THE J O U R N A L OF BIOLOGICAL CHEMISTRY Vol. 258, No. 11, Issue of June 10, pp. 6756-6761,1983 Printed in U. S. A.

A Role of the B-Oligomer Moiety of Islet-activating Protein, Pertussis Toxin, in Development of the Biological Effects on Intact Cells*

(Received for publication, December 6, 1982)

Makoto TamuraS, Katsumi NogimoriS, Motoyuki YajimaS, Katsuhiko AseS, and Michio UiSn From the $Kyoto Research Laboratories, K a h n Pharmaceutical Co., Ltd., Yamashina, Kyoto 607 and the §Department of Physiological Chemistry, Faculty of Phnrmaceutical Sciences, Hokkaido University, Sapporo 060, Japan

Islet-activating protein (IAP), pertussis toxin, is an oligomeric protein (Tamura, M., Nogimori, K., Murai, S., Yajima, M., Ito, K., Katada, T., Ui, M., and Ishii, S . (1982) Biochemistry 21,5516-5522), the biggest sub- unit (M. = 28,000, referred to as the A-protomer) of which catalyzes transfer of the ADP-ribose moiety of NAD to the membrane M, = 41,000 protein. The pen- tamer, termed the B-oligomer, consisting of the resid- ual subunits was the moiety of IAP that was responsible for binding to the cell surface, as revealed by compet- itive inhibition of the development of the IAP actions on intact rat C6 glioma cells and rat adipocytes. The binding of the B-oligomer to its receptor proteins was divalent via the constituent two dimers; it stimulated mitosis of lymphocytes and caused an insulin-like ac- tion to enhance glucose oxidation in adipocytes, just as did concanavalin A, presumably as a result of cross- linking or aggregation of the membrane proteins. The A-promoter displayed its biological action on adipo- cytes only when the B-oligomer had been bound to the cells. Thus, IAP is a typical A-B toxin in which the B- oligomer is first bound to the cell surface proteins to enable the A-protomer to reach to the site of its action within the cell. Diverse biological actions of pertussis toxin may be accounted for by the mitogenic action of the B-oligomer as well as ADP-ribosyltransferase ac- tivity of the A-promoter.

Islet-activating protein, pertussis toxin, exerts its unique influence on the membrane receptor-adenylate cyclase system in a variety of cell types. IAP' catalyzes transfer of the ADP- ribose moiety of the intracellular NAD to an M, = 41,000 protein of cell membranes (1-5), thereby reversing receptor- mediated or GTP-dependent inhibition of adenylate cyclase (3-5). Conceivably, the M, = 41,000 protein is one of the subunits of the guanine nucleotide regulatory protein (Ni) that transduces a signal of receptor stimulation to the cyclase catalytic component in an inhibitory fashion; ADP-ribosyla- tion of this protein by IAP results in loss of the Ni function (3,4). Exposure of some cell types to IAP enhanced receptor- mediated or GTP-dependent activation of membrane adenyl- ate cyclase as well as receptor-mediated accumulation of intracellular CAMP (6-10). It has recently been argued that this enhancement also emerges from IAP-induced impairment

* The costs or pubiication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ll To whom correspondence should be addressed. ' The abbreviations used are: IAP, islet-activating protein; Hepes,

4-(2-hydroxyethyl)-l-piperazineethanesulfonate; IgG, immunoglo- bin G.

of the capability of Ni to attenuate activation of adenylate cyclase (4).

IAP is an oligomeric protein consisting of six subunits, five of which are dissimilar to each other (11). These subunits were first named S-1 (M, = 28,000), S-2 (23,000), S-3 (22,000), S-4 (11,700), and S-5 (9,300) according to the order of their molecular weights (11). IAP was readily dissociated to S-1 and a pentamer, an association product of the other subunits, in a concentrated urea solution (11). The S-1 was referred to as an A(Active)-protomer, since it displayed ADP-ribosyl- transferase or NAD-glycohydrolase activity in vitro (11, 12). The pentamer then should be a B(Binding)-oligomer if IAP is really one of the A-B toxins (13) as has previously been proposed (11). The purpose of the present paper is to show that IAP is bound to the cell surface at certain sites on its B- oligomer moiety to enable the A-protomer to reach the site of its action inside the cell membrane, thereby affording exper- imental supports to our previous proposal (11).

EXPERIMENTAL PROCEDURES

Preparation of IAP, Its Constituent Peptides, and Their Antibod- ies-IAP was purified from the 2-day culture supernatant of Borde- tella pertussis (Tohama strain, Phase I) according to the procedure described elsewhere (14, 15). The purified IAP was dissociated into its constituent peptides, i.e. the A-protomer, Dimer-1 (S-2 plus S-4). Dimer-2 (S-3 plus S-4), and C-subunit (S-5) by being exposed to 5 M urea for 4 days at 4 "C; these peptides were separated from each other by means of a column of carboxymethyl-Sepharose CL-GB, as re- ported in detail previously (11). The B-oligomer was prepared by combining Dimer-1, Dimer-2, and the C-subunit at the 1:l:l molar ratio in 2 M urea. Alternatively, the B-oligomer was separated as such from the A-protomer, when the exposure time of IAP to 5 M urea was shortened to 5 h. Separation of these two fragments was readily undertaken by using haptoglobin-Sepharose 4B as an affinity column for the B-oligomer (11). Unless otherwise specified, these IAP sub- units as well as IAP itself were stored in a vehicle consisting of 2 M urea and 0.1 M potassium phosphate buffer (pH 7.0) a t 4 "C until use. The vehicle containing no peptide was used for control experiments.

Anti-A-protomer or anti-B-oligomer rabbit antiserum was pre- pared by essentially the same procedure as that for anti-IAP anti- serum (14, 16). The IgG fraction was isolated from these antisera by affinity column chromatography using protein A-Sepharose (Phar- macia Fine Chemicals). The Ouchterlony double immunodiffusion reaction was carried out with these antibodies on 0.6% agar in 0.1 M Verona1 buffer (pH 8.6) for 12 h at 37 "C.

Adenylate Cyclase Assay and ADP-ribosylation Reaction with the Crude Membrane Preparation from C6 Glioma Cells-Rat C6 glioma cells were cultured and exposed to 100 ng/ml (or various concentra- tions in Fig. 1B) of IAP (IAP-treated) or to the vehicle for nontreated cells for the final 3 h of culture at 37 "C as described previously (10). The crude membrane fraction prepared from these cells (2, 10) was suspended in 25 mM Tris-HCI (pH 7.5) containing 2.5 mM MgCL at a concentration of 5 mg of protein/ml and stored in liquid nitrogen until use. Protein was determined by the method of Lowry et al. (17) using bovine serum albumin as standard.

The reaction mixture for adenylate cyclase assay was 50 mM Tris- HCl (pH 7,5), 5 mM MgC12, 1 mM ATP, 10 mM phosphocreatine, 100

6756

by guest on Decem

ber 9, 2020http://w

ww

.jbc.org/D

ownloaded from

Role of B-Oligomer of Islet-activating Protein 6757

units/ml of creatine phosphokinase, 10 pM GTP, 2 mM ethylene glycol bis(@-aminoethyl ether)-N,N,N’,N’-tetraacetic acid, 0.4 mM dithiothreitol, 1 mM 3-isobutyl-l-methylxanthine, 2 mg/ml of bovine serum albumin in a total volume of 200 pl. Incubation was started by the addition of crude membranes (40 pg of protein) and continued for 5 min at 37 “C. The CAMP generated during incubation was estimated as described previously (10) by a radioimmunoassay method (18).

The reaction mixture for ADP-ribosylation of membranes (120- 160 pg of protein) was 50 mM Tris-HC1 (pH 7.5), 5 mM MI#&, 1 mM ATP, 1 mM dithiothreitol, 20 mM thymidine, 10 pM [a-”P]NAD (5 Ci/mmol), and 50 pg/ml of IAP. After a 10-min incubation at 37 “C, membranes were washed, dissolved, and submitted to sodium dodecyl sulfate-polyacrylamide gel electrophoresis as described elsewhere (1, 2). The radioactive band (M, = 41,000) of the dried gels was excised and counted for its 32P content.

Mitogenic Actiuify-Splenic cells obtained from the male ICR-JCL mouse were suspended in RPMI 1640 culture medium containing 10% fetal calf serum and 5 mM Hepes at the density of 2 X lo6 cells/ ml. The cell suspension was cultured with additions in multiwell plates (Linbro) at 37 “C for 24-96 h under a humidified atmosphere of 5% co2/95% air. [3H]Thymidine (0.2 pci) was added to each well 24 h before the end of culture. The cells were then collected on a glass fiber filter paper (Whatman, GF/A), washed with saline, 5% trichloroacetic acid, and methanol, and dried to be counted for radio- activity.

Glycerol Release and Glucose Oxidation in Rat Adipocytes-Rat adipocytes (1-10 X IO6 cells) prepared by the collagenase digestion method of Rodbell (19) were suspended in 1 ml of Krebs-Ringer bicarbonate solution containing 2% bovine serum albumin. This cell suspension was incubated at 37 “C under an atmosphere of 95% 0 2 /

5% C02 for either prior treatment with the B-oligomer or its constit- uent peptides or estimation of glycerol production or glucose oxidation in the presence of IAP or other additions. Where indicated, preincu- bation was carried out at 27 “C. The pretreatment was followed by twice washing with the fresh medium containing no addition before the further incubation for estimation of the metabolic activities. Glycerol in the medium was determined enzymatically (20) after deproteinization by the Ba(OH)2-ZnSO, method. In the case of glu- cose oxidation assay, the incubation medium was supplemented with 0.2 mM [1-“C]glucose (0.025 pCi/ml) and 14C02 production was measured after deproteinization with 0.2 ml of 2 M perchloric acid (19).

Materials-[l-“C]Glucose (0.25 Ci/mmol) and [6-3H]thymidine (15 Ci/mmol) were obtained from New England Nuclear. Concana- valin A and enzymes and coenzymes for glycerol assay were purchased from Sigma. The sources of all other materials used are those de- scribed in the previous papers (1-12).

Presentation of Data-In the present communication, most of the data are shown as changes of values dependent on concentrations of additions such as IAP, its constituent peptides, or antibodies. These experiments were repeated two or three times to determine the appropriate concentration ranges, and the representative results have been presented as figures.

RESULTS

Competitive Inhibition by B-Oligomer of the Actions of Na- tive IAP on Intact Rat C6 Glioma Celkr-Membranes prepared from rat C6 glioma cells that had been exposed to 100 ng/ml of IAP for 3 h exhibited GTP-dependent adenylate cyclase activity much higher than the activity of the membranes from nontreated cells (the left end of plots in Fig. l A ; see Ref. 10). Replacement of IAP by its B-oligomer, just as was the case with its A-protomer (12), did not result in any enhancement of membrane adenylate cyclase assayed with GTP (open cir- cles in Fig. lA). The B-oligomer added 15 min prior to the IAP addition, however, inhibited the action of IAP on intact cells to enhance GTP-dependent adenylate cyclase in mem- branes prepared therefrom (solid circles in Fig. LA). In con- trast, the action of IAP was not affected by the B-oligomer which was added at 20 min (or later) after the onset of the exposure of intact cells to IAP (data not shown).

The action of IAP on intact C6 cells was due to ADP- ribosylation of the membrane M, = 41,000 protein with intra-

0 1 5 40 0 1 10 100 1000 8-oligomer added (rg/ml) IAP added (ng/rnl)

FIG. 1. Competitive inhibition by B-oligomer of the actions of IAP on intact C6 cells. Rat C6 glioma cells were exposed to IAP (100 ng/ml in A or at concentrations as indicated on abscissa in B ) and/or its B-oligomer (at concentrations as indicated on abscissa in A or 10 pg/ml in B ) at the final stage of cell culture. The time of exposure to IAP was 3 h, and the B-oligomer was added to the culture medium 15 (A) or 10 ( B ) min before the addition of IAP (or the addition of vehicle in the case of IAP omission). IAP was added (0) or not (0) in A, while the B-oligomer was added (0, A) or not (0, A) in E . A batch of the membranes prepared from these cells was assayed for GTP-dependent adenylate cyclase activity (0,O). Another batch of the membranes was incubated to measure the incorporation of ADP-ribose from [a-32P]NAD into the M, = 41,000 protein (A, A). See under “Experimental Procedures” for other details.

cellular NAD as substrate (1, 2). The fraction of the mem- brane proteins that remained to be ADP-ribosylated in intact cells was then radiolabeled with [CY-~~PINAD during the sub- sequent incubation of the membranes prepared therefrom again with IAP (2) or its A-protomer (12). Thus, the plot of membrane GTP-dependent adenylate cyclase activities as a function of concentrations of IAP during incubation of the membrane-donor cells was a mirror image of the plot of the 32P-content of the M, = 41,000 protein after radiolabeling of the same membranes (Fig. 1B). The decreasing [32P]ADP- ribosylation in membranes reflects the ADP-ribosylation that has occurred in intact cells making the M, = 41,000 substrate unavailable to the subsequent reaction. The combined addi- tion of the B-oligomer with IAP during cell incubation caused the same degree of shift of both plots to the right, reflecting the competitive inhibition of both IAP actions on intact cells by the B-oligomer (Fig. 1B).

The 32P-content of the M, = 41,000 protein was inversely correlated with GTP-dependent adenylate cyclase activity of the same membranes (r = -0.986; p < 0.001) regardless of whether membrane-donor cells had been exposed to IAP in the presence or absence of its B-oligomer (Fig. 2). Thus, the ADP-ribosylation of the 1M, = 41,000 protein is responsible for IAP-induced enhancement of GTP-dependent adenylate cyclase even in the presence of its B-oligomer. It is very likely that IAP is bound to the cell surface of intact C6 cells via particular sites on its B-oligomer moiety before the onset of its action to catalyze ADP-ribosylation of a membrane pro- tein.

Failure of the Anti-B-oligomer Antibody to Suppress the Action of IAP on Isolated Membranes-Antisera were raised in rabbits against native IAP, its A-protomer and B-oligomer. The double immunodiffusion technique using the IgG frac- tions prepared from these antisera showed that the anti-IAP antibody precipitated with IAP or its constituent peptides, A- protomer and B-oligomer (Fig. 3). In the case of the anti-A- protomer or anti-B-oligomer antibody, there was a sharp precipitation arc with its respective antigen similarly whether the antigen was present in an isolated state or in an associa-

by guest on Decem

ber 9, 2020http://w

ww

.jbc.org/D

ownloaded from

6758 Role of B-Oligomer of Islet-activating Protein

0 40 80 120 160 200 Adenylate cyclase (pmol/rng/min)

FIG. 2. Correlation of adenylate cyclase activity with ADP- ribosylation of the M , = 4 1,000 protein in membranes of cells treated with IAP in the presence or absence of B-oligomer. The ADP-ribosylation data in Fig. 1B are plotted against the adenyl- ate cyclase data obtained with the same membranes. 0, without B- oligomer; O, with B-oligomer.

I

FIG. 3. Specific immunoprecipitation of IAP. its A-proto- mer and B-oligomer with their respective antibodies. Double immunodiffusion was carried out with 180 pg of anti-IAP (left), anti- A-protomer (center) or anti-B-oligomer (right) I& in each center well as described under "Experimental Procedures." The contents of surrounding wells: IAP, 6 pg; A , 3.2 pg of A-protomer; B, 5.6 pg of B- oligomer.

tion state as IAP. The A-protomer and B-oligomer displayed different immunological characteristics in the sense that nei- ther of them was recognized by the antibody raised against the other (Fig. 3).

Both of the anti-A-protomer and anti-B-oligomer antibod- ies were effective in suppressing the action of IAP on intact C6 cells; the GTP-dependent adenylate cyclase activity of membranes isolated from IAP-treated cells was progressively lowered by increasing the concentration of the antibodies in the IAP-containing incubation medium of cells (Fig. 4A). In sharp contrast, the direct effect of IAP on cell membranes causing ADP-ribosylation of the M, = 41,000 protein was antagonized by the anti-A-protomer antibody only; the anti- B-oligomer antibody was without effect in this regard (Fig. 4B) in accord with our previous reports (12) that the A- protomer was as effective as the native IAP when directly added to the membrane preparation. Thus, the B-oligomer moiety of the IAP molecule is essential for the whole IAP to exert its influence on intact cells.

Mitogenic Action of ZAP and Its B-Oligomer on Mouse Lymphocytes-The incorporation of [3H]thymidine into DNA of mouse splenic cells was accelerated by IAP as well as by concanavalin A (Fig. 5). The strongest action was observed when cells were exposed to these proteins at concentrations of 1-2 pg/106 cells (or 2-4 pglml) for 48 h. IAP resembles the lectin in that it is also a T-cell mitogen; it was not effective on splenic cells prepared from a thymus-deficient nude mouse.2 Fig. 5 also shows that the B-oligomer of IAP was as

* Y . Tsuchiya, K. Nogimori, K. Hosoda, and M. Ui, unpublished observations.

effective as the native IAP in promoting DNA synthesis, whereas the A-protomer was without effect.

We have reported (11) that the B-oligomer is a pentamer in which two dimers, referred to as Dimer-1 and Dimer-2, are associated by means of a connecting subunit (C-subunit or S- 5). None of these constituent peptides, Dimer-1, Dimer-2, and C-subunit, exhibited mitogenic activity when added to splenic cells each by itself (Fig. 6), though both dimers antagonized, but C-subunit did not, the action of the B-oligomer (data not shown, but see Fig. 7 later). Thus, a multiple attachment of the B-oligomer to the cell surface via its different constituent peptides, probably Dimer-1 and Dimer-2, appears to be essen- tial for the B-oligomer (or IAP) to exhibit its mitogenic activity.

Interference with IAP-induced Glycerol Release from Rat

01 10 100 0 20 60 180 500 lgG added (pg/rnl) lgG added (rg/ml)

FIG. 4. Suppression of the IAP actions on intact cells and isolated membranes by the antibody against its A-protomer or B-oligomer. In A , anti-A-protomer (0) or anti-B-oligomer (A) IgG was added to C6 cell cultures 15 min before the start of the usual 3-h IAP treatment. Temperature of culture was lowered to 27 "C during 15-min exposure to IgG only. GTP-dependent adenylate cy- clase activity of membranes prepared from these cells is plotted against concentrations of IgG added. In B, membranes prepared from nontreated cells were submitted to the usual ADP-ribosylation reac- tion with IAP in the presence of anti-A-protomer (0) or anti-B- oligomer (A) I&. ADP-ribosylation of the membrane M, = 41,000 protein is plotted against concentrations of IgG used.

FIG. 5. Mitogenic action of IAP and its B-oligomer on mouse lymphocytes in comparison with concanavalin A. Mouse splenic lymphocytes were cultured with IAP (O), A-protomer (A), B-oligomer (W), or concanavalin A (X) a t concentrations indicated on abscissa to estimate their mitogenic activities by [3H]thymidine incorporation as described under "Experimental Procedures." Culture time is shown in each panel.

by guest on Decem

ber 9, 2020http://w

ww

.jbc.org/D

ownloaded from

Role of B-Oligomer of Islet-activating Protein 6759

0 0.5 1 5 Protein added (rg/106cells)

FIG. 6. Failure of constituent peptides of B-oligomer to stimulate mitosis of lymphocytes. Splenic lymphocytes were cul- tured with IAP (O), B-oligomer (I), Dimer-l (A), Dimer-2 (V), or C- subunit (X) for 48 h as described in the legend to Fig. 5.

Adipocytes by B-Oligomer and Its Constituent Peptides-Di- rect addition of IAP into the incubation medium of rat adi- pocytes results in enhanced glycerol release from the cells associated with intracellular accumulation of cAMP after 2- to 4-h incubation? This action of IAP, too, may reflect mod- ification of the membrane receptor-adenylate cyclase system just as has been observed so far with other cell types. Mem- brane receptors for endogenous agonists such as adenosine are linked to adenylate cyclase in an inhibitory fashion in adipocytes, and reversal of this receptor-mediated inhibition, which was actually observed with membranes from IAP- treated adipocytes (4), must lead to cAMP accumulation and glycerol release. We have employed the glycerol release from adipocytes as an assay system for the IAP action on intact cells in Figs. 7 and 9 which is much simpler to study than membrane adenylate cyclase activity or ADP-ribosylation of a membrane protein as was measured with C6 cells in Figs. 1 and 4.

Preincubation of adipocytes with the B-oligomer for a while as short as 5 min produced the cells with this protein attached so firmly that IAP-induced glycerol release was suppressed in a competitive manner even after thorough washing (Fig. 7A). The suppression by the B-oligomer was specific to IAP- induced lipolysis; there was no difference in glycerol release between the B-oligomer-treated cells and the nontreated cells when it was induced by cholera toxin or glucagon (Table I). Conceivably, the B-oligomer interfered with the development of the IAP action as a result of competitive occupation of the same binding sites on the cell surface.

Not only the B-oligomer itself but also any of its constituent dimers, Dimer-1 and Dimer-2, was capable of suppressing IAP-induced glycerol release; the order of their efficiencies to interfere with the IAP action was B-oligomer > Dimer-1 > Dimer-2 (Fig. 7B). The C-subunit was without effect at all in this regard. Thus, it is very likely that two dimeric proteins comprised in the B-oligomer moiety are responsible for bind- ing of IAP to the cell surface of adipocytes leading to devel- opment of its biological action to cause glycerol release.

Insulin-like Action of IAP and Its B-oligomer-Lectins such as concanavalin A are known to mimic the action of insulin on white adipocytes in uitro (21-23). Since IAP and its B- oligomer were similar to concanavalin A in inducing mitosis of lymphocytes (Fig. 5), a possibility was next studied if these proteins also stimulate glucose oxidation in isolated fat cells. Fig. 8 shows that glucose oxidation was promoted to a similar extent by concanavalin A and insulin in adipocytes, though the lectin was much less potent than the hormone on the

M. Yajima, M. Tamura, K. Nogimori, and M. Ui, manuscript in preparation.

molar basis. IAP and its B-oligomer also mimicked the action of insulin to stimulate glucose oxidation; they were roughly equipotent to, but less efficient than, concanavalin A. More glucose was oxidized by the B-oligomer than by the native IAP at their maximal concentrations employed (Fig. 8).

The B-Oligomer Confers the A-Protomer Responsiveness on Intact Cells-The A-protomer of IAP was without effect on intact C6 cells, despite its capability to catalyze ADF-ribo- sylation of the membrane protein and to enhance GTP- dependent adenylate cyclase when added directly to the cell- free membrane preparation (12). In order to afford evidence for the indispensable role of the B-oligomer in occurrence of the A-protomer-catalyzed reactions in intact cells, we used the B-oligomer-bound adipocytes, which were prepared as in Fig. 7, again in experiments shown in Fig. 9.

In accord with our previous observation with C6 cells ( E ) , the A-protomer of IAP failed to stimulate glycerol release during incubation of the fresh fat cells. When fat cells bound with the B-oligomer were incubated, however, the addition of the A-protomer increased glycerol release therefrom in a concentration-dependent manner (Fig. 9). It is conceivable that the A-protomer was allowed to gain access to the site of

TABLE I Specific inhibition of IAP-induced glycerol release from adipocytes by

B-oligomer A d i p y t e s preincubated without (-) or with (+) 20 pg/ml of B-

oligomer for 10 min were further incubated without (none), or with IAP (30 ng/ml), cholera toxin (300 ng/ml) or glucagon (1 pg/ml) for 3 h to measure glycerol release. The data are means f S.E. from four observations.

Preincu- Glycerol release during incubation with

B-oligomer None bated with

IAP Cholera toxin Glucagon

pnwl l lOs cells - 0.20 f 0.03 2.81 & 0.15 1.90 * 0.10 1.02 k 0.05 + 0.25 f 0.04 0.30 & 0.03" 1.82 & 0.13 1.11 f 0.10

Effect of B-oligomer pretreatment is significant ( p < 0.001).

I'o

;\ e") 0 2 10 100 0 1 10 50

IAP added (ng/rnl) Protein added (rglrnl)

FIG. 7. Inhibition of IAP-induced glycerol release from rat adipocytes by B-oligomer or its constituent peptides. In A, adipocytes (4 X lo6 cells/ml) were preincubated without (0) or with 0.3 (O), 1 (A), or 3 (H) pg/ml of B-oligomer for 5 min at 37 "C before 3-h incubation at the same cell density with various concentrations of IAP as described under "Experimental Procedures." In B, adipo- cytes (2 X 10' cells/ml) were preincubated with various concentrations of B-oligomer (O), Dimer-1 (A), Dimer-2 (V), or C-subunit (H) for 10 min at 27 "C before 210-min incubation with 30 ng/ml of IAP. Glycerol released during incubation with IAP is expressed as a per- centage of the control value (obtained with adipocytes preincubated with vehicle only) in B.

by guest on Decem

ber 9, 2020http://w

ww

.jbc.org/D

ownloaded from

6760 Role of B-Oligomer of Islet-activating Protein

100 “4

Protein added -Log M A-Protomer added (ng/ml)

FIG. 8 (left). Insulin-like action of IAP, B-oligomer, and con- canavalin A on adipocyte glucose oxidation. Adipocytes (2 x lo5 cells/ml) were, after 30-min preincubation without additions, incu- bated with various concentrations of insulin (0), concanavalin A (A), IAP (O), B-oligomer (A), or C-subunit (0) in the presence of [l-”C] glucose for 2 h as described under “Experimental Procedures.” The maximal glucose oxidation obtained with more than 1 nM insulin was 4.5 times the control value obtained without any addition. Glucose oxidation on the ordinate is expressed as a percentage with the maximal and control values taken as 100 and 0, respectively. The molar concentrations of proteins were calculated with the M, value of 6,000 (insulin), 100,000 (concanavalin A), 117,000 (IAP), 80,000 (B-oligomer), and 9,300 (C-subunits).

FIG. 9 (right). Effect of B-oligomer to permit A-protomer to induce glycerol release from adipocytes. Adipocytes (6 X lo5 cells/ml) preincubated without (0) or with (0) 10pg/ml of B-oligomer for 5 min at 27 “C were washed, resuspended at the cell density of 3 X lo6 cells/ml, and further incubated with increasing concentrations of A-protomer for 4 h to measure glycerol release.

its action within the cell, provided the B-oligomer served as its acceptor on the cell surface.

DISCUSSION

The biggest subunit (S-1), referred to as the A-protomer, of IAP has been found to be an enzyme that catalyzes transfer of the ADP-ribose moiety of NAD to a membrane protein of rat C6 glioma cells (11, 12). The unique action of IAP on various cell types is reasonably accounted for in terms of this enzymic activity (4), despite failure of the A-protomer by itself to act on intact cells (12). Thus, the oligomeric protein consisting of the residual subunits of IAP was termed the B- oligomer with the expectation that it would bind to particular sites on the cell surface thereby rendering the associated A- protomer to traverse the plasma membrane to reach to the site of its action within the cell (11). The experimental data presented above are in accordance with this expectation; the B-oligomer competed with IAP for the same site(s) on C6 cells (Fig. 1). The biological action of IAP on intact adipocytes was also inhibited competitively by the B-oligomer (Fig. 7). The B-oligomer was not inhibitory in either cell type unless it was added before the onset of IAP action. The indispensable role of the B-oligomer moiety in the cellular action of IAP was verified by the inhibition of the action of IAP on intact C6 cells by the anti-B-oligomer antibody which was ineffective on the action of IAP on the cell-free membranes (Fig. 4).

Radioassay for binding of IAP or its B-oligomer to cells was not feasible because the subunit assembly of these proteins was readily impaired by iodination of their constituent amino acids. Instead, mitosis of lymphocytes and glucose oxidation in adipocytes have been employed in the present study as alternative indices of protein binding to the cell surface. It is well known that a lectin such as concanavalin A exhibits mitogenic activity as a result of its multipoint attachment to glycoproteins on the cell surface; cross-linking or clustering of these receptor proteins is probably involved in mitogenicity (24-27). Thus, increased [3H]thymidine incorporation into

splenic cells caused by the B-oligomer of IAP may reflect its multivalent binding to membrane proteins. The subunit struc- ture of the B-oligomer (11) is such that it consists of two dimers (Dimer-1 and Dimer-2) which are connected by the C- subunit (S-5). It is very likely that the B-oligomer is bound divalently to the surface of the cell by means of the two dimers, since any of the dimers was not mitogenic by itself (Fig. 6) despite their binding to the cells as revealed by their competition with IAP on adipocytes (Fig. 7). The C-subunit does not appear to possess a binding site; it was without effect on the biological activity of IAP in adipocytes (Fig. 7). IAP must be bound to the cell surface in the same manner as its B-oligomer, because both proteins were equally effective as mitogens (Fig. 5 ) . The B-oligomer is the moiety that is re- sponsible for the binding of IAP to the cell membrane.

The insulin-like activity of IAP and its B-oligomer to enhance glucose oxidation in adipocytes (Fig. 8) affords ad- ditional evidence for divalent binding of the B-oligomer to the cell surface. Cross-linking or aggregation of receptor pro- teins achieved by multipoint attachment to the proteins has been suggested to be responsible for the insulin-like action of the antibodies to the insulin receptors (28) or intrinsic mem- brane proteins (29) as well as concanavalin A (30, 31). The same mechanism must be involved in the insulin-like action of the B-oligomer of IAP. The native IAP was less effective than its B-oligomer in stimulating glucose oxidation in adi- pocytes. Presumably, the insulin-like action displayed by the B-oligomer moiety of IAP attached to the fat cell surface was partially cancelled by the subsequent entry of its A-protomer moiety which results in tremendous increases in the cellular CAMP content? In accord with this notion, glucagon or epi- nephrine, which acts on adipocytes via generation of CAMP, is somewhat antagonistic to insulin, e.g. lipolysis is increased by glucagon or epinephrine while decreased by insulin.

Pertussis toxin exhibits such diverse biological activities as to be termed lymphocytosis-promoting factor, histamine-sen- sitizing factor, hemagglutinin, apart from IAP (32, 33). In addition, the toxin acts as a potent adjuvant. Our findings that not only the A-protomer but also the B-oligomer was biologically active when IAP interacted with various cell types will contribute to elucidation of such diversity of the toxin action; we would propose that mitogenicity of the B-oligomer shown in the present study plays an important role in devel- opment of lymphocytosis-promoting, hemagglutinating, and adjuvant activities of pertussis toxin. The in vitro mitogenicity of pertussis toxin for mouse T-lymphocytes was previously described by Kong and Morse (34, 35).

In summary, the B-oligomer is the moiety of IAP directly involved in the IAP binding to the cell surface. The binding is the first step required for the A-protomer, ADP-ribosyl- transferase, of IAP to enter the cell. Actually, the A-protomer was ineffective on adipocytes unless the B-oligomer had been bound to the cells (Fig. 9). Since the binding of the B-oligomer to the cell surface is divalent via the two constituent dimers (Dimer-1 and Dimer-2) connected with each other by means of another subunit (C-subunit), it stimulates the cell, probably as a result of cross-linking or aggregation of the binding proteins on the cell membrane, in such a fashion as to cause mitosis of lymphocytes or insulin-like action on adipocytes. Further studies are now in progress in our laboratory for a possible role of the B-oligomer of IAP in diverse biological actions of whooping cough bacteria.

REFERENCES 1. Katada, T., and Ui, M. (1982) Proc. Natl. Acad. Sci. U. s. A. 79,

2. Katada, T., and Ui, M. (1982) J. BWl. Chern. 257,7210-7216 3129-3133

by guest on Decem

ber 9, 2020http://w

ww

.jbc.org/D

ownloaded from

Role of B-Oligomer of Islet-activating Protein 6761

3. Murayama, T., Katada, T., and Ui, M. (1983) Arch. Biochem.

4. Murayama, T., and Ui, M. (1983) J. Biol. Chem. 258,3319-3326 5. Kurose, H., Katada, T., Amano, T., and Ui, M. (1983) J. Biol.

6. Katada, T., and Ui, M. (1979) J. Biol. Chem. 254,469-479 7. Hazeki, O., and Ui, M. (1981) J. BioZ. Chem. 2 5 6 , 2856-2862 8. Katada, T., and Ui, M. (1981) J. Biochem. (Tokyo) 89,979-990 9. Katada, T., and Ui, M. (1981) J. Biol. Chem. 256,8310-8317

10. Katada, T., Amano, T., and Ui, M. (1982) J . Biol. Chem. 257,

11. Tamura, M., Nogimori, K., Murai, S., Yajima, M., Ito, K., Katada, T., Ui, M., and Ishii, S. (1982) Biochemistry 21 , 5516-5522

12. Katada, T., Tamura, M., and Ui, M. (1983) Arch. Biochem. Biophys., in press

13. Gill, D. M. (1978) in Bacterid Toxins and Cell Membranes (Jel- jaszewicz, J., and Wadstrom, T., eds) pp. 291-332, Academic Press, New York

14. Yajima, M., Hosoda, K., Kanbayashi, Y., Nakamura, T., Nogi- mori, K., Nakase, Y., and Ui, M. (1978) J. Biochem. (Tokyo)

15. Yajima, M., Hosoda, K., Kanbayashi, Y., Nakamura, T., Taka-

16. Katada, T., and Ui, M. (1980) J. Biol. Chem. 255 , 9580-9588 17. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J.

18. Honma, M., Satoh, T., Takezawa, J., and Ui, M. (1977) Biochem.

19. Rodbell, M. (1964) J. Biol. Chem. 2 3 9 , 375-380 20. Pinter, J. K., Hayashi, J. A., and Watson, J. A. (1967) Arch.

Biophys. 22 1 , 381-390

Chem. 258,4870-4875

3739-3746

83,295-303

hashi, I., and Ui, M. (1978) J. Biochem. (Tokyo) 8 3 , 305-312

(1951) J. Biol. Chem. 193,265-275

Med. 18,257-273

Biochem. Biophys. 121, 404-414

21. Cuatrecasas, P., and Tell, G. P. E. (1973) Proc. Natl. Acad. Sci.

22. Czech, M. P., and Lynn, W. S. (1973) Biochim. Biophys. Acta

23. Lawrence, J. C., and Lamer, J. (1978) J. Biol. Chem. 253,2104- 2113

24. Gunther, G. R., Wang, J. L., Yahara, I., Cunningham, B. A., and Edelman, G . M. (1973) Proc. Natl. Acad. Sci. U. S. A. 70,1012- 1016

25. Ruddon, R. W., and Weisenthal, L. M., Lundeen, D. E., Bessler, W., and Goldstein, I. J . (1974) Proc. Natl. Acad. Sci. U. S. A.

26. Wands, J. R., Podolsky, D. K., and Isselbacher, K. J. (1976) Proc.

27. Wang, J. L., and Edelman, G. M. (1978) J . Biol. Chem. 253 ,

28. Kahn, C. R., Baird, K. L., Jarrett, D. B., and Flier, J. S. (1978)

29. Pillon, D. J., Grantham, J. R., and Czech, M. P. (1979) J. Biol.

30. Kahn, C. R., Baird, K. L., and Van Obberghen, E. (1981) FEBS

31. Suya, H., Abe, Y., Tanaka, I., Ishii, S., and Itaya, K. (1982) J. Biochem. (Tokyo) 9 2 , 1251-1257

32. Munoz, J. J., and Bergman, R. K. (1978) in International Sym- posium on Pertussis (Manclark, C. R., and Hill, J. C., eds) DHEW Publication No. (NIH)79-1830, pp. 143-150, National Institutes of Health, Bethesda, Maryland

U. S. A. 70, 485-489

297,368-377

71,1848-1851

Natl. Acad. Sci. U. S. A. 73, 2118-2122

3000-3007

Proc. Natl. Acad. Sei. U. S. A. 75,4209-4213

Chem. 254,3211-3220

Lett. 129 , 131-134

33. Pittman, M. (1979) Reo. Infect. Dis. 1,401-412 34. Kong, A. S., and Morse, S. I. (1977) J. Exp. Med. 145, 151-162 35. Kong, A. S., and Morse, S. I. (1977) J. Exp. Med. 145,163-174

by guest on Decem

ber 9, 2020http://w

ww

.jbc.org/D

ownloaded from

M Tamura, K Nogimori, M Yajima, K Ase and M Uidevelopment of the biological effects on intact cells.

A role of the B-oligomer moiety of islet-activating protein, pertussis toxin, in

1983, 258:6756-6761.J. Biol. Chem. 

  http://www.jbc.org/content/258/11/6756Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/258/11/6756.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on Decem

ber 9, 2020http://w

ww

.jbc.org/D

ownloaded from