Vol. July
178, No. 31, 1991
2, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Pages 474-479
IDENTIFICATION OF THE +WBUNIT INTERACTION RETINAL CYCLIC-GMP PHOSPHODIESIERASE
SITES IN THE jWXJBUNIT
Brenda Oppert and Dolores J. Takemoto Department of Biochemistry Kansas State University Manhattan, KS 66506 Received
June
7, 1991
Using synthetic peptides, the identification of the retinal cyclic-GMP phosphodiesterase (cGMP PDE) interaction sites for the inhibitory r-subunit in the catalytic t-z-subunitwere recently localized to residues #16-30 and 78-90 in the o-subunit (1). In this study, a binding radioimmunoassay (RIA) showed a weak interaction between PDE7 and PDEB subunits in PDE/3 residues #15-34, and stronger interaction sites were found in residues W-110 and 211-230. Sequence comparison between PDEa and PDEB illustrate some differences in these regions, particularly in PDEo 16-30 and PDEB 15-34 regions. Differences in interaction sites in PDEcr and PDE/3 for PDEr may account for the differences 0 1991Academic Press, Inc. in affinities observed between PDE? and the catalytic subunits.
Bovine retinal cyclic-GMP phosphodiesterase (cGMP PDE) affects the level of cGMP in the outer segments, controlling the response of ion channels in the plasma membrane that are associated with cGMP (2,3). Rod outer segment PDE consists of an (Y (88 kDa) (4) and @ (5) (84 kDa) catalytic subunit, and two y (11 kDa) (6,7) inhibitory subunits. The two PDEy’s exhibit different binding affinities for PDEcra (8). In the bovine system, partial activation (5-1796) occurs due to loss of a low affinity PDE? when it complexes with the o-subunit of the rod outer segment G-protein, transducin, resulting in a soluble transducin-cr-PDE-y complex (8). Removal of the initial PDEy requires only stoichiometric amounts of transducin (7). Removal of the high affinity PDE7, resulting in full activation of the PDE, requires FM concentrations of transducin-cr (7). This activation does not result in removal of the high affinity PDE-y from the membrane and is probably due to the displacement of PDEy from the catalytic complex (8,9). Crosslinking studies have shown that PDEy can be linked to either PDEa or PDE/3 (15). The site on PDEy that interacts with PDEa/3 is located within residues 24-45 and at the C-terminus (residues 54-87) of PDEy (10-14). Based on sequence analysis comparisons between related phosphodiesterases, the N-termini of PDEcr and PDEj3 are ABBREVIATIONS: cGMP - guanosine 3’,5’-cyclic monophosphate, KLH - Keyhole limpet hemocyanin, PDE - phosphodiesterase, PMSF - phenylmethylsulfonyl fluoride, RIA radioimmunoassay, ROS - rod outer segment. 0006-291X/91 Copyright All rights
$1.50
0 1991 by Academic Press, Inc. of reproduction in any form reserved.
474
Vol.
178,
No.
2, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
suspected to contain regulatory interaction sites (4). Interaction regions for PDE-y on PDEo! were identified within PDEcY residues 16-30 and 78-90 (1) by several biochemical techniques, including binding radioimmunoassay (RIA).
Using similar binding assays and selected synthetic fl-peptides,
regions for PDEy in PDE@ are localized to PDE-beta residues #15-34, 91-110, and 211-230. MATERIALS
AND METHODS
Bovine eyes were obtained immediately following slaughter from Iowa Beef Packers (Emporia, KS). Carrier free “‘I was purchased from Amersham International. Amino acids (t-butoxycarbonyl and resin forms) were obtained from Vega Biochemicals, United States Biochemicals Corp., or from Sigma. Reagents used for peptide synthesis were HPLC grade from Fisher or Sigma, or Sequenal grade from Pierce. All other buffers and reagents were from Sigma. Rod outer segments (ROS) were prepared by the method of Papermaster and Dreyer (16). Soluble PDE (c$+y2) was light eluted from ROS membranes in a buffer containing 10 mM Tris, pH 7.4, 0.1 mM dithioerythritol, 0.2 mM phenylmethylsulfonyl fluoride (PMSF), 10 FM leupeptin, and 1 PM pepstatin. PDE-y was purified from soluble PDE by heat and acid treatment according to the method of Hurley and Stryer (6). Peptides corresponding to indicated PDE/3 sequences were synthesized manually by the method of Merrifield (17) as modified by Gorman (18) and as previously described (1). Peptide antisera were produced by crosslinking to keyhole limpet hemocyanin (KLH) by the procedure of Takemoto et al (19). The solid-phase RIA (20) was adapted for a binding assay by a previously described procedure (1). Briefly, polystyrene tubes (12 x 75 mm) were incubated for 3 h with 0.2 96 glutaraldehyde in 0.5 ml of buffer A (0.1 M sodium phosphate, pH 5.0). (All incubations were done at room temperature, unless otherwise noted). After washing three times with buffer A, peptides were added at 10 pg/tube in 200 ~1 of buffer B (0.1 M sodium phosphate, pH 8.0) and incubated at 37 “C for 1 h. Tubes were then washed twice with buffer C (0.15 M NaCl, 0.5% Tween 20) and once with buffer B. Purified PDE? (0.1 fig) was added to the tubes in 300 ~1 in buffer B and agitated for 1 h. After rinsing twice with buffer B, tubes were incubated with 0.2% glutaraldehyde in buffer B for 15 min, with agitation, followed by two washes with buffer C and one wash with buffer D (0.1 M sodium phosphate, pH 7.4, 0.15 M NaCl, 0.05% Tween 20). Antiserum was added (1:lOO) in buffer D, and tubes were agitated overnight. Following three washes with buffer E (10 mM Tris, pH 8.0, 0.05% Tween 20) and two washes with water, 0.5 ml of buffer E containing 1251-labeledprotein A (1 x 106 cpmltube) was added. Tubes were agitated for 1 h, then washed twice with buffer E and twice with water, and counted in a Beckman y counter. Peptide and PDEy concentrations were chosen to be saturating. Different lots of peptide were tested in cases of positive binding to PDEy. RESULTS AND DISCUSSION Antiserum was made against PDE-y peptide sequence l-49 (1). The entire PDEa sequence, in approximately 15 amino acid segments, has been tested for ability to bind PDEr (1). PDE/3 peptides of approximately
15 amino acids (selected as those regions differing significantly
in sequence
between PDEa and PDEfl) were synthesized for use in binding studies. These experiments were designed to identify the regions on PDEB that interact with PDEy.
Binding RIA using the PDE/3
peptides to bind PDE-y directly show weak PDE-y binding to @-peptide 15-34, and stronger binding to fi-peptides 91-110, and 211-230 (Table 1). Controls of polylysine and polyglutamate discount a 475
Vol.
178, No. 2, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
Table1. Bindingof PDE+y to selected PDB/3peptides Sample B l-10
cpm(-PDEY) 105f 29
/311-22
75 f 156
-31 f 56
/3 15-34
43 f
366 f 73
/S23-36
53 f 48
39
cpm(+PDW 354 f 191
58f
45
fi 28-47
7f
59
-12 f 102
/934-44
-31 f
21
-58 f 43
/345-60
3f
11
-93 f 50
/361-75
13 f 21
25 f 61
/376-90
453 f 279
407 f 180
@91-110
107f
1559f 408
/all-130
139f 62
36 f 91
/3151-171
27 f 36
1 f 91
8191-210
19f
30
22 f 174
821 l-230
78*
3
1624f 147
8451-471
84*
13
-44& 58
(3471-490
67 f 61
-25 f 81
8491-510
135f
145 f 87
0684-705
79 f 36
69f
8791-810
108f 47
178 f 102
8811-830
76 + 62
127 f 97
/3826-841
88 f 39
221 f 274
f3842-852
270 f 134
443 f 182
pulylysine
158f 50
204* 40
pulyglutamate
Of
17
50
20
88
-47 f 118
Radioimmunoassays (seeexperimental procedures) wereperformedwith 10 Fg of eachpeptide fued to polystyrenetubes(maximumbindingis 1.0c(gpeptide)andincubatedwith or without0.1 pg of purifiedPDEy. After glutaraldehyde fsing of anybound PDEy, antiserum y-l-49 (1:100) wasreactedwith boundPDEy. PDE-yonly (PDE-yaddedbeforeblockingwith Tween-20)cpm =3,655 f 128. PDE-ybackground (PDEy addedafter first blockingwith Tween-20)cpm = 272 f 53. Background cpm= 168f 7. Datavaluesarethe meanof triplicatesf standard deviation. non-specific charge interaction.
p-peptide 15-34 overlaps the sequence in PDEcr (16-30) that
exhibited PDEr binding (1). This interaction was not as strong as the other identified interaction sitesin PDEB (91-110and 211-230)or PDEcu(78-90). The PDEr binding siteson PDW are close to a proposed PDEr interaction region in PDEar(78-90). It is possiblethat diffennt interaction regionson PDEly and PDEB for PDEr may account for the different aftinites observedin PDEy binding to PDE@. 476
Vol.
178,
No.
2, 1991
BIOCHEMICAL
AND
P:
RESEARCH
COMMUNICATIONS
a16-30
1 a:
BIOPHYSICAL
50
GEVTAEEVSK FLDSNVSFAK NSPSEGQVHRFLDQNPG?AD l
l
**
l
l2Ym
VISDLLCPRE
MVDPSNYHA
QYFGRRLSPEDyAEIAcEM;X l
*
l
PE-GCTSRIE
*
615-34
a: P:
a78-90
51 LNSVEESEII LCQVEESAAL * ****
FDLLRDFQDN FELVQDHQEN *
*
l
m VNRSRVVFKI
l
l
l
100 DRMSLFMYRA
NXXLLRRLCSILHA
*t
l
*
l
*
s-
l
l
***
PSl-110-r
a: p:
101 RNGIAELATR WGVAELATR l
**
LFNVHXDAVL &FSVQPDSVL
et****
l
*
l
*
l
EECLVAPDSE EDCLVPPDSE
IVFPLDRGW IVFPLDIGW
l
l
*
l
**
l
***
*****
l
150 GRVALSXXIV GHVAQTKKNV **
l
***
l
*
*
+p91-110
a: p:
a:
P:
151 NVPNTEEDER NVQDVXECPH **
l
201 HFTENDEEIL CFTVNDEDVF l
*
l
**
*
PCDFVDTLTE PSSPADELTD
YQTRNILASP YVTRNILATP
*
*
.
l
**
LKYLNPANLI LFYLNPGTLN *t**t*
l
*tt*
IHNGKDWAV
200 InAv?mmGP IXAVNXLDGP
l
l
IRNGXDVVAI l
********
*
l
*****
l
*******
l
**
250 LNSGSXVPEE LNSANXVFEE
MXVFRLSYLRNCETRRGQIL LXIYRLSYLH NCETRRG& l
*****
*
l
**
l ***
6211-230 Fiq. 1. Sequence program SsQALIGN. with PDEy. Scoring p: PDEp subunit).
comparison between N-termini Underlined regions correspond symbol used: * = identity.
of PDEa and PDEp, using the to the peptides, thet interact (a: PDEa subunit;
Regions of greatest dissimilarity between PDEa and PDE/3 sequences are the N- and C-termini (21). Using the computer program SEQALIGN’,
comparisons were made of the N-terminal regions
of PDEa and PDEB (Fig. 1). PDEa 16-30 differs significantly from PDEO 15-34. The differences between PDEa 78-90 and a corresponding &sequence are more conservative. &sequences 91-110 and 211-230 are more similar to their corresponding PDEa sequences. Key differences (such as the initial proline-cysteine sequence in 891-l 10) from the a sequence may contribute to a very different three-dimensional peptide structure in solution. Different overlapping peptides in each sequence are currently being synthesized in order to investigate this possibility. Since only selected B-peptides were used in this assay(based on their unique sequence compared to corresponding PDEa sequences), other interaction regions may exist. In addition, this assay is focused to look at in vitro direct binding of PDE? to the peptides, and weak interactions may not ‘SEQALIGN is a program authored by Clark, K L., Teller, D. C., and Reeck, G. R. (manuscript in preparation), copyright (c) 1989, Kansas State University Research Foundation. 477
Vol.
178,
No.
2, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
be identified by this procedure. PDE/3 peptides that interacted with PDEr in this binding assay will be altered, either by molecular biology techniques or peptide synthesis procedures, to investigate the effect of these mutations on PDEr interactions. Two binding sites have been identified on PDEy (10,13). These are an inhibitory site that includes residues 63-87 and a binding site that includes residues 24-45 of PDE-y. Since two interaction sites have been identified on PDEy, two corresponding sites may be present on PDEcr and/or PDEP. Based on these binding studies, strong interaction sites for PDE-y on PDE@ are within residues 91-110 and 21 l-230 of PDEP, while a weaker site may be present in residues 15-34.
ACKNOWLEDGMENTS This research was supported by NIH/NEI grant EYO6490 and by grant G28 from the American Heart Association, Kansas Affiliate, to DJT. This is contribution #91-479J from the Kansas Agriculture Experiment Station. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.
Oppert, B., Cunnick, J. M., Hurt, D., and Takemoto, D. J. (1991) J. Biol. Chem. (accepted with revision) Stryer, L. (1985) Biopolymers 24, 29-47. Fung, B. K. -K., Hurley, J. B., and Stryer, L. (1981) Proc. Natl. Acad. Sci. U.S.A. 78, 152-156. Ovchinnikov, Y. A., Gubanov, V. V., Khramtsov, N. V., Ischenko, K. A., Zagranichny, V. E., Muradov, K. G., Shuvaeva, T. M., and Lipkin, V. M. (1987) FEBS Lett. 223, 169-173. Lipkin, V. M., Gubanov, V. V., Khramtsov, N. V., Vasilevskaya, I. A., Atabekova, N. V., Muradov, K. G., Shuvaeva, T. M., Surina, E. A., Zagranichny, V. E., and Li, T. (1990) Bioorganicheskaya Kimia 16, 118-120. Hurley, J. B., and Stryer, L. (1982) J. Biol. Chem. 257, 11094-11099. Deterre, P., Bigay, J., Forquet, F., Robert, M., and Chabre, M. (1988) Proc. Natl. Acad. Sci. U.S.A. 85, 2424-2428. Whalen, M. M., and Bitensky, M. W. (1989) Biochem. J. 259, 13-19. Wensel, T. G., and Stryer, L. (1986) Proteins Struct. Funct. Genet. 1, 90-99. Morrison, D. F., CuMick, J. M., t&pert, B., and Takemoto, D. J. (1989) J. Biol. Chem. 264, 11671-11681. Cunnick, J. M., Hurt, D., Oppert, B., Sakamoto, K., and Takemoto, D. J. (1990) Biochem. J. 271, 721-727. Lipkin, V. M., Dumler, I. L., Muradov, K. G., Artemyev, N. 0. and Etingof, R. N. (1988) FEBS Len 234, 287-290. Brown, R. L., and Stryer, L. (1989) Proc. Natl. Acad. Sci. U.S.A. 86, 4922-4926. Lipkin, V. M., Udovichenko, I. P., Bondarenko, V. A., Yuorvskaya, A. A., Telnykh, E. V., and Skiba, N. P. (1990) Biomedical Sciences 1, 305-313. Fung, B., Young, J., Yamane, H., and Griswold-Prenner, I. (1990) Biochem. 29, 26572664. Papermaster, D. S., and Dreyer, W. J. (1974) Biochemistry 13, 2438-2444. 478
Vol.
17. 18. 19.
20. 21.
178,
No.
2, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Merrifieid, R. B. (1963) J. Amer. Chem. Sot. 85, 2149-2154. Gormann, J. J. (1984) Anal. Biochem. 136, 397406. Takemoto, D. J., Spooner, B., and Takemoto, L. J. (1985) Biochem. and Biophys. Res. Commun. 132,438-444. Suter, M. (1982) J. Immunol. Methods 53, 103-108. Lipkin, V. M., Khramtsov, N. V., Vasilevskaya, I. A., Atabekova, N. V., Muradov, K. G., Gubanov, V. V., Li, T., Johnston, J. P., Volpp, K. J., and Applebury, M. L. (1990) J. Biol. Chem. 265, 12955-12959.
479