*Department of Chemistry, Faculty of Science, and tResearch Laboratory for Genetic Information, Kyushu ... first, second, and third positions of the triplet codons.
Proc. Nad. Acad. Sci. USA Vol. 89, pp. 8557-8561, September 1992 Biochemistry
Unusually high conservation of untranslated sequences in cDNAs for Trimeresurusflavoviridis phospholipase A2 isozymes TOMOHISA OGAWA*, NAOKO ODA*, KIN-ICHI NAKASHIMA*, HIROYUKI SASAKIt, MASAHIRA HATrORIt, YOSHIYUKI SAKAKItt, HIROSHI KIHARA§, AND MOTONORI OHNO*1 *Department of Chemistry, Faculty of Science, and tResearch Laboratory for Genetic Information, Kyushu University, Fukuoka 812, Japan; *Institute of Medical Science, University of Tokyo, Tokyo 108, Japan; and §Biotechnology Research Laboratories, Takara Shuzo Co. Ltd., Shiga 520-21, Japan
Communicated by Christian B. Anfinsen, May 21, 1992
ABSTRACT As a step toward understanding the structure and function of phospholipases A2 (PLA2s), we isolated and sequenced several cDNAs encoding Trimeresurus flavoviridis venom PLA2 isozymes including two [Lys49JPLA2s called basic proteins I and II, [Thr37]PLA2, and PLX'-PLA2. Comparison of the nucleotide sequences of these cDNAs with the previously isolated [Asp49'PLA2 cDNA revealed some interesting findings from the viewpoint of evolution. First, the homologies of the 5' and 3' untranslated regions (98% and 89%, respectively) were much higher than that ofthe protein-coding regions (67%). The predicted secondary structure showed the characteristic stemloop structures for both the untranslated regions of the mRNAs, suggesting that these regions play some functional role(s) in translation or stability of mRNAs. Second, base substitutions appeared to have occurred at similar rates for the three positions of codons among these PLA2s. The results are discussed in terms of evolution of PLA2s. Northern blot analysis showed that these PLA2s are specific to venom gland.
base substitutions have occurred at similar rates among the first, second, and third positions of the triplet codons.
EXPERIMENTAL PROCEDURES Materials. Restriction endonucleases and other modifying enzymes were obtained from Takara Shuzo (Kyoto). Taq DNA polymerase was from Promega. [a-32P]dCTP (3000 Ci/mmol; 1 Ci = 37 GBq), [y-32P]ATP (5000 Ci/mmol), and [a-[35S]thio]dATP (1000 Ci/mmol) were from Amersham. Other reagents were of reagent grade. Oligonucleotide Synthesis. Oligonucleotides for polymerase chain reaction (PCR) or hybridizations (Fig. 1) were prepared on a DNA synthesizer (Applied Biosystems model 380A) and purified on a C18 Sep-Pak cartridge (Waters). Venom Gland cDNA Library. All specimens of T. flavoviridis were collected on Tokunoshima Island, Kagoshima Prefecture, Japan. Construction of the T. flavoviridis venom gland cDNA library in Agt 10 was reported previously (19). PCR. DNAs were amplified by PCR using the total cDNA library as a template and the two oligonucleotide primers corresponding to the N- and C-terminal sequences of [Asp491PLA2 (primers 1 and 3 in Fig. 1) or basic proteins I and II (primers 4 and 6 in Fig. 1). After subcloning of the PCR products into pUC119 vector, the amplified DNAs were confirmed both by Southern blotting with internal probes (probe 2 or 5 in Fig. 1) and by partial sequencing. They were employed for screening of cDNA library after being radiolabeled with the Multiprime DNA labeling system (Amersham). Cloning of PLA2 cDNAs. For screening of cDNAs for basic proteins I and II, about 9 x 105 plaques from the T. flavoviridis venom gland AgtlO cDNA library were screened with the PCR-derived probe specific for basic proteins I and II. Hybridization was performed overnight at 600C and the final wash was carried out with 0.1 x standard saline citrate (SSC) plus 0.1% SDS at 650C to give 431 positive clones (BP-1 to -431). Dot blot analysis for 91 clones (BP-1 to -91) with the N-terminal oligonucleotide probe (primer 4 in Fig. 1) yielded no positive clones, indicating that full-length cDNA clones coding for basic proteins I and II were very few. Therefore, PCR was employed for more efficient screening. The residual 340 clones (BP-92 to -431) were divided into 34 sets and PCR was carried out for each set of clones, using the N- and C-terminal primers (primers 4 and 6 in Fig. 1) for basic proteins I and II. The sets containing clones that were not amplified were excluded. Each clone in the sets containing a clone or clones that were amplified was further subjected to
Phospholipase A2 (EC 3.1.1.4) (PLA2) catalyzes the hydrolysis of the 2-acyl ester linkage of 3-sn-phosphoglycerides with the requirement of Ca2+. To date, the primary structures of >80 PLA2s have been determined (1-9). Aspartate-49 (numbered according to the aligned numbering of PLA2s of various sources) was found to constitute a part of the Ca2+ binding site (10). The PLA2s are structurally classified into two groups, I and 11 (11). The group II enzymes can be separated into two subgroups: [Asp49]PLA2, with high activity, and [Lys49]PLA2, with extremely low activity (5, 6, 12). A variety of physiological activities such as neurotoxicity (13), cardiotoxicity (14), and anticoagulant activity (15) have been reported for PLA2s. It has become evident that PLA2s play a crucial role in the pathogenesis of inflammation (16-18). Thus, structure-function studies of PLA2s are of importance. An [Asp49]PLA2 (19) and two [Lys49]PLA2s called basic proteins I and 11 (5, 6), which belong to group II, have been isolated from Trimeresurus flavoviridis (Habu snake) venom and sequenced. Since these T. flavoviridis PLA2 isozymes are structurally homologous (5, 6) and thus provide a useful system for understanding the structurefunction relationships of PLA2s, cloning of cDNAs for these PLA2s has been conducted. Here we report the cloning of cDNAs encoding four T. flavoviridis PLA2 isozymesl including basic proteins I and II and two other PLA2s. Comparison of these cDNAs with [Asp49]PLA2 cDNA (19) revealed that the nucleotide sequences of the untranslated regions (UTRs) of PLA2 mRNAs are much more conserved than those of the protein-coding regions. Roles of the highly conserved UTRs of mRNAs are discussed based on prediction analysis of their secondary structures. We also show that
Abbreviations: PLA2, phospholipase A2; UTR, untranslated region. ITo whom reprint requests should be addressed at: Laboratory of Biochemistry, Department of Chemistry, Faculty of Science, Kyushu University 33, Fukuoka 812, Japan. "The sequences reported in this paper have been submitted to the GenBank/EMBL/DDBJ data base (accession nos. D01232, D01233, D01234, and D01239).
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact. 8557
8558
Proc. Natl. Acad. Sci. USA 89 (1992)
Biochemistry: Ogawa et al. amino acid
base
3W-9I
20
47H-52G
17
11N9_21p
23
3' 6W-_3T
23
47H-53K
20
oligonucleotide
No.
sequence
Each sample (10 jug) was denatured by the glyoxal/dimethyl sulfoxide method (22) and electrophoresed in a 1.1% agarose gel. After blotting to Hybond-N+ nylon membrane (Amersham), the electrophoretically separated RNAs were hybridized with labeled cDNA fragments (probes 3, 5, 6, 7, and 4 in Fig. 2) in 50% formamide hybridization buffer at 500C, washed with 0.1x SSC/0.1% SDS at 650C, and exposed to x-ray film for 3 days at -800C or to the imaging plate of a Fuji Bio Image analyzer, BAS 2000, for 1 hr at room temperature. Data Analysis. The DNASIS package developed by Hitachi Software Engineering was used for analysis of DNA sequences. The GENAS system (23) at Kyushu University Computer Center was employed to search the EMBL/ 1990) for data bank genetic sequence GenBank the(November, secondary structures and to analyze sequence homology of mRNAs according to the method of Zukker and Steigler
Asp-49-phospholipase A2 1
sense
primer
5'-TGGCAGTTTGAGAATATGAT-3' A
2
internal probe
antisense primer
A
C
5'-CCGTAGCAGCAGTCGTG-3' A
3
C
A
A
A
A
5'-GGTTCGCTGTCTTCTTGGCAGTT-3 C A A C C A A T C
basic proteins I and II 4
sense primer
5'-TGGAAGATGATATTTCAGGAGAC-3 T C A A A C
5
internal probe
5'-TTTTTGTAGCAGCATTTGTG-3' C
6
antisense primer
A
A
A
C
A
5' -CAGGTGTCGGCTTTTTTGCA-3' A A A C C A T T C C
16Cc-.22 c
20
(24).
FIG. 1. Structures of the oligonucleotides us ed as probes and PCR primers. The partial amino acid sequences ch4osen for syntheses of the oligonucleotides are indicated. The third boases of the triplet The amino codons for the C-terminal amino acids were rem acid sequences of [Asp49]PLA2 and basic proteins in Fig. 3.
RESULTS Cloning of cDNAs Coding for Basic Proteins I and HI. The first screening of about 9 x 105 plaques from the T. flavoviridis venom gland cDNA library with the PCR-derived probe for basic proteins I and II provided 431 positive clones. When further selection was carried out with the PCR technique, three nearly full-length clones (BP-109, -174, and -201) were obtained. Clone BP-109 contained a 5' UTR of 297 base pairs (bp) followed by a single open reading frame of 414 bp, which codes for basic protein II, and a 3' UTR of 108 bp (Fig. 3). The nucleotide sequence of clone BP-201 encoded basic protein I and contained a 5' UTR of 67 bp, a single open reading frame of 414 bp, and a 3' UTR of 108 bp (Fig. 3). The nucleotide sequence of clone BP-174 was identical to that of clone BP-109 (encoding basic protein II) except that the 5' UTR was shorter by 126 bp. The amino acid sequences of basic proteins I and II predicted from the cDNA sequences were completely in agreement with those of the proteins determined previously (7, 8). Thus, Leu (CTG) at position -5, which is located in the signal sequence, and Asp (AC) at position 58 ofbasic protein I are replaced by Val (GTG) and Asn (AAC), respectively, in basic protein II. It is evident that substitution at position 58 (Asp/Asn) based on the protein
ioved.
PCR. Thus, only the clones to be amplifiied were finally selected. We call this an "exclusion PCR te chnique." An attempt was made to obtain cDNrAs coding for [Asp49]PLA2s other than the one already obitained (19). The cDNA library was screened with the PCR--amplified DNA probe specific for [Asp49]PLA2 under mild wiashing condition (lx SSC/0.1% SDS, 650C). In addition, atbout 2.5 x 105 phage clones were screened with cDNA fragiments (probes 1, 2, and 7 in Fig. 2) with the sequences of tlhe conserved 5' UTRs of T. flavoviridis PLA2 mRNAs. Sequence Analysis. The cDNA inserts exe:ised from positive phage DNAs with EcoRI were subclone4d into pUC119. cDNA restriction maps are shown in Fig. 2. The nucleotide sequences were analyzed by the dideoxy c}hain-termination method after preparation of template as described by Hattori and Sakaki (20), using M13 universal and ra eversed primers and synthetic oligonucleotide primers. RNA Blot Analysis. Total RNA was exitracted from T. flavoviridis tissues by the method of Chirggwin et al. (21). EO
A
PLA2
-L probe I
pr -e
AAA-
probe 4
probe 3
probe 2
ISPI
I __,!0,\\\
-4--T--
BP2
F-
BP2 (clone BP-25) probe 5 B
PLX
\AAAA-
probe 6 EO
A
T-PL
K' AAAAA-
..
probe 7
FIG. 2. Restriction maps of cDNA clones for T.flavoviridis PLA2 isozymes. PLA2, [Asp49]PLA2 cDNA; BP1, basic protein I cDNA; BP2, basic protein II cDNA; PLX', PLX'-PLA2 cDNA; and T-PL, [Thr37]PLA2 cDNA. Protein-coding regions and UTRs are shown by open box and line, respectively. Stippled boxes represent the signal-sequence domains. Restriction sites: EO, EcoO109; A, Acc I; K, Kpn I; P, Pst I; B, Bc! I. The EO and B sites were found by EcoO109 and Bc1 I digestion, respectively, of the cDNA inserts that were excised from the pUC119 plasmids with EcoRl. Probes 1-7 employed for screening of the cDNA library and for Northern and Southern blot analyses are indicated.
Biochemistry: Ogawa et al.
Proc. Nat!. Acad. Sci. USA 89 (1992)
OP2
AGCCCAAGCCCCCCA
BP2
GTACTTTTGTCACTAATA.ACAGAAACCTGCCAMCT7CAAGCCCAACATGGCC...AAAGAA....ACC.C..A -A
PLA2
T-PLA77-C7CCC BPI BP2
GTGACCTCGAGGAGCAGAGGGAGCCTGACAGGCCCGAATAGCCACATcCC:,TC ...C....
AAC CCC AAA
CC
CC'~ CCC AAC TAC
CAC ACC TCC AAC AAC CCC CAT ACC GCC
Cys Assi Pro Lys tLeu C~y Lys Cyr Chr Tyr Ser Tsp Asn Asn Cly Asp CIle Val
T-PL
.C..........................
..
BPI
G.-
..
A.C .A. .CC
C. ...
...
AA.
.CC.
Ser Lys Ps et Asp Ser :.ys A~a ....T F2 .........A.GCA. C.CC .. C... AA. CC .... Met A. T.C G..C... puI Lye SerC.CG.G. As-pLi ..C Csp Asp Tyr .C........ Clu ...
Asp
T-PL
ACC
8559
Ser
Ser
Lys
Ala
TTCTCCTTCTGATCCTTCCCTACAGGTCATCCCTC-ACTTACMCACCCAACC~mTT GGCGTC:C:ACAC
BPI
PLA2
CCC
CAC
CCC
CCC
TCC
AA
Asp
Cly
Pro
Cys
Lys
CAA
CIlo
Cys
Cly
CA
Glu
Cys
VaC
CA
CCC
CAC
Cls
Cys
Asp
A
CCC
Arg Ala
T-PL
5'UT~PLA2 T-PL
CCCCTCACCAGCGGAGCCAACC-ACCCGGCCCCGCCA:TCCCCAZ:7C-CGA7:CCCGAG
G
A
A. A.A AA. .A CCA C.. CCC ..C. ..
BPI
Cly
CACTC-TCCAA ...............................
Clu
.A.
Lye
Cly
MA.
Lys
Asn
Clu
G PLX
Ass
A.A
BP2
A.
Gly
.A
GCI
Pro
Leo,
CCA
..CCC..
Pro
Leu
... ..ACC
Lys
C
.C..........A
Cls Lys ...............A..
Thr
Asn
A
Lye
coding region ATG ACC Met Arq
PLA2
signal sequence ACT
CCC
Thr- L.eo,
CGC rp
ATA
Ile
ACC GCC
GCC TCC CCCG
Me, Ala Val
L-e-
Le-.
CCC
Vat'
CGG
GGC CTC CAT
GGC
CCG
GCA Ala
CCA
PLA2
Ct-y Va- As-- Cly Gly L-e.-
Ala
CCC
CCC
Cys
PIhe
ACC
Cle
CGA GAC
AAT
Arg
Ass
Asp
CTG GAC Leu Asp
ACA
TAC
GAC
ACA
AAC
AMA
TAC
Thr
Tyr
Asp
Arg
Ass
Lye
Tyr
Tsp
T-PL
TCC
BF1
.C
C..
Vat
Val
.C
T
CA.
...
Cly
A..
.A.
Ase
Lye
.AC. .
Thr
.C......G...
C.....A A.. A. .. .AC. .
SF2
XT....C.
.C..G
.A
Clu
Lou
Ser PIE'
Cbu Ase Cly Lye .C .............A.G A.C A. .CG. Leu
Lye
Lys
Lye
C
MA TCC
Cln
Tsp
Phe
CAC
MA
ATC
ACC
Clu
Ass
Met
Itoe
ATC Cie
A
CCCG
MA
Lye
Vat
A
Val :tys
MA
ACC
ACA
Gly
Lye
CCC
tile
CCC
PLA2
Sos
CCC
CAC
CCC
Arg
Cyr
Pro
CCC
Ala
CCC
Ser
AAC
TCC
CAC
CAT
TCA
Ass
Cys
Gln Glu Asp
Ser
CAC
CMA Cls
AT.
Arg
coding region
pLA2
Thr Met
UT CCA Pro
CCC
TAACTCTCTGCAGGC
C yeses
T-PL
Bpi
TCC
.G
Val
Leu
CTCC
SF2
Leu
VaC
PLC'C
G .....
Crp
Lye
C.
.-
Crp
...
Arg
TCt Cyr
A.
C..
Phe
C. C.. C Phe Lye
AC...C.G
Lou
PLA2
-GCC
AC
Lye
C..
CCC
CAC
CCA
CCC
TAC
CCC
Ses
Ala
Cyt
Cty
Cys
Cyr
Cys
-A.
Clu
C~r
CC.
Cty
AC.
CC.
Tht
Cly
Met
Thr
CCC
AA
.C.
Clui
Ala
AA
.C.
Cbu
...A.-AC.CCGA
Lye
ACT
AC.
Cin GC-.;
CCC
CCA
CCC
Arg
Cly
Lys
MA
P.MA Ala L~ys
CC.
Clu
Cly Tsrp Cly Gly
A
Ata
CCC
Ata
G.AC
Cly
GCC
AAC
SF2 Ass
Vat
MCS Lye
FIX.
CCA
AAC
CAC
Pro
Lye
Asp
S.C
Cly .C
SLy
CCC
FIX'
Ala
TTA.......A.
T.
Asn
Leu
..A.GC............
PLA2
Leu
A.C
Ass
Asp
Pro
Pro
AT.
Cle
CT...
A..
CT...
A..
Lye
Lye
A..
A..
Phe
.T
Ala Asp .T
Ala
Phe
Lye
TCC
AC.
..C
CC.
A..
Phe
Thr
Asp
Pro
Tihr
Lye
Asp
...T..... A.C Thr
...T..... A.C Thr
G A..M. Lye
CAGGAAAA.ACTCCTrCAATTACACMATCGTAGTTTGTGTTACTCTATTTTTTTGAATGCAATACTGAGTAAT
FLIX'
CC.....C...........A... CC.....C...........A... G... C.CC0............
..A..C.A........A--....
..A.C ........A.... ....
........A.-C....
Arg
Vat
TTA.......A. Vat
............
FLA2
Arg
.A................. CT .M........ Phe
Pie
C..
C..
T-PLI
SF2
CP2
Lye
ACTr.....AMA Cle Lye ACT A. T.C Thr
Bpi Bpi
ACT.....AMA Cbe
AAC
.C..C
Pro
Dpi
Ass
Lye
AMCAClAGTGCCACCTCTCCACTCAAAAAMAAA
T-PL Bp..F. F2
FIX'.
PLA2 T-PL
GCC
ACC
CAC
CCCG
Ala
Thr
Asp
Arg
CCG
TG C
Cys
Cys
CCC
CAC
Phe
Val Hic
Cys
TAC
CCA
Cyr
Cly
Cys
A
MA
Lye
ACC Vat
Thr
GCC Sly
..
BI
..T
..T
A..
.C
.A
SP
.T
..T
A..
.C
.A
Ses
.....
MA.........
Tyr
Lye
.C.......
FIX'
AA.........
.... .... A.........
[Asp49]PLA2
FIG. Nucleotide sequences of cDNAs and the corresponding amino acid sequences for Tflavoviridis PLA2 isozymes. PLA2, cDNA; T-PL, [Thr37]PLA2 cDNA; BP1, basic protein I cDNA; BP2, basic protein H cDNA; PLX', PLX'-PLA2 cDNA. Nucleotide residues identical to those of [Asp49]PLA2 cDNA are indicated by dots. The 5' UTRs of PLA2 cDNAs that were longer than the 5' UTR of cDNA were compared with the 5' UTRs of [Thr37]PLA2 cDNA, basic protein I cDNA, and basic protein cDNA in this order in the upstream
[Asp491PLA2
Ca2+
direction. Underline indicates the putative polyadenylylation signal, AATAAA. Hatched box shows the binding site at position 49 (48 in the above sequences) in the proteins. Stippled box shows the amino acid residues at position 58 that differ between basic proteins I and II. sequences of basic
proteins
I and
II1(7,
8) is due
to
dent mRNAs but not modification after translation
indepen-
or
artifact
during purification. Cloning of cDNAs Coding for Other PLA2 Isozymes. Comparison of the nucleotide sequences of the cDNAs encoding [Asp49]PLA2 and basic proteins I and II showed that the UTRs are much more homologous than the coding regions. This is of great interest because usually the coding regions are more
conserved than the UTRs (25-28).
observation further,
an
attempt
was
To confirm the
made to obtain cDNAs
encoding [Asp49]PLA2 isozymes other than the one already obtained (19). When the cDNA library was screened with the PCR-amplified [Asp4PLA2 DNA probe under mild washin conditions, 136 positive clones (PL-1 to -136) were obtained. Clones
were
further selected based
on
two criteria:
(i) the
fragment excised with EcoRI is longer than the [Asp49IPLA2 cDNA obtained previously (19) and (ii) hybridization profiles against the three oligonucleotide probes (primers 1 and 3 and probe 2 in Fig. 1) are different from those of cDNA. Thus, cDNAs encoding two more PLA2 isozymes
[Asp491PLA2
were
obtained. Clone PL-14 encoded
deduced amino acid sequence PLX PLA2 isolated from T.
was
a
PLA2
isozyme
whose
similar to that of TEY-
flavoviridis
by Kini
et a!.
(29) but with substitutions of 9 amino acid residues. It
was,
venom
therefore, named PLX'-PLA2. Clone PL-53 encoded
an
(Asp49]PLA2
isozyme identical to the reported previously at position (19) except for one base substitution: lysine 37 of was replaced by threonine This The cDNA library was isozyme is designated also screened with DNA fragments corresponding to the 5' UTRs (probes 1, 2, and 7 in Fig. 2). The resulting 58 positive clones (5'UT-1 to -58) were selected with three oligonucleotide probes (primers 1 and 3 and probe 2 in Fig. 1) and with
(MAG) (ACG).
[Asp491PLA2
[Thr37]PLA2.
the
same
criteria
as
described above. Clone 5'UT-3 encoded
PLX'-PLA2 but the 5' UTR when were
compared with clone
was
shorter
by 18 nucleotides
PL-14. Clones 5'UT-6 and -19
identical with clone PL-53 but with different
poly(A)+
7PLA2 exist in venom gland. Clone 5'UT-22 390-bp insert coding for [Asp4]PLA2 but with a
PLA2 and [Thr
contained
lengths of
tail. These results suggest that mRNAs for PLX'-
a
8560
Biochemistry: Ogawa et al.
Proc. Natl. Acad. Sci. USA 89 (1992)
silent mutation for the Leu residue at position 61 (CTC to
CT@). Comparison of cDNAs Coding for T. flavovridis PLA2 Isozymes. The nucleotide sequences of cDNAs for five T. flavoviridis PLA2 isozymes were compared. As shown in Fig. 3, the UTRs are much more homologous than the protein-coding regions: nucleotide identities are 98% (59/60) for the 5' UTRs, 67% (281/414) for the coding regions, and 89%o (97/109) for the 3' UTRs. Thus, the unusually high conservation of the 5' and 3' UTRs was confirmed. Since the 5' and 3' UTR sequences are generally less conserved than the coding regions (25-28), the results suggest that the UTRs in PLA2 genes have some functional roles. Northern Blot Analysis. The tissue distribution of mRNAs for basic proteins I and II was examined with basic protein I cDNA. Northern blot analysis showed no positive signal in the samples from the tissues other than venom gland (Fig. 4A). Thus, mRNAs for basic proteins I and II are considered to be specific only to venom gland as in the case of [Asp49]PLA2 mRNA (19). Fig. 4B shows autoradiograms illustrating hybridization profiles of venom gland and panb
A
0 ~~~~b,
S
0~~~~~~~~~~ ~
~
4
dP
z -e%4WC? r
28s
_
rRNA
1s8s rRNA
B V P
V P
V P
V P
V P
creas RNAs with the DNA probes from the coding regions of [Asp49]PLA2, basic protein II, and PLX'-PLA2 cDNAs and from the 5' and 3' UTRs (probes 3, 5, 6, 7, and 4 in Fig. 2, respectively). The major positive signals were at about 1.1 kb and were similar to one another.
DISCUSSION cDNAs encoding T. flavoviridis PLA2 isozymes were isolated and their nucleotide sequences were determined. Two cDNAs coding for isozymes we have named [Thr37]PLA2 and PLX'-PLA2 were found in addition to those of known PLA2s. The most interesting feature found in this study is the extremely high conservation of the 5' and 3' UTRs in the cDNAs (Fig. 3). The percent nucleotide sequence identities in the 5' UTRs, the coding regions, and the 3' UTRs are 98%, 67%, and 89o, respectively. To our knowledge, such greater sequence homology in the UTRs than in the coding regions has not been reported in cDNAs for other isozyme systems. Usually, less homology has been observed in the UTRs, as in various isoforms of the G-protein a-subunit family (25, 26) and the protein kinase C family (27, 28). The great homology in the 5' UTRs of cDNAs for T. flavoviridis PLA2 isozymes suggests that the 5' UTRs of mRNAs could play significant roles. It has been reported that secondary structures (hairpin, stem-loop, and pseudoknot structures) of the UTRs of mRNAs for the transferrin receptor and ferritin play roles in regulation of translation through interactions with protein factors and in mRNA stability (30, 31). The secondary structure prediction (24) for mRNA of [Thr37]PLA2 indicated that the 5' and 3' UTRs can form stem-loop structures (Fig. 5). mRNAs of the other isozymes were also predicted to form the similar structures. Such structures of mRNAs for T. flavoviridis PLA2 isozymes may be favorable for interaction with the elements necessary to initiate or terminate protein synthesis. For example, a loop located close to the initiation codon on a stem may permit formation of a stable initiation complex. The stem-loop structures of the 5' and 3' UTRs appear to contribute to stability. From the nucleotide sequences of cDNAs (Fig. 3), we note that the signal sequences consist of 16 amino acid residues and that the domains coding for these sequences are exceptionally highly conserved when compared with other portions of the coding regions. The high conservation may suggest that
.01
n codon
28s
18S rRNA
~~~~/_
1
rRNA
O
signal sequence
1
0
4x-
i
FIG. 4. (A) Northern blot analysis of RNAs extracted from the indicated tissues with basic protein I cDNA. (B) Northern blot analysis of RNAs extracted from venom gland (V) and pancreas (P) with DNA probes indicated. All probes are shown in Fig. 2. PLA2, cDNA for the coding region of [Asp49]PLA2 (probe 3); BP, cDNA for the coding region of basic protein H (probe 5); PLX', cDNA for the coding region of PLX'-PLA2 (probe 6); 5'UT and 3'UT, the 5' and 3' UTRs (probes 1 and 4), respectively. Arrows indicate the mobilities of 28S and 18S rRNAs as size markers.
poly A signal
3UTR
5'UTR
5'
FIG. 5. Predicted secondary structures of 5' and 3' UTRs of mRNAs coding for T. flavoviridis [Thr37]PLA2. Analysis was conducted with the GENAS program (23) according to the method of Zukker and Steigler (24). The 5' UTR and the N-terminal sequence domain including the signal sequence domain, as well as the 3' UTR, form stem-loop structures. No significant secondary structure is noted in the other parts of the coding region.
Proc. Natl. Acad. Sci. USA 89 (1992)
Biochemistry: Ogawa et al. the signal peptides share common roles in membrane penetration and susceptibility to a common signal peptidase. Examination of the nucleotide sequences of the coding regions of cDNAs revealed that substitutions have occurred in the first, second, and third positions of the triplet codons at similar rates: 45/140 (32.1%), 42/140 (30.0%), 40/140 (28.6%), respectively. This seems unusual compared with the other cases in which the silent (third) site of codons are more variable than the first and second positions, where changes cause amino acid changes (32). The above observation suggests that T. flavoviridis PLA2 isozymes have evolved without any evolutionary restrictions. Generally, nucleotides encoding amino acids essential for protein functions tend to be conserved, but this is not necessarily the case for genes coding for T. flavoviridis PLA2 isozymes. It is likely that the genes encoding basic proteins I and II, which have extremely low lipolytic activity, were produced from the gene encoding the highly active [Asp49]PLA2 through base substitutions including converting the Asp49 codon to a lysine codon. Such substitutions, otherwise, could afford the potential for generating proteins with new physiological activity. It is well known that PLA2s are widely distributed in various tissues and organs. In mammals, group I PLA2s occur in pancreas, lung, spleen, and stomach, while group II PLA2s are found in liver, spleen, intestine, and platelet. In contrast, basic proteins I and II and [Asp49]PLA2 (19) from T. flavoviridis were found to be expressed only in venom gland (Fig. 4). It can be assumed that control mechanisms for expression of PLA2s differ between venom gland and other tissues. The unique features of cDNAs for T. flavoviridis PLA2 isozymes prompted us to analyze the organization of the PLA2 family in the genome. Southern blot analysis of genomic DNAs suggested that genes for T. flavoviridis venom PLA2 isozymes were formed by gene duplication to generate a gene family (data not shown). This has been confirmed by recent work showing that two genes encoding [Asp49]PLA2s are arranged in tandem in the genome (unpublished data). This work was supported in part by Grant-in-Aids for Scientific Research (nos. 03453165, 03554021, and 9018) from the Ministry of Education, Science, and Culture of Japan. 1. Dufton, M. J., Eaker, D. & Hider, R. C. (1983) Eur. J. Biochem. 137, 537-544. 2. Davidson, F. F. & Dennis, E. A. (1990) J. Mol. Evol. 31, 228-238. 3. Gomez, F., Vandermeers, A., Vandermeers-Piret, M.-C., Herzog, R., Rathe, J., Stievenart, M., Winand, J. & Christophe, J. (1989) Eur. J. Biochem. 186, 23-33. 4. Schmidt, J. J. & Middlebrook, J. L. (1989) Toxicon 27, 805818. 5. Yoshizumi, K., Liu, S.-Y., Miyata, T., Saita, S., Ohno, M., Iwanaga, S. & Kihara, H. (1990) Toxicon 28, 43-54. 6. Liu, S.-Y., Yoshizumi, K., Oda, N., Ohno, M., Tokunaga, F.,
7. 8. 9. 10. 11.
12. 13. 14. 15. 16. 17. 18.
19. 20. 21. 22.
23. 24. 25. 26. 27.
28. 29. 30. 31.
32.
8561
Iwanaga, S. & Kihara, H. (1990) J. Biochem. (Tokyo) 107, 400-408. Takasaki, C., Yutani, F. & Kajiyashiki, T. (1990) Toxicon 28, 329-339. Oda, N., Nakamura, H., Sakamoto, S., Liu, S.-Y., Kihara, H., Chang, C.-C. & Ohno, M. (1991) Toxicon 29, 157-166. Pungercar, J., Kordis, D., Strukelj, B., Liang, N. S. & Gubensek, F. (1991) Toxicon 29, 269-273. Dijkstra, B. W., Drenth, J., Kalk, K. H. & Vandermaalen, P. J. (1978) J. Mol. Biol. 124, 53-60. Heinrikson, R. L., Krenger, E. T. & Keim, P. S. (1977) J. Biol. Chem. 252, 4913-4921. Maraganore, J. M., Merutka, G., Cho, W., Welches, W., Kezdy, F. J. & Heinrikson, R. L. (1984) J. Biol. Chem. 259, 13839-13843. Yang, C. C., King, K. & Sun, T. P. (1981) Toxicon 19,645-659. Barrington, P. L., Condrea, E., Soons, K. R., Yang, C.-C. & Rosenberg, P. (1984) Toxicon 22, 743-758. Verheij, H. M., Boffa, M. C., Rothen, C., Bryckaert, M. C., Verger, R. & deHaas, G. H. (1980) Eur. J. Biochem. 112, 25-32. Seilhamer, J. J., Pruzanski, W., Vadas, P., Plant, S., Miller, J. A., Kloss, J. & Johnson, L. K. (1989) J. Biol. Chem. 264, 5335-5338. Kramer, R. M., Hession, C., Johansen, B., Hayes, G., McGray, P., Chow, E. P., Tizard, R. & Pepinsky, R. B. (1989) J. Biol. Chem. 264, 5768-5775. Suwa, Y., Kudo, I., Imaizumi, A., Okada, M., Kamimura, T., Suzuki, Y., Chang, H. W., Hara, S. & Inoue, K. (1990) Proc. Natl. Acad. Sci. USA 87, 2395-2399. Oda, N., Ogawa, T., Ohno, M., Sasaki, H., Sakaki, Y. & Kihara, H. (1990) J. Biochem. (Tokyo) 108, 816-821. Hattori, M. & Sakaki, Y. (1986) Anal. Biochem. 152, 232-238. Chirgwin, J. M., Przybyla, A. E., MacDonald, R. J. & Rutter, W. J. (1979) Biochemistry 24, 5294-5299. McMaster, G. K. & Carmichael, G. G. (1977) Proc. Nati. Acad. Sci. USA 74, 4835-4838. Kuhara, S., Matsuo, F., Futamura, S., Fujita, A., Shinohara, T., Takagi, T. & Sakaki, Y. (1984) Nucleic Acids Res. 12, 89-99. Zukker, M. & Steigler, P. (1981) Nucleic Acids Res. 9, 133-148. Matsuoka, M., Itoh, H., Kozasa, T. & Kaziro, Y. (1988) Proc. Natl. Acad. Sci. USA 85, 5384-5388. Strathmann, M. & Simon, M. I. (1990) Proc. Natl. Acad. Sci. USA 87, 9113-9117. Ohno, S., Kawasaki, H., Imajoh, S., Suzuki, K., Inagaki, M., Yokokura, H., Sakoh, T. & Hidaka, H. (1987) Nature (London) 325, 161-166. Kubo, K., Ohno, S. & Suzuki, K. (1987) FEBS Lett. 223, 138-142. Kini, R. M., Kawabata, S. & Iwanaga, S. (1986) Toxicon 24, 1117-1129. Hentze, M. W., Caughman, S. W., Rouault, T. A., Barriocanal, J. G., Dancis, A., Harford, J. B. & Klausner, R. D. (1987) Science 238, 1570-1573. Leibold, E. A., Laudano, A. & Yu, Y. (1990) Nucleic Acids Res. 18, 1819-1824. Nei, M. (1987) Molecular Evolutionary Genetics (Columbia Univ. Press, Irvington-on-Hudson, NY).
