Identification of APC2, a homologue of the adenomatous polyposis coli tumour suppressor

Identification of APC2, a homologue of the adenomatous polyposis coli tumour suppressor

Brief Communication 105 Identification of APC2, a homologue of the adenomatous polyposis coli tumour suppressor J.H. van Es*, C. Kirkpatrick†, M. va...

328KB Sizes 8 Downloads 136 Views

Brief Communication

105

Identification of APC2, a homologue of the adenomatous polyposis coli tumour suppressor J.H. van Es*, C. Kirkpatrick†, M. van de Wetering*, M. Molenaar‡, A. Miles*, J. Kuipers*, O. Destrée‡, M. Peifer† and H. Clevers* The adenomatous polyposis coli (APC) tumoursuppressor protein controls the Wnt signalling pathway by forming a complex with glycogen synthase kinase 3b (GSK-3b), axin/conductin and b-catenin. Complex formation induces the rapid degradation of b-catenin. In colon carcinoma cells, loss of APC leads to the accumulation of b-catenin in the nucleus, where it binds to and activates the Tcf-4 transcription factor (reviewed in [1,2]). Here, we report the identification and genomic structure of APC homologues. Mammalian APC2, which closely resembles APC in overall domain structure, was functionally analyzed and shown to contain two SAMP domains, both of which are required for binding to conductin. Like APC, APC2 regulates the formation of active b-catenin–Tcf complexes, as demonstrated using transient transcriptional activation assays in APC–/– colon carcinoma cells. Human APC2 maps to chromosome 19p13.3. APC and APC2 may therefore have comparable functions in development and cancer. Addresses: *Department of Immunology, University Hospital, P.O. Box 85500, 3508 GA, Utrecht, The Netherlands. †Department of Biology, Coker Hall, CB#3280, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599–3280, USA. ‡Hubrecht Laboratory, Netherlands Institute for Developmental Biology, Uppsalalaan 8, 3584 CT, Utrecht, The Netherlands. Correspondence: H. Clevers E-mail: [email protected] Received: 27 October 1998 Revised: 18 November 1998 Accepted: 4 December 1998 Published: 18 January 1999 Current Biology 1999, 9:105–108 http://biomednet.com/elecref/0960982200900105 © Elsevier Science Ltd ISSN 0960-9822

Results and discussion The tumour-suppressor gene APC is conserved from man to fly [2,3]. A search of mammalian expressed sequence tag (EST) databases for sequences showing homology to the human APC sequence revealed a human foetal brain EST clone (H50183), encoding a protein fragment with significant similarity to Armadillo repeats 2–6 of all known APC proteins. We tentatively termed the corresponding gene APC2. We then searched a Drosophila EST database, which revealed the existence of two clones (LD18122 and LD24920) related to the known Drosophila APC gene (dAPC). The sequencing of these nearly full-length cDNA

clones confirmed the existence of a second APC gene in the fruit fly (see Supplementary material published with this article on the internet). The tissue-specific expression of APC2 was compared to that of APC by probing dot blots of normalized poly(A) mRNA (RNA Master BlotTM 7770–1, Clonetech). Like APC, the highest levels of expression of APC2 were found throughout the central nervous system (see Supplementary material). We subsequently screened human foetal kidney and brain cDNA libraries, a human Pac (P1-derived artificial chromosome library) and a mouse genomic P1 library. Compilation of sequences revealed an open reading frame of 2274 amino acids for mouse APC2 and 2302 amino acids for human APC2. The mouse APC2 gene spans 14 kb and consists of 14 exons (see Supplementary material). In Figure 1a, the domain structures of human APC and mouse APC2 are compared. The highest similarity occurred in the amino-terminal 728 amino acids. Carboxy-terminal to the Armadillo repeat region, the homology dropped steeply. Despite this, protein sequence motifs that are believed to allow APC to interact with β-catenin, the so-called ‘20 amino acid repeats’ [2], could readily be identified. Somewhat surprisingly, the ‘15 amino acid repeats’, which in APC have been proven to bind β-catenin [2], appeared to be absent in APC2. In APC, three SAMP domains are interspersed between the 20 amino acid repeats and mediate binding to axin/conductin [4–7], whereas only two SAMP domains were found in APC2. The region of APC containing the 20 amino acid repeats and SAMP domains has been shown to harbour the functional regions required for the control of β-catenin in colorectal cancer cells and in Xenopus axis induction [8,9]. Because APC2 is an obvious candidate disease gene, we determined the chromosomal localization of the human APC2 gene by fluorescence in situ hybridization (FISH) analysis of metaphase chromosomes from human leukocytes. The gene could be unambiguously assigned to chromosome 19p13.3 (Figure 2). The murine P1 genomic clone contained, next to the complete mouse APC2 gene, the proprotein convertase 4 gene, which has been mapped to chromosome 10 in mouse and chromosome 19 in man [10]. APC controls the Wnt signalling pathway through its ability to form a complex with GSK-3β, axin/conductin and β-catenin. In the currently held scenario, β-catenin is believed to be phosphorylated by GSK-3β when it participates in this complex [1,5–7]. This modification of

106

Current Biology, Vol 9 No 2

Figure 1 (a)

Conserved domain Armadillo repeats Oligomerizatation hAPC

15 amino acid repeats

57% identity

20 amino acid repeats Basic domain 2843 amino acids SAMP repeats

2274 amino acids

mAPC2 (b) 20 amino acid repeats

DLG-binding site

SAMP repeats

hAPC rpt1 EDTP I CFS RCSSLSS LSSAED mAPC2 rpt1 QDGP M SLS RCSSLSS LSSTGH hAPC2 rpt1 QEGP L SLS RCSSLSS LSSAGR

APC

SDDDDIE ILEECIISAMPTKSS RKAKKLA

APC

NKAEEGD ILAECINSAMPKGKS HKPFRVK

hAPC rpt2 QETP L MFS RCTSVSS LDSFES mAPC2 rpt2 QETP L VLS RCSSVSS LGSFES hAPC2 rpt2 QETP L VLS RCSSVSS LGSFES

APC

SIDSEDD LLQECISSAMPKKK

hAPC rpt3 ESTP D GFS CSSSLSA LSLEEP mAPC2 rpt3 EKPD E NFS CASSLSA LALHEL hAPC2 rpt3 EKPD E NFS CASSLSA LALHEH

A novel mammalian APC homologue. (a) Comparison of human APC with mouse APC2. Domains conserved between the APC relatives are indicated by boxes. DLG, Discs large protein. (b) The homologies between the 20 amino acid repeats and SAMP domains of human APC, human APC2 and mouse APC2 are shown. Genbank accession numbers for the APC2 genes are as follows: mouse APC2, AJ130783–AJ130796; human APC2 cDNA (corresponding to amino acids 1–705), AJ012652; and exon 14, AJ131187.

RPSRLK

mAPC 2 ATDKELE ALRECLGAAMPARLR KVASALV hAPC 2 AADQELE LLRECLGAAVPARLR KVASALV mAPC 2 SPRAEEE LLQRCISLAMPRRRY QVPGSRR hAPC 2 SPRAAEE LLQRCISSALPRRRP PVSGLRR

hAPC rpt4 EGTP I NFS TATSLSD LTIEEP mAPC2 rpt4 EGTP V NFS SAASLSD ETLQGP hAPC2 rpt4 EGTP V NFS SAASLSD ETLQGP hAPC

rpt5 EGTP Y CFS RNDSLSS LDFDDD

hAPC rpt6 ENTP V CFS HNSSLSS LSDIDQ mAPC2 rpt5 DETP P CYS LTSSASS LSEPEA hAPC2 rpt5 DETP P CYS LTSSASS LSEPEH hAPC

rpt7 EDTP V CFS RNSSLSS LSIDSE

β-catenin induces its ubiquitination and subsequent destruction by the proteasome [11]. Wnt signalling decreases the activity of GSK-3β. The consequent alteration of the phosphorylation state of β-catenin rescues it from destruction and allows it to travel to the nucleus, where it associates with T-cell factor (Tcf) transcription factors [12–17]. This association results in the transcriptional activation of Tcf target genes. In colon cancer and melanoma, mutations in APC or β-catenin induce the constitutive formation of nuclear β-catenin–Tcf complexes, resulting in constitutive transcription of target genes [18–20], such as c-myc [21]. The activity of Tcf-4 is essential for the maintenance of stem cells in the intestine [22]. In the yeast two-hybrid assay, the APC2 SAMP domains interacted with conductin. But it should be noted that both SAMP domains were necessary for the interaction Figure 2 Human APC2 maps to chromosome 19p13.3. FISH was performed on metaphase chromosomes derived from phytohaemagglutininstimulated human blood cells using standard procedures. The metaphase ‘spreads’ of human leukocytes were hybridized with a 6 kb APC2-specific probe. The APC2 signal appears in red. The inset shows an overlay of the APC2 signal with a green chromosomal staining of chromosome 19.

Current Biology

(Figure 3). This contrasted with the observation that a single APC SAMP domain suffices for the interaction with

Brief Communication

107

Figure 3 Conductin binds to the SAMP domains of APC2 in the yeast two-hybrid assay. The yeast strain HF7c was co-transformed with one plasmid encoding the indicated APC2 fragment fused to the VP16 activation domain (AD) and another encoding the RGS domain of conductin fused to the GAL4 DNA-binding domain (PGBT9-RGS) or the empty vector (PGBT9). The presence of the two plasmids allows growth on selective LT plates, and a positive interaction between the two fusion proteins allows growth on selective LTH plates which contain 20 mM 3-amino-1,2,4triazole (3-AT). As a positive control, APC fragments 1 and 2 were used with PGBT9RGS [4]. A positive interaction is indicated by +++ and no interaction by –––. LT, Leu–Trp–; LTH, Leu–Trp–His–.

VP16 AD

PGBT9

20 20 SAMP amino amino SAMP acids acids

PGBT9-RGS

LT +++

LTH –––

LT +++

LTH - –– –

mAPC2 1275–1634

+++

–––

+++

+++

mAPC2 1275–1585

+++

–––

+++

–––

mAPC2 1275–1371

+++

–––

+++

-–––

mAPC2 1351–1585

+++

–––

+++

–––

mAPC2 1351–1634

+++

–––

+++

–––

APC #1 1464–1604

+++

–––

+++

+++

APC #2 1516–1595

+++

–––

+++

+++

20 amino SAMP acids

Current Biology

conductin (Figure 3) [4]. This suggests differences either in the affinity of the interactions between APC2 or APC and conduction, or in the local protein structure of the SAMP domains. In APC2, the presence of conserved interaction domains for β-catenin and for axin/conductin predicted its involvement in the control of β-catenin-mediated Tcf target gene transcription. To test this idea directly, we applied a well-established functional assay: the transient Tcf transcription assay [12,13]. Transfections of reporter plasmids that contain either three optimal (TOP) or three mutant (FOP) Tcfbinding sites have demonstrated that Tcf factors activate transcription only when bound to β-catenin [12,13]. In colon carcinoma cells lacking APC or expressing a mutant

form of β-catenin, the deregulation of β-catenin results in the inappropriate transcription of Tcf reporter plasmids [18,19]. The transfection of wild-type APC restores this control in APC–/– cells, but not in cells expressing mutant βcatenin [18,19]. Using this assay, we tested whether APC2 could complement APC function in colon carcinoma cells. Indeed, transfection of either APC2 or APC in the APC–/– mutant SW480 (Figure 4a) or DLD-1 (data not shown) cells inhibited the inappropriate activation of the Tcf reporter plasmid. As expected, no effects of APC2 or APC were seen in the HCT116 cell line, which expresses mutant β-catenin (Figure 4b) or in several control cell lines (data not shown). Importantly, APC2 and APC did not downregulate the activity of the mutant reporter plasmid or the co-transfected Renilla luciferase control plasmid (pRL-TK; Figure 4). In

Figure 4 Control of β-catenin signalling by APC2 in colorectal cancer cells. (a) The APC–/– SW480 cells or (b) the HCT116 cells containing mutant β-catenin were transfected with the Tcf reporter plasmid pTOPTKFLASH (TOP; 200 ng) or the mutant reporter pFOPTKFLASH (FOP; 200 ng), in combination with the indicated amounts of APC2 or APC expression vectors, the corresponding amount of empty ‘stuffer’ expression vector (pCDNA3), and the internal transfection control Renilla luciferase plasmid (pRL-TK; Promega), essentially as described in [18]. For each sample, values are given as the ratio of luciferase activity to Renilla activity. Although pTOPTKFLASH is actively transcribed in both cell lines, inhibition of its transcription by APC2 is observed only in the APC–/– cells.

(a) SW480

(b) HCT116

pCDNA3

pCDNA3

40 ng APC

100 ng APC

200 ng APC

300 ng APC

1000 ng APC

1000 ng APC

pCDNA3

pCDNA3

40 ng APC2

100 ng APC2

200 ng APC2

300 ng APC2

1000 ng APC2

1000 ng APC2

0

10,000

TOP

20,000

30,000

40,000

FOP

0

10,000 20,000 30,000 40,000

TOP

FOP Current Biology

108

Current Biology, Vol 9 No 2

Figure 5 (a)

(b)

Current Biology

Downregulation of endogenous β-catenin by APC2 in SW480 cells. SW480 cells were (a) mock-transfected with an empty vector or (b) transiently transfected with APC2. Endogenous β-catenin was visualized 24 h later by immunofluorescence using antibodies from Transduction laboratories.

addition, the downregulation of endogenous β-catenin in SW480 cells transiently transfected with APC2 was visualized by immunofluorescence. Control cells showed speckled nuclear staining, whereas nuclear staining was absent in cells transfected with APC2 (Figure 5). Its similarities with APC would predict APC2 to be a tumour-suppressor protein. It has always been puzzling why the overwhelming majority of neoplasms in familial adenomatous polyposis patients occur in the intestine, while APC, β-catenin and Tcf factors are much more broadly expressed. It is possible that APC and APC2 perform a redundant tumour-suppressor function in many tissues. The identification of APC2 and dAPC2 will allow the creation of mouse and fly strains that are mutant for these genes. Such animal models will shed more light on the unique functions of these genes and on their interactions with other components of the Wnt pathway. While this manuscript was being reviewed, the sequence of the human APC2 gene was published by H. Nakagawa et al. [23]. Supplementary material Further details of dAPC and dAPC2 and the mouse APC2 gene, and the expression levels of human APC and APC2 in a variety of tissues are published with this article on the internet.

Acknowledgements We thank Amy Malterer and Dave Hayes of Genome Systems Inc. for their help, J. Behrens for the conductin two hybrid plasmids, Tom O’Toole for technical assistance and Nick Barker for critically reading this manuscript.

References 1. Cadigan KM, Nusse R: Wnt signalling: a common theme in animal development. Genes Dev 1997, 11:3286-3305. 2. Polakis P: The adenomatous polyposis coli (APC) tumor suppressor. Biochim Biophys Acta 1997, 1332:F127-F147. 3. Hayashi S, Rubinfeld B, Souza B, Polakis P, Wieschaus E, Levine AJ: A Drosophila homolog of the tumor suppressor gene adenomatous polyposis coli down-regulates b-catenin but its zygotic expression is not essential for the regulation of Armadillo. Proc Natl Acad Sci USA 1997, 94:242-247.

4. Behrens J, Jerchon BA, Wurtele M, Grimm J, Asbrand C, Wirtz R, et al.: Functional interaction of an axin homolog, conductin, with beta-catenin, APC and GSK3beta. Science 1998, 280:596-599. 5. Sakanaka C, Weiss JB, Williams LT: Bridging of b-catenin and glycogen synthase kinase-3b by Axin and inhibition of b-catenin mediated transcription. Proc Natl Acad Sci USA 1998, 95:3020-3023. 6. Ikeda S, Kishida S, Yamamoto H, Murai H, Koyama S: Axin, a negative regulator of the Wnt signaling pathway, forms a complex with GSK-3b and b-catenin and promotes GSK-3b-dependent phosphorylation of b-catenin. EMBO J 1998, 17:1371-1384. 7. Hart MJ, de los Santos R, Albert IN, Rubinfeld B, Polakis P: Downregulation of b-catenin by human Axin and its association with the APC tumor suppressor, b-catenin and GSK3b. Curr Biol 1998, 8:573-581. 8. Munemitsa S, Albert I, Souza B, Rubinfeld B, Polakis P: Regulation of intracellular b-catenin levels by the adenomatous polyposis coli (APC) tumor-suppressor protein. Proc Natl Acad Sci USA 1995, 92:3046-3050. 9. Vleminckx K, Wong E, Guger K, Rubinfeld B, Polakis P, Gumbiner BM: Adenomatous polyposis coli tumor suppressor protein has signaling activity in Xenopus laevis embryos resulting in the induction of an ectopic dorsoanterior axis. J Cell Biol 1997, 136:411-420. 10. Mbikay M, Seidah NG, Chretien M, Simpson EM: Chromosomal assignment of the genes for proprotein convertases PC4, PC5 and PACE 4 in mouse and human. Genomics 1995, 26:123-129. 11. Aberle H, Bauer A, Stappert J, Kispert A, Kemler R: b-Catenin is a target for the ubiquitin-proteasome pathway. EMBO J 1997, 16:3797-3804. 12. Molenaar M, van de Wetering M, Oosterwegel M, Peterson-Maduro J, Godsave S, Korinek V, et al.: XTcf-3 transcription factor mediates bcatenin induced axis formation in Xenopus embryos. Cell 1996, 86:396-401. 13. van de Wetering M, Cavello R, Dooijes D, van Beest M, van Es J, Loureiro J, et al.: Armadillo coactivates transcription driven by the product of the Drosophila segment polarity gene dTCF. Cell 1997, 88:789-799. 14. Behrens J, von Kries JP, Kuhl M, Bruhn L, Wedlich D, Grosschedl R, Birchmeier W: Functional interaction of b-catenin with the transcription factor LEF-1. Nature 1996, 382:638-642. 15. Huber O, Korn R, McLaughlin J, Ohsugi M, Herrmann BG, Kemler R: Nuclear localization of b-catenin by interaction with transcription factor LEF-1. Mech Dev 1996, 59:3-10. 16. Brunner E, Peter O, Schweizer L, Basler K: Pangolin encodes a LEF-1 homologue that acts downstream of armadillo to transduce the Wingless signal in Drosophila. Nature 1997, 385:829-833. 17. Riese J, Yu X, Munnerlyn A, Eresh S, Hsu SC, Grosschedl R, Bienz M: LEF-1 a nuclear factor coordinating signaling inputs from wingless and decapentaplegic. Cell 1997, 88:777-787. 18. Korinek V, Barker N, Morin PJ, van Wichen D, de Weger R, Kinzler KW, et al.: Constitutive transcriptional activation by a bcatenin-Tcf complex in APC–/– colon carcinoma. Science 1997, 275:1784-1787. 19. Morin PJ, Sparks AB, Korinek V, Barker N, Clevers H, Vogelstein B, Kinzler KW: Activation of b-catenin-Tcf signaling in colon cancer by mutations in b-catenin or APC. Science 1997, 275:1787-1790. 20. Rubinfeld B, Robbins P, El-Gamil M, Albert I, Porfiri E, Polakis P: Stabilization of b-catenin by genetic defects in melanoma cell lines. Science 1997, 275:1790-1792. 21. He TC, Sparks AB, Rago C, Herweking H, Zawel L, da Costa LT, et al.: Identification of c-Myc as a target of the APC pathway. Science 1998, 281:1509-1512. 22. Korinek V, Barker N, Moerer P, van Donselaar E, Huls G, Peters PJ, Clevers H: Depletion of epithelial stem-cell compartments in the small intestine of mice lacking Tcf-4. Nat Genet 1998, 19:379-383. 23. Nakagawa H, Murata Y, Koyama K, Fujiyama A, Miyoshi Y, Monden M, et al.: Identification of a brain-specific APC homologue, APCL, and its interaction with beta-catenin. Cancer Res 1998, 15:5176-5181.

S1

Supplementary material Identification of APC2, a homologue of the adenomatous polyposis coli tumour suppressor J.H. van Es, C. Kirkpatrick, M. van de Wetering, M. Molenaar, A. Miles, J. Kuipers, O. Destrée, M. Peifer and H. Clevers Current Biology 18 January 1999, 9:105–108 Figure S1 A novel Drosophila APC homologue. (a) Comparison of dAPC with dAPC2. Domains conserved between the APC homologues are indicated by boxes. (b) The homologies between the 15 amino acid and 20 amino acid repeats of dAPC and dAPC2 are shown. The dAPC2 accession number is AF091430.

(a)

Conserved domain Armadillo repeats

15 amino 20 amino acid acid repeats repeats

Basic domain 2416 amino acids

dAPC 73% identity

>1137 amino acids

dAPC2 (b)

20 amino acid repeats

15 amino acid repeats

dAPC rpt1 YCEEGTPGYFSRYDSLSSLD dAPC2 rpt1 YCEEGTPGSFSRFDSLNSLT

dAPC rpt1 TEEQPIDYSVKYSEN dAPC2 rpt1 TEEQPIDYSMKYMEH

dAPC rpt2 NSALETPLMFSRRSSMDSLV dAPC2 rpt2 DSALETPLMFSRRSSMDSLV

dAPC rpt2 DLDQPTDFSLRYAEN dAPC2 rpt2 DLDQPTDFSARYKER

dAPC rpt3 FNVEHTPAQFSTATSLSNLS dAPC2 rpt3 FHVEHTPAAFSCATSLSNLS

dAPC

rpt3 TEDTPYVISNAASVT

dAPC rpt4 YCTEDTPALLSKVPSNTNLS dAPC2 rpt4 YCTEDTTAVLSKAPSNSDLS dAPC rpt5 FLVEDSPCNFSVVSGLSNLT dAPC2 rpt5 YYVEDSPCTFSVISGLSHLT

Current Biology

Figure S2

APC 50 40 30 20 10

APC2 0 0 10 20 30 40 50 Bladder Whole brain Spinal cord Heart Foetal kidney Bone marrow Lung Foetal brain Stomach Small intestine

128

Current Biology

Expression of human APC2. Dot blots of 50 normalized poly(A) mRNA samples isolated from a large variety of tissues (and several controls) were hybridized with probes corresponding to the 5′ end of the APC2 and APC genes, according to the manufacturer’s instructions (RNA Master Blot 7770-1, Clonetech). Although all tissue RNA samples yielded positive signals with both probes when compared to control RNA samples supplied on the blot, high levels of expression for both genes were observed throughout the central nervous system. Relative differences between APC and APC2 expression in a selected set of tissues are shown. The signals on individual dots were quantified by phosphorimaging; the signals obtained for the bladder mRNA samples were arbitrarily set at 1 for each of the individual probes and the background obtained with human Cot1 DNA was set at 0.

S2

Supplementary material

Table S1 Genomic organisation of mouse APC2. Intron

Exon

Exon number

Exon length

Exon

Intron

1

141

g aag

GTGAGAGCCTGCATGGAG

AGACCATGCCCGCAG

gag g

2

91

aaa g

GTGAGTGACTTCCAGAAC

CCTCCCTCCCTCCAG

ct ct

3

181

ag ag

GTGAGAGGGCGTGGGAAA

CATCCCCTCCACCAG

a tgc

4

109

c acg

GTGAGCTCCCATCGTACC

GCCCCCTCCCAGCAG

ttt t

5

114

g cag

GTGCGGGCAGTGCACACC

TTCTCTGTTCTCTAG

atc c

6

78

t cag

GTACCAAGACAAGCAGGG

ATGTTCTTCCTGCAG

gct c

7

99

c aag

GTAAGAAGGGAAAATCTG

CGTGTTCCCCTGCAG

gtg g

8

373

gaa a

GTGGTACAGGAGACAGTG

CACGCCTGCCCACAG

ct cc

9

96

cta g

GTGGGGTGTCCCCATCT

TTCCCACACCTACAG

gg gg

10

140

c aag

GTTCCTAGGGTGGGAAGA

TGTCAATCCCAACAG

gcc a

11

78

t cag

GTACACAGGGCAGGGGAG

TCCCACCCTGGTCAG

gtt g

12

117

c aag

GTGAGCATGGGCAGCCTG

TCCCTCCCACCCCAG

gag t

13

215

ac ag

GTCAGCTCCCATCACGAT

TCTGGGTTTCCCCAG

g cag

14

4990

Genomic organization of the mouse APC2 gene. A genomic fragment of mouse APC2 was obtained by PCR using primers based on human sequence in the Armadillo repeat region. A mouse genomic P1 library was subsequently screened by Genome Systems using primers based on the cloned mouse fragment. The P1 clone obtained was used to clone and sequence mouse APC2 using standard procedures.