Oene, 96 (1990) 213-218 Elsevier
213
GENE03~6
Cloning and characterization of c D N A s encoding a novel cyclic AMP-binding protein in D i c t y o s t e l i u m discoideum (Recombinant DNA; cellular slime moulds; antibody; gene fusion; proline-rich; tellurite-resistance determinants; expression in Escherichia coli)
Caroline E. Grant* and Adrian Tsang Department of Biology, McGill University,Montreal, Quebec HJA IBI (Canada) Receivedby D.T. Denhardt: 16 May 1990
Revised: 10 July 1990 Accepted: 11 July 1990
SUMMARY
The cellular slime mould, Dictyostelium discoideum, contains a novel cyclic AMP-binding protein, CABP1, wh/ch is composed of two subunits. Using anti-CABP1 monoclonal antibody as a probe, a cDNA clone was isolated from a Agtll expression library. By hybrid selection of the complementary mRNA and its translation in vitro, we demonstrated that the cDNA hybridized to mRNAs encoding both CABPI polypeptides. With the positive cDNA as a probe, we isolated a series ofoverlapping cDNA clones covering the coding region of both CABP 1 mRNAs. Expression ofthe cloned cDNAs in bacteria and sequence analysis showed that the CABPI subunits are identical in amino acid (aa) sequence, except that the small subunit is missing 37 aa near its N terminus. Genomic analysis suggested that the two CABP1 transcripts are derived from a single gene. The N-terminal half of each subunit is rich in proline, glutamine and giycine residues and contains a large block of aa repeats. The C-terminal half has an approx. 47 % aa id,mtity (86% with fuactionally conservative substitutions) with two polypeptides encoded by a plasmid determinant for tellurium anion resigtance.
INTRODUCTION
Dictyostelium discoideum grows vegetatively as individual amoebae. When deprived of their bacterial food source, the Correspondence to: Dr. A. Tsang, Department of Biology,McGill University, 1205 Ave. Docteur Penfield, Montreal, Que. H3A IBl (Canada) Tel. (514)398-6435; Fax (514)398-5069. * Present address: Cancer Research Labs, Botterell Hall, Queen's University, Kingston, Ont. K7L 3N6 (Canada) Tel. (613)545~2979.
Abbreviations: aa, amino acid(s); bp, base pair(s); t.ABPI, cAMPbinding protein 1; CABPI, gene (DNA) encoding CABP1; cAMP, cyclic AMP; cDNA, DNA complementary to RNA; D, Dictyosteiium; Denbardt's, 0.2% each of bovine serum albumin, Ficoll and polvvinylpyrrolidone; IPTG, isopropyl-p-D-thiogalactopyranoside;kb, kilobase(s) or 1000 bp, c~Ab, monoclonal antibody; nt, nucleotide(s); oligo, oligodeoxyribonucieotide; ORF, open reading frame; PAGE, polyacrylamidegel electrophoresis; g, resistance/resistant; SDS, sodium dodecyl sulfate; SSC, 0.15 M NaCI/0.015 M Naj" citrate pH 7.6; Te, tellurium anion. 0378-1119/90/$03.50 © 1990ElsevierSciencePublishersB.V.(BiomedicalDivision)
cells aggregate and eventually develop into a fruiting body composed of two basic cell types, stalk and spore cells (reviewed in Loomis, 1982). During the developmental process, cAMP acts as the chemotactic signal for aggregation (reviewed in Gerisch, 1987), induces cellular differentiation (Kay, 1982)~and modulates the expression of many genes (Williams et al., 1980; Chung et al., 1981; Barklis and Lodish~ 1983). We have previously identified a novel cAMP-binding protein in D. discoideum cells, termed CABPI (Tsang and Tasaka, 1986). This plotein can be detected in the nucleus of developing cells suggesting that CABP1 may have a role in the cAMP-mediated regulation ofgene expression (Kay et al., 1987; Tsang et al., 1988). In D. discoideum strain NC4, CABPI is composed of two subunits, 43-kDa CABP1A and 38-kDa CABP1B (Tsmtg and Tasaka, 1986; Tsang et al., 1987). The aim of the present study is to identify and characterize cDNAs which encode the CABP1 subunits.
214 1
2
3
5
4
6
7
8
...... t 7
-1A
~
'[
-1B -34 m
-31
Fig. 1. Isolation and identification of CABPI cDNA. A mixture of poly(A) ÷ RNA~ prepared from NC4 cells that had developed for 12 h and 20 h was used to construct the eDNA library in the expression vector 2gtl I (Maniatis et al., 1982; Huynh et al., 1985). The phage were used to infect E. coliY 1090 and the resulting plaques were transferred in duplicate onto nitrocellulose filters (BA85, Schleicher & Schuell) and probed with three anti-CABPl mAbs (Tsang and Tasaka, 1986) as described by Huynh et al. ~1985). A 0.67-kb eDNA from a positive phage was subcloned into the Eco RI site of the plasmid pAT 153. The identity of the 0.67-kb eDNA was examined by hybrid selection (Maniatis et al., 1982) as follows. Poly(A)+RNA was hybridized to filter-bound pATI53 carrying the 0.67-kb eDNA or pBR322. After hybridization for 3 h at 45 °C, the filters were washed extensively to remove nonspecific RNAs. The bound RNAs were eluted by boiling and concentrated by ethanol precipitation. The eluted RNA was then used to direct translation in vitro in a rabbit reticulocyte lysate system (Bethesda Research Laboratories). Each 30-#1 translation reaction contained the eluted RNA from a single filter and 80 #Ci [32S]methionine (Amersilam, 1400 Ci/mmol). Proteins from 2 #1 of the 30-#1 translation reactions were immediately solubilized in SDS sample buffer (lanes I-4) while the remaining sample was immunoprecipitated with anti-CABPl mAb (Tsang and Tasaka, 1986) before solubilization (lanes 5-8). The translation products were resolved in 0.1% SDS-12% PAGE and visualized by fluorography (Skinner and Griswold, 1983). Lanes: 1 and 5, 1 #g of total poly(A)+ RNA; 2 and 6, no exogenous RNA added; 3 and 7, RNA selected with pBR322; 4 and 8, RNA selected with pATI53 carrying the 0.67-kb eDNA. The two subunits of CABPI are marked as 1A and IB and the 34- and 31-kDa polypeptides which coprecipitated with CABP1 (Tsang and Tasaka, 1986) as 34 and 31, respectively.
CABP1 mAbs as probes. The identity of the eDNA was examined by hybrid selection. Fig. 1 shows that the mRNA species selected with the 0.67-kb eDNA were translated into two major polypeptides of 43 and 38 kDa, the sizes identicaI to those ~f the two CABP1 subunits (lanes 4 and 5). Both of these polypeptides reacted positively with the anti-CABP1 mAb (Fig. 1, lane 8). These data suggest that the 0.67-kb eDNA was complementary, to mRNA populations which encode the two CABP1 subunits. We have isolated three additional eDNA clones using the 0.67-kb eDNA as a probe. Thent sequence ofthese cDNAs were determined and compared (Fig. 2). From the nt sequence we deduced two composite cDNAs that are identical, except that the larger cDNA has an additional 111 nt in the coding region. The origins of the two cDNA populations were examined by RNA blotting. As shown in Fig. 3, the probe sequences that are present in both composite CABPI cDNAs hybridized with two mRNA species of 1.25 and 1.15 kb. The probe sequence that is specific for tile large composite cDNA, however, hybridized only with the 1.25-kb mRNA. Thus, the two classes ofcDNA appear to have derived from different mRNA populations and are not the result of cloning artifacts.
RESULTS AND DISCUSSION
(b) The cloned cDNAs represent the entire coding sequence of the CABPI mRNAs The sizes of CABP 1 subunits in the D. discoideum strain NC4 are 43 and 38 kDa in SDS-PAGE (Tsang et al., 1987). The large composite eDNA, however, encodes a 35-kDa polypeptide of 333 aa, and the small composite eDNA encodes a 31-kDa polypeptide of 296 aa. To determine the apparent M r of the polypeptides encoded by the composite cDNAs, we cloned the cDNAs into appropriate pIMS expression vectors and synthesized the gene products in Escherichia coil cells (Simon et al., 1988). Fig. 4A shows that the construct carrying the large eDNA produces a polypeptide which is 4 aa larger than its ORF. The plasmid carrying the small eDNA encodes the same number of aa as its ORF. The mobility ofthe fusion proteins and the endogenous CABP1 subunits were compared on SDS-PAGE (Fig. 4B). In each case, the plasmid construct produced an immunoreactive polypeptide with a mobility equivalent to that of the CABP1 polypeptide produced in the D. discoideum cells. These results confirmed that the cloned cDNAs represent the entire coding regions of the CABPI mRNAs, and that the large eDNA encodes CABP1A and the small eDNA encodes CABPIB.
(a) Identification of cDNAs complementary to CABPI mRNAs A cDNA clone containing a 0.67-kb insert was isolated from a 2gtl 1 expression library using a mixture of anti-
(c) Genomic organization of the CABPl-encoding sequences To determine if the two CABPI sequences are derived from a single gene, we probed a blot of genomic DNA with
215 CABPIB
CABPIA Met Tyr Asn Pro Pro Pro Pro Ser G!y Ser G l n ~ l y Asn Ash AsnJl5 A T A T ~ T ~ ATG TAT ~ C CCA CCA CCA CCA TCT GGT TCA C ~ CIGT ~ T ~ T A A T i ~
i
(Ii)
i
I
61
jTyr Tyr Arg Cln Pro Ser Ser ~ r Pro Gly Val Ser Asn Pro Asn Pro Gln Ala Ash 34 ITAT TAT AGA C ~ CCA TCA TCC ACA CCG GGT GTA TCA ~ C CCA ~AC ACCAT C ~I GCT ,
118
JGln Phe Leu Pro Pro Gin Pro Ser Asn Thr Thr Gin Thr Pro Gly Gly Tyr Pro Pro 53 C ~ TTT TTA CCA CCA CAA CCA TCT AAT ACT ACA CAA ACA CCA GGA GGC TAT CCA CCA
(35)
175
Gin Gin Gin Gin Arg Pro Pro Thr Gly Ala Pro Gin Gin Pro Gly Gly Tyr Pro Thr 72 CAA CAA CAA C A A A C A CCA CCA ACT CCA GCA CCA CAb. CAA CCA CCA GCT TAT CCA ACT Pro Pro Pro Pro Gly Ala Pro Cly Gly Tyr Pro Pro Gln Gln Gln Pro Ala Gly Gln 91 CCA CCA CCA CCA GGA GCA CCA GGA GGT TAT CCA CCA CAA CAA CAA CCA GCC GGT CAA ...................... I .......
(54)
232
289
j
(16)
!
Tyr Gly Ala Pro Pro Gln Gln Gln Pro Ala Gly Gln Tyr Cly Ala Pro Gln Pro Ala II0 (73) TAT CGA GCA CCA CCA C ~ C ~ C ~ CCA GCC GGT C ~ TAC GGA GCA CCA C ~ CCA GCA ..............
II
......................
2 ......................
i
l ..........
Gly Gln Tyr Gly Ala Pro Gln Pro Ala Gly Gln Tyr Cly Ala Pro Gln Pro Ala Gly 129 (92) 346
403
GGA CAA TAT G(;T GCA CCA CAA CCA GCA GGA CAA TAT GGT GCA CCA CAA CCA GCA GGA ..... 3 ................ II ................ 4 ................ I I. . . . . . . . . . . . . . Gln Tyr Gly Ala Pro Pro Pro Pro Pro Gly Gly Ala Gly Ile Ser Leu Val Lys Ash 148 (Iii) CAA TAT GGC GCA CCA CCA CCA CCA CCA GGA GGT GCA GGT ATT TCA TTA GTA AAG AAT
,s
. . . . . . . . . . . . . . . . .
I
A
517
Gln Gln Ile Ser Leu Thr Lys Glu Asp Pro CAA C A A A T T TCA TTA ACT AAA G A A G A T CCA Xho II Gly Trp Asp Val Asn Thr Thr Pro Thr Ala GGT TGG GAT GTA AAC ACC ACA CCA ACT GCA
574
Met Leu Asn Ala GIn Gly Arg Val Arg Thr Ser GI~ Asp Vhe Ile Phe Tyr Asn Ash 205 (168) ATG TTG AAT GCA CAA GGT AGA GTT AGA ACC TCA CAA GAT TTC ATT TTC TAC AAT AAC
631
Lys Val Ser Arg Asp Asn Ser Val Ser His Gin Gly Asp Ash Leu Thr Gly Gln Gly 224 (187) AAG GTA TCA AGA GAT AAC TCT GTT TCT CAT CAA GGT CAT AAT TTA ACA GGT CAA GGT
688
Glu Gly Asp Asp Glu Val Vai Leu Val Asn Leu Gin Ala Val Ser Pro Asp Val Thr 243 (206) GAA GGT GAT GAT GAA GTT GTC CTC GTA AAC TTA CAA GCA GTT TCA CCA GAC GTC ACT
745
Arg Leu Val Phe Ala Val Thr Ile His Leu Ala Asp Glu Arg Arg Gln Asn Phe Thr 262 (225) CGT TTG GTT TTC GCT GTC ACC ATT CAT TTA CCC CAT GAA ACA ACA CAA AAC TTT ACA
802
Met Val Pro Arg Ala Phe Ile Arg Val Ala Ann ~ln Glu Thr Gly Arg Asn Ile Cys 281 (244) ATG GTA CCA ACA GCT TTC ATT CGT GTT GCC AAT C~tA GAA ACT GGT ACA AAT ATC TGT
859
Arg Tyr Asp Leu Ser Gln Glu Gly Gly Pro Asn Thr Ala Leu Ile Ala Cly Glu Val 300 (263) CGT TAC CAT CTC TCT CAA GAA GGT GGT CCA AAC ACT GCC CTC ATT CCT CGT CAA GTT
460
Thr Leu Arg Lys Leu Thr Ile Gly Leu 167 (130) ACT CTT A G A A A A TTA ACA AAT GGT TTA Pro Phe Asp Leu Asp Ala Val Val Phe 186 (149) CCA TTC GAT TTG CAT GCA GTT GTT TTC
Tyr Arg Asp Pro Set Asn Pro Ash Ash Trp Set Phe Val Ala Val Gly Lys Gly Met 319 (282)
916
TAT CGT CAT CCA TCA AAT CCA AAC AAT TGG TCT TTT GTT GCT GTT CGT AAA GGT ATG
973
Gin Gly Ala Leu Pro Gly ~ u Leu Gln lle Phe CLy Cys Cln C ~ GGC GCT CTT CCA GGT TTA CTC C ~ ATC TTT ~GT TGT CA& T ~ T T T ~ T A T T A T T A T E T
1035
~
T
T
L
~
333 (29b)
TT~T~C~C~T~T~CTATTTGT~TcTTAC~T'CABPIA
(A)I5 CABPIB (A)30
Fig. 2. Nucleotide sequence of CABPIA and CABPIB, and predicted aa sequen,3e. Additional cDNA clones were isolated from a second 2gtl I library (Klein et al., 1988) using the 0.67-kb cDNA as a probe. The 0.67-kb cDNA wa~ nick-translated (Bethesda Research Laboratories kit) and hybridized to the cDNA library at 42°C in a solution containing 50% formamide/5 × SSC/5 x Denhardt's/0.5% SDS/100 #g per ml salmon sperm DNA. Washing was done at 65°C in 2 × SSC/0.1 ~ SDS. 28 positive plaques were obtained from t['e approx. 60000 phage plaques screened. Three positives with inserts of approx. I kb were subcloned into the plasmid vectors pBluescript + or - (Stratagene). Sequencing was done using the chain-termination method of Sanger et al. (1977) as modified by Stambaugh and Blakesley (1988). Staggered deletions were obtained for the MI3 clones (the 0.67-kb cDNA was subcloned into MI3 phage for sequence analysis) using T4 DNA polymerase (Dale et al., 1985) and for the plasmid clones using DNAse I (Laughon and Scott, 1984). Approximately 70% to 80% of the sequence of each cDNA was determined on both strands. The nt sequences of the CABPIA and CABPIB composite cDNAs are identical and, therefore, a single nt sequence is shown. Downward arrows at the beginning of the sequence denote the 5' ends, and the downward and horizontal arrows at the end of the sequence show the 3' ends of the composite cDNAs. The nt are numbered on the left of the figure. The aa are numbered on the right of the figure with those referring to the CABP1B polypeptide m t~rackets. The 111 nt (37 aa) not present in CABPIB are boxed. The position of aa 143 of CAB PiA ( A ) and the XhoII cleavage site are indicated below the nt sequence. A 5-aa repeat is double-underlined and the 12- and 9-aa repeats are underlined with dashed lines and numbered. The nt sequence reported here has been submitted to GenBank Data Library with acession number M36176.
216
1
2
3
A
1350 -1250
5' END OF CABPIA AND IB
-
Met Tyr Asn PrJ Pro Pro Pro S e r Gly ATG TAT AAC CCA CCA CCA CCA TCT G G T
LARGE SUBUNIT (1A) CONSTRUCT- i . 1
- 960
Fig. 3. Northern-blot ar~alysisofthe CABP1 transcript,,. Poly(A)÷RNA (2.5 #g/lane) from NC4 cells was fractionated on a 1.5% agarose-2.2 M formaldehyde gel (Maniatis et al., 1982) and transfr~rred to Nytran membrane (Schleicher & Schueii). Lanes: 1, the RN,~, was probed with an end-labelled oligo (probe B in Fig. 5A; corresponding to nt 197-173 of Fig. 2, 5'-TGTTGTGGTGCTCCAGTTGGTGG' fC-3') which is common to both CABPI cDNAs. The blot was hybridized m~d washed in 2 x SSC at 60~C; 2, the RNA was probed with an et~d-labelledoligo (probe A in Fig. 5A; corresponding to nt 146-122 of Fig. 2, 5'-CCTGGTGTTTGTAGTATTAGATG-3') whic~ is specific for the larger cDNA. The blot was hybridized and washed in 2 x SSC at 50°C; 3, the RNA was hybridized with probe C (see Fig. 5A). Probe C is a cDNA restriction fragment of approx. 600 nt that contains the common 3' end of the CABPI cDNAs (from the XholI site to the 3' EcoRl linker). Hybridization and washing conditions for probe C are the same as those described in the legend to Fig. 2 except that the blots were washed at 68°C.
... ...
k b cDNA i n pIMS5
Met G l y G l u P h e A r g T . y K A s n P r o P r o ~ r o P r o S e r G I ~ . . . a t g g g g aAA TTC CGG TAT AAC CCA CCA CCA CCA TCT GGT . . . SMALL SUBUNIT ( 1 B ) CONSTRUCT- 1 . 0 k b cDNA i ~ pIMS1 M e t ASh S e t A l a P r o Pr~o ~ _ q S e t G_~ . . . a t g AAT TCC GCA CCA CCA CCA TCT GGT . . .
1
4
2
3
4
~ p*
67-
43the same c D N A and oligo sequences used in the analysis of the CABP1 transcripts shown in Fig. 3 (see Fig. 5A for a map). Fig. 5B shows that all probes reacted with the same genomic D N A bands except for the HinfI-digested D N A (there is a HinfI cleavage site between the 5'- and 3'-specific probes), and the KpnI-digested D N A (the 3'-specific c D N A probe contains a KpnI cleavage site). These data suggest that there is probably only one copy of the C A B P 1 gene per haploid genome. We now have evidence that the two CABP1 trans,=ripts are the products o f alternative splicing ( G r a n t et al., submitted). The d a t a may also explain why in some D. discoideum genetic variants both C A B P I subunits have altered Mrs (Tsang et al., 1987).
(d) Predicted aa sequences of the C A B P I polypeptides Based on their aa composition, the C A B P 1 polypeptides can be divided into two distinct regions. The C-terminal 191 aa have no unusual feature. This region contains a roughly equal number of acidic and basic aa. On the other hand, the N terminus of the C A B P 1 poiypeptides has several distinctive characteristics. The aa composition here is very limited, and the entire region is hydrophilic. Pro, Glu, and Gly make up 29~o, 18 ~o, and 17 ~o, respectively. M o s t of the remaining aa are either Asp or residues containing an hydroxyl group, especially Tyr. Charged or strongly hydrophobic aa constitute only 2~o. In addition, there are five repeated sequences of 9 aa, Q P A G Q Y G A P . Each of the repeats is symmetrical
w
4
30W
w
20Fig. 4. Expression ol the CABPI gene in bacteria. (Panel A) Strategy for cloning the CABPI cDNAs in-fro,me with the ATG codon of the appropriate plMS vector (Simon et ai., 1988). The nt .;equences underlined are the portion of the EcoRl linker at the ~' ~.d of the cDNA that becomes part of the coding sequen:~ h~ the construct. Underlined aa show where the deduced aa sequenc~ of the construct and that of the CABPI composite eDNA are the same. Each of the eDNA sequences cloned into the expression vectors has several in-frame TAA stop codons at the c~,~ of the coding sequence. (Panel B) Expression of the !acZ -cD~'~A fusion gene was induced by IPTG and the bacterial cells lysed by sonicati~n. The cell lysate was cleared by centrifugation and the proteinx im~unoprecipitated with the anti-CABPl mAb (Tsang aad Tasaka, 1986). Following resolution in 0.1~ SDS-12% PAGE, the proteins were stained with Coomassie blue. Lanes: I, extract from bacteria containing the plMSl vector only; 2, extract from bacteria containing the small eDNA construct', 3, extract from l0 s NC4 cells that had developed for 12 h; 4, extract from bacteria containing the large eDNA construct. The position of the CABP1A polypeptide is marked by the blackened triangle and the 1B polypeptide by the open triangle. The intensely stained bands (50 and 25 kDa) are the heavy and ~ight chains of the precipitated immunoglobulins. Protein size markers on left margin are in kDa.
217
A
CABP1 TerORF4 TerORk5
143 190 ..ISLVKNQQ ISLTKEDPTLRKLTIGLGWDVNTTPTAPFD LDAWFMLNA MSV--S-GGN V--S-TA-SM ENVLV ..... ARS-DGQD ..... SA-L-AMAV .... GGN V ..... A-SMNVALV ..... TRV-DGQA ..... S--LVGE
CABPI TerORF4 TerORF5
191 QGRVRTSQDF IFYNNKVSRD N-K--GDA ....... LK-AN-K-LSDSH- V .... TT-P-
CABPI TerORF4 TerORF5
DVTRLVFAVT IHLADERRQN --DKII-V .... D-QA---S --El{ . . . . . . . . E - - S - K - -
FTMVPRAFIR VANQETGRNI -GQ-SG .... LV-DDNQTEV -G--SNS-M- -V-NDN-SE-
290 CRYDLSQEGG A .... TEDAS A-F---EDAS
CABPI TerORF4 TerORF5
291 PNTAL~AGEV TE--MLF--L TE--M-F--L
SFVAVGKGHQ K-R---L-¥A K-K---Q-FA
GCQ -INAS -INI
Kpl iI
Hinfl X ~oll CABPIA
240 NSVSHQGDNL TGQGEGDDEV VLVNLQAVSP G--T-T---R --E-D .... S LKIK-D--PG GA-Q ..... R --E-D .... Q -KID-T~-AA
241
CABPIB
PROBES A
G
B
E X~EH
PROBE C K
;= X Y v E H
GALPGLLQIF -G-ASVCAQ¥ -G-AA-ATQH
Fig. 6. Comparison of the C terminus of the CABPI polypeptides and the plasmid encoded polypeptides for Tea. The aa 142-333 of the CABPIA polypeptides (CABPI) are compared to the aa sequence of ORFs 4 and 5 (TerORF4 and TerORFS) of the Te" determinant from the bacterial plaamid pMER610 (Jobling and Richie, 1988). The dots indicate the gaps placed hLthe aa sequence~ to optimize the alignment. The dashes in the sequence ofORFs 4 and 5 signify aa that are identical to those in CABP1.
100 bp
PROBE A
333 YRDPSNPNNW --HNGA...--HGAE...-
K
Fig. 5. Oenomic analysis by Southern blotting with the CABPl-specific probes. CA)Cartoon showing the homologous regions ar,d restriction sites in the CABPI composite cDNAs. The blackened boxes represent the 5'-noncoding region and the first 34 nt of the coding region. The crosshatched boxes represent 852 nt of coding as well as the Y-noncoding regions. The open box represents the 111 nt found only in the CABPIA cDNAs. The location of the sequences hybridizing to the probes are indicated by the vertical lines. Detailed description of the probes and the hybridization conditions are outlined in the legend to Fig. 3. Note that the scale is only approximate. (Panel B) A single Southern blot of genomic DNA was probed sequentially with each of the :hree probes. Probes A and B gave identical res~dts and consequently only the results of probe A are shown. Before each hybridization reaction, the previous probe was removed from the blot and its removal monitored by autoradiography. Each lane contains 2.5/~g ofAX2 genomic DNA digested with EcoRl (E), Xbal iX), EcoRI + Xbal (E,X), Hinfl (H), and KpnI (K), and resolved in a 1% agarose gel. Markers are at 21.2, 5.15, 4.97, 4.28, 3.53, 2.03, 1.90, 1.58, 1.33, 0.98, 0.83, and 0.56 kb.
around the central Glu (Q) and Tyr (Y) residues. Further, the first two copies of these repeats are part of a 12-at, repeat, PQQQPAGQYGAP. In general, the N-terminal regions contain a high concentration of residues capable of forming hydrogen bonds which may imply that they are involved in interactions with other cellular constituents. (e) H~,mology of CABPI to other proteins A search for other aa sequences with si~milarities to the CABP 1 polypeptides was made using the Genbank, EMBL and NBRF data bases and the program TFASTA (Pearson and Lipman, 1988). When the 142 aa of the N-terminal of CABP1A were used as the query sequence, only proteins with high Pro content were extracted from the databases. The similarity between CABP1 and most of these proteins was restricted to Pro residues. Of interest, however, is the similarity between the 9-aa repeat of CABP1, QPAGQYGAP, and the aa repeat of the circumsporozoite protein from Plasmodium cynomolgi, PAGDGAPAA (Galinski et al., 1987). When the C-terminal 191 aa were used as a query sequence in the database search, we detected 45 % and 49 % aa identities (84 % and 86 % including conservative substitutions) with two polypeptides encoded by a plasmid determinant for resistance to tellurium anions (Fig. 6; Jobling and Richie, 1988). This plasmid, pMER610, was originally isolated from an Alcaligenes sp., and CABP1 exhibits homology to two of five ORFs residing in the resistance determinant on the plasmid. The two homologous ORFs encode 191- ,~ad i92-aa polypeptides. While they can function separately in conferring resistance to Te, the
218 presence of both O R F s is required for optimal activity (Jobling and Ritchie, 1987; 1988). The C A B P 1 protein has two subunits, both of which contain the Te R domain. It is possible that C A B P I also requires two copies of this domain for full function. The high degree of conservation between CABP i and a Te R determinant suggests that they perform a similar f,,nction. Unfortunately, the mechanism by which pMER610 confers resistance to the Te is unknown. Thus, this information does not provide additional insight into the function of C A B P I . The close similarity between the Te R sequences of pMER610 and the C terminus of C A B P I indicates that they are related by descent. A simple explanation is that these sequences have evolved from a common ancestral gene. The relative stability of these sequences is presumably maintained by strong selection pressure, the nature of which is currently unclear. Recently, Syvanen (1987) proposed that horizontal gene flow plays a role in evolution. Since in nature amoebae of Dictyostelium feed on bacteria of decaying forest vegetation, it is conceivable that the existence of these sequences in the plasmid and in Dictyostelium may have been the result of a horizontal gene exchange event. At present, however, we do not have additional evidence to support the idea of inter-kingdom gene transfer. (g) Conclusions (1) We have cloned the entire coding regions of the CABP1 m R N A s and demonstrated that the two subunits of this novel D. discoideum c A M P binding protein are identical in aa sequence except for an additional 37 aa present in the larger polypeptide C A B P I A . (2) The C A B P I m R N A s are approx. 1.25 and 1.15 kb and are encoded by a single gene. (3) A significant homology was found between the C termini of the CABP1 subunits and two polypeptides encoded by a bacterial plasmid determinant t~lat confers resistance to Te.
ACKNOWLEDGEMENTS We thank Dr. Peter N. Devreotes for the gift of the AX3 c D N A 2gtl 1 library and Dr. Michel Veron for the plMS expression vectors. This work was supported by the National Cancer Institute of Canada.
REFERENCES Barklis, E. and Lodish, H.F.: regulation of Dictyostelium discoideum mRNAs specific for prespore or prestalk cells. Cell 32 (1983) 1139-1148.
Chung, S., Landfear, S.M., Blumberg, D.D., Cohen, N.S. and Lodish, H.F.: Synthesis and stability of developmentally regulated Dictyostelium discoideum mRNAs are affected by cell-cell contact and cAMP. Cell 24 (1981) 785-797.
Dale, R.M.K.,McClure, B.A. and Houchins, J.P.: A rapid single-stranded cloning strategy for producing a sequential series of overlapping clones for use in DNA sequencing: application to sequencing the corn mitochondrial 18s rDNA. Plasmid 13 (1985) 31-40. Galinski, M.R., Arnot, D.E., Cochrane, A.H., Barnwell, J.W., Nussenzweig, R.S. and Enea, V.: The circumsporozoite gene of the Plasmodium cynomoigi complex. Cell 48 (1987) 311-319. Gerisch, G.: Cyclic AMP and other signals controlling cell development and differentiation in Dictyostelium. Annu. Rev. Biochem. 56 (1987) 853-879. Huynh, T.V., Young, R.A. and Davis, R.W.: Construction and screening cDNA libraries in 2gtl0 and 2gtl 1. In Glover, D.M. (Ed.), DNA Cloning - - A Practical Approach, Vol. 1. IRL Press, Washington, DC, 1985, pp. 49-78. Jobling, M.G. and Ritchie, D.A.: Genetic and physical analysis ofplasmid genes expressing inducible resistance of tellurite in Escherichia coll. Mol. Gen. Genet. 208 (1987) 288-293. Jobling, M.G. and Ritchie, D.A.: The nucleotide sequence of a plasmid determinant for resistance to tellurium anions. Gene 66 (1988) 245-258. Kay, R.R.: cAMP and spore differentiation in Dictyostclium discoideum. Proc. Natl. Acad. Sci. USA 79 (1982) 3228-3231. Kay, C.A., Noce, T. and Tsang, A.S.: Transloeation of an unusual cAMP receptor to the nucleus during development of Dictyostelium discoideum. Proc. NatL Acad. Sci. USA 84 (1987) 2322-2326. Klein, P.S., Sun, T.J., Saxe III, C.L., Kimmel, A.R., Johnson, R.L. and Devreotes, R.N.: A chemotactic receptor controls development in Dictyostelium discoideum. Science 241 (1988) 1467-1472. Laughon, A. and Scott, M.P.: sequence of a Drosophila segmentation gene: protein structure homologywith DNA-bindingproteins. Nature 310 (1984) 25-31. Loomis, W.F. (Ed.): The Development of Dictyostelium discoideum. .Academic Press, New York, 1982. Maniatis, T., Fritsch, E.F. and Sambrook, J.: Molecular Cloning. A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1982. Pearson, W.R. and Lipman, DJ.: Improved tool |or biological sequence analysis. Proc. Natl. Acad. Sci. USA 85 (1988) 2444-2448. Sanger, F., Nicklen, G. and Coulson, A.R.: DNA sequencing with chainterminating inhibitors. Proc. Natl. Acad. Sci. USA 74 (1977) 5463-5467. Simon, M.N., Mutzel, R., Mutzel, H. and Veron, M.: Vectors for expression of truncated coding sequences in Escherichia coll. Plasmid 19 (1988) 94-102. Skinner, M.K. and Griswold, M.D.: Fiuorographic detection of radioactivity in polyacrylamidegels with 2,5-diphenyloxazole in acetic acid and its comparison with existing procedures. Biochem. J. 209 (1983) 281-284. Stambaugh, K. and Blakesley, R.: Extended DNA sequencing with Klenow fragment: the kilobase sequencing system. Focus 10 (1988) 29-31. Syvanen, M.: Molecular clocks and evolutionary relationships: possible distortions due to horizontal gene flow.J. Mol. Evol. 26 (1987) 16-23. Tsang, A., Grant, C., Kay, C., Bain, G., Greenwood, M., Note, T. and Tasaka, M.: Characterization of an unusual cAMP receptor and its related polypeptides in Dictyostelium discoideum. Dev. Genet. 9 (19[;8) 237-245. Tsang, A.S., Kay, C.A. and Tasaka, M.: Expression o~an altered cAMP binding proteir, by rapid-developing strains of Dictyostelium discoideum. DeveLop. Biol. 120 (1987) 294-298. Tsang, A.S., and Tasaka M.: Identification of multiple cyclic AMPbinding proteins in oeveloping Dictyostelium discoideum ceils. J. Biol. Chem. 26i (1986) 10753--10759. Williams, J.G., Tsang, A.S. and Mahbubani, H.: A change in the rate of transcription of a eukaryotic gene in response to cyclic AMP. Proc. Natl. Acad. Sci. USA 77 (1980) 7171-7175.
213
GENE03~6
Cloning and characterization of c D N A s encoding a novel cyclic AMP-binding protein in D i c t y o s t e l i u m discoideum (Recombinant DNA; cellular slime moulds; antibody; gene fusion; proline-rich; tellurite-resistance determinants; expression in Escherichia coli)
Caroline E. Grant* and Adrian Tsang Department of Biology, McGill University,Montreal, Quebec HJA IBI (Canada) Receivedby D.T. Denhardt: 16 May 1990
Revised: 10 July 1990 Accepted: 11 July 1990
SUMMARY
The cellular slime mould, Dictyostelium discoideum, contains a novel cyclic AMP-binding protein, CABP1, wh/ch is composed of two subunits. Using anti-CABP1 monoclonal antibody as a probe, a cDNA clone was isolated from a Agtll expression library. By hybrid selection of the complementary mRNA and its translation in vitro, we demonstrated that the cDNA hybridized to mRNAs encoding both CABPI polypeptides. With the positive cDNA as a probe, we isolated a series ofoverlapping cDNA clones covering the coding region of both CABP 1 mRNAs. Expression ofthe cloned cDNAs in bacteria and sequence analysis showed that the CABPI subunits are identical in amino acid (aa) sequence, except that the small subunit is missing 37 aa near its N terminus. Genomic analysis suggested that the two CABP1 transcripts are derived from a single gene. The N-terminal half of each subunit is rich in proline, glutamine and giycine residues and contains a large block of aa repeats. The C-terminal half has an approx. 47 % aa id,mtity (86% with fuactionally conservative substitutions) with two polypeptides encoded by a plasmid determinant for tellurium anion resigtance.
INTRODUCTION
Dictyostelium discoideum grows vegetatively as individual amoebae. When deprived of their bacterial food source, the Correspondence to: Dr. A. Tsang, Department of Biology,McGill University, 1205 Ave. Docteur Penfield, Montreal, Que. H3A IBl (Canada) Tel. (514)398-6435; Fax (514)398-5069. * Present address: Cancer Research Labs, Botterell Hall, Queen's University, Kingston, Ont. K7L 3N6 (Canada) Tel. (613)545~2979.
Abbreviations: aa, amino acid(s); bp, base pair(s); t.ABPI, cAMPbinding protein 1; CABPI, gene (DNA) encoding CABP1; cAMP, cyclic AMP; cDNA, DNA complementary to RNA; D, Dictyosteiium; Denbardt's, 0.2% each of bovine serum albumin, Ficoll and polvvinylpyrrolidone; IPTG, isopropyl-p-D-thiogalactopyranoside;kb, kilobase(s) or 1000 bp, c~Ab, monoclonal antibody; nt, nucleotide(s); oligo, oligodeoxyribonucieotide; ORF, open reading frame; PAGE, polyacrylamidegel electrophoresis; g, resistance/resistant; SDS, sodium dodecyl sulfate; SSC, 0.15 M NaCI/0.015 M Naj" citrate pH 7.6; Te, tellurium anion. 0378-1119/90/$03.50 © 1990ElsevierSciencePublishersB.V.(BiomedicalDivision)
cells aggregate and eventually develop into a fruiting body composed of two basic cell types, stalk and spore cells (reviewed in Loomis, 1982). During the developmental process, cAMP acts as the chemotactic signal for aggregation (reviewed in Gerisch, 1987), induces cellular differentiation (Kay, 1982)~and modulates the expression of many genes (Williams et al., 1980; Chung et al., 1981; Barklis and Lodish~ 1983). We have previously identified a novel cAMP-binding protein in D. discoideum cells, termed CABPI (Tsang and Tasaka, 1986). This plotein can be detected in the nucleus of developing cells suggesting that CABP1 may have a role in the cAMP-mediated regulation ofgene expression (Kay et al., 1987; Tsang et al., 1988). In D. discoideum strain NC4, CABPI is composed of two subunits, 43-kDa CABP1A and 38-kDa CABP1B (Tsmtg and Tasaka, 1986; Tsang et al., 1987). The aim of the present study is to identify and characterize cDNAs which encode the CABP1 subunits.
214 1
2
3
5
4
6
7
8
...... t 7
-1A
~
'[
-1B -34 m
-31
Fig. 1. Isolation and identification of CABPI cDNA. A mixture of poly(A) ÷ RNA~ prepared from NC4 cells that had developed for 12 h and 20 h was used to construct the eDNA library in the expression vector 2gtl I (Maniatis et al., 1982; Huynh et al., 1985). The phage were used to infect E. coliY 1090 and the resulting plaques were transferred in duplicate onto nitrocellulose filters (BA85, Schleicher & Schuell) and probed with three anti-CABPl mAbs (Tsang and Tasaka, 1986) as described by Huynh et al. ~1985). A 0.67-kb eDNA from a positive phage was subcloned into the Eco RI site of the plasmid pAT 153. The identity of the 0.67-kb eDNA was examined by hybrid selection (Maniatis et al., 1982) as follows. Poly(A)+RNA was hybridized to filter-bound pATI53 carrying the 0.67-kb eDNA or pBR322. After hybridization for 3 h at 45 °C, the filters were washed extensively to remove nonspecific RNAs. The bound RNAs were eluted by boiling and concentrated by ethanol precipitation. The eluted RNA was then used to direct translation in vitro in a rabbit reticulocyte lysate system (Bethesda Research Laboratories). Each 30-#1 translation reaction contained the eluted RNA from a single filter and 80 #Ci [32S]methionine (Amersilam, 1400 Ci/mmol). Proteins from 2 #1 of the 30-#1 translation reactions were immediately solubilized in SDS sample buffer (lanes I-4) while the remaining sample was immunoprecipitated with anti-CABPl mAb (Tsang and Tasaka, 1986) before solubilization (lanes 5-8). The translation products were resolved in 0.1% SDS-12% PAGE and visualized by fluorography (Skinner and Griswold, 1983). Lanes: 1 and 5, 1 #g of total poly(A)+ RNA; 2 and 6, no exogenous RNA added; 3 and 7, RNA selected with pBR322; 4 and 8, RNA selected with pATI53 carrying the 0.67-kb eDNA. The two subunits of CABPI are marked as 1A and IB and the 34- and 31-kDa polypeptides which coprecipitated with CABP1 (Tsang and Tasaka, 1986) as 34 and 31, respectively.
CABP1 mAbs as probes. The identity of the eDNA was examined by hybrid selection. Fig. 1 shows that the mRNA species selected with the 0.67-kb eDNA were translated into two major polypeptides of 43 and 38 kDa, the sizes identicaI to those ~f the two CABP1 subunits (lanes 4 and 5). Both of these polypeptides reacted positively with the anti-CABP1 mAb (Fig. 1, lane 8). These data suggest that the 0.67-kb eDNA was complementary, to mRNA populations which encode the two CABP1 subunits. We have isolated three additional eDNA clones using the 0.67-kb eDNA as a probe. Thent sequence ofthese cDNAs were determined and compared (Fig. 2). From the nt sequence we deduced two composite cDNAs that are identical, except that the larger cDNA has an additional 111 nt in the coding region. The origins of the two cDNA populations were examined by RNA blotting. As shown in Fig. 3, the probe sequences that are present in both composite CABPI cDNAs hybridized with two mRNA species of 1.25 and 1.15 kb. The probe sequence that is specific for tile large composite cDNA, however, hybridized only with the 1.25-kb mRNA. Thus, the two classes ofcDNA appear to have derived from different mRNA populations and are not the result of cloning artifacts.
RESULTS AND DISCUSSION
(b) The cloned cDNAs represent the entire coding sequence of the CABPI mRNAs The sizes of CABP 1 subunits in the D. discoideum strain NC4 are 43 and 38 kDa in SDS-PAGE (Tsang et al., 1987). The large composite eDNA, however, encodes a 35-kDa polypeptide of 333 aa, and the small composite eDNA encodes a 31-kDa polypeptide of 296 aa. To determine the apparent M r of the polypeptides encoded by the composite cDNAs, we cloned the cDNAs into appropriate pIMS expression vectors and synthesized the gene products in Escherichia coil cells (Simon et al., 1988). Fig. 4A shows that the construct carrying the large eDNA produces a polypeptide which is 4 aa larger than its ORF. The plasmid carrying the small eDNA encodes the same number of aa as its ORF. The mobility ofthe fusion proteins and the endogenous CABP1 subunits were compared on SDS-PAGE (Fig. 4B). In each case, the plasmid construct produced an immunoreactive polypeptide with a mobility equivalent to that of the CABP1 polypeptide produced in the D. discoideum cells. These results confirmed that the cloned cDNAs represent the entire coding regions of the CABPI mRNAs, and that the large eDNA encodes CABP1A and the small eDNA encodes CABPIB.
(a) Identification of cDNAs complementary to CABPI mRNAs A cDNA clone containing a 0.67-kb insert was isolated from a 2gtl 1 expression library using a mixture of anti-
(c) Genomic organization of the CABPl-encoding sequences To determine if the two CABPI sequences are derived from a single gene, we probed a blot of genomic DNA with
215 CABPIB
CABPIA Met Tyr Asn Pro Pro Pro Pro Ser G!y Ser G l n ~ l y Asn Ash AsnJl5 A T A T ~ T ~ ATG TAT ~ C CCA CCA CCA CCA TCT GGT TCA C ~ CIGT ~ T ~ T A A T i ~
i
(Ii)
i
I
61
jTyr Tyr Arg Cln Pro Ser Ser ~ r Pro Gly Val Ser Asn Pro Asn Pro Gln Ala Ash 34 ITAT TAT AGA C ~ CCA TCA TCC ACA CCG GGT GTA TCA ~ C CCA ~AC ACCAT C ~I GCT ,
118
JGln Phe Leu Pro Pro Gin Pro Ser Asn Thr Thr Gin Thr Pro Gly Gly Tyr Pro Pro 53 C ~ TTT TTA CCA CCA CAA CCA TCT AAT ACT ACA CAA ACA CCA GGA GGC TAT CCA CCA
(35)
175
Gin Gin Gin Gin Arg Pro Pro Thr Gly Ala Pro Gin Gin Pro Gly Gly Tyr Pro Thr 72 CAA CAA CAA C A A A C A CCA CCA ACT CCA GCA CCA CAb. CAA CCA CCA GCT TAT CCA ACT Pro Pro Pro Pro Gly Ala Pro Cly Gly Tyr Pro Pro Gln Gln Gln Pro Ala Gly Gln 91 CCA CCA CCA CCA GGA GCA CCA GGA GGT TAT CCA CCA CAA CAA CAA CCA GCC GGT CAA ...................... I .......
(54)
232
289
j
(16)
!
Tyr Gly Ala Pro Pro Gln Gln Gln Pro Ala Gly Gln Tyr Cly Ala Pro Gln Pro Ala II0 (73) TAT CGA GCA CCA CCA C ~ C ~ C ~ CCA GCC GGT C ~ TAC GGA GCA CCA C ~ CCA GCA ..............
II
......................
2 ......................
i
l ..........
Gly Gln Tyr Gly Ala Pro Gln Pro Ala Gly Gln Tyr Cly Ala Pro Gln Pro Ala Gly 129 (92) 346
403
GGA CAA TAT G(;T GCA CCA CAA CCA GCA GGA CAA TAT GGT GCA CCA CAA CCA GCA GGA ..... 3 ................ II ................ 4 ................ I I. . . . . . . . . . . . . . Gln Tyr Gly Ala Pro Pro Pro Pro Pro Gly Gly Ala Gly Ile Ser Leu Val Lys Ash 148 (Iii) CAA TAT GGC GCA CCA CCA CCA CCA CCA GGA GGT GCA GGT ATT TCA TTA GTA AAG AAT
,s
. . . . . . . . . . . . . . . . .
I
A
517
Gln Gln Ile Ser Leu Thr Lys Glu Asp Pro CAA C A A A T T TCA TTA ACT AAA G A A G A T CCA Xho II Gly Trp Asp Val Asn Thr Thr Pro Thr Ala GGT TGG GAT GTA AAC ACC ACA CCA ACT GCA
574
Met Leu Asn Ala GIn Gly Arg Val Arg Thr Ser GI~ Asp Vhe Ile Phe Tyr Asn Ash 205 (168) ATG TTG AAT GCA CAA GGT AGA GTT AGA ACC TCA CAA GAT TTC ATT TTC TAC AAT AAC
631
Lys Val Ser Arg Asp Asn Ser Val Ser His Gin Gly Asp Ash Leu Thr Gly Gln Gly 224 (187) AAG GTA TCA AGA GAT AAC TCT GTT TCT CAT CAA GGT CAT AAT TTA ACA GGT CAA GGT
688
Glu Gly Asp Asp Glu Val Vai Leu Val Asn Leu Gin Ala Val Ser Pro Asp Val Thr 243 (206) GAA GGT GAT GAT GAA GTT GTC CTC GTA AAC TTA CAA GCA GTT TCA CCA GAC GTC ACT
745
Arg Leu Val Phe Ala Val Thr Ile His Leu Ala Asp Glu Arg Arg Gln Asn Phe Thr 262 (225) CGT TTG GTT TTC GCT GTC ACC ATT CAT TTA CCC CAT GAA ACA ACA CAA AAC TTT ACA
802
Met Val Pro Arg Ala Phe Ile Arg Val Ala Ann ~ln Glu Thr Gly Arg Asn Ile Cys 281 (244) ATG GTA CCA ACA GCT TTC ATT CGT GTT GCC AAT C~tA GAA ACT GGT ACA AAT ATC TGT
859
Arg Tyr Asp Leu Ser Gln Glu Gly Gly Pro Asn Thr Ala Leu Ile Ala Cly Glu Val 300 (263) CGT TAC CAT CTC TCT CAA GAA GGT GGT CCA AAC ACT GCC CTC ATT CCT CGT CAA GTT
460
Thr Leu Arg Lys Leu Thr Ile Gly Leu 167 (130) ACT CTT A G A A A A TTA ACA AAT GGT TTA Pro Phe Asp Leu Asp Ala Val Val Phe 186 (149) CCA TTC GAT TTG CAT GCA GTT GTT TTC
Tyr Arg Asp Pro Set Asn Pro Ash Ash Trp Set Phe Val Ala Val Gly Lys Gly Met 319 (282)
916
TAT CGT CAT CCA TCA AAT CCA AAC AAT TGG TCT TTT GTT GCT GTT CGT AAA GGT ATG
973
Gin Gly Ala Leu Pro Gly ~ u Leu Gln lle Phe CLy Cys Cln C ~ GGC GCT CTT CCA GGT TTA CTC C ~ ATC TTT ~GT TGT CA& T ~ T T T ~ T A T T A T T A T E T
1035
~
T
T
L
~
333 (29b)
TT~T~C~C~T~T~CTATTTGT~TcTTAC~T'CABPIA
(A)I5 CABPIB (A)30
Fig. 2. Nucleotide sequence of CABPIA and CABPIB, and predicted aa sequen,3e. Additional cDNA clones were isolated from a second 2gtl I library (Klein et al., 1988) using the 0.67-kb cDNA as a probe. The 0.67-kb cDNA wa~ nick-translated (Bethesda Research Laboratories kit) and hybridized to the cDNA library at 42°C in a solution containing 50% formamide/5 × SSC/5 x Denhardt's/0.5% SDS/100 #g per ml salmon sperm DNA. Washing was done at 65°C in 2 × SSC/0.1 ~ SDS. 28 positive plaques were obtained from t['e approx. 60000 phage plaques screened. Three positives with inserts of approx. I kb were subcloned into the plasmid vectors pBluescript + or - (Stratagene). Sequencing was done using the chain-termination method of Sanger et al. (1977) as modified by Stambaugh and Blakesley (1988). Staggered deletions were obtained for the MI3 clones (the 0.67-kb cDNA was subcloned into MI3 phage for sequence analysis) using T4 DNA polymerase (Dale et al., 1985) and for the plasmid clones using DNAse I (Laughon and Scott, 1984). Approximately 70% to 80% of the sequence of each cDNA was determined on both strands. The nt sequences of the CABPIA and CABPIB composite cDNAs are identical and, therefore, a single nt sequence is shown. Downward arrows at the beginning of the sequence denote the 5' ends, and the downward and horizontal arrows at the end of the sequence show the 3' ends of the composite cDNAs. The nt are numbered on the left of the figure. The aa are numbered on the right of the figure with those referring to the CABP1B polypeptide m t~rackets. The 111 nt (37 aa) not present in CABPIB are boxed. The position of aa 143 of CAB PiA ( A ) and the XhoII cleavage site are indicated below the nt sequence. A 5-aa repeat is double-underlined and the 12- and 9-aa repeats are underlined with dashed lines and numbered. The nt sequence reported here has been submitted to GenBank Data Library with acession number M36176.
216
1
2
3
A
1350 -1250
5' END OF CABPIA AND IB
-
Met Tyr Asn PrJ Pro Pro Pro S e r Gly ATG TAT AAC CCA CCA CCA CCA TCT G G T
LARGE SUBUNIT (1A) CONSTRUCT- i . 1
- 960
Fig. 3. Northern-blot ar~alysisofthe CABP1 transcript,,. Poly(A)÷RNA (2.5 #g/lane) from NC4 cells was fractionated on a 1.5% agarose-2.2 M formaldehyde gel (Maniatis et al., 1982) and transfr~rred to Nytran membrane (Schleicher & Schueii). Lanes: 1, the RN,~, was probed with an end-labelled oligo (probe B in Fig. 5A; corresponding to nt 197-173 of Fig. 2, 5'-TGTTGTGGTGCTCCAGTTGGTGG' fC-3') which is common to both CABPI cDNAs. The blot was hybridized m~d washed in 2 x SSC at 60~C; 2, the RNA was probed with an et~d-labelledoligo (probe A in Fig. 5A; corresponding to nt 146-122 of Fig. 2, 5'-CCTGGTGTTTGTAGTATTAGATG-3') whic~ is specific for the larger cDNA. The blot was hybridized and washed in 2 x SSC at 50°C; 3, the RNA was hybridized with probe C (see Fig. 5A). Probe C is a cDNA restriction fragment of approx. 600 nt that contains the common 3' end of the CABPI cDNAs (from the XholI site to the 3' EcoRl linker). Hybridization and washing conditions for probe C are the same as those described in the legend to Fig. 2 except that the blots were washed at 68°C.
... ...
k b cDNA i n pIMS5
Met G l y G l u P h e A r g T . y K A s n P r o P r o ~ r o P r o S e r G I ~ . . . a t g g g g aAA TTC CGG TAT AAC CCA CCA CCA CCA TCT GGT . . . SMALL SUBUNIT ( 1 B ) CONSTRUCT- 1 . 0 k b cDNA i ~ pIMS1 M e t ASh S e t A l a P r o Pr~o ~ _ q S e t G_~ . . . a t g AAT TCC GCA CCA CCA CCA TCT GGT . . .
1
4
2
3
4
~ p*
67-
43the same c D N A and oligo sequences used in the analysis of the CABP1 transcripts shown in Fig. 3 (see Fig. 5A for a map). Fig. 5B shows that all probes reacted with the same genomic D N A bands except for the HinfI-digested D N A (there is a HinfI cleavage site between the 5'- and 3'-specific probes), and the KpnI-digested D N A (the 3'-specific c D N A probe contains a KpnI cleavage site). These data suggest that there is probably only one copy of the C A B P 1 gene per haploid genome. We now have evidence that the two CABP1 trans,=ripts are the products o f alternative splicing ( G r a n t et al., submitted). The d a t a may also explain why in some D. discoideum genetic variants both C A B P I subunits have altered Mrs (Tsang et al., 1987).
(d) Predicted aa sequences of the C A B P I polypeptides Based on their aa composition, the C A B P 1 polypeptides can be divided into two distinct regions. The C-terminal 191 aa have no unusual feature. This region contains a roughly equal number of acidic and basic aa. On the other hand, the N terminus of the C A B P 1 poiypeptides has several distinctive characteristics. The aa composition here is very limited, and the entire region is hydrophilic. Pro, Glu, and Gly make up 29~o, 18 ~o, and 17 ~o, respectively. M o s t of the remaining aa are either Asp or residues containing an hydroxyl group, especially Tyr. Charged or strongly hydrophobic aa constitute only 2~o. In addition, there are five repeated sequences of 9 aa, Q P A G Q Y G A P . Each of the repeats is symmetrical
w
4
30W
w
20Fig. 4. Expression ol the CABPI gene in bacteria. (Panel A) Strategy for cloning the CABPI cDNAs in-fro,me with the ATG codon of the appropriate plMS vector (Simon et ai., 1988). The nt .;equences underlined are the portion of the EcoRl linker at the ~' ~.d of the cDNA that becomes part of the coding sequen:~ h~ the construct. Underlined aa show where the deduced aa sequenc~ of the construct and that of the CABPI composite eDNA are the same. Each of the eDNA sequences cloned into the expression vectors has several in-frame TAA stop codons at the c~,~ of the coding sequence. (Panel B) Expression of the !acZ -cD~'~A fusion gene was induced by IPTG and the bacterial cells lysed by sonicati~n. The cell lysate was cleared by centrifugation and the proteinx im~unoprecipitated with the anti-CABPl mAb (Tsang aad Tasaka, 1986). Following resolution in 0.1~ SDS-12% PAGE, the proteins were stained with Coomassie blue. Lanes: I, extract from bacteria containing the plMSl vector only; 2, extract from bacteria containing the small eDNA construct', 3, extract from l0 s NC4 cells that had developed for 12 h; 4, extract from bacteria containing the large eDNA construct. The position of the CABP1A polypeptide is marked by the blackened triangle and the 1B polypeptide by the open triangle. The intensely stained bands (50 and 25 kDa) are the heavy and ~ight chains of the precipitated immunoglobulins. Protein size markers on left margin are in kDa.
217
A
CABP1 TerORF4 TerORk5
143 190 ..ISLVKNQQ ISLTKEDPTLRKLTIGLGWDVNTTPTAPFD LDAWFMLNA MSV--S-GGN V--S-TA-SM ENVLV ..... ARS-DGQD ..... SA-L-AMAV .... GGN V ..... A-SMNVALV ..... TRV-DGQA ..... S--LVGE
CABPI TerORF4 TerORF5
191 QGRVRTSQDF IFYNNKVSRD N-K--GDA ....... LK-AN-K-LSDSH- V .... TT-P-
CABPI TerORF4 TerORF5
DVTRLVFAVT IHLADERRQN --DKII-V .... D-QA---S --El{ . . . . . . . . E - - S - K - -
FTMVPRAFIR VANQETGRNI -GQ-SG .... LV-DDNQTEV -G--SNS-M- -V-NDN-SE-
290 CRYDLSQEGG A .... TEDAS A-F---EDAS
CABPI TerORF4 TerORF5
291 PNTAL~AGEV TE--MLF--L TE--M-F--L
SFVAVGKGHQ K-R---L-¥A K-K---Q-FA
GCQ -INAS -INI
Kpl iI
Hinfl X ~oll CABPIA
240 NSVSHQGDNL TGQGEGDDEV VLVNLQAVSP G--T-T---R --E-D .... S LKIK-D--PG GA-Q ..... R --E-D .... Q -KID-T~-AA
241
CABPIB
PROBES A
G
B
E X~EH
PROBE C K
;= X Y v E H
GALPGLLQIF -G-ASVCAQ¥ -G-AA-ATQH
Fig. 6. Comparison of the C terminus of the CABPI polypeptides and the plasmid encoded polypeptides for Tea. The aa 142-333 of the CABPIA polypeptides (CABPI) are compared to the aa sequence of ORFs 4 and 5 (TerORF4 and TerORFS) of the Te" determinant from the bacterial plaamid pMER610 (Jobling and Richie, 1988). The dots indicate the gaps placed hLthe aa sequence~ to optimize the alignment. The dashes in the sequence ofORFs 4 and 5 signify aa that are identical to those in CABP1.
100 bp
PROBE A
333 YRDPSNPNNW --HNGA...--HGAE...-
K
Fig. 5. Oenomic analysis by Southern blotting with the CABPl-specific probes. CA)Cartoon showing the homologous regions ar,d restriction sites in the CABPI composite cDNAs. The blackened boxes represent the 5'-noncoding region and the first 34 nt of the coding region. The crosshatched boxes represent 852 nt of coding as well as the Y-noncoding regions. The open box represents the 111 nt found only in the CABPIA cDNAs. The location of the sequences hybridizing to the probes are indicated by the vertical lines. Detailed description of the probes and the hybridization conditions are outlined in the legend to Fig. 3. Note that the scale is only approximate. (Panel B) A single Southern blot of genomic DNA was probed sequentially with each of the :hree probes. Probes A and B gave identical res~dts and consequently only the results of probe A are shown. Before each hybridization reaction, the previous probe was removed from the blot and its removal monitored by autoradiography. Each lane contains 2.5/~g ofAX2 genomic DNA digested with EcoRl (E), Xbal iX), EcoRI + Xbal (E,X), Hinfl (H), and KpnI (K), and resolved in a 1% agarose gel. Markers are at 21.2, 5.15, 4.97, 4.28, 3.53, 2.03, 1.90, 1.58, 1.33, 0.98, 0.83, and 0.56 kb.
around the central Glu (Q) and Tyr (Y) residues. Further, the first two copies of these repeats are part of a 12-at, repeat, PQQQPAGQYGAP. In general, the N-terminal regions contain a high concentration of residues capable of forming hydrogen bonds which may imply that they are involved in interactions with other cellular constituents. (e) H~,mology of CABPI to other proteins A search for other aa sequences with si~milarities to the CABP 1 polypeptides was made using the Genbank, EMBL and NBRF data bases and the program TFASTA (Pearson and Lipman, 1988). When the 142 aa of the N-terminal of CABP1A were used as the query sequence, only proteins with high Pro content were extracted from the databases. The similarity between CABP1 and most of these proteins was restricted to Pro residues. Of interest, however, is the similarity between the 9-aa repeat of CABP1, QPAGQYGAP, and the aa repeat of the circumsporozoite protein from Plasmodium cynomolgi, PAGDGAPAA (Galinski et al., 1987). When the C-terminal 191 aa were used as a query sequence in the database search, we detected 45 % and 49 % aa identities (84 % and 86 % including conservative substitutions) with two polypeptides encoded by a plasmid determinant for resistance to tellurium anions (Fig. 6; Jobling and Richie, 1988). This plasmid, pMER610, was originally isolated from an Alcaligenes sp., and CABP1 exhibits homology to two of five ORFs residing in the resistance determinant on the plasmid. The two homologous ORFs encode 191- ,~ad i92-aa polypeptides. While they can function separately in conferring resistance to Te, the
218 presence of both O R F s is required for optimal activity (Jobling and Ritchie, 1987; 1988). The C A B P 1 protein has two subunits, both of which contain the Te R domain. It is possible that C A B P I also requires two copies of this domain for full function. The high degree of conservation between CABP i and a Te R determinant suggests that they perform a similar f,,nction. Unfortunately, the mechanism by which pMER610 confers resistance to the Te is unknown. Thus, this information does not provide additional insight into the function of C A B P I . The close similarity between the Te R sequences of pMER610 and the C terminus of C A B P I indicates that they are related by descent. A simple explanation is that these sequences have evolved from a common ancestral gene. The relative stability of these sequences is presumably maintained by strong selection pressure, the nature of which is currently unclear. Recently, Syvanen (1987) proposed that horizontal gene flow plays a role in evolution. Since in nature amoebae of Dictyostelium feed on bacteria of decaying forest vegetation, it is conceivable that the existence of these sequences in the plasmid and in Dictyostelium may have been the result of a horizontal gene exchange event. At present, however, we do not have additional evidence to support the idea of inter-kingdom gene transfer. (g) Conclusions (1) We have cloned the entire coding regions of the CABP1 m R N A s and demonstrated that the two subunits of this novel D. discoideum c A M P binding protein are identical in aa sequence except for an additional 37 aa present in the larger polypeptide C A B P I A . (2) The C A B P I m R N A s are approx. 1.25 and 1.15 kb and are encoded by a single gene. (3) A significant homology was found between the C termini of the CABP1 subunits and two polypeptides encoded by a bacterial plasmid determinant t~lat confers resistance to Te.
ACKNOWLEDGEMENTS We thank Dr. Peter N. Devreotes for the gift of the AX3 c D N A 2gtl 1 library and Dr. Michel Veron for the plMS expression vectors. This work was supported by the National Cancer Institute of Canada.
REFERENCES Barklis, E. and Lodish, H.F.: regulation of Dictyostelium discoideum mRNAs specific for prespore or prestalk cells. Cell 32 (1983) 1139-1148.
Chung, S., Landfear, S.M., Blumberg, D.D., Cohen, N.S. and Lodish, H.F.: Synthesis and stability of developmentally regulated Dictyostelium discoideum mRNAs are affected by cell-cell contact and cAMP. Cell 24 (1981) 785-797.
Dale, R.M.K.,McClure, B.A. and Houchins, J.P.: A rapid single-stranded cloning strategy for producing a sequential series of overlapping clones for use in DNA sequencing: application to sequencing the corn mitochondrial 18s rDNA. Plasmid 13 (1985) 31-40. Galinski, M.R., Arnot, D.E., Cochrane, A.H., Barnwell, J.W., Nussenzweig, R.S. and Enea, V.: The circumsporozoite gene of the Plasmodium cynomoigi complex. Cell 48 (1987) 311-319. Gerisch, G.: Cyclic AMP and other signals controlling cell development and differentiation in Dictyostelium. Annu. Rev. Biochem. 56 (1987) 853-879. Huynh, T.V., Young, R.A. and Davis, R.W.: Construction and screening cDNA libraries in 2gtl0 and 2gtl 1. In Glover, D.M. (Ed.), DNA Cloning - - A Practical Approach, Vol. 1. IRL Press, Washington, DC, 1985, pp. 49-78. Jobling, M.G. and Ritchie, D.A.: Genetic and physical analysis ofplasmid genes expressing inducible resistance of tellurite in Escherichia coll. Mol. Gen. Genet. 208 (1987) 288-293. Jobling, M.G. and Ritchie, D.A.: The nucleotide sequence of a plasmid determinant for resistance to tellurium anions. Gene 66 (1988) 245-258. Kay, R.R.: cAMP and spore differentiation in Dictyostclium discoideum. Proc. Natl. Acad. Sci. USA 79 (1982) 3228-3231. Kay, C.A., Noce, T. and Tsang, A.S.: Transloeation of an unusual cAMP receptor to the nucleus during development of Dictyostelium discoideum. Proc. NatL Acad. Sci. USA 84 (1987) 2322-2326. Klein, P.S., Sun, T.J., Saxe III, C.L., Kimmel, A.R., Johnson, R.L. and Devreotes, R.N.: A chemotactic receptor controls development in Dictyostelium discoideum. Science 241 (1988) 1467-1472. Laughon, A. and Scott, M.P.: sequence of a Drosophila segmentation gene: protein structure homologywith DNA-bindingproteins. Nature 310 (1984) 25-31. Loomis, W.F. (Ed.): The Development of Dictyostelium discoideum. .Academic Press, New York, 1982. Maniatis, T., Fritsch, E.F. and Sambrook, J.: Molecular Cloning. A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1982. Pearson, W.R. and Lipman, DJ.: Improved tool |or biological sequence analysis. Proc. Natl. Acad. Sci. USA 85 (1988) 2444-2448. Sanger, F., Nicklen, G. and Coulson, A.R.: DNA sequencing with chainterminating inhibitors. Proc. Natl. Acad. Sci. USA 74 (1977) 5463-5467. Simon, M.N., Mutzel, R., Mutzel, H. and Veron, M.: Vectors for expression of truncated coding sequences in Escherichia coll. Plasmid 19 (1988) 94-102. Skinner, M.K. and Griswold, M.D.: Fiuorographic detection of radioactivity in polyacrylamidegels with 2,5-diphenyloxazole in acetic acid and its comparison with existing procedures. Biochem. J. 209 (1983) 281-284. Stambaugh, K. and Blakesley, R.: Extended DNA sequencing with Klenow fragment: the kilobase sequencing system. Focus 10 (1988) 29-31. Syvanen, M.: Molecular clocks and evolutionary relationships: possible distortions due to horizontal gene flow.J. Mol. Evol. 26 (1987) 16-23. Tsang, A., Grant, C., Kay, C., Bain, G., Greenwood, M., Note, T. and Tasaka, M.: Characterization of an unusual cAMP receptor and its related polypeptides in Dictyostelium discoideum. Dev. Genet. 9 (19[;8) 237-245. Tsang, A.S., Kay, C.A. and Tasaka, M.: Expression o~an altered cAMP binding proteir, by rapid-developing strains of Dictyostelium discoideum. DeveLop. Biol. 120 (1987) 294-298. Tsang, A.S., and Tasaka M.: Identification of multiple cyclic AMPbinding proteins in oeveloping Dictyostelium discoideum ceils. J. Biol. Chem. 26i (1986) 10753--10759. Williams, J.G., Tsang, A.S. and Mahbubani, H.: A change in the rate of transcription of a eukaryotic gene in response to cyclic AMP. Proc. Natl. Acad. Sci. USA 77 (1980) 7171-7175.
