JOURNAL OF RECEPTOR RESEARCH, 13(1-4), 541-558 (1993)

HIGH LEVEL FUNCTIONAL EXPRESSION OF HUMAN P l - ADRENERGIC RECEPTOR IN BACULOVIRUS-INFECTED CELLS

SCREENED BY A RAPID IN SITU PROCEDURE

Viviane RavetY, Nathalie Blinq, Jean-Luc Guilfaumeq, Franqoise Petitjean 7, Lucien Cabanie 9 and A. Donny Strosbergr

YLaboratoire dlmmuno-Pharmacologie Moleculaire, Centre National de la Recherche Scientifique and Universite Pans VII at lnstitut Cochin de

Genetique Moleculaire, Paris, France § lnstitut Pasteur, 25 rue du Dr. Roux, 75724 Paris cedex 15, France

ABSTRACT

A novel screening assay for the identification of baculovirus infected cells expressing membrane receptors was developed by using a replica transfer technique. Sf9 cells were cotransfected with wild type baculoviral DNA and the transfer vector pVL941-Pl containing the coding region of the human p l -adrenergic receptor gene. Infected cells embedded in agarose were incubated with [l ~~l]-iodocyanopindolol and transferred onto filters that were subsequently autoradiographed. This procedure resulted in the isolation of recombinant baculoviruses that expressed P l -adrenergic receptors. Binding assays carried out with [1251]-ICYP indicated that more than 600,000 receptors were expressed per cell, the highest level noted so far for this receptor in genetically engineered cells. Sf9 cells expressing the pl -AR were analysed by ligand binding, competition experiments, adenylyl cyclase stimulation and photoaffinity labeling. These cells express a homogenous population of receptors and display the known pharmacological properties of p1 -AR in human tissues.

54 1

Copyright 0 1993 by Marcel Dekker, Inc.

Jour

nal o

f R

ecep

tors

and

Sig

nal T

rans

duct

ion

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f B

ritis

h C

olum

bia

on 1

0/29

/14

For

pers

onal

use

onl

y.

542 RAVET ET AL.

Abbreviations:

P-AR: P-adrenergic receptor; ICYP: iodocyanopindolol; AcNPV: Autograph californica nuclear polyhedrosis virus; Sf9 cells: Spodoptera frugiperda cells; CHO cells: Chinese Hamster Ovary cells; PMSF: phenylmethylsulfonyl fluoride; SDS-PAGE: sodium dodecyl sulfate polyacrylamide gel electrophoresis; pfu: plaque forming unit; cyclic AMP: cyclic adenosine 3'3' monophosphate; PBS: Phosphate Buffer Saline.

A number of eukaryotic plasma membrane receptors mediate their effects through coupling to GTP-binding regulatory proteins (G proteins). These receptors are integral membrane glycoproteins that contain a hydrophobic core of seven transmembrane a-helical regions. Prominent among the members of this family are the P-adrenergic receptors (P-AR) (1). The primary sequence of these receptors has been elucidated by molecular cloning of the cDNAs or genes. Detailed structure-activity relationship analysis of purified receptors has been prevented thus far by the lack of available protein in sufficient quantities. To increase the amount of protein available for analysis, P-AR have been expressed in mammalian cells but large scale expression in such cells remains difficult, expensive and time consuming (2). Expression of P-AR in E. coli

presents considerable technical advantages, but in all systems examined so far, levels of receptor expression remain low and proteins are neither glycosylated nor palrnitoylated(3-6). An attractive alternative for obtaining high level expression of recombinant proteins is the baculovirus-Sf9 cell system in which processing of membrane glycoproteins is apparently similar to that in mammalian cells (7).

Recombinant baculoviruses are generated through homologous recombination in vivo following co-transfection of permissive insect cells with wild type viral DNA and an appropriate transfer vector. The transfer vector contains the foreign gene, inserted within the viral polyhedrin gene downstream from the polyhedrin promoter.

Jour

nal o

f R

ecep

tors

and

Sig

nal T

rans

duct

ion

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f B

ritis

h C

olum

bia

on 1

0/29

/14

For

pers

onal

use

onl

y.

BACULOVIRUS-INFECTED CELLS 543

The simplest and most direct method of screening recombinant viruses is by direct visualization; wild-type virus-infected Sf9 cells contain occlusion bodies, due to the synthesis of the polyhedrin protein, while recombinant viruses do not lead to the formation of these bodies, since the foreign gene is cloned in place of the polyhedrin gene. However, it is difficult to identify one or more recombinant plaques, with an occlusion- negative phenotype, among several hundred wild-type phenotypes possessing the occlusion- positive morphology.

Several procedures have been developed to detect recombinant virus either by dot blot or direct plaque hybridization, using nucleic acid probes. However, these techniques are time consuming and do not indicate whether the hybridizing recombinants are able to express the desired gene product, or whether the protein products retain their functional properties.

We now report the development of a rapid straightforward procedure for the identification of Sf9 cells expressing P-adrenergic receptors. The technique consists of a direct in situ visualization of recombinant viral plaques, using the [~~~l ] - iodocyanopindolol binding properties of such receptors.

Human pl-ARs expressed in high numbers in Sf9 cells retain their pharmacological binding properties and are able to couple to the resident insect adenylyl cyclase.

Construction of the transfer vector pVL941-PIAR: The plasmid pUCpl-AR, containing the coding region of the human pl-AR, was digested with restriction enzymes EcoRl and Bglll and bluntened by filling in with Klenow fragment of DNA polymerase with deoxynucleotides. The 1446 bp fragment (Pl -AR coding region) was separated on an agarose gel, blotted on NA45 paper and subsequently eluted. This fragment was cloned into the transfer vector pVL941 which had been linearized with BamHl and bluntened as before. The ligated plasmid was used to transform competent E. coli cells, JMl 01 , and the bacterial clones

Jour

nal o

f R

ecep

tors

and

Sig

nal T

rans

duct

ion

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f B

ritis

h C

olum

bia

on 1

0/29

/14

For

pers

onal

use

onl

y.

544 RAVET ET AL.

carrying plasmid, with the insert in the correct orientation, were identified by restriction analysis. The pVL941-plAR plasmid was amplified in a 100 ml culture and purified using a Qiagen column. The AcNPV DNA was isolated according to Smith and Summers (8).

Transfer of p l -AR gene into AcNPV genome: Transfer of human p1- AR gene into the AcNPV genome was achieved by co-transfection of pVL941-PlAR and AcNPV DNA into exponentially growing Sf9 cells by electroporation (1mF under 220 V). Sf9 cells (3 x 106) were transfected with 10 pg of AcNPV DNA and 20 pg of pVL941 -pl AR DNA. After 6 days the supernatant from the transfection culture was serially diluted and used to infect fresh Sf9 cells.

Production of cells and screening for recombinant virus: Fresh Sf9 cells seeded into 60 mm dishes at an initial density of 4 x 106 per plate were infected with between 10-4 to 10-6 pfu. The plates were overlayed with 1.5% agarose and incubated at 28°C until plaques were well formed. Cells embedded in agarose were overlayed with a 300 pM [ I 25l]-ICYP (2000 Ci/mmole, Amersham) solution in Hank's Hepes, ascorbic acid buffer (pH 6.8) and incubated for 1 h at 37°C. The radioactive solution was then removed. The agar overlay was carefully removed from plates and placed plaque-side up in a sterile culture dish. A dry Hybond C filter disk was placed on top of the agar, and was orientated by stabbing with a needle dipped in ink. The filter was left in place for 5 minutes until uniformly wet and then carefully peeled off before autoradiography.

The classical method of screening for recombinant virus by limiting dilution dot blot hybridization using the i3pl probe (third intracytoplasmic loop) has been performed as described by Fung et a/. (9).

Preparation of membranes: Membranes from Sf9 cells were prepared as follows. Harvested cells were washed with PBS and resuspended in lysis buffer (10 mM Tris-HCI (pH 7.4), 1 mM EDTA, 0.5 mM PMSF) at a concentration of 108 cells/ml. Cells were then ruptured using a homogenizer Potter S (Braun) at maximum speed (1500 rpm) with 10-15 strokes, and centrifuged at 1090 g (3000 rpm Beckman JA-20 rotor) for 8 min at 4°C. The supernatant was kept on ice and the pellet was homogeneized two more times using a Potter S. All supernatants

Jour

nal o

f R

ecep

tors

and

Sig

nal T

rans

duct

ion

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f B

ritis

h C

olum

bia

on 1

0/29

/14

For

pers

onal

use

onl

y.

BACULOVIRUS-INFECTED CELLS 545

were combined, and the centrifugation at 1090 g for 8 min was repeated. The resulting supernatant was centrifuged at 58000 g (20000 rpm Beckman JA-20 rotor) for 1 hour at 4 "C. This pellet was resuspended in storage buffer (25 mM Tris-HCI (pH 7.4) containing 25 mM magnesium chloride, 10% glycerol, 0.5 mM PMSF, 5 pg/ml leupeptine, 7 pg/ml pepstatine and 4 pg/ml benzamidine). Membranes (2-7 mg protein/ml) were divided into 100 pl aliquots and stored at -80°C.

Immunoblorring: Sf9 membrane proteins separated by SDS-PAGE (1 0% acrylamide) were transferred onto nitrocellulose by electroblotting using a Trans-Blot SD apparatus (Biorad) for 45 min at a current intensity of 0.8 mA/cm2. After 12 hours of preincubation at 4°C for saturation in PBS containing 3% skimmed milk and 0.1% Tween 20 (PMT buffer), proteins on nitrocellulose were submitted to 24 hours incubation at 4°C with a rabbit polyclonal antibody serum specific for human pl -AR (Rap 3- 1) (10). All subsequent steps were done at room temperature. The nitrocellulose membrane was then washed several times with PMT buffer and biotinylated Goat-AntiRabbit antibody (Jackson Immuno-Research) was added for 45 min, at a dilution of 1/1000 in PMT. After washing 3-4 times in PMT buffer, Streptavidine-peroxydase (Jackson Immuno- Research) was added for 60 min at a 1/1000 dilution in PMT buffer. The membrane was then extensively washed, first by PMT buffer then by PBS, and revealed by a TMB kit (Kirkegaard and Perry Laboratories) following the recommendations of the manufacturer. For specific displacement, the specific antibody incubation step was done after preincubation of the antibody with the peptide used for immunization (10pg/ml) in PMT buffer for 2 hours at room temperature.

Binding assays: Sf9 cells (2 x 104) expressing p l -AR were incubated at 37°C for 1 h in 1 ml of Hank's buffer containing 20 mM Hepes (pH 6.2), 0.1% bovine serum albumin, and 1 to 1000 pM [1251]- ICYP as appropriate. Non-specific binding level was determined in the presence of 2 pM (k) propranolol. Separation of bound from free ligand was achieved by filtration on glass filters which were soaked for 10 min in 0.3% polyethyleneimine. Competition binding experiments were performed on membranes (10 pg/ml) as in direct binding assays using

Jour

nal o

f R

ecep

tors

and

Sig

nal T

rans

duct

ion

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f B

ritis

h C

olum

bia

on 1

0/29

/14

For

pers

onal

use

onl

y.

546 RAVET ET AL.

16 pM [l 251]-ICYP as radiolabeled ligand and various concentrations of co m pet i t o rs .

Photoaffinity labeling: Photoaffinity labeling was performed by incubating membranes (60 pg per ml) with 200 pM [125l]-ICYP diazirine (2000 Ci/mmole, Amersham) for 1 hour at 37°C in a buffer containing 10 mM Tris-HCI (pH 7.4), 150 mM NaCI, 5 mM MgC12 and 2 mM CaC12. Non- specific displacement of ligand was obtained using 2 pM (k) propranolol. Following incubation, the unbound ligand was removed by a single rapid wash. The membrane suspensions were irradiated with ultraviolet light for 3 min (6 cm from a 200 W super-pressure HBO mercury lamp fitted with a 31 0 nm filter). Photoaffinity labeled membranes were subjected to electrophoresis on a 10% SDSIPAGE gel, and then autoradiographed.

Adenylyl Cyclase Stimulation Assay: Sf9 cells in liquid culture, infected for 2 days with a pl-AR recombinant baculovirus, were washed in 1 ml culture medium supplemented with 1 mM ascorbic acid, 1 mM 3- isobutyl-methyl-xanthine, buffered with 20 mM Hepes (pH 6.8). 1 x 106 cells were incubated for 1 h at 28°C in 1 ml buffer, in the absence or presence of 2 x 10-5 M forskolin or 10-12 to 10-4 M of (-) isoproterenol, (-) norepinephrine and (-) epinephrine. The reaction was stopped by boiling for 5 min, and following centrifugation (4000 rpm, 10 min, 4"C), the total cyclic AMP level contained in 50 1.11 of supernatant, was determined, using the Amersham (3HI-cAMP assay kit. Beta radiations were quantified for 1 min with a LKB-Wallac 141 0 scintillation counter.

Data Analysis : The computer program EBDA (Biosoft) was used for Scatchard analysis, and EC50 values obtained from adenylyl cyclase stimulation was determined using a non-linear regression analysis to a sigmoid curve with the Graph-PAD Software program (copyright 1987 by H.J. Motulsky).

RESUI TS

Screening for recombinant viruses: The screening for recombinant baculoviruses, by hybridization of Sf9 RNA with a pl-AR specific probe (i3pl), allowed the selection of 6 recombinants. This method of screening

Jour

nal o

f R

ecep

tors

and

Sig

nal T

rans

duct

ion

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f B

ritis

h C

olum

bia

on 1

0/29

/14

For

pers

onal

use

onl

y.

BACULOVIRUS-INFECTED CELLS 547

provide no indication as to whether recombinant viruses are able to synthesize the foreign gene product. We developed a novel replica transfer procedure involving ligand binding, based on a modification of the technique employed to detect P2-AR expressing bacterial colonies reported by Breyer et al. (6). Infected cells embedded in agarose were overlayed with [1251]-ICYP as described in materials and methods. A replica imprint of the monolayer was made by placing a filter directly on the agarose. Most of the cells and viruses remain in the agarose overlay to allow subsequent recovery. Autoradiography revealed labeled plaques in cells infected with virus generated by cotransfection of pVL941-pl and AcNPV DNA (fig. 1 B) but not in cells infected with wild- type AcNPV (fig.lA). Positive plaques were recovered from the agarose after realignment of the template filter and were plaque purified several times. Figure 1C shows an example of the binding screen of one recombinant virus (X3H9) after a single round of plaque purification. This method was found to be quite efficient to determine, at the initial screening stage, whether recombinant viruses were able to synthesize a functional protein.

Molecular nature of the receptor expressed by Sf9 cells: The molecular nature of the receptor expressed by Sf9 cells infected with a p1 -AR recombinant baculovirus was investigated by photoaffinity labeling. A membrane fraction was treated with [1251]-ICYP diazirine and analysed following SDS-PAGE and autoradiography. Radiolabel was specifically incorporated into Sf9 cell membranes, resulting in the iodination of two proteins of 51 and 69 kDa as visualized on the autoradiographed SDS-PAGE gel (figure 2A). Labeling of both proteins was abolished by prior incubation of the Sf9 cell membranes with 2 pM (k) propranolol (figure 28). The theoretical molecular mass of the peptide moiety of the pl-AR, based solely on amino acid content, is 51 kDa. The photoaffinity labeled 51 kDa component in Sf9 cell membranes may correspond to a partially glycosylated form of the receptor and the 69 kDa component to a fully glycosylated form, equivalent to that made in transfected CHO cells (2). When immunoreactive p1 -AR was monitored by Western-blot of infected cells, a major band of 69 kDa was observed (figure 3A). This band had a mobility similar to the upper band ICYP

Jour

nal o

f R

ecep

tors

and

Sig

nal T

rans

duct

ion

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f B

ritis

h C

olum

bia

on 1

0/29

/14

For

pers

onal

use

onl

y.

548 RAVET ET AL.

Fia. 1, A panel of filters autoradiographed after [I 251j-ICYP in situ binding on Sf9 cells infected with wild type baculovirus (A), on Sf9 cells infected with progeny virus resulting from cotransfection of wild type baculoviral DNA and pVL941 pl (B), on Sf9 cells infected with purified recombinant baculovirus X3H9 (arrow) (C).

Jour

nal o

f R

ecep

tors

and

Sig

nal T

rans

duct

ion

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f B

ritis

h C

olum

bia

on 1

0/29

/14

For

pers

onal

use

onl

y.

BACULOVIRUS-INFECTED CELLS 549

F U 2 . Autoradiograph of SDS-PAGE showing photoaffinity labeling of the Pl-AR expressed in Sf9 cells with [1251]-ICYP diazirine in the absence (A) or in the presence of 2 pM (2) propranolol (B).

Fia. & lmmunoblot of the pl-AR in membranes from Sf9 cells. Filters were probed with an antiserum directed against a peptide derived from the second extracellular loop of the receptor (A), or with this antiserum preincubated with the peptide (8).

Jour

nal o

f R

ecep

tors

and

Sig

nal T

rans

duct

ion

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f B

ritis

h C

olum

bia

on 1

0/29

/14

For

pers

onal

use

onl

y.

RAVET ET AL.

0 A5G7

X3H9

loo0 -

800 -

600-

400 -

I

0 100 200

time (hours)

Fia. 4, Time course of pl -AR expression in Sf9 cells. Sf9 cells (2x1 06 cells/ml) in shaker culture were infected at a multiplicity of infection of 1 with recombinant baculoviruses (A5G7 and X3H9). The concentration of p1 -AR was measured at various times according to [125l]-ICYP binding activity.

diazirine labeled receptor. Detection of this band was inhibited by preincubation of the antibody with the peptide against which it was raised (1 0) (figure 36).

Kinetics of expression and ligand binding characteristics of /3l -AR expressed in Sf9 cells: Production of the pl -AR was maximal 4 days after infection as shown in figure 4. The initial lag was approximately 1 day post-infection. This time course is typical for genes under the control of the polyhedrin promoter (1 1). This lag could be decreased somewhat by increasing the multiplicity of infection, as cells infected at a lower multiplicity continue to grow until all are infected.

Among the three recombinant viruses selected by RNA hybridization, A5G7 expressed the pl-AR at a high level (10 pmol/mg membrane protein). The expression of human Pl-AR with A5G7 was 2 fold greater than that seen previously for the turkey P-AR in Sf9 cells (1 2). The recombinant virus X3H9, selected by binding screening, displayed a

Jour

nal o

f R

ecep

tors

and

Sig

nal T

rans

duct

ion

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f B

ritis

h C

olum

bia

on 1

0/29

/14

For

pers

onal

use

onl

y.

BACULOVIRUS-INFECTED CELLS 551

slightly lower expression level than A5G7 (466,180 versus 632,420 sites per cell, 100 hrs post-infection) (figure 5). The P-AR antagonist [125l]- ICYP bound to the p l -AR in Sf9 cells with Kd values of 38 and 56 pM for A5G7 and X3H9 recombinant viruses. Kd values for the Pl-AR expressed in Sf9 cells are in close agreement with those published for p l - A R expressed in transfected cells such as CHO cells or L cells (2; Tate, personal communication). Saturable [125l]-ICYP binding to p l -Sf9 cells was competitively inhibited by a range of P-AR agonists (figure 6A) and antagonists (figure 6B). As predicted, (-) norepinephrine was found to have a greater affinity (Ki = 1.79 pM) than (-) epinephrine (Ki = 4.3 pM). The p1 -AR selective agonist prenalterol had a calculated Ki of 322 nM for pl-AR. Salbutamol, a P2-AR selective agonist, had a much lower affinity as expected (23 pM). The non selective P-AR agonist, (-) isoproterenol, had a calculated Ki of 179 nM for pl-AR. The [125l]-ICYP binding to p1- AR expressed in Sf9 cells was stereospecific since (-) propranolol was a more potent competitor than (+) propranolol (Ki = 4.8 nM versus Ki = 582 nM). The p1 -AR selective antagonist, CGP-20712A, was a better competitor than the p2-AR selective antagonist ICI-118551, again as expected (Ki = 23 nM versus Ki = 480 nM). The P-AR agonists and antagonists bound in the appropriate order of affinity: (-) isoproterenol > prenalterol > (-) norepinephrine > (-) epinephrine > salbutamol for agonists and (-) propranolol > CGP-20712A > ICI-118551 > (+) propranolol for antagonists as expected for p l -AR.

Activation of adenylyl cyclase in Sf9 cells by the p 7-AR: I n recombinant baculovirus-infected Sf9 cells, the human p1 -AR was able to promote the activation of resident Gs and to stimulate insect adenylyl cyclase. Two days post-infection Sf9 cells possess an endogenous adenylyl cyclase which could be activated 13-fold by 25 pM forskolin (339 f 26 pmoles of cyclic AMP accumulated per 106 cells). A two-day infection period allows sufficient expression of functional p1 -ARs in the cytoplasmic membrane of Sf9 cells and was short enough to prevent total viral disruptions. The agonist (-) isoproterenol increased the p1 -Sf9 intracellular concentration of cAMP 4-fold (figure 7) whereas wild-type baculovirus Sf9 cells maintained their basal level of cAMP (not shown). However, it is worthnoting that in both p l turkey-Sf9 cells (12) and

Jour

nal o

f R

ecep

tors

and

Sig

nal T

rans

duct

ion

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f B

ritis

h C

olum

bia

on 1

0/29

/14

For

pers

onal

use

onl

y.

552 RAVET ET AL.

a * U m

0.1 0.2 0.3

ICYP (nM)

A

7 1

5 -

- 0.0

3-

1- 0 1 3 4 5 6 8

0 Bound (pmoles)

I I I

0 . I 0.2 0.3

K Y Y (nM)

B

Fia. LSaturation isotherm of the specific binding of [1251]-ICYP to Sf9 cells infected with p l -AR recombinant baculoviruses A5G7 (A) and X3H9 (B). Data represent the mean of three independent experiments. Inset, Scatchard plot of the same data.

Jour

nal o

f R

ecep

tors

and

Sig

nal T

rans

duct

ion

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f B

ritis

h C

olum

bia

on 1

0/29

/14

For

pers

onal

use

onl

y.

BACULOVIRUS-INFECTED CELLS 553

-10

25 -

20 - h W -

15- g

8’ a lo-

Y

L, -0 C

$.

5 -

Log [antagonist] (M)

B

-8 -6 -4 -2

Log [agonist] (M) A

o (+) Propranolol 0 lC1118551 a CGP20712A m (-) Propranolol

0 - 10 -8 -6 -4 -2

Fia. B Competition binding experiments performed on Sf9 cell membranes expressing p l -AR with P-adrenergic agonists (A) and antagonists (B). Data represent the mean of three independent experiments performed on two membrane preparations.

Jour

nal o

f R

ecep

tors

and

Sig

nal T

rans

duct

ion

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f B

ritis

h C

olum

bia

on 1

0/29

/14

For

pers

onal

use

onl

y.

554

C 0 .- d 2 100-

.- z d v1

p;: 80 - n u h U

- 2 60 - .a x C

U

0

- n 4 0 - L

RAVET ET AL.

1 2 0 7

seI 20 -10 -8 -6 -4 -12

Fia. 7, Stimulation of adenylyl cyclase in p l -AR recombinant baculoviruses infected Sf9 cells. Cyclic AMP accumulation is measured in the presence of (-) isoproterenol ( ), (-) norepinephrine (A), and (-) epinephrine (0). Data represent the mean of three independent experiments performed in duplicate.

human p1 -Sf9 cells (-) isoproterenol activates insect adenylyl cyclase with a two order lower potency compared to the pl-CHO cells (2). This difference stem from either local membrane disruptions induced by viral infection, or from a lower intrinsic rate of coupling in Sf9 cells. Parker et a/. (12) reported the same range of adenylyl cyclase activation constants in pl turkey-Sf9 cells. The order of potency of three pl -AR agonists to activate adenylyl cyclase was determined as (-) isoproterenol > (-) norepinephrine > (-) epinephrine with Kact values of 39 k 9 nM, 391 f 87 nM, and 2910 f 960 nM, respectively. This order of potency is in agreement with that published for p1 -AR expressed in CHO cells (2).

Expression in baculovirus-infected Sf9 insect cells constitutes an efficient and reliable system for producing high levels of proteins,

Jour

nal o

f R

ecep

tors

and

Sig

nal T

rans

duct

ion

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f B

ritis

h C

olum

bia

on 1

0/29

/14

For

pers

onal

use

onl

y.

BACULOVIRUS-INFECTED CELLS 555

however screening for the appropriate recombinant viruses remains laborious. Several procedures have previously been developed to detect recombinant viruses by dot-blot hybridization with nucleic acid probes (9) or by in siru screening with antibody probes (13). However, these techniques provide no indication as to whether recombinant viruses encode active functional protein. We have expressed the human p1 -AR in baculovirus system and have developed an efficient replica transfer screening assay based on production of functional receptor. This method may actually constitute an useful tool for the expression cloning of other membrane receptor genes.

Moreover, using this procedure, we isolated human p l - A R

recombinant baculoviruses which yielded a significantly higher level of expression in Sf9 cells as compared with other systems. Indeed, the number of receptors expressed is between 3 and 4 fold higher in Sf9 cells than in CHO cells (2).

From the results reported here, it appears that the Kd, Ki and Kact values for the pl-AR expressed in Sf9 cells are in agreement with those reported for p1-AR expressed in Xenopus laevis oocytes (14), in CHO cells (2) and in human myocardium (15). Taken together, these findings indicate that when expressed in Sf9 cells, the human pl-AR maintains its intrinsic binding properties although the molecular forms of the photoaffinity labeled receptor are different from those of the receptor expressed in other types of cells. In Sf9 cells, two forms of pl-ARs of 51 and 69 kDa were photoaffinity labeled, while in CHO cells expressing pl - AR, 59 and 72 kDa polypeptides were labeled (2) Pl -ARs from mammalian tissues display Mr of approximatively 65-67 kDa (1 6) The heterogeneity of pl-AR Mr reported in these systems is most likely owing to different cell-specific or species-specific glycosylations.

Much remains to be discerned concerning the nature of protein glycosylation in insect cells; N-linked glycosylation has been shown to occur in insect cells, although its subsequent processing appears to be different. Butters and Hughes (17) have shown that N-linked oligosaccharides in mosquito cell lines are deficient in sialic acid,

Jour

nal o

f R

ecep

tors

and

Sig

nal T

rans

duct

ion

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f B

ritis

h C

olum

bia

on 1

0/29

/14

For

pers

onal

use

onl

y.

556 RAVET ET AL.

galactose and fucose. Thus, the conversion of high-mannose to complex N-linked oligosaccharides does not appear to take place in those insect cells. The presence of sialic acid, galactose and fucose has been reported in the oligosaccharide moiety of the turkey Pl-AR (18). These differences between mammalian and insect glycosylation processes could explain the differences between pl-AR Mr’s reported in Sf9 and CHO cells. It is also not surprising that the pl-AR expressed in Sf9 cells maintains its binding properties since previous studies have shown that the absence or removal of the carbohydrate moiety from turkey pl-AR or human P2-AR does not affect the pharmacological and functional properties of the receptor (18, 19).

In conclusion, the present study demonstrates the utilization of the baculovirus expression system for the highly efficient production of the human p1-AR. Moreover, the Sf9 cells express a homogenous population of receptors that exhibit the known ligand binding properties of p1-AR in human tissues and in addition couple to G-protein, and activate adenylyl cyclase. The baculovirus expression system constitutes a valuable tool that will facilitate the approach to structural studies of the 0-AR family.

ACKNOWLEDGEMENTS:

We thank Dr L. J. Emorine for helpful discussions and for kindly providing us with the human pl-AR coding region, Drs Y. Magnusson and J. Hoebeke for anti p l antibody, M.F. Drumare for excellent technical assistance. Support for our work is primarily from the Centre National de la Recherche Scientifique, the lnstitut National de la Sante et de la Recherche Medicale, University Paris VII, the Ministry for Research and the Bristol Meyers Squibb Company. We are also grateful for the help from the Ligue Nationale contre le Cancer, the Fondation pour la Recherche Medicale Franqaise and the Association pour la Recherche sur le Cancer.

Jour

nal o

f R

ecep

tors

and

Sig

nal T

rans

duct

ion

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f B

ritis

h C

olum

bia

on 1

0/29

/14

For

pers

onal

use

onl

y.

BACULOVIRUS-INFECTED CELLS 557

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

REFERENCES

Strosberg, A.D. Structure/function relationship of proteins belonging to the family of receptors coupled to GTP-binding proteins. Eur. J. Biochem. 196, 1-10, 1991.

Tate, K.M.; Briend-Sutren, M.-M.; Emorine, L.J.; Delavier-Klutchko, C.; Marullo, S.; and Strosberg, A.D. Expression of three human p- adrenergic-receptor subtypes in transfected Chinese hamster ovary cells. Eur. J. Biochem. 196, 357-361, 1991.

Marullo, S.; Delavier-Klutchko, C.; Eshdat, Y.; Strosberg, A.D.; and Emorine, L. Human p2 -adrenergic receptors expressed in Escherichia coli membranes retain their pharmacological properties. Proc. Natl. Acad. Sci. USA 85, 7551-7555, 1988.

Marullo, S.; Delavier-Klutchko, C.; Guillet, J.-G.; Charbit, A.; Strosberg, A.D.; and Emorine, L. Expression of human p1 and p2 adrenergic receptors in E. coli as a new tool for ligand screening. Bioflechnology 7, 923-927, 1989.

Chapot, M.-P.; Eshdat, Y.; Marullo, S.; Guillet, J.-G.; Charbit, A.; Strosberg, A.D.; and Delavier-Klutchko, C. Localization and characterization of three different 0-adrenergic receptors expressed in Escherichia coli. Eur. J. Biochem.l87,137-144, 1990.

Breyer, R.M.; Strosberg, A.D.; and Guillet, J.-G. Mutational analysis of ligand binding activity of p2 adrenergic receptor expressed in Escherichia coli. EMBO J. 9, 2679-2684, 1990.

Luckow, V.A.; and Summers, M.D. Trends in the development of baculovirus expression vectors. Bioflechnology 6, 47-55, 1988.

Smith, G.E.; and Summers, M.D. DNA homology among subgroup A, 8, and C baculoviruses. Virology 123, 393-406, 1982.

Fung, M.-C.; Chiv, K.Y.M.; Weber, T.; Chang, T.-W.;and Chang, N.T. Detection and purification of a recombinant human B lymphotropic virus (HHV-6) in the baculovirus expression system by limiting dilution and DNA dot-blot hybridization. J. Virol. Methods 19, 33-42,1988.

Magnusson, Y.; Marullo, S.; Hoyer, F.; Waagstein, B.; and Hoebeke, J. Mapping of a functional autoimmune epitope on the beta-1 -adrenergic receptor in patients with idiopathic dilated cardiornyopathy. J. Clin. Invest. 86, 1658-1 663, 1990.

Jour

nal o

f R

ecep

tors

and

Sig

nal T

rans

duct

ion

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f B

ritis

h C

olum

bia

on 1

0/29

/14

For

pers

onal

use

onl

y.

558 RAVET E T AL.

1 1 .

12.

13.

14.

15.

16.

17.

18.

19.

Miller, L.K. Baculoviruses as gene expression vectors. Annu. Rev. Microbiol. 42, 177-1 99, 1988.

Parker, E.M.; Kameyama, K.; Higashijima, T.; and Ross, E.M. Reconstitutively active G protein-coupled receptors purified from baculovirus-infected cells. J. Biol. Chem. 266, 1 , 519-527, 1991.

Capone, J. Screening recombinant baculovirus plaques in situ with antibody probes. Gene Anal. Tech. 6, 62-6, 1989.

Frielle, T.; Collins, S.; Daniel, K. W.; Caron, M.; Lefkowitz, R. J.; and Kobilka, B. K. Cloning of the cDNA for the human pl - adrenergic receptor. Proc. Natl. Acad. Sci. USA 84, 7920-7924, 1987.

Bristow, M. R.; Hershberger, R. E.; Port, D. J.; Minobe, W.; and Rasmussen, R. pl- and p2- adrenergic receptor-mediated adenylate cyclase stimulation in nonfailing and failing human ventricular myocardium. Mol. Pharmacol. 35, 295-303, 1989.

Moxham, C. P.; George, S. T.; Graziano, M. P.; Brandwein, H. J.; and Malbon, C. C. Mammalian p1- and p2- adrenergic receptors: immunological and structural comparisons. J. Biol. Chem. 261, 31,

Butters, T. D.; and Hughes, R. C. Isolation and characterization of mosquito cell membrane glycoproteins. Biochim. Biophys. Acta

14562-1 4570, 1986.

640, 655-671, 1981.

Cervantes-Olivier, P.; Durieu-Trautmann, 0.; Delavier-Klutchko, C.; and Strosberg, A.D. The oligosaccharide moiety of the p1- adrenergic receptor from turkey erythrocytes has a biantennary, N- acetyllactosamine-containing structure. Biochemistry 24, 3765- 3770, 1985.

Cervantes-Olivier, P.; Delavier-Klutchko, C.; Durieu-Trautmann, 0.; Kaveri, S.; Desmandril, M.; and Strosberg, A.D. The p2- adrenergic receptors of the human epidermoid carcinoma cells bear two different types of oligosaccharides which influence expression on the cell surface. Biochem. J., 250, 133-1 43, 1988.

Jour

nal o

f R

ecep

tors

and

Sig

nal T

rans

duct

ion

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f B

ritis

h C

olum

bia

on 1

0/29

/14

For

pers

onal

use

onl

y.