Vol. 22, No. 2

INFECTION AND IMMUNITY, Nov. 1978, p. 430-434

0019-9567/78/0022-0430$02.00/0 Copyright i 1978 American Society for Microbiology

Printed in U.S.A.

Bordetella pertussis Does Not Induce fl-Adrenergic Blockade E. HEWLETT,t * A. SPIEGEL,' J. WOLFF,' G. AURBACH, AND C. R. MANCLARK2 National Institute of Arthritis, Metabolism, and Digestive Diseases, National Institutes of Health, Bethesda, Maryland 20014,' and Division of Bacterial Products, Bureau of Biologics, Food and Drug Administration, Bethesda, Maryland 200142

Received for publication 15 August 1978

Bordetella pertussis organisms induce histamine sensitivity and diminish the normal hyperglycemic response to epinephrine in experimental animals. These effects have been attributed to ,8-adrenergic blockade. However, under conditions in which the decrease in epinephrine-induced hyperglycemia after B. pertussis administration was demonstrable, there was no change in rat reticulocyte ,8adrenergic receptor number or affinity measured by iodohydroxybenzylpindolol binding or in isoproterenol-stimulated adenylate cyclase activity. Therefore, there was no generalized,-adrenergic blockade induced by B. pertussis. The observed effects can be explained by the hypersecretion of insulin resulting from B. pertussis administration. Administration of Bordetella pertussis organisms, or a protein from the supernatant culture medium, to rats or mice causes a decrease in the hyperglycemic response to epinephrine and an increase in sensitivity to the lethal effects of histamine (9, 11, 15, 16, 21). Because both of these effects are mimicked by ,B-adrenergic antagonists, it has been postulated "that the pertussis organism possesses, or causes the host to elaborate a specific substance with a steric configuration complementary to the ,B-adrenergic receptors," resulting in a generalized and sustained f.8-adrenergic blockade (20). Both experimental administration of B. pertussis and clinical whooping cough are reported to cause /3adrenergic blockade (2, 10). Since the development of high-affinity specific ligands such as iodohydroxybenzylpindolol (IHYP)(1) and dihydroalprenolol (12), it has become possible to evaluate ",B blockade" directly by measurement of 83-adrenergic receptor number and affinity. In the present study, IHYP binding and catecholamine-responsive adenylate cyclase were measured in reticulocyte membranes from rats given B. pertussis vaccine. The data show that under conditions in which the characteristic decrease in epinephrine-induced hyperglycemia is demonstrable, there is no evidence for blockade of the rat reticulocyte /8 receptor. MATERIALS AND METHODS

organisms were given in the form of whole-cell B. pertussis vaccine (Bureau of Biologics lot 7b) which was detoxified by mild heating and contained thimerosal (1:10,000). Each animal received a single intraperitoneal injection of 0.5 ml containing approximately 2.4 x 1010 organisms at the time indicated. Soluble material from B. pertussis vaccine was prepared by sonic disruption of Lot 7b vaccine (30-s pulse x 6 at an output setting of 50, with a Sonifier Cell Disruptor, model W140D, Heat Systems-Ultrasonics, Inc., Plainview, N.Y.). Disrupted cell fragments were removed by centrifugation at 100,000 x g for 40 min. The supernatant material was evaluated for: (i) ability to reproduce untreated vaccine effects on hyperglycemic response to epinephrine; and (ii) effect of IHYP binding to reticulocyte membranes in vitro. Hemolysis was induced in groups III, IV, and V by intramuscular injection of 1-acetyl-2-phenylhydrazine (APH; Sigma Chemical Co.) on days 2, 3, and 4. This treatment resulted in reticulocytosis of >90% in each group. Each treatment group (Table 1) was subdivided before sacrifice. Half of each group (five to seven animals) was used to obtain reticulocytes and determine basal serum glucose (hexokinase method, Statzyme reagents, Worthington Diagnostics, Freehold, N.J.). The other half was challenged with 0.5 ml of 1epinephrine bitartrate (Sigma) subcutaneously 30 min before sacrifice to determine the response of serum glucose to epinephrine (Table 2). Preparation of reticulocyte membranes, measurement of specific IHYP binding, and assay of adenylate cyclase were performed as previously described (3). The specificity and reversibility of IHYP binding and the effects of ,Badrenergic agonists and antagonists were established for the rat reticulocyte system in that study (3).

Male Sprague-Dawley rats (Taconic Farms, Inc., Germantown, N.Y.), weighing 250 + 40 g, were treated according to the protocol in Table 1. B. pertussis

RESULTS AND DISCUSSION Control animals (group I) responded normally to epinephrine, with an 81% increase in serum glucose (Table 2). The APH-treated, unvaccinated animals (group III) also showed a normal

t Present address: Division of Geographic Medicine, Department of Medicine, Case Western Reserve University School of Medicine, Cleveland, OH 44106.



VOL. 22, 1978

TABLE 1. Protocol for B. pertussis administration and induction of reticulocytosis Treatment on day: Group 1






Sacrifice I NSSa Sacrifice II VACb Sacrifice APHCAPH APH III Sacrifice IV VAC APH APH APH V APH APH APH VAC Sacrifice a NSS, Sterile saline intraperitoneally in 0.5 ml. bVAC, B. pertussis lot 7b vaccine, approximately 2.4 x 1010 organisms intraperitoneally in 0.5 ml. cAPH, 70 mg/kg intramuscularly in 0.5 ml.


As shown in Fig. 1, there was no difference in the ability of isoproterenol to displace specifically bound (3) '25IHYP from membranes of control versus vaccinated animals. There was also no reduction in specific 125IHYP binding sites in the membranes from the two groups of vaccinated animals as compared with unvaccinated (group IV, 49 fmol/mg of protein; group V, 44 fmol/mg of protein versus group III, 33 fmol/mg of protein). Similarly, the activity of adenylate cyclase and the response to isoproterenol were the same for reticulocyte membranes from control and pertussis-treated animals (Fig. 2). Thus, both the affinity and number of reticulocyte f8-adrenergic receptors as measured by direct binding studies and the adenylate cyclase response to isoproterenol are unaltered in rats receiving B. pertussis. To evaluate the direct effect of a soluble extract of pertussis vaccine on IHYP binding, the whole-cell vaccine was disrupted by sonic treatment, and the supernatant (100,000 x g) material was used for study. The extract from 2.4 x 1010 organisms produced the same effect on the glucose response to epinephrine as did the whole vaccine (Table 3). The vaccine extract was then assayed for its ability to alter IHYP binding. As shown in Table 4, pertussis vaccine extract had no effect on IHYP binding in vitro in either the absence or presence of isoproterenol or excess unlabelled HYP. These findings, that B. pertussis does not cause generalized /3-adrenergic blockade, in vivo or in vitro, make it imperative that we reexamine the evidence upon which that hypothesis was based. One of the arguments for f-blockade was based on the observation that the response of adenylate cyclase activity or cyclic AMP accumulation to isoproterenol is diminished in cer-

response (98% increase). Vaccination alone caused the expected blunting of hyperglycemia after epinephrine (35% increase; different from -control, P < 0.005). When vaccine was given either before (group IV) or after (group V) APH treatment, the glycemic response to epinephrine was completely eliminated. The reason for increased vaccine effect in the hemolyzed animals is not known, but it is apparent from the response of group III that APH alone has no effect on epinephrine responsiveness. Because the hemolysis did not interfere with the expected effects of pertussis administration and because the development of reticulocytosis was not affected by pertussis treatment, this model system was considered a reasonable one in which to examine pertussis effects on the /8 receptor and adenylate cyclase. Also, the rat reticulocyte /8-adrenergic receptor has the characteristics of a,82 receptor (3), which is the same as that in liver, bronchi, and lymphocytes, the targets of the apparent blocking effects of pertussis in vitro (10) and in clinical whooping cough (2). The generation of a massive reticulocytosis after APH administration allowed the study of the effect of pertussis vaccination on TABLE 2. Response of blood glucose to epinephrinea Blood glucose (mg/dl ± SEM) the beta receptor under two different conditions. In group IV, the pertussis vaccine was adminis% Change Basal Group Post-epinephtered just before hemolysis, and thus before the rine treatment production, differentiation, and release of the +81 228 ± 15 126 ± 6 I reticulocytes into the circulation. Group V ani161 ± 14 +35b 119 ± 5 II mals received the pertussis dose after hemolysis, +98 ±5 232 ± 11 117 III near the peak of reticulocytosis. Because kinetic 108 ± 10 120 ± 4 -10 IV data indicate that reticulocytes remain in the -6 115 ± 30 V 122 ± 8 massive a response circulation for 3 days during treated were rats male according Sprague-Dawley such as this (6), the cells from group V animals to the protocol in Table 1. Food was removed 16 h which were circulating at the time of vaccination before sacrifice on day 8. Half of each group (five to were essentially the same cells as those collected seven animals) were given 0.5 mg of epinephrine bifor membrane preparation and assay. By study- tartrate subcutaneously 30 min before sacrifice. Blood ing reticulocytes generated both before (group was removed for glucose assay by cardiac puncture. V) and after (group IV) vaccination, we would Glucose was assayed by the hexokinase method. SEM, error of the mean. detect any potential fl-blocking effects of B. Standard b Different from group I, P < 0.005. pertussis.





0r a z D 0o




5 4 [Isoproterenol I - LOG M FIG. 1. '25IHYP binding as a function of isoproterenol concentration. Reticulocyte membranes from group III (control) (@), group IV (vaccinated, then hemolyzed) (A), and group V (hemolyzed, then vaccinated) (A) animals were incubated with increasing concentrations of (-)-isoproterenol. The radioligand concentration was 35 pM. Results are expressed as percentage of maximal specific binding (group III, 33 fmol/mg ofprotein, group IV, 49 fmol/mg ofprotein, group V, 44 fmol/mg ofprotein) and are the mean of triplicate determinations. ' """I



'"""I 'J """

100 co







6 5 4 [Isoproterenol ]-LOG M FIG. 2. Adenylate cyclase activity as a function of isoproterenol concentration. Reticulocyte membranes from group III (0), IV (A), and V (A) animals were incubated with increasing concentrations of (±)isoproterenol and adenylate cyclase activity was assayed as described previously (11). The results are expressed as percentage of maximal isoproterenol-stimulated enzyme activity (group III, 1,650 pmol of cyclic AMP per mg ofprotein per 10 min; group IV, 1,890 pmol of cyclic AMP per mg ofprotein per 10 min; group V, 1, 750 pmol of cyclic AMP per mg of protein per 10 min) and are the mean of triplicate determinations.

tain tissues by B. pertussis (14, 17). However, this decrease was also demonstrable in the response to PGE, and fluoride, and hence is not an effect at the fl-receptor level, but rather at some distal site (14, 17). In addition, our study with rat reticulocyte membranes and other studies (7, 13) show that this altered response is not generalized, but is restricted to certain tissues. The other major studies proposing pertussisinduced fl-adrenergic blockade were based on the initial observation that epinephrine-induced hyperglycemia is decreased or absent after B.

pertussis treatment (20). It was subsequently discovered that pertussis-vaccinated animals have augmented glucose tolerance with increased peripheral utilization, enhanced glycogen deposition, and decreased free-fatty acid release in response to epinephrine (4). Children with clinical whooping cough also have a decreased blood sugar response to subcutaneous epinephrine (2). Despite the fact that each of these effects could also be explained by hyperinsulinemia, it was concluded that the phenomena were due to ,8 blockade with loss of the ,f-


VOL. 22, 1978

TABLE 3. Response of blood glucose to epinephrine after soluble B. pertussis extract Blood glucose (mg/dl + SEM)'




% Change

rine treatment 251 ± 3 180 ± 13

+151% 100 ± 6 I +76%c 102 ± 18 II a SEM, Standard error of the mean. b Animals received 100,000 x g supernatant from ultrasonic disruption of 2.4 x 100' organisms intraperitoneally in 0.5 ml. c Different from group I, P < 0.01.

TABLE 4. Effect of soluble B. pertussis extract on IHYP binding to control reticulocyte membranes in

vitro a


None 1-Isoproterenol (10-5 M) 1-Isoproterenol (10-4 M) HYP (10-7 M)

Pertussistreated membranes


8,944 ± 320 6,918 ± 643 4,136 ± 403 2,622 ± 599

9,218 ± 482 7,105 ± 451 3,722 ± 108 2,228 ± 54


a Reticulocyte membranes (0.2 ml) were combined with either an equal volume of (i) soluble B. pertussis extract (pertussis-treated) containing 0.76 mg of protein per ml, or (ii) bovine serum albumin (1.0 mg/ml). The mix was then incubated at 22°C for 10 min. Portions (60 Al) of each membrane solution were then incubated with '25IHYP (250,000 cpm) and additions as shown above in a total volume of 0.5 ml at 37°C for 10 min. Aliquots (100,ul) in triplicate were filtered by suction through Gelman A/E glass filter. Results shown are mean counts per minute ± standard deviation of triplicate.

adrenergic antagonism to insulin action (4). It is now known, however, that B. pertussis administration is associated with profound alterations in the response of insulin secretion to assorted secretagogues (5, 8, 19). Normally, epinephrine has a predominantly a-adrenergic effect, inhibiting insulin secretion and lowering circulating insulin levels (5, 8, 18, 19). In contrast, rats or mice given B. pertussis vaccine develop marked hyperinsulinemia and respond to epinephrine with a further increase in circulating insulin levels (5). Pertussis treatment also enhances insulin secretion in response to glucose, isoproterenol, isobutyl methylxanthine and arginine (8). Because our present results rule out generalized f?-adrenergic blockade, the effects on glucose metabolism induced by B. pertussis treatment are best explained by this enhanced insulin secretion. It is not yet clear whether the active component from B. pertussis exerts its effect directly on the f8 cell of the pancreas or


whether the hyperinsulinemia and enhanced secretion represent secondary responses. In any case, these data illustrate that adrenergic responses cannot be reliably interpreted simply on the basis of effects on blood glucose. LITERATURE CITED 1. Aurbach, G. D., S. A. Fedak, C. J. Woodward, J. S. Palmer, D. Hauser, and F. Troxler. 1974. ,B-Adrenergic receptor: stereospecific interaction of iodinated ,8-blocking agent with high affinity site. Science 186:1223-1224. 2. Badr-El-Din, M. K., G. H. Aref, H. Mazloum, Y. A. ElTowesy, A. S. Kassem, M. A. Abdel-Moneim, and A. Amr Abbassy. 1976. The beta-adrenergic receptors in pertussis. J. Trop. Med. Hyg. 79:213-217. 3. Bilezikian, J. P., A. M. Spiegel, E. M. Brown, and G. D. Aurbach. 1977. Identification and persistence of beta adrenergic receptors during maturation of the rat reticulocyte. Mol. Pharmacol. 13:775-785. 4. Fishel, C. W., and A. Szentivanyi, 1963. The absence of adrenaline-induced hyperglycemia in pertussis-sensitized mice and its relation to histamine and serotonin hypersensitivity. J. Allergy 34:439-454. 5. Gulbenkian, A., L. Schobert, C. Nixon, and I. I. A. Tabachnick. 1968. Metabolic effects of pertussis sensitization in mice and rats. Endocrinology 83:885-892. 6. Hillman, R. S. 1969. Characteristics of marrow production and reticulocyte maturation in normal man in response to anemia. J. Clin. Invest. 48:443-453. 7. Katada, T., and M. Ui. 1976. Accelerated turnover of blood glucose in pertussis-sensitized rats due to combined actions of endogenous insulin and adrenergic beta-stimulation. Biochim. Biophys. Acta 421:57-69. 8. Katada, T., and M. Ui. 1977. Perfusion of the pancreas isolated from pertussis-sensitized rats: potentiation of insulin secretory responses due to /1-adrenergic stimulation. Endocrinology 101:1247-1255. 9. Kind, L. S. 1958. The altered reactivity of mice after inoculation with Bordetella pertussis vaccine. Bacteriol. Rev. 22:173-182. 10. Morse, S. I. 1977. Whooping cough, p. 277-280. In P. D. Hoeprich (ed.), Infectious diseases. Harper and Row,

Hagerstown. 11. Morse, S. I., and J. H. Morse. 1976. Isolation and prop-

erties of the leukocytosis- and lymphocytosis-promoting factor of Bordetella pertussis. J. Exp. Med. 413:1483-1502. 12. Mukherjee, C., M. G. Caron, M. Coverstone, and R. J. Lefkowitz. 1975. Identification of adenylate cyclasecoupled ,8-adrenergic receptors in frog erythrocytes with (-)-[:'H] alprenolol. J. Biol. Chem. 250:4869-4876. 13. Ortez, R. A. 1976. Effect of histamine on lung cyclic AMP levels in normal and pertuasis-vaccinated mice. Biochem. Pharmacol. 25:523-526. 14. Ortez, R. A., D. Seshachalam, D., and A. Szentivanyi. 1975. Alterations in adenyl cyclase activity and glucose utilization of Bordetella pertussis-sensitized mouse spleen. Biochem. Pharmacol. 24:1297-1302. 15. Parfentjev, I. A., and M. A. Goodline. 1948. Histamine shock in mice sensitized with Hemophilu.s pertussis vaccine. J. Pharmacol. Exp. Ther. 92:411-413. 16. Parfentjev, I. A., and W. L. Schleyer. 1949. The influence of histamine on the blood sugar level of normal and sensitized mice. Arch. Biochem. 20:341-346. 17. Parker, C. W., and S. I. Morse. 1973. The effect of Bordetella pertussis on lymphocyte cyclic AMP metabolism. J. Exp. Med. 137:1078-1090. 18. Porte, D., Jr., A. L. Graber, T., Kuzuya, and R. A.



Williams. 1966. The effect of epinephrine on immunoreactive insulin levels in man. J. Clin. Invest. 45:228-236. 19. Sumi, T., and M. Ui. 1975. Potentiation of the adrenergic beta-receptor-mediated insulin secretion in pertussissensitized rats. Endocrinology 101:1247-1255. 20. Szentivanyi, A., C. W. Fishel, and D. W. Talmadge. 1963. Adrenaline mediation of histamine and serotonin

INFECT. IMMUN. hyperglycemia in normal mice and the absence of adrenaline-induced hyperglycemia in pertussis-sensitized mice. J. Infect. Dis. 113:86-98. 21. Yajima, M., K. Hosoda; Y. Kanbayashi, T. Nakamura, I. Takahashi, and M. Ui. 1978. Biological properties of islets-activating protein (IAP) purified from the culture medium of Bordetellapertussis. J. Biochem. (Tokyo) 83:305-312.

Bordetella pertussis does not induce beta-adrenergic blockade.

Vol. 22, No. 2 INFECTION AND IMMUNITY, Nov. 1978, p. 430-434 0019-9567/78/0022-0430$02.00/0 Copyright i 1978 American Society for Microbiology Prin...
649KB Sizes 0 Downloads 0 Views