YEAST

0

VOL. 7: 971-979 (1991)

0 0 0

0

0

111

0 0 0

0o

Yeast Sequencing Reports

0 0 0 0

The Nucleotide Sequence of a Third CyclophilinHomologous Gene from Saccharomyces cerevisiae L. FRANCO, A. JIMENEZ*, J. DEMOLDER', F. MOLEMANSt, W. FIERSt AND R. CONTRERAS'

Centro de Biologia Molecular (C.S.I.C.IUA.M.),Universidad Autdnoma, Canto Blanco, 28049, Madrid, Spain tLahoratorium voor Moleculaire Biologie, Rijksuniversiteit-Gent, K.L. Ledeganckstraat 35,B-9000 Gent, Belgium. Received 10 June 1991; revised 14 July 1991 The nucleotide sequence of a 1558 bp DNA fragment from the right arm of chromosome 111 of Saccharomyces cerevisiae contains an open reading frame of 954 nucleotides with coding potential for a protein with high similarity to the ubiquitous cyclophilins which are both peptidyl-prolyl cis-trans isomerases and cyclosporin A-binding proteins. It should, therefore, represent the third gene (SCC3) of this kind from S. cerevisiae. SCC3 is present in a single copy in the genome of S. cerevisiae and results in a constitutively expressed 1.2 kb transcript during cell growth. Its putative protein product (Scc3) contains two hydrophobic cores, one at the amino terminal, 20 amino acids long, which could serve as a signal peptide, and the other one at the carboxyl end with a structure similar to a transmembrane helix. These findings suggest that Scc3 could be a secretory or, more likely, a transmembrane protein. The only cyclophilin with similar structure to that of Scc3 is ninaA from Drosophila melanogasrer, a transmembrane protein which seems to be implicated in the correct folding and/or intercalation of rhodopsin in the endoplasmic reticulum of the fly photoreceptors (Stamnes, M.A. er al., Cell 65,219-227, 1991). In addition, the amino and the carboxy regions of Scc3 and ninaA share a significant level of homology, which suggests that they have a similar function, albeit for different target proteins.

INTRODUCTION Cyclophilins (CyPs) are an important group of proteins, present in all living organisms, which catalyse in vitro the cis-trans isomerization of peptidyl-prolyl peptide bonds (PPIase activity) (Handschumacher et al., 1984). It has been proposed that i n v i v o CyPs increase the rate of proper folding/unfolding of specific cellular proteins, an action which they do perform 'in vitro' (Lang et al., 1987; Kiefhaber et al., 1990). In addition, CyPs from eukaryotic cells bind with high affinity, and their PPIase activity is inhibited by, the clinically important immunosuppressive agent cyclosporin A (CsA), a fungal cyclic undecapeptide produced by Tolypocladium inflatum (for a review see Stamnes and Zuker, 1990). In contrast, the periplasmic rotamase of Escherichia coli, a PPIase homolog to CyP, is not inhibited by CsA (Liu and Walsh, 1990). Because it has been shown that CsA specifically binds to CyP in *Corresponding author.

0749-503X/9 1/09097149$05.00 01991 by John Wiley & Sons Ltd

T-cells, this primary action may interfere with the isomerization of X-P bonds catalysed by CyP on cytokine transcriptional factors at the early steps of T-cell activation (Siekierka et al., 1989). Similarly to CsA, FK 506, a macrolide antifungal drug produced by Streptomyces tsukabaensis, is a potent immunosuppressive agent, which also prevents initial stages of T-cell activation. FK 506 binds to specific cellular proteins, the FK 506-binding proteins (FKBP) (Kino et al., 1987), which have been found in human, calf, Neurospora crassa and yeast cells (Siekierka et al., 1990; Tropschug et al., 1989). Even though CyPs and FKBPs share PPIase activity, they clearly differ in molecular weight, amino acid sequence and drug-binding specificity (Stamnes and Zuker, 1990). Chemically related to FK 506 is rapamycin, a macrolide antifungal agent which is also a potent immunosuppressive compound. Although these two drugs compete for binding to FKBPs, they differ in that rapamycin acts at the late steps of T-cells activation (Dumont et al., 1990).

972 In yeasts, a CyP has been identified, purified to homogeneity and partially sequenced from Saccharomyces cerevisiae (Haendler et al., 1989; Dietmeier and Tropschug, 1990). Moreover, two genes (named herein SCCI and SCC2, for S. cerevisiae cyclophilin) encoding CyPs (herein Sccl and Scc2, respectively) (Haendler et al., 1989; Dietmeier and Tropschug, 1990; Koser et al., 1990b) and two other genes (FKBI and RBPI) which determine a FKBP (Wiederrecht et al., 1991) and a rapamycin-binding protein (Koltin et al., 1991), respectively, from S. cerevisiae have been cloned and sequenced. In addition, the nucleotide sequence of a gene encoding a CyP from Candida albicans is also known (Koser et al., 1990a). The deduced gene products from SCCl and SCC2 are proteins of = 17 and 18 kDa, respectively, which share > 50% identity to human and bovine CyPs. Sccl and Scc2 differ mainly in the presence of a hydrophobic 34 amino acid N-terminal extension in the latter, which may function as a signal sequence for subcellular compartmentalization (Koser et al., 1990b). This work reports the nucleotide sequence of a novel gene (SCCS) from chromosome I11 of S. cerevisiae with potential CyP-coding capacity and a comparison of its derived amino acid sequence to those of other known and putative CyPs. MATERIALS AND METHODS

Yeast strains and phages Saccharomyces cerevisiae S288C a, mal, mel, ga12, SUCI, CUPI was obtained from the Yeast Genetic Stock Center. Phages 1PM5307 (Gent) and hPM6589 (Madrid) were provided by S. Oliver (UMIST). DNA methodologies Sequencing of the relevant h DNA inserts was performed by the ‘dideoxy ’ chain-termination method of Sanger et al. (1977), following the modification of Hattori and Sakaki (1 986) for doublestranded DNA. For the right region of the DNA insert from hPM5307 (Figure 1), ordered deletions were obtained by subcloning specific restriction fragments in plasmid pUC18. In addition, other deletions were obtained by shearing plasmid DNA with glass beads, followed by restriction with Smal and self-ligation. To eliminate plasmids deleted in the vector, the ligation mixture was restricted with KpnI. Compressions and other significant difficulties were solved by including 7-deaza-dGTP in the sequencing reactions, which were performed under polymerase chain reaction

L. FRANC0 E T M .

Figure1 Restriction maps and sequencing strategy. (A) Restriction maps of the overlapping DNA regions of lPM5307 (top) and hPM6589 (bottom) encoding SCC3. (B) Sequencing strategy. The bold arrow indicates the length of coding sequence and direction of transcription of SCC3. Thin arrows indicate the length and direction of sequencing reactions. Dashed lines indicate the total sequenced DNA fragment (1558 bp).

conditions at 70°C with Taq DNA polymerase. The bulk of the sequencing reactions were carried out with the sequencing kit from Pharmacia. For the left region of the DNA insert from hPM6589 (Figure l), the Sequenase 2.0 kit (U.S.B.) and plasmid bluescript SK were used throughout. Polymerization of DNA was primed with the universal and reverse oligodeoxynucleotides (Boehringer Manheim GMBH) or by other oligodeoxynucleotides which were synthesized to complement previously determined nucleotide sequences. The DNA coding strand was sequenced on nuclease Ex0111 deleted DNA fragments which were obtained with the Nested Deletion kit from Pharmacia. To perform Southern blot hybridizations, total DNA from S. cerevisiae S288C was restricted with several restriction endonucleases, developed on 0.7% agarose gel electrophoresis, transferred to a nylon membrane and then hybridized to a 1.7 kb Hind1 DNA fragment containing the SCC3 gene (Figure 1) which was labeled with 32P by nick translation (Rigby et al., 1977). To perform a Northern blot hybridization, total RNA from s. cerevisiae S288C was developed on 0.8% agarose gel electrophoresis under denaturing conditions, transferred to a nylon sheet and then hybridized to a 1.3 kb EcoRI-HindII DNA fragment containing part of the SCC3 sequence (Figure 1) wich was labeled with 32P as indicated above.

Informatic support Nucleotide and amino acid sequences were analyzed with the aid of the University of Wisconsin group of programs (Devereux et al., 1984) on a Micro Vax R 215F station in conjunction with sequence data from NBRF-protein (release 25) and SwissProt (release 15).

973

NUCLEOTIDE SEQUENCE OF THIRD CYCLOPHILIN-HOMOLOGOUS GENE FROM S. CEREVfSfAE

RESULTS AND DISCUSSION

In addition, transmembrane domains were sought with the programs HELIXMEM and from the PCGENE package (Intelligenetics). Putative cleavage sites of signal peptides were determined according to von Heijne (1986) from the PSIGNAL program (Intelligenetics).

Sequencing and sequence analysis ofthe scc3 gene The yeast DNA inserts from chromosome I11 of hPM5307 and hPM6589 overlap as schematized in Figure 1. According to the map locations provided by

-300

CTGCACCAA ATACGAGTT GTAAGACCC ATACATCTA TTATGACCC TAGTGTTGT TTTAGGGCC TATGCGCCC

-228

TTTTTTTCT TCCATTGTC TGATAAATA ATGTATCAT TTTGACAAC TTAAAGGTT CAGCTAAAC GACACGAGA DraI A-ATATT A P @ A C T G C r m T G A AAGGCGTCA GTCAGGTGC AATTAATAG GTCACATAT TCTTTGCTT

-156

TC-AA~.I

-229 -157 -85

-84

CAATTG$TC

-12

TTCAGCAAC AACATGTGG TTGAAATCC TTGCTGCTC TGCCTGTAC TCCTTAGTA CTCTGCCAA GTCCACGC M W L K S L L L C L Y S L V L C Q V H A

59

60

GCACCTTCA TCAGGGAAG CAGATTACC TCCAAGGAT GTTGATCTT CAGAAAAAA TATGAGCCC AGTCCCCCC A P S S G K Q I T S K D V D L Q K K Y E P S P P

131

132

GCCACACAT CGTGGAATA ATCACTATC GAATACTTT GATCCCGTT TCGAAGTCG ATGAAAGAG GCGGATCTG A T H R G I I T 1 E Y F D P V S K S M K E A D L

203

204

ACTTTTGAG TTGTACGGT ACTGTCGTG CCCAAAACT GTGAACAAC TTTGCTATG CTGGCCCAT GGTGTTAAG T F E L Y G T V V P K T V N N P A M L A H G V K

215

216

GCAGTTATC GAAGGGAAA GATCCCAAT GATATACAT ACTTACTCG TACCGTAAG ACCAAAATC AACAAGGTT A V I E G K D P N D I H T Y S Y R K T K I N K V

341

348

TACCCTAAC AAGTATATC CAGGGTGGT GTGGTTGCC CCAGATGTG GGTCCTTTC ACCGTCTAT GGGCCCAAA Y P N K Y I Q G G V V A P D V G P F T V Y G P K

419

4 20

TTTGATGAC GAAAACTTT TACTTAAAA F D D E N F Y L K EcoRI TCTAACACC TCGG CATCATCACC S N T S E F I I T

CATGACAGG CCTGAAAGA CTCGCAATG GCCTATTTT GGACCTGAT H D R P E R L A M A Y F G P D

491

ACTAAAGCC GATGGAAAT GAGGAATTG GATGGCAAA AGTGTCGTG T K A D G N E E L D G K S V V

563

564

TTTGGTCAA ATAACTTCT GGTCTAGAT CAACTAATG GATGCTATT CAATACACA GAAACAGAC GAATATGGA F G Q I T S G L D Q L M D A I Q Y T E T D E Y G

635

636

AAGCCTCAG CATGAATTA CGGTTCCTG TATTTCGTT CTAGAAATC TTAAAAATT AGTAACATC TTAGATTTG K P Q H E L R F L Y F V L E I L K I S N I L D L

701

708

CACGCTGCG TACACAGAA AAAGTCGAG AAGTTTAGA AATGGCGAT GTGTCTGTT GGCTCCACT TTGGAAAAC H A A Y T E K V E K F R N G D V S V G S T L E N

I79

C % r w

TCGTAGATT GGAAGACCG AAGAAAATA GATAGACTT TCGAATAAG

a

+1

4 92

-----.

Xbal

-13

Xbal

DraI

780 ATCTTCCGT AACGATAAA GCCTACACA CCTTTAACC ACCTCCACT GGAACCACC I

852

F

R

N

D

K

A

Y

T

P

L

T

T

S

T

G

T

T

GCCTAT 851 A

Y

D

L N H SphI CCAATTTCC AGAGCCTTG ATGTGTTTA ACTGTTCTT GGCCTTTGT TTCATTGCC TACAAGGGC ATGCACGAA P I S R A L M C L T V L G L C F I A Y K G M H E

923

924

AAGCCTCAT ACGGTTTCA TTAAGACAC AAGTAAGCA ACTTCTTTT CGATGGCTT GATGACCAC ATGGCTATG K P H T V S L R H K *

995

996

ATTTT~X~GXXATATAT ACATGACAA ~ T T A T T ~ T TTT A T ~ T T AGCGATACCF-GXF~TTCT

TGGTTAATA

1061

1068

CCCTTTTGA CAGCCTCTA CTGCATTGC CCTTGATAA TAATTTTTT TCGATTGCA TCACCACAG CCATGATTT

1139

1140

TTTGGGTAT AGAAAGGCA GTTGCTCCT GCAAGTCAT TTCTTAGTT GAATTACAT CTCCTTCAA TCTTTCTGA

1211

1212

TCTCCGTGA TAATCTTGT TACCTCCAT TTTTATAAG CGGAATACA CC

1258

Figure 2 Complete sequence of the 1558 bp segment containing SCC3. Putative TATA box, CT block, CAPyACAC motive and polyadenylation sites are boxed. The dashed underline indicates a putative glycosylation site. The deduced amino acid sequence for Scc3 is presented in the one letter code. The underline at the 5’ coding region indicates the length of the predicted hydrophobic signal peptide and the vertical arrow its possible cleavage site. The underline at the 3’ coding region indicates a putative hydrophobic transmembrane domain.

L. FRANC0 ETAL.

974 Dr S.G. Oliver, this DNA region lies within the SUP61 and tsm5 genetic loci (Mortimer and Child, 1985). The restriction map and sequencing strategy of the sequenced 1558 bp DNA region containing a putative CyP gene are also indicated in Figure 1. A computer analysis of this sequence showed the presence of an open reading frame (ORF) of 954 nt which could determine a polypeptide of 3 18 amino acids (Figure 2), MW 35800, and an isoelectric point of 6.9. Its deduced amino acid sequence indicated a high similarity to those of known and presumptive cyclophilins (see below). This ORF may, therefore, correspond to a third cyclophilin gene (SCC3) from S. cerevisiae. It is present as a single copy in the genome of

A

S. cerevisiae (Figure 3) and its codon usage corres-

ponds to those of genes which are expressed with medium frequency (data not shown). Analysis of the 5' non-coding region of SCC3 showed a TATA box at -135, a putative CAAA (at -70) transcription initiation site, a CT box (at -78) and a CAPyACA (at -64) promoter motive. N o upstream activating sequences were detected, which could indicate that this region represents a constitutive promoter. Indeed, Northern analysis indicated that SCC3 renders an, apparently abundant, 1.2 kb transcript during yeast exponential growth (Figure 3). The 3' non-coding region of SCC3 shows a potential sequence (TTTTTATA) at + 1054 for polyadenylation.

B

23.79.5 6.6-

-

4.26

-

2.25 1.96

-

0.59 -

Figure 3 Southern and Northern blot analysis. Southern (A) and Northern (B) blots hybridizations were performed as indicated in Materials and Methods. Numbers at the left hand side indicate the size (in kb) of the DNA (A) and RNA (B) markers. The number at the right hand side (B) indicates the size (in kb) of the hybridizing transcript.

NUCLEOTIDE SEQUENCE OF THIRD CYCLOPHILIN-HOMOLOGOUS GENE FROM S. CEREVISIAE

975

This motif is flanked by four other ones (Figure 2) which could act as either modulators of polyadenylation or polyadenylation sites themselves. If transcription initiates close to -70 and terminates at about +1060, it would produce a transcript of near 1130 nt, a figure close to that determined experimentally.

NH2-terminal hydrophobic region. As indicated above, only Scc3 has a potential transmembrane domain at the carboxy end. The degrees of similarity between the sequences of these proteins (segments I to X) are: 90 (55%),58 (35%) and 71 (43%) identical residues for Sccl/Scc2, SCCl/Scc3 and Scc2/Scc3, respectively.

Analysis of the deduced amino acid sequence of Scc3

Alignment of the rotamasefrom E. coli with cyclophilins. Functional significance

The deduced amino acid sequence of Scc3 presents two highly hydrophobic domains at its amino- and carboxy-termini. The former is probably a secretion signal peptide of 20 amino acids with a possible cleavage site between A20 and A21 (Figure 2). The sequence of the latter fits with a transmembrane helix domain of 17 amino acids. These findings suggest that Scc3 is secreted or membrane-anchored. In addition, this protein has a potential S-A-T site for glycosylation (Figure 2).

Amino acid comparison with known and putative cyclophilins The deduced amino acid sequence of Scc3 shows a high degree of similarity to those from other known or presumptive CyPs (Figure 4). The presence of specific insertions and variable regions allows us to define 10 segments (numbered I to X) of significant homology. The highest variability is found at the amino termini, where a hydrophobic region (signal peptide) is found in 5 of the CyPs and in the rotamase of E. coli (see below) and at the carboxyl end where only Scc3 and D . melanogaster ninaA share a putative transmembrane helix domain (Figure 4). Including the amino acid sequence of Scc3 in the comparison reduces the number of amino acid residues which are totally preserved i n all the considered CyPs sequences. This, which could be related to a high divergent evolution of Scc3, simplifies previously described consensus sequences (see for example, Stamnes and Zuker, 1990). Thus, from a total of approximately 165 residues of the aligned sequences (segments I to X ) , 42 are invariable if the Scc3 sequence is not considered and only 26 if included. In addition, two novel CyP specific motifs (A: LxxxxxPxTxxNF and B: GxxxxxYGxxFxDE; Figure 3) could be defined, which was proved by comparison with all the amino acid sequences present in the NBRF (25814) and SwissProt (1 8364) data banks. Considering the three CyPs from S. cerevisiae, only Scc2 and Scc3 present an apparently non-related

Most of the invariant residues of CyPs are present in the rotamase of E . coli. Segment I has similar homology among all CyPs and rotamase. Segments 11, VI and VIII are relatively conserved, although in motif A (segment 11) rotamase contains a S instead of a T (Figure 4). Segment IV is specially well conserved, which could suggest that it is implicated in PPIase activity. Segments 111, V, X and XI share low homology. Concerning segment V, it is well conserved between all CyPs, but has low homology with the corresponding region of rotamase. This might suggest that segment V is implicated in a CyP specific function (i.e. CsA-binding andlor recognition). Segment VII deserves special attention. It is well conserved, although in the alignment an insertion of 8 amino acids (FGYAVFG) has been allowed in the E . coli rotamase. This could result from a duplication, followed by sequence degeneracy. In addition, it has been shown (Liu et af., 1991) that mutation of the W121 residue (which is present in all CyPs excepting Scc3) of the human CyP to F or A diminishes its susceptibility to CsA by 75- to 200-fold, whereas its PPIase activity is lowered only 2- to 13-fold. In contrast, if the F136 residue of the E . coli (Figure 4) rotamase is mutated to W, its sensitivity to CsA is increased by 23-fold, whereas its PPIase activity remains unchanged with respect to wild type (Liu et al., 1991). These results suggest an involvement of W121 in the binding of CsA and, in addition, a structural separation of this site with that responsible for PPIase activity (Liu et al., 1991). It therefore appears interesting that the considered W residue is not present in Scc3, where it is substituted by an E (Figure 4). It remains to be seen if Scc3 is insensitive to CsA. SCC3 and ninaA The deduced amino acid sequences of Scc3 and ninaA present intriguing similarities. At their amino termini both proteins have a potential signal peptide which, according to von Heijne (1986), could be

/I3 1 /

*

*

IGD EDVCRVI

ADNWH (8 ~ - - K =

. ...

A - D K I

SQVPTHD--VGPYQNVPSKPWIISAKVLPK

*

~IT--SCLDQLMDAI~YTETDE-YCKEJQHELRFLYFVLEIM~SNIL (2s)

--DOLDIV-KKIgSFGSGS-~~IVIEESOeG ' --LQIEW-~STKTDS-RIQIXDVIIADCGKIEVEKPFAIAKE *

--LQIDW-KKVZSLGSNS-6IVIDKOCTV

--DCYDIV-KKVESLGSPS-CARULRIWAKSCEL

/llS/

/116/ /116/ /116/

~~W--LQLIPI-KKAIAVGSSS-CRCSKPWIADCG~L * --rcrrrW-FtDI1KvGSDS-CR~wopL --~-KAIZKVGTFU?--CSTSXWKV~S --SOETRII-KD~VGSSS--CKTSQEVLJ~

w

...

I(FGPRHFSVLKTTGSLVSSrrSSSS~TATFSQTS---------

I(KVLLAAALIAGSVFFLGPSAADEKKKGP------------------

/124/ /124/

/96/ /74/ /65/

/41/ /38/ /39/ /39/ /39/ /73/

/41/

/41/ /41/ /41/ /41/

/1/ /1/

i;;

/1/ /I/

/1/ /1/

/1/ /1/ ./I/ -. /1/

I

Il

.

NUCLEOTIDE SEQUENCE OF THIRI) CYCLOPHILIN-HOMOLOGOUS GENE FROM S.CEREVISIAE

cleaved producing peptides of 20 amino acids of significant similarities (Figure 4). Moreover, both Scc3 and ninaA present a possible transmembrane helix domain, which also shares significant similarities, at their carboxy termini (Figure 4). This suggests that they could be anchored in a membrane structure. Indeed, ninaA is processed in vitro and translocated into canine microsomal membranes, possibly with its carboxy terminus anchored in the membrane (Stamnes et al., 1991). In these respects, it is worth mentioning that one of the two forms of the CyP from N . crassa contains a signal peptide which is processed by proteolysis and then translocated in the mitochondria (Tropschug et al., 1988). All these findings suggest the transmembrane sequence at the carboxy end of Scc3 (Figure 2) would anchor the molecule in the membrane. The catalytic domain would then be localized outside of this membrane. In addition to those homologies, ninaA is known to be glycosylated in vitro (Stamnes et al., 1991) and Scc3 has a potential glycosylation site (Figure 2). A model for ninaA (a component of the Drosophila photoreceptor cells) function is that it promotes isomerization of a(some) X-P bond(s) of rhodopsin which may assist the correct membrane intercalation of the latter (Singer, 1990). Whether Scc3, has a related function with a specific substrate to that of ninaA remains to be studied. A final, and obvious, question is whether SCC3 and ninaA could functionally replace each other in a heterologous organism. ACKNOWLEDGEMENTS We thank Dr Steve G. Oliver for the gift of h clones and to Mrs A. Martin for expert technical assistance. This work was supported by the EEC under the BAP programme and research grants from the CAICYT (BIO 88 - 0614 - E) and Fundacidn Ramdn Areces (Madrid), and BAP-0348-B (GDF) (Gent). R.C. is a research director at the Belgian Fund for Scientific Research. L.F. holds a Fellowship for Doctores y

977

Tecnologos Extranjeros en Espaiia from the Ministerio de Educacion. REFERENCES Bergsma, D.T. and Sylvester, D . (1990). A Chinese hamster ovary cyclophilin cDNA sequence. Nucl. Acids Res. 18,200. Danielson, P.E., Forss-Petter, S., Brow, M.A., Calavetta, L., Douglass, J., Milner, R.J. and Sutcliffe, J.G. (1988). plB15: a cDNA clone of the rat mRNA encoding cyclophilin. DNA 7,261-267. de Martin, R. and Philipson, L. (1990). The gene for cyclophilin (peptidyl-prolyl cis-trans isomerase) from Schizosaccharomyces pornbe. Nucl. Acids Res. 18, 4917. Devereux, J . , Haeberli, P. and Smithies, 0. (1984). A comprehensive set of sequence analysis programs for the VAX. Nucl. Acids Res. 12,38-395. Dietmeier, K. and Tropschug, M. (1990). Nucleotide sequence of a full-length cDNA coding for cyclophilin (peptidyl-prolyl cis-trans isomerase) of Saccharomyces cerevisiae. Nucl. Acids Res. 18, 373. Dumont, F.J., Staruch, M.J., Koprak, S.L., Melino, M.R. and Sigal, N.H. (1990). Distinct mechanisms of suppression of murine T cell activation by the related macrolides FK-506 and rapamycin. J. Immunol. 144, 251-258.

Gasser, C.S., Gunning, D.A., Budelier, K.A. and Brown, S.M. (1990). Structure and expression of cytosolic cyclophilin/peptidyl-prolylcis-trans isomerase of higher plants and production of active tomato cyclophilin in Escherichia coli. Proc. Nut. Acad. Sci. USA 87, 95 19-9523. Haendler, B., Hofer-Warbinek, R. and Hofer, E. (1987). Complementary DNA for human T-cell cyclophilin. EMBO J. 6,947-950. Haendler, B., Keller, R., Hiestand, P.C., Kocher, H.P., Wegmann, G . and Movva, N.R. (1989). Yeast cyclophilin: isolation and characterization of the protein, cDNA and gene. Gene 8 3 , 3 9 4 6 .

Figure 4 Alignment of the amino acid sequences of known and putative cyclophilins. The alignment of amino acid sequences was performed by hand, based on multiple binary comparisons from the COMPARE program (Devereux et al., 1984). Compared CyPs were: HUCI, human T-cells (Haendler et al., 1987); CHC, Chinese hamster ovary (Bergsma and Sylvester, 1990); RAC, rat (Danielson er a / . , 1988); TOC, BNC and ZMC, tomato, Brassica mpus and maize, respectively (Gasser et al., 1990); EGC, Echinococcus granulosus (Lightowlers er al., 1989); Scc 1, S.cerevisiae (Dietmeier and Tropschug, 1990); SPC, Schizosaccharornyces pornbe (de Martin and Philipson, 1990); CAC, Candida albicans (Koser er al., 1990a); HUC2, human endothelid cells (Price et al., 1991); NCC, Neurospora crassa (Tropschug, 1990); Scc2, S.cerevisiae (Koser er al., 1990b); DMC, ninaA from Drosophila rnelanogaster (Schneuwly et al., 1989; Shie er a/., 1989). In addition, ROT is the rotamase from E . coli (Kawamukai et al., 1989). Open letters on black background indicate invariant amino acid residues in all CyPs. Bold letters indicate highly conserved residues. (W) Amino acid residues of E. coli rotamase which are invariant with respect to the CyPs. Numbers in brackets indicate the number of amino acid residues which have been removed from the relevant sequences. Conserved and homologous amino acids from the putative amino signal peptide and the carboxy transmembrane domain of DMC and Scc3 are boxed.

L. FRANC0 ETAL..

Handschumacher, R.E., Harding, M.W., Rice, J. and Drugge, R.J. (1984). Cyclophilin: a specific cytosolic binding protein for cyclosporin A. Science 226, 544-546. Hattori, Y. and Sakaki, Y. (1986). Dideoxy sequencing method using denatured plasmid templates. Anal. Biochem. 152,232-238. Kawamukai, M., Matsuda, H., Fujii, W., Utsumi, R. and Komano, T. (1989). Nucleotide sequence of f i c and fic-1 genes involved in cell filamentation induced by cyclic AMP in Escherichia coli. J. Bacteriol. 171, 4525-4529. Kiefhaber, T., Quaas, R., Hahn, U. and Schmid, F.X. (1989). Folding of ribonuclease T1.l. Existence of multiple unfolded states created by proline isomerization. Biochemistry 29,3053-3061. Kino, T., Hatanaka, H., Miyata, S., Inamura, N., Nishiyama, M., Yajima, T., Goto, T., Okuhara, M., Kohsaka, M., Aoki, H. and Ochiai, T. (1987). FK-506, a novel immunosuppressant isolated from Streptomyces. J. Antibiotics 40, 1256-1265. Koltin, Y., Faucette, L., Bergsma, D.J., Levy, M.A., Cafferkey, R., Koser, P.L., Johnson, R.K. and Livi, G.P. (1991). Rapamycin sensitivity in Saccharomyces cerevisiae is mediated by a peptidyl-prolyl cis-trans isomerase related to human FK506-binding protein. Mol. Cell. Biol. 11, 1718-1723. Koser, P.L., Livi, G.P., Levy, M.A., Rosemberg, M. and Bergsma, D.J. (1990a). A Candida albicans homolog of a human cyclophilin gene encodes a peptidyl-prolyl cis-trans isomerase. Gene 96, 189-195. Koser, P. L., Sylvester, D., Livi, G. P. and Bergsma, D.J. (1990b). A second cyclophilin-related gene in Saccharomyces cerevisiae. Nucl. Acids Res. 18, 1643. Lang, K., Schmid, F.X. and Fischer, G. (1987). Catalysis of protein folding by prolyl isomerase. Nature 329, 268-270. Lightowlers, M.W., Haralambous, A. and Rickard, M.D. (1989). Amino acid sequence homology between cyclophilin and a cDNA-cloned antigen of Echinococcus granulosus. Mol. Biochem. Parasitol. 36,287-289. Liu, J. and Walsh, C.T. (1990). Peptidyl-prolyl cis-trans isomerase from Escherichia coli: A periplasmic homolog of cyclophilin that is not inhibited by cyclosporin A. Proc. Nat. Acad. Sci. USA 87,40284032. Liu, J., Chen, C-M and Walsh, C.T. (1991). Human and Escherichia coli cyclophilins: sensitivity to inhibition by the immunosuppressant cyclosporin A correlates with a specific tryptophan residue. Biochemistry 30, 2306-2310. Mortimer, R.K. and Child, D. (1985). Genetic map of Saccharomyces cerevisiae, Edition 9. Microbiol. Rev. 49,181-212.

Price, E.R., Zydowsky, L.D., Jin, M., Baker, C.M., Mckeon, F.D. and Walsh, C.T. (1991). Human cyclophilin B: A second cyclophilin gene encodes a peptidyl-prolyl isomerase with a signal sequence. Proc. Nat. Acad. Sci. USA 88, 1903-1907. Rigby, P.W., Dieckmann, M., Rhodes, C.J. and Berg, P. (1 977). Labelling deoxyribonucleic acid to high specific activity in vitro by nick translation with DNA polymerase I. J. Mol. Biol. 113,327-351. Sanger, F., Nicklen, S. and Coulson, A.R. (1977). DNA sequencing with chain-terminating inhibitors. Proc. Nat. Acad. Sci. USA 74,5463-5467. Schneuwly, S . , Shortridge, R.D., Larrivee, D.C., Ono, T., Ozaki, M. and Pak, W.L. (1989). Drosophila ninaA gene encodes an eye-specific cyclophilin (cyclosporin A binding protein). Proc. Nat. Acad. Sci. USA 86, 5390-5394. Shieh, B.-H., Stamnes, M.A., Seavello, S., Harris, G.L. and Zuker, C.S. (1989). The ninaA required for visual transduction in Drosophila encodes a homologue of cyclosporin A-binding protein. Nature 338,67-70. Siekierka, J. J., Hung, S.H.Y., Poe, M., Lin, C.S. and Sigal, N.H. (1989). A cytosolic binding protein for the immunosuppressant FK506 has peptidyl-prolyl isomerase activity but is distinct from cyclophilin. Nature 341,755-757. Siekierka, J.J., Wiederrecht, G., Greulich, H., Boulton, D., Hung, S.H.Y., Cryan, J., Hodges, P.J. and Sigal, N.H. (1990). The cytosolic-binding protein for the immunosuppressant FK-506 is both a ubiquitous and highly conserved peptidyl-prolyl cis-trans isomerase. J. Biol. Chem. 265,2101 1-21015. Stamnes, M.A. and Zuker, C.S. (1990). Peptidyl-prolyl cistrans isomerases, cyclophilins, FK 506-binding protein, and ninaA: four of a kind. Curr. Op. Cell Biol. 2, 1104-1107. Singer, S.J. (1990). The structure and insertion of integral proteins in membranes. Ann. Rev. Cell Biol. 6, 247-296. Stamnes, M.A., Shieh, B.-H., Chuman, L., Harris, G.L. and Zuker, C.S. (1991). The cyclophilin homolog ninaA is a tissue-specific integral membrane protein required for the proper synthesis of a subset of Drosophila rhodopsins. Cell 65,219-227. Takahashi, N., Hayano, T . and Sukuki, M. (1989). Peptidyl-prolyl cis-trans isomerase is the cyclosporinA-binding protein cyclophilin. Nature 337,476-478. Tropschug, M. (1990). Nucleotide sequence of the gene coding for cyclophilin/peptidyl-prolylcis-trans isomerase of Neurospora crassa. Nucl. Acids. Res. 18, 190. Tropschug, M., Nicholson, D.W., Hartl, F.W., Kohler, H., Pfanner, N., Wachter, E. and Neupert, W. (1988).

NUCLEOTIDE SEQUENCE OF THIRD CYCLOPHILIN-HOMOLOGOUS GENE FROM S. CEREVISIAE

Cyclosporin A-binding protein (cyclophilin) of Neurospora crassa. One gene codes for both the cytosolic and mitochondria1 forms. J . Biol. Chem. 263, 14433-14440. Tropschug, M., Barthelmess, I.B. and Neupert, W. (1989). Sensitivity to cyclosporin is mediated by cyclophilin in Neuropora crassa and Saccharomyces cerevisiae. Nature 342,953-955.

979

von Heijne, G. (1986). A new method for predicting signal sequence cleavage sites. Nucl. Acids R e s . 14, 46834690. Wiederrecht, G., Brizuela, L., Elliston, K., Sigal, N.H. and Siekierka, J.J. (1991). FKBl encodes a nonessential 506-binding protein in Saccharomyces cerevisiae and contains regions suggesting homology to the cyclophilins. Proc. Nat. Acad. Sci. USA 88, 1029-1033.