transcriptional start site. Upstream of this site are three consensus sequences for the transcription factor AP-1 (TGAXMA (47)) around positions 87-92, 113-1 18, and 150- 155, Myb-binding a site (CAACTG (48))overlapping with a n AP-1 site at 91-96, and a CCAAT box, recognized by the CP2 protein (AGCCACT(49))at position 353-360. Around positions23-53 and 289-331, two purine-rich regions are located. These two regions share
Ill
IV
v
8.5 kb
l
I
1
R I
3'
VI
homology with a variety of sequences in the database, but a s yet no particular function can be ascribed to them. Chromosomal localization of the CD27 gene. FISH using biotinylated CD27 DNA a s probe gave a clear hybridization signal on the short arm of chromosome 12, on band 1 2 ~ 1 3based , on 4',6-diamino-2-phenylindole banding. Thechromosome assignment wasconfirmed by cohybridization of CD27 DNA and an alphoid repeat probe, specific for the centromere of chromosome 12 (Fig. 4). About 50 metaphases were scored and 10 were photographed. In 48 metaphases, both chromosomes 12 showed convincing double hybridization spots that were clearly brighter than the background hybridization to other chromosomal regions. In two cells, only one chromosome 12p waslabeled. DISCUSSION
We have presented the sequence of the human CD27 gene. Southern blot analysis of human DNA with a variety of restriction enzymes had previously indicated that CD27 is a single copy gene (13) (data not shown).EcoRI digestion resulted in a single hybridizing fragment of about 8 kb. This fragment was cloned in phage-X and analyzed in detail. By FISH, using this fragment a s a probe, the gene has been assigned to chromosome 12 and has tentatively been mapped to band 1 2 ~ 1 3One . other member of the family, the TNFR-I gene, has also been mapped to the shortarm of chromosome 12 [ 12pter-cen). though no precise data areavailable so far (50).The only other member of the family, of which the gene has been cloned and mapped, the N G F R gene, is located on chromosome 17q21-22 (35).Several other genes that encode leukocyte Ag. Von Willebrand factor (12pter-p12), CD4 (12pter-pl2).its relative LAG3 (12pl3.3),and CD9 (12pl3), have beenmapped between band 12pl2 and the terminus (50).Other genes located in this region include the K-rasgene (p12.1)and some genes encoding salivary proline-rich proteinsof the BstNI and HaeIII families (50). which show some homology with the proline-rich STP domain of CD27. The possible significance of the latter is not obvious at present. Changes in chromosome 12p areobserved in a variety of cancers: a classicalexample is the i( lap) in germ line tumor (50).In hematologic malignancies, i.e., myelodysplasic syndrome, myeloid leukemia, and lymphoid leukemia, deletionsor translocations involving 12p arealso regularly found (50). The precise breakpoints and gene targets of these rearrangements are not yet known, but are expected to be quite heterogeneous. The CD27 gene is a n additional tool to investigate cytogenetic changes in the 12p 13 region and reciprocally to obtain refined mapping of the CD27 gene relative to other 12p13 markers and to various cancer "breakpoints." The 5' region of the CD27 gene contains consensus
3940
...............................
.
x
.
301 AAACACCCACMTAGAGATTCTGCCTTCAAAGGTTGGCl 'TCCCACCTGAACCACCCACTGCCCAGGGGGTGC~GAAGAGACAGCAGCGCCAGCTTGGA
_""
801 TCGTGMCCACTGTGACCACCATAGAAAGACTGCTCAGTGTGATCCTTGCATACCGGGGGTCTCCTTCTCTCCTGACCACCACACCCGGCCCCACTGTGA
(30)
Figure2. Sequence of the 5' and 3' regions of the human CD27 gene. A. The 5' region between the left handEcoRI and Hind111 site containing the first two exons: E. the 3' region between the right hand Hind111 and RamHI sites (as indicated by two bars in Fig, 1). #, Location of the first base of the CD27 cDNA sequence andpotentialtranscriptionstartsite. Nucleoleftmargin, tides are numbered in the starting from the EcoRI site. Consensus sequences for CP2 (353-360).Myb (9196),andAP-l(87-92.113-118.and150155) binding are underlined (- - - -). Dotted linesaboue the sequence, Purinerich regions discussed in the text, those underneath the Alu repeats. The poly(A) adenvlation site i s underlined. A. End of exon VI. Nucleotide ambiguities are explained in Materials and Methods. The
V
K
D
C
D
Q
H
R
K
T
A
P
C
O
~
C
I
P
C
V
S
F
S
P
D
H
H
T
R
P
H
C
E
601 G ~ C C ~ C T O C A C C T T ~ G C T C T C C A C A C G T C T C A C T C ~ C C T A T C C T C ~ C T G G C C C T C ~ C T T C C C C C A ~ T C T T C T C G T T C G C M C T O C A C C A T U C ~ G (91) L L V P Y C I L J A
lL0l (183)
1501 U U T A l A C A ~ C U I ~ T ~ C A ~ C A C C A C ~ ~ G C C l G G C G ~ T C C T C C C C C ~ C C A C C A C A C ~ C C A C T G G C T C ~ C C C C A C T ~ C C C A ~ G C C T ~ ~ I C Y R S N
(216) Y
1601 GAAACCCACO\CCTCCACTTCACCAGCCTTGGTCCTACCCCTTCTCTCCCCTTAGCTGGCGTGCTCCTGACACCCCTCCCCC~GCGC~CCCGCCTCTAC
1701 C C A l C T C C T ~ C T C C C G T C T ~ C C C C T G C C C ~ C A C T C C T G G ~ C A A G A C T C A ~ C G G A T C T C C ~ T C T G C A ~ C ~ G G A G M A ~ T C C T G T G G A ~ C C l G C A G A G (221) K C E S P V E P A E P
1801 C T T G T C G T T ; C A G C T ~ C ~ C ; A G G ~ G G A G ~ A ~ G G C A G C I A A A C ~ G ~ A ~ ~ ~ ~ ~ ~ ~ I G C T C C C C C ~ G A G C (232) C R Y S C P R E E E G E T I P I O E O Y R K F E P A C S P 1901 ClCCGGkAGCTCCAClACAGCCCTCGCCTCCACCCCCACCCCGCCGACCATCCAAGGGAGAGl~CACCTGGCAGC~CMCTGCACTCCCATCCTCTTO 2001 TCltGGCCC~TTCCTCTGT~CACGTG*CPIGIGIGCCTTI~ C C A D I C I G C ~ A C G G A C C A C ~ A ~ ~ T A T G ~ T ~ G G T G ~ ~ C T G G G M C C A G G A G ~ C C A
2101
_"
G C C A G C T ~ C ~ . C C T G C ~ C T G ~ * ~ G A G ~ G ~ ~ ~ ~ ~ ~ ~ ~ C T G G ~ T ~ ~ M A A C A ~ A C T ~ C C ~ G C ~ C C_"" ~ ~ ~ ~ A C ~ ~ ~ ~ A ~ ~ C I A ~
2201 A C A C A ~ C A C C A C C C T ~ C A C E C T ~ C A G C T ~ C C T C T C ~ C A M T C C C A C T C C C A A M A T C C A C T T ~ C C T G G C C G C C ~ C C ~ G G ~ T C A G G T C A C ~ C T G G C A G A G ~
2301 ACGlTlCTCAGCCGGTTATGTTGGCGCClGCT~AGGGCGGGGGlG~G~AGlCCAGGCTAClGTG~TCC
B a MI
sequences forMyb, AP-1, and CP2 binding, but noTATA box. Many genes lack the TATA box, such as CD3-6 (51) and Thy-I (52), but also the NGFR gene (35).Myb h a s beenshowntobe involved in the G,/S transitionin
normal human T cells after Ag stimulation (53).A s with CD27 mRNA (54).levels of c-rnyb mRNA increase around 14 to 16 h after T cell activation (54, 56). Therefore, it will be of interest to investigate whetherMyb is involved
ORGANIZATION HUMAN OF THE
394 1
CD27 G E N E
TABLE I Nucleotide sequenceoJ infron/exon boundaries in t h e h u m a nCD27 g e n e Exon
5’ End cccttaccctctcctcccag ctgggcctctcttccccccag ttctgtctctgtttttccag gttgtcccctcctattacag catcggatctccttctgcag
1 GAACAT GTCCTT AGATGC CCCAAA ACAAAG
2
Exon Exon 3 Exon 4 Exon 5 Exon 6
AGCCAG ACTCTG TCAGTG GGCCAC GATCAA CACCCT
gtaagaggsg’ gtgaggtgsg gtaagttcca gtgagttttc gtaagagaca gcagcctcca’
Exon sequcncc is shown in capital letters: intron sequence. in lower case. 3‘-End sequence: sequencc common to3’ end of the CD27 mRNA (exeluding poly[A) tail) is underlined and shownin eapitals: downstream sequenceis shown in lower case. a
L Figure3.Comparison of t h e splice sites within the human CD27 a n d NCFR genes. The different domains of the proteinsare indicated (L. Leaderpeptide: TM. transmembrane region: IN. intracellular
CYs
STP
TM
IN
CD27
Exon
cated by oerfical bars’.
Exon
I
I
I1
F i g u r r 4 . Metaphase rhromosomes from a rultureof PBL derived from a healthy female volunteer. hybridized with biotinylatedCD27 and pm12H8 probes a s indicated in Maferialsand Mefhods.CU27 shows hybridization to lhe short arm of rhromosome 12. whileprr12H8 identifies the rentromeric region of 12. A. FlTC fluoresrenre: H. 4’.6-dia1nino-2-phenylindolc fluorescenceshowingQ-bands.Rightpanel.Partialmetaphases.Partial a fine background bul also increased specific hybridization. Arrows. Chromosome 12. metaphases after less stringent washing show
in the activation-related increase in CD27 expression. cance, if any. of these islands remainsto beinvestigated. The CD27 upstream region has theusual low CpG dinu- No CpG islands havebeen found in itsrelative, the NGFR cleotide content (3%).In contrast to the NGFR gene. gene, within either the gene or its flanking sequences which has an upstream CC-rich region (73% G+C). the (57). The intronlexon organization of the CD27 gene does 5’ region of CD27 has no increased level of GC vs AT bp (56%C+C for the upstream region vs 57% for the com- not coincide with the distinctive protein domains, nor plete region). This would be consistent with the finding with that of the NGFR gene. A division of a protein into that GC-rich regions are often associated with house- functionaldomains has been described for. e.g.. the keeping genes or genes with low constitutive expression members of the Ig supergene family (36).The Ig domain in different cell types, such as the NGFR gene. In con- appears in a wide variety of proteins and is linked to trast, CD27 expression appears to be strictly limited to many other different protein domains. Splicing always location in a codon, allowing variation lymphocytes at specific stages of differentiation and ac- occurs at the same in the number of Ig domains without loss of reading tivation. The largest intron in the CD27 gene between the sec- frame. The members of the NGFR family also have a ond and thirdexon contains numerous repeats, including variablenumber of cysteine-richrepeats. Though the a n Alu repeat 21 0 to 510 bases upstream of exon Ill splice sites within the Cys domain of the CD27 and NGFR (indicated in Fig. 2B). A s yet, we have no complete se- genes differ. all sites split a codon between the first and quence of thisintron.The library of deletion clones second base, which allows a similar variation in the number of repeats. lacked clones that should link the various contiguous A s outlined in the introduction, we were particularly sequences. Due to the presence of highly repetitive DNA, no specific oligonucleotide primers could be designed to interested in the location of the splice sites in the ligandcross the remaining gaps. This intron contains two CpG binding Cys domain in relation to the previously identiislands, with regions of 24% and 15%CpG. The signifi- fied cysteine repeats (15, 16). A comparison of the pos-
3942
ORGANIZATION OF THE HUMAN CD27 GENE
tulated repeats in NGFR and CD27 is shown in Figure 5A. Each repeat contains six cvsteines with conserved spacing. Accbrding to this cD27 has two plete and one incomplete repeat. However, splice sites divide the CD27 Cys domain in three parts with four, six, and six cysteines* the sing1e site within the NGFR Cys region generates two parts with an uneven IllJmber Of cysteines (7 and 17)' Therefore' the Observed 'PIice sites in the NGFR and cD27genes do not coincide with the previouslv identified cvsteine reDeats in the ligand-binding domain. If one aligns the Cys domain of CD27 with that of NGFR and reconsiders the location of the splice sites with respect to the 'ysteine it becomes that they do occur in a particular pattern (Fig. 5 B ) . Both splice sites in the ~ ~ Cys 2 domain 7 occur at exactly the Same position within the first and second postulated repeat, Le., between the first andsecond base of a triplet coding for glycine, the third amino acid after a cxxc sequence. Exon I contains one CX,CXXC (C,) motif, whereas exon I1 and exon I11 contain two C3 motifs. This the possibility Of without loss Of reading frame. The splice site that separatesexons I1 and 111 in the NGFR genetiesinwith this partitioning of the Cys repeats in C3 motifs: it occurs at a position that is comparable to a site exactly in between the two C3 motifs
in exon I' Of the cD27 gene' Exon 'I Of the NGFR gene contains two C3 motifs, whereas exon I11 encodes the remainder of the Cys domain, with five complete C3 motifs anda finalregion containing only CX,C. The presence Of the five (CD27)or seven (NGFR) c3 motifs adds up to an uneven number of cysteines. However*in both proteins a n extra cysteine residue is present in the firstC3 motif, which may be of importance for the formation of disulfide bonds, involved in folding this part of the protein into a globular domain. Most other members of the family identified so far also possess this additional cysteine residue. This division in C3 motifs, based on archetypal units, explains the puzzling splice site after the cysteine in the NGFR gene' Proof for this new model for the evolution of the Cys domain of this receptor family via duplication of this C3 motif will await analysis of the gene organization of other members of the family. Acknowledgments. We thank E.M.E.%lit for performing the in situ hybridization experiments, M.A. van for in preparing the manuscript, and Dr. N.E. Murray for XNM607 and E. colt DL538.
FLgure5. A , Comparison of the repeats of inthe NGFR Cys domainwiththose CD27. using thealignment reviewed by Mallett and Barclay ( 1 5).This alignment is based on the position of the six cysteine residues in each repeat. CD27 two hascomplete and one incomplete repeat. Splice sites (*)eitherabove (CD27)or below (NGFR)the sequence do notcoincide with this division. E. Alternative division of the Cys domains of CD27 and NGFR in Cs motifs, based on intron/exonboundaries.Each motif contains the sequence CX,CXXC. and all splice B sites (''1 elther above (CD27Jor below CD27 (NGFRJthe sequence occur at the end of such a motif.
REFERENCES 1. Bigler, R. D.. and N. Chiorazzi. 1986. A novel disulfide-linked cell surface molecule present on resting and activated human T lymphocytes. In Leukocyte Typlng I I . E. L. Reinherz. B. F. Haynes. L. M. Nadler. and I. D. Bernstein. eds. Sorinser-Verlae.New York. D. 503. 2. Stockinger. H.. 0.Majdic. K. Liszka.-U.Kolle;, W. Holter; 'c. Peschel. P. Bettelheim,H. Gisslinger, andW. Knapp. 1986.T14. a n o n modulating 150-Kd T cell surface antigen, In Leukocyte Typing 11, E. L. Reinherz. B. F. Haynes. L. M. Nadler, and 1. D. Bernstein. eds. Springer Verlag, New York. p. 513. 3. Van Lier. R. A. W., J. Borst. T. M. Vroom, H. Klein, P. van Mourik, W. P. Zeylemaker. and C. J. M. Melief. 1987. Tissue distribution and biochemical andfunctionalproperties of Tp55 ICD271, a novel T cell differentiation antigen. J. irnmunol. 139:i589: 4. Bigler, R. D.. Y. Bushkin, and N. Chiorazzi. 1988. S152 (CD27J.A modulating disulfide-linked T cell activation antisen. J. Irnrnunol. v 141:21. 5. Maurer, D.. W. Holter, 0. Majdic, G. F. Fischer. and W. Knapp. 1991. CD27 expression by distinct a subpopulation of human B lymphocytes. Eur. J. Irnrnunol. 20:2679. 6. Van Lier. R.A. W.. M. Oudkerk POOl. P. Kabel. S. MOuS. F.Terpstra, M. A. De Rie, C. J. M. Melief, and F. Miedema. 1988. Anti-CD27 monoclonal antibodiesidentifytwofunctionallydistinctsub-populations within theCD4+T cell subset. Eur. J. Irnrnunol. 18:81I . 7. De Jong. R.,M. Brouwer,B.Hooibrink, T. vanderPouw-Kraan. F. Miedema. andR. A. W. van Lier. 1992. TheCD27- subset of peripheral blood memory CD4+ lymphocytes contains functionally differentiatedTlymphocytes that develop by persistentantigenicstimulation in vivo. Eur. J. Irnrnunol. 22:993. 8. Sugita, K.. Y.Torimoto, Y. Nojima, J. F. Daley, S. F. Schlossman. and C. Morimoto. 1991. The 1A4 molecule (CD27) is involved inT activation. J . rrnrnunoz.147:1477. 9. Borst, J., C. Sluyser, E. de Vries, H. Klein, C. J.M. Melief, and R. A. W. Van Lier. 1989. Alternative molecular form of human T cellspecificantigen CD27 expresseduponT cell activation. Eur. J. Irnrnunol. 19:357. 10. Loenen. W. A. M., E. de Vries, L. A. Gravestein, R.Q. Hintzen. R. A. W. vanLier,and J. Borst. 1991.The CD27 membranereceptor, a lymphocyte-specific member of thenervegrowthfactorreceptor family, gives rise to a soluble form by protein processing that does not involve receptor endocytosis.Eur. J. Irnrnunol. 22447. 11. Hintzen, R. Q.. R. de Jong, C. E. Hack, M. Chamuleau. E. F. R. de Vries, I. J. M. ten Berge, J. Borst. and R. A. W. van Lier. 1991. A soluble form of the human T cell differentiation antigen CD27 is released after triggeringof the T cell receptor/CD3 complex. J. Irnrnunol. 147:29. 12. Hintzen. R. Q.. R. A. W. vanLier.K. C. Kuypers, P. A. Baars, W. Schaarsberg, C. J. Lucas, and C. H. Polman. 1991. Elevated levels of a soluble form of the T cell activation antigen CD27 in cerebral spinal fluid of multiplesclerosispatients. J. Neuroirnrnunol. 35:211. 13. Camerini. D., G. Walz, W. A. M. Loenen, J. Borst, and B. Seed. 1991. The T cell activation antigen CD27 is a member of the nerve growthfactor/tumornecrosisfactorreceptorgenefamily. J. Irnrnunol. 147:3165. 14. Loenen, W. A. M.. R. de Jong, L. A. Gravestein,R. A. W. vanLier, and J. Borst. 1991. The T cell activation molecule CD27 is a member of the nerve growth factor receptor gene family. In BiologicalSignal Transduction. NATO Series. Vol. H 52, E. M. Ross and K. W. A. Wirtz, eds. SpringerVerlag. Berlin/Heidelberg. p. 131. 15. Mallett, S . . and A. N. Barclay. 1991. A new superfamily of cell surface proteins related to the nerve growth factor receptor. Irnmunol. T o d a y 12:220. 16. Johnson, D., A. Lanahan, c. R. Buck, A. Sehgal, c. Morgan, E. Mercer, M. Bothwell,and M. Chao. 1986.Expressionandstructure of the human NGF receptor. Cell 47:545. 17. Radeke, M. J.. T.P.Misko, C. Hsu, L. A. Herzenberg.and E. M. Shooter. 1987. Gene transfer and the molecular cloning of the rat
-
A CD27CIS 1 1 NGFR CYS
SCPERHYAa-----GKLC--C---QMCEP--GTFLVKDCDQHRK~OCD ACPTGLYTH-----SGRC--C---KACNL--CEGVAQPCG--ANQTVCE
C D 2 7C I S 2 NGFR C I S 2
PClPGGVSFSPDHHTRPCESC---RHCNS~-GLLVRIIWCTI-TANAECPCLDNVTFSDVVSATEPCKPC---TECLG--L~SHSAPCVE-ADDAVC-
C D 2C7 I S 3 NGFR CIS 3
ACRNGYPCRD--"-KECTEC
NGFR C I S
RCAYGYYODEET---GHCEAC---SVCEV--CSGLVFSCPD-KPNTVCE
4
ECPEGTYSDEANHV-DPCLPC---TVCED--TERQLRECTP-YADAECE
1 2 3 4 SCPERHYYAQCKLCCOWCEPG TFLVKDCDPHRKAAQCDPCIP GVSFSPDHHTRPHCESCRHCNSG -LLVRNCTITANAECACRN CY""-"QCRDKECTECDPL
1
2
3
4
NGFR ACPTGLY-THSCECCKACNLG ECVAOPCCANQ-TV-CEPCLD SVTFSDWSATEPCKPCTECVCL PSWSAPCVEADDAVCRCAV GVY0DETTGRC"EACRVCEAC NGFR . . . . . . . . . . . . . . . . . . . . . SGLVFSCQDKONTV-CEECPD G-TYSDEANHVDPCLPCTVCEDT EROLRECTRYADAECE 6 7 a *
5
5
ORGANIZATION OF THE HUMAN CD27 GENE 37. nerve growth factor receptor. Nature 325593. 18. Loetscher, H., Y. E. Pan, H. Lahm. R. Gentz, M. Brockhaus, H. Tabuchi, and W. Lesslauer. 1990. Molecular cloning and expression Cell 61 :351. of the human55 kd tumor necrosis factor receptor. 38. 19. Schall, T. J., M. Lewis, K. J. Koller, A. Lee, G. C. Rice, G. H. W. Wong, T. Gatanaga, G. A. Granger, R. Lentz, H. Raab, W. J. Kohr, and D. V. Goeddel. 1990. Molecular cloning and expression of a 39. receptor for human tumor necrosis factor. Cell 61 :361. 20. Smith, C. A.. T. Davis, D. Anderson, L. Solam. M. P. Beckman, R. Jerzy, S. K. Dower, D. Cosman, and R. G. Goodwin. 1990. A receptor 40. for tumor necrosis factor defines an unusual family of cellular and viral proteins. Science 248: 1019. 21. Kohno, T., M. T. Brewer, S. L. Baker, P. E. Schwartz, M. W. King. K. K. Hale, C. H. Squires, R. C. Thompson, and J.L. Vannice. 1990. 41, A second tumor necrosis factor receptor gene product can shed a naturallyoccurring tumor necrosis factor inhibitor. Proc. Acad. Natl. Sci. USA 87:8331. 42. 22. Itoh, N.. S. Yonehara, A. Ishii. M. Yonehara, S.4. Mizushima, M. Sameshima, A. Hase, Y.Seto. and S. Nagata. 1991. Thepolypeptide encoded by the cDNA for human cell surface antigen fas can mediate 43. apoptosis. Cell 66:233. 23. Stamenkovic, I., E. A. Clark. and B. Seed. 1989. A B-lymphocyte activation molecule related to the nerve growth factor receptor and 44. induced by cytokines in carcinomas.EMBO J . 8:1403. 24. Durkop, H.. U. Latza, M. Hummel, F. Eitelbach, B. Seed, and H. 45. Stein. 1992. Molecular cloning and expression of a new member of the nerve growth factor receptor family that is characteristic for Hodgkin's disease. Cell 68:421. 46. 25. Mallett, S., S. Fossum, and A. N. Barclay. 1990. Characterization of the MRC OX40 antigen of activated CD4 positive T lymphocytes-a molecule related to nerve growth factor receptor.EMBO 47. . I 0. IT)fi? -. Y."". 26. Kwon, B. S., and S. M. Weissman. 1989. cDNA sequences of two 48. inducible genes. Proc.Natl. Acad. Sci. USA 86:1963. 27. Levi-Montalcini, R. 1987. The nerve growth factor 35 years later. Science 237:I 154. 28. Beutler, B., and A. Cerami. 1989. Thebiology of the cachectin/TNF- 49. 01 primary mediator of the host response. Ann. Rev. Immunol. 7:625. 29. Liu, Y.-J., D. E. Yoshua. G.T. Williams. C. A. Smith, J. Gordon, and 1. C. M. MacLennan. 1989. Mechanismsof antigen-driven selec- 50. tion In germinal centers. Nature 342:929. 30. Pauli, S., A. Rosen. B. Ehlin-Henriksson. S. Braesch-Andersen, E. 51. Jakobson, H. Koho, and P. A. Perlmann. 1989.Thehuman B lymphocyte and carcinoma antigen CDw40 is a phosphoprotein involved in growth signal transduction. J . Immunol. 142:590. 31. Yonehara. s., A. Ishii. and M. Yonehara. 1989. A cell-killing mono52. clonal antibody (Anti-Fas) to a cell surface antigenco-downregulated with the receptorof tumor necrosis factor.J . Exp. Med. 169: 1747. 53. 32. Trauth, B.C., C . Klas, A. M. J. Peters, S. Matzku, P. Moller, W. Falk. K.-M. Debatin. and P. H. Krammer. 1989. Monoclonal antibody-mediated tumor regression by induction of apoptosis. Science 245:301. 54. 33. Welcher, A. A.. C. M. Bitler, M.J. Radeke, and E. M. Shooter. 1991. Nerve growthfactorbindingdomain of thenervegrowthfactor receptor. Proc. Natl. Acad. Sci.USA 88:159. 34. Evans, R. M.. and S. M. Hollenberg. 1988. Zinc fingers: guilt by 55. association. Cell 52: I . 35. Sehgal, A., N. Patil, and M. Chao. 1988. A constitutive promoter 56. directs expression of the nerve growth factor receptor. Mol. Cell. Biol. 8:3160. 36. Williams. A. F.,and A. N. Barclay. The immunoglobulin superfamily-domains for cell surface recognition. Ann. Reu. Immunol. 57. 6:381.
3943
Murray, N. E. 1983.Phage lambdaand molecular cloning. In Lambda 11. Hendrix, R. W., J. W. Roberts. F. W. Stahl. and R. A. Weisberg, eds. Cold Spring Harbor Laboratory. Cold Spring Harbor. NY, p. 395. Whitaker, P.A., A. J. B. Campbell, E. M. Southern, and N. E. Murray. 1988. Enhanced recovery and restriction mapping ofDNA fragments cloned in a new X vector. Nucleic Acids Res. 16:6725. Sambrook, J., E. F. Fritsch, and T. Maniatis (eds.). 1989. Molecular Cloning. A Laboratory Manuul, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor. NY. Imboden, M. A.,P. W. Laird. M. Affolter, and T. Seebeck. 1987. Transcription of the intergenic regions of the tubulin gene cluster of Trypanosoma brucei: evidencefor a polycistronic transcription unit in a eukaryote. Nucleic Acids Res. 157357. Tabor, S.. and C. C. Richardson. 1987. DNA sequence analysis with a modified bacteriophage T7 DNA polymerase. Proc. Natl. Acad. Sci. USA 84:4767. Devereux. J. P. Haeberli, and 0.Smithies. 1984. A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res. 12:387. Pearson, W. R., and D. J. Lipman. 1988. Improved tools for biological sequence comparison. Proc. Natl. Acad. Sci. USA 852444. Higgins, D. G.. and P. M. Sharp. 1988. Clustal: a package for peron a microcomputer. Gene formingmultiplesequencealignment 73:237. Pinkel, D.,T. Straume, and J. Gray. 1986. Cytogenetic analysis using quantitative, high-sensitivity, fluorescence hybridization. Proc. Natl. Acad. Sci. USA 83:2934. Shapiro. M. B., and P. Senapathy. 1987. RNA splice junctions of different classes of eukaryotes: sequence statistics and functional implications in gene expression. Nucleic Acids Res. 15:7155. Curran, T., and B. R. Franza, Jr. 1988. Fos and Jun: the AP-1 connection. Cell 5539.5. Biedenkapp. H., U.Borgmeyer, A. E.Sippel, and K. H. Klempnauer. 1988. Viral myb oncogene encodes a sequence specific DNA binding activity. Nature335:835. Chodosh, L.A., A. S. Baldwin, R. W. Carthew, and P. A. Sharp. 1988. Human CCAAT-binding proteins have heterologous subunits. Cell 52: 1 1 . Solomon, E., and C. Rawlings (eds.). 1991. Human gene mapping 1 1. Proceedings of the 1 1th InternationalWorkshop on Human Gene Mapping. Cytogenet. Cell Genet. 58640. Van den Elsen. P., K. Georgopoulos, B.-A. Shepley, S. Orkin, and C. Terhorst. 1986.Exonlintron organization of the genes coding for the 6 chains of the human and murine T-cell receptor/T3 complex. Proc. Natl. Acad. Sci. USA 83:2944. Giguere. V.,K.4. Isobe, and F. Grosveld. 1985. Structure of the murine Thy-1 gene.EMBO J. 4r2017. Gewirtz, A. M., G. Anfossi, D. Venturelli, S. Valpreda, R. Sims, and B. Calabretta. 1989. G , / S transition in normal human T-lymphocytesrequiresthenuclearprotein encoded by c-myb.Science 245: 180. De Jong, R., W. Loenen, M. Brouwer, L. van Emmerik. E. F. R. de Vries, J. Borst, and R. A. W. van Lier. 1991. Regulation of expression of CD27. a T-cell specific memberof novel a familyof membrane receptors. J . Immunol. 146:2488. Crabtree. G. R. 1989.ContingentgeneticregulatoryeventsinT lymphocyte activation. Science 243:355. Reed, J. C.. J. D. Alpers, P. C. Nowel. and R. G. Hoover. 1986. Sequential expression of protooncogenes during lectin-stimulated mitogenesis of normal human lymphocytes. Proc. Natl. Acad. Sci. U S A 83:3982. Bird, A. P. 1986. CpG-rich islands and the functionof DNA methylation. Nature321:209.
