EXPERIMENTAL
PARASITOLOGY
T~panosoma
74,69-76 (1992)
cruzi: Alteration of CAMP Metabolism Infection of Human Endothelial Cells
STEPHEN A. MORRIS,*Y'HERBERTTANOWITZ,* VICTOR B. HATCHER,* JOHN P. BILEZIKIAN,~AND
following
MAYNARDMAKMAN, MURRAY WITTNERt
*Departments of Pathology and Medicine and Biochemistry, Albert Einstein College of Medicine, Bronx, New York, New York 10461 USA; and tDepartments of Medicine and Pharmacology, College of Physicians and Surgeons, Columbia University, New York, New York 10032 USA
S. A., TANOWITZ, H., MAKMAN, M., HATCHER, V. B., BILEZIKIAN, J. P., AND (1992). Trypanosoma cruzi: Alteration of CAMP metabolism following infection of human endothelial cells. Experimental Parasitology 74, 69-76. We have previously reported that Trypanosoma cruzi infection of endothelial cells results in alterations in the metabolism of Cazf , inositol triphosphate (IP,), and prostacycline (PC&). In this report, we demonstrate that infection also alters the metabolism of CAMP. Infection of endothelial cells does not significantly alter P-adrenergic receptor density or affinity, adenylate cyclase activity, and whole-cell CAMP levels. However, incubation of infected endothelial cells with the phosphodiesterase inhibitor isobutylmethylxanthine (IBMX) resulted in less than a 60% increase in cell CAMP in contrast to the greater than a 100% increase observed in uninfected endothelial cells under otherwise identical reaction conditions. Infected endothelial cells demonstrated a twofold increase in phosphodiesterase activity when measured directly. Moreover, homogenates prepared from infected endothelial cells previously incubated with isoproterenol for 20 min showed little or no change in PDE activity. In contrast, homogenates prepared from uninfected endothelial cells treated under otherwise identical reaction conditions showed a 5.7-fold increase in PDE activity. In the presence of IBMX, isoproterenol-dependent stimulation of CAMP levels in infected endothelial cells reached a maximum level at 5 min of incubation, and thereafter rapidly declined. In contrast, CAMP levels in uninfected endothelial cells reached a maximum at 2 min of incubation, and thereafter remained elevated throughout the duration of the incubation. Infection-associated changes in isoproterenol dependent stimulation of CAMP accumulation appear to relate, in part, to changes in PDE activity. 8 19% Academic press. IK. Hemoflagellates; Trypanosoma cruzi; EndoINDEX DESCRIPTORS AND ABBREVIATIONS: thelial cells: Cvclic AMP (CAMP); Isobutylmethylxanthine (IBMX); Phosphodiesterase activity (PDE). MORRIS,
WITTNER,
M.
and in vitro are consistent with abnormalities in the microvasculature, resulting in foThe pathophysiological basis for cardiac cal pathology. In support of this hypothesis involvement in Chagas’ disease remains un- we have demonstrated that direct infection clear. We (Morris et al. 1989,199O)and oth- of endothelial cells in vitro results in the ers (Rossi et al. 1984) have hypothesized a perturbation of a number of biochemical central role for the microvasculature in processes thought to be critical to the maincontributing to the cardiac pathology ob- tenance of microvasculature perfusion served in both the chronic and acute stages (Morris et al. 1988). For example, infection of the disease. Pathological studies in vivo of endothelial cells has been shown to interfere with Ca2’ homeostasis (Morris et ’ To whom correspondence and reprint requests al. 1988), inositol triphosphate (IP3) genershould be addressed at Department of Medicine, College of Physicians and Surgeons, 630 West 168th ation (Morris et al. 1989) and the synthesis of prostaglandin I, (Tanowitz et al. 1990). Street, New York, NY 10032. 69 0014-4894/92$3.00 Copyright Q 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
70
MORRIS ET AL.
More recent investigations by others (Luckhoff et al. 1990; Schafer et al. 1984; Stelzner et al. 1989) into the biochemical basis by which endothelial cells participate in the regulation of microvascular perfusion have focused on the role of CAMP metabolism. In this regard, a number of observations have demonstrated areas in which CAMP may play a critical role. Cyclic AMP was shown to influence the concentration of intracellular Ca2+ in response to administration of ATP (Luckhoff et al. 1990) and play a protective role in defending endothelial cells from direct or indirect damage caused by adhesion and aggregation of circulating platelets. In addition, CAMP has been assigned an important role in the regulation of angiotensin converting enzyme synthesis, an enzyme involved in the formation of the vasoconstrictor angiotensin II (Lloyd et al. 1987). In this report, we show that infection of endothelial cells with Trypanosoma cruzi results in an alteration of CAMP metabolism. MATERIALS
AND METHODS
"'1 , [w~~P]ATP, and [3H]cAMP were Materials. obtained from New England Nuclear-duPont (Boston, MA). S’Guanylylimidodiphosphate (Gpp(NH)p) was obtained from Sigma Corporation (St. Louis, MO). All other reagents were of the highest purity commercially available. Culture and maintenance of endothelial cells. Human umbilical vein endothelial cells were isolated according to the method of Jaffe (Jaffe 1980)and grown as described by Gordon et al. (1985). Parasites. The Tulahuen strain of T. cruzi was passaged by syringe in A/J mice (Jackson Laboratories, Bar Harbor, ME; Trischmann et al. 1978;Morris et al. 1988). Trypomastigotes used for infection of endothelial cells were maintained in L,E, myoblasts as previously described (Rowin et al. 1983;Morris et al. 1984). Infection of endothelial cells was performed as described (Morris et al. 1984). Cells were harvested after 2 to 3 days of infection when the percentage of parasitism ranged from 5 to 50% as determined by May Grunwald-Giemsa staining (Tanowitz et al. 1990). Determination of P-receptor sites. The binding assay used [‘*51]iodocyanopindolol (ICYP). Cyanopindolol was iodinated and purified to a specific activity of 2200 Ci/mmol according to methods previously pub-
lished (Morris et al. 1988). Aliquots of endothelial cell membranes (100-300 up) were distributed to tubes containing [“‘IIICYP in a final volume of 1 ml whole cell buffer (0.15 M NaCl, 0.01 M Tris at pH 7.5), 0.01 M KCl, and 1 mg/ml bovine serum albumin with 2 mgiml dextrose. The binding assay was carried out in a shaking water bath for 20 min at 37°C. By 20 min of incubation under these conditions, binding of [“‘I]ICYP had reached equilibrium (results not shown) and values obtained from longer incubation periods (30-60 min) were identical to those reported here. The reaction was terminated by filtering the suspension over Gelman A/E glass fiber filters. The filters were washed with 10 ml 0.01 M Tris at room temperature, and radioactivity was determined in a Packard Auto-Gamma scintillation counter. Specific binding was defined as the difference between total binding and binding inhibited by (-)propranolol (0.1 PM). Maximal binding capacity was determined by Scatchard analysis (Morris et al. 1988) using the LIGAND binding analysis program (Munson and Rodbard 1980). Adenylate cyclase assay. The reaction mixture contained 50 mM Tris-HCI (pH 7.4), 2.5 mM magnesium unless otherwise noted, 0.143 mM ATP, an ATP regenerating system (creatine phosphate/creatine phosphokinase), ATP ([w~*P]ATP, 1-2 X lo6 cpm/assay tube) and 6 mM theophylhne in a final volume of 75 ~1 (Bilezikian et al. 1978). Agents dissolved in 50 mM Tris-HCl (pH 7.5) were added in concentrations noted in individual experiments. The adenylate cyclase reaction was terminated by the addition of an ATP-cyclic AMP stopping solution. Isolation of [32P]cyclic AMP was accomplished by sequential Dowex and alumina chromatography using [3H]cyclic AMP as a recovery marker (Salomon et al. 1974). Measurement of CAMP levels. For the measurement of CAMP levels in intact cells, the medium was removed rapidly and an aqueous acetic acid solution (1: I v/v) was added. The resulting suspensions were heated for 3 min at 90°C and centrifuged to remove cellular debris. Aliquots of the supernatant fluid were dried at 80°C in small glass tubes and assayed for CAMP by the modified binding assay of Gilman (Brown et a[. 1975). Determination of protein levels. Protein concentration was determined using the Bio-Rad method (BioRad Laboratories; Richmond, CA). Phosphodiesterase activity. Cyclic nucleotide phosphodiesterase activities were determined in a two-step incubation method previously reported (Duttagupta et al. 1974). Plates of endothelial cells prepared as described were washed two times in ice-cold 0.9% saline and scraped into a buffer containing 0.25 M sucrose and 50 mM Tris. The suspension was homogenized using a Polytron homogenizer. For the first incubation, aliquots (100-300 wg protein) of this homogenate were distributed to tubes containing [‘HIcAMP (25 Ci/ mmole, approximately 25,000 CPM@, MgCl, [5 mM],
T. cruzi: INFECTION
AND CELL PHOSPHODIESTERASE
2 mercaptoethanol [3.75 mM], and Tris buffer [40 mM], pH 8; incubated at 30°C for 10 min in a shaking water bath; and terminated by boiling for 3 min. In the second incubation, snake venom nucleotidase was added to the suspensions, which were then incubated for 10 min at 30°C. The second reaction was stopped by the addition of a 1:3 slurry of Bio-Rad resin AGlX2, followed by centrifugation and counting of [3H]adenosine in the supematant in a Packard liquid scintillation counter. RESULTS
The possibility that infection of endothelial cells may alter the density or affinity of the p-adrenergic receptors responsible for the activation of adenylate cyclase was first examined. As shown in Fig. 1, analysis of endothelial cell membranes prepared from uninfected or endothelial cells infected with T. cruzi revealed no significant differences. In membranes prepared from uninfected or infected endothelial cells, P-adrenergic receptor density was found to be 81 * 6 fmol/ mg protein with a l3-adrenergic affinity of 50 +6pM. As reported previously, 6 days after infection of mice we were able to demonstrate organisms in the endothelium of the coronary microvasculature (Morris et al.
71
ACTIVITY
1988). At this same time, hormone-sensitive myocardial adenylate cyclase activity was depressed. Accordingly, we examined whether infection of endothelial cells in vitro may result in a similar depression of adenylate cyclase activity despite the absence of a change in the density of j3-adrenergic receptors. As shown in Fig. 2, infection of cultured endothelial cells did not substantially alter adenylate cyclase activity whether measured alone or in response to isoproterenol, forskolin, Gpp(NH)p, or the combinations of isoprotereno1 plus Gpp(NH)p or forskolin plus Gpp(NH)p. Hence, under the reaction conditions employed here, infection did not substantially alter the capacity of endothe-
20 B/'Wk
IS0 FORSK Gpp
ISO FORSK GYP
FIG. 1. P-Adrenergic receptors in endothelial cells: Influence of Trypanosoma cruzi infection. Homogenates of endothelial cells were prepared as described under Materials and Methods from uninfected (solid circles) and infected (open circles) cells, and l3-adrenergic receptor density and affinity determined as described using [‘*‘I]ICYP. Individual points are the means of three separate determinations. The standard deviations are less than 10% of the means and, for purposes of clarity, are not included.
+ Gbp
FIG. 2. Endothelial cell adenylate cyclase activity: Influence of infection. Homogenates were prepared from uninfected (open bars) and infected (hatched bars) endothelial cells, and adenylate cyclase activity was determined in response to no agent (Basal), isoproterenol (1 ~JW), forskolin (1 pm, Gpp(NH)p (1 FM), or the combination of isoproterenol and Gpp(NH)p or forskolin and Gpp(NH)p in an adenylate cyclase assay for 15 min performed as described. There are no significant differences between the activities measured in uninfected or infected endothelial cells. Results shown are means and standard deviations of six separate experiments.
72
MORRIS
lial cells to generate CAMP. We therefore ascertained whether infection altered the levels of whole cell CAMP, since cellular steady-state levels of CAMP reflect not only CAMP synthesis, but also its metabolism by phosphodiesterase (PDE). Despite interexperimental variation, in a given preparation of endothelial cells whole cell basal levels of CAMP were not substantially altered by infection over a range of parasitism (252 + 32 pmol cAMP/mg total protein in uninfected endothelial cells vs 246 -+ 25 pmol cAMP/mg total protein in infected endothelial cells; results not shown). However, the addition of isobutylmethylxanthine (IBMX), an inhibitor of phosphodiesterase activity, revealed a specific infection associated change. When infected endothelial cells were incubated with IBMX (2 rniV) for 45 min, the increase in whole-cell CAMP levels (354 + 12 pmol cAMP/mg total protein) reflected a 1.4 * 0.2-fold increase in CAMP levels due to the presence of the inhibitor. In contrast, under otherwise identical reaction conditions, IBMX incubation increased cell CAMP in uninfected endothelial cells to 554 5 38 pmol cAMP/mg total protein, or an increase of 2.2 ? 0.2-fold. The difference in response to IBMX between uninfected and infected endothelial cells was significant (P < .Ol; results not shown). These results suggested that infection may have altered the metabolic disposition of CAMP by virtue of an influence on PDE activity. To explore this possibility, we determined PDE activity directly. When PDE activity was determined in whole endothelial cell homogenates, infection was observed to induce several important changes. First, as shown in Fig. 3, basal phosphodiesterase activity was markedly increased (>2.5-fold increase). Second, infection influenced PDE activity in response to prolonged hormone-dependent elevation of cell CAMP. An increase in PDE activity has been consistently observed following prolonged accumulation of cell
ET AL.
lE4 lE4 lE4 9000 z
6000
g
7000
E”
6000
2i i 6
4000
5000 3000 2000 1000 0 Unmfected
Infected
3. Phosphodiesterase activity in endothelial cells: Influence of infection. Whole-cell homogenates were prepared for uninfected and infected endothelial cells previously incubated without (open bar) or with (hatched bar) isoproterenol (10 pA4) for 20 min, and phosphodiesterase activity was determined as described under Materials and Methods. Results are expressed as CPMs adenosine per milligram total protein and represent the means and standard deviations determined from five separate experiments. *Significantly different from corresponding value in uninfected cells, P < .Ol. FIG.
CAMP levels, in response to either hormone stimulation or exposure to CAMP analogues (Duttagupta et al. 1974). When we determined PDE activity in uninfected endothelial cells previously incubated with isoproterenol for 20 min, we observed the expected marked increase in PDE activity. In marked contrast, however, there was no significant change in the PDE activity measured in infected endothelial cells previously incubated with isoproterenol. Thus, infection resulted not only in an increase in basal PDE activity but also a reduction in sensitivity to increases in cellular CAMP following accumulation of cellular CAMP. To ascertain the consequences of the infection-associated changes in phosphodiesterase activity on hormone-stimulated CAMP accumulation in intact cells, we determined the time course of CAMP accumulation in response to isoproterenol in uninfected and infected endothelial cells. In many cell culture systems known to possess a B-adrenergic adenylate cyclase complex, the addition of isoproterenol may not result in a large or significant increase in
T. cruzi: INFECTION AND CELL PHOSPHODIESTERASE ACTIVITY CAMP by virtue of an inhibitory influence of serum on hormone-sensitive adenylate cyclase activation (Morris and Makman, 1976). Consistent with this observation, (results not shown) we determined that addition of isoproterenol to either uninfected or infected endothelial cells grown in complete medium with serum did not substantially alter the levels of whole-cell CAMP. Nonetheless, as shown in Fig. 4A, despite the removal of serum from uninfected endothelial cells, a small increase in CAMP was observed after 10 min exposure to isoproterenol (from 307 + 28 pmol CAMP/ whole cell protein at the start of the incubation to 484 + 36 pmol CAMP/whole-cell protein after 10 min of incubation, P < 0.05; Fig. 4A). This increase in CAMP in response to isoproterenol returned to baseline by 20 min incubation with the agonist, suggesting a high turnover of CAMP by these cells as reported by others (Luckhoff ef al. 1990). In contrast, there was no significant change in CAMP levels in infected endothelial cells exposed to isoproterenol under otherwise identical reaction conditions (Fig. 4B). When uninfected endothelial cells are exposed to isoproterenol in the presence of IBMX (Fig. 4A), a threefold increase in cell CAMP occurs at 2 min of incubation; with longer time periods of isoproterenol exposure, whole-cell CAMP declined slightly to a level about twofold above the unstimulated CAMP levels. Infection altered several aspects of this response to isoproterenol. In the presence of IBMX, isoproterenol stimulation of infected endothelial cells resulted in a rise in CAMP that reached a maximal level similar to that observed in uninfected endothelial cells (Fig. 4A) but only after 5 min of incubation. Thereafter, in contrast to uninfected cells, whole-cell levels of CAMP in infected endothelial cells rapidly returned to basal unstimulated levels within 10 min of incubation despite the continued presence of isoproterenol and IBMX. These differences in CAMP generation in infected
1400
B
73
+
B 1200 s 1000 E ii s
A
TIME (minutes)
4. Isoproterenol-dependent stimulation of CAMP levels in endothelial cells: Influence of infection. Uninfected (A) or infected (B) endothelial cells were washed free of their growth medium and fed PBS without (open circles) or with IBMX (closed circles; 2 rnhf) and incubated for 45 min, at the end of which isoproterenol (1 pkf) was added. At the indicated times after the addition of the hormone, total cell CAMP levels were determined. In (C), the ratio of CAMP determined in the presence of isoproterenol and IBMX at time f over the level of CAMP in the presence of IBMX but in the absence of isoproterenol is plotted as a function of time of incubation with isoproterenol for uninfected (open circles) and infected (closed circles) endothelial cells. Results shown are means and standard deviations of three separate determinations. *Significantly different from corresponding values obtained in uninfected endothelial cells, P < .Ol. FIG.
and uninfected endothelial cells in response to the presence of isoproterenol are more clearly revealed when plotted as the ratio of CAMP in the presence of isoproterenol at time t over the amount of CAMP in the whole cell at time zero (Fig. 4C). Clearly, the peak in CAMP accumulation in re-
74
MORRIS
sponse to isoproterenol in infected endothelial cells is not only delayed in onset, but rapidly attenuated despite the continued presence of both isoproterenol and IBMX.
ET AL.
myocardial dysfunction (Morris et al. 1990). Infection of endothelial cells resulted in an increase in PDE activity. While the physiological role of CAMP-dependent phosphodiesterase remains to be defined, DISCUSSION the metabolism of CAMP may well be an In this report we have documented the integral part of the total CAMP accumulated consequences of T. cruzi infection on the in response to hormone-dependent stimulasynthesis and metabolism of whole cell tion of adenylate cyclase. In this regard, it CAMP. In contrast to the influence of in- has been reported that expression of human fection in the animal, direct infection of cul- recombinant CAMP-dependent PDE isotured human endothelial cells did not ap- form IV in PDE-deficient yeast reversed parently influence CAMP generation in the growth arrest phenotype in this organresponse to isoproterenol in whole homoge- ism, notably without a significant change in nates. Thus, infection altered neither p-ad- basal levels of CAMP (McHale et al. 1991). renergic receptor density and affinity nor Hence, the physiological consequences of adenylate cyclase activity per se. However, the infection-associated changes in PDE acinfection altered the metabolism of endo- tivity may impact on the modulation of thelial cell CAMP. Infection increased basal CAMP accumulation in response to physioPDE activity and abolished inducible PDE logical stimulation. Of direct relevance to activity observed in uninfected endothelial our study there has been a report that cells incubated with isoproterenol. The CAMP levels positively correlate with the consequences of these infection-assoexpression of thrombomodulin receptors in ciated changes in PDE activity may impact cultured human umbilical vein endothelial on the infection-associated change ob- cells (Ishii et al. 1990). Hence, the marked served in the accumulation of CAMP in re- attenuation in CAMP accumulation in response to isoproterenol. sponse to isoproterenol in infected endoThe observation that direct infection of thelial cells may negatively influence the endothelial cells in vitro did not alter B-ad- expression of thrombomodulin receptors renergic adenylate cyclase activity con- and, consistent with our hypothesis of a trasts with in viva studies. Six days after role of the microvasculature in the pathoinfection in the mouse, when parasites can genesis of Chagas’ cardiomyopathy (Morris be demonstrated only in the endothelial et al. 1990), foster a procoagulant state. cells of the coronary microcirculation and Infection also abolished the usual pattern levels of parasitism are extremely low, of PDE induction observed after prolonged myocardial adenylate cyclase activity in re- isoproterenol stimulation which requires elsponse to isoproterenol is decreased (Mor- evation of cellular CAMP (Duttagupta et al. ris et al. 1988). It would appear therefore 1974). However, despite the infectionthat the depression in myocardial adenylate associated increased in PDE activity, we cyclase activity may be related in part to can still demonstrate an increase in CAMP alterations in the function of infected endo- levels in infected cells exposed to isoprothelial cells with a compromise in the mi- terenol. Hence, the absence of inducible crocirculation of the myocardium. This ob- PDE activity in infected endothelial cells servation is consistent with our hypothesis reflects additional, as yet to be identified implicating an important and early role for infection-associated alterations. The physithe microvasculature, and particularly the ological role for the induction of CAMPendothelial cell, in the pathogenesis of dependent PDE activity following the in-
T. cruzi:
INFECTION
AND
CELL
PHOSPHODIESTERASE
ACTIVITY
75
crease in cytosolic CAMP remains to be es- BROWN, J. H., MISRA, R. K., AND MAKMAN, M. H. 1975. A modified method for the analysis of CAMP. tablished. Moreover, it would appear that Advances in Cyclic Nucleic Research 5, 365-372. the absence of inducible PDE activity fol- DE CASTRO, S. L., MEIRELLES, M. DE N. L., AND lowing infection renders it unlikely that OLIVEIRA, M. M. 1987. Trypanosoma cruzi: AdrenPDE activity substantially contributes to ergic modulation of cyclic AMP role in proliferation and differentiation of amastigotes in vitro. Experithe rapid decline in CAMP levels following mental Parasitology 64, 369-375. isoproterenol stimulation per se. In this reDUTTAGUPTA, C., RIFAS, L., AND MAKMAN, M. H. gard, we reported that infection greatly ac1974. Phosphodiesterase activity in cultured glial celerated the rate of P-adrenergic desensicells. In Vitro 10, 347-351. tization (loss of responsiveness to isopro- GORDON, P. V., CONN G., AND HATHCER, V. B. 1985. Glycosaminoglycan synthesis in cultures of terenol-dependent stimulation of adenylate early and late passage human endothelial cells: The cyclase activity) in L,E, myoblasts (Morris influence of an anionic endothelial cell growth factor et al. 1984). Hence, infection-associated acand the extracellular matrix. Journal of Cellular celeration of p-adrenergic desensitization Physiology 133, 671. may be an additional contribution to the ISHII, H., KIZAKI, K., UCHIYAMA, H., HORIE, S., AND KAZAMA, M. 1990. Cyclic AMP increases precipitous decline in whole-cell CAMP levthrombomodulin expression on membrane surface els following prolonged incubation with isoof cultured human umbilical vein endothelial cells. proterenol (Morris et al. 1984). Thrombosis Research 59, 841-850. We have not determined as yet the spec- JAFFE, E. 1980. Culture of Human endothelial cells. ificity of the infection associated changes in Transplant. Proceedings 12, 49-53. PDE activity, i.e., whether such changes LLOYD, C. J., CARY, D. A., AND MENDELSOHN, F. A. 0. 1987. Angiotensin converting enzyme incan be demonstrated in other tissues. Deduction by cyclic AMP and analogues in cultured spite the fact that during acute infection, a endothelial cells. Molecular and Cellular Endocrinumber of diverse tissues are involved, it nology 52, 219-225. may turn out that the specificity of the con- LUCKHOFF, A., MULSCH, A., AND BUSSE, R. 1990. sequences of infection with regard to PDE CAMP attenuates autocoid release from endothelial cells: relation to internal calcium. American Journal activity relates more to the unique imporof Physiology 258, H960-H966. tance of this particular pathway in endothelial cell physiology. Nonetheless, the MCHALE, M. M., CIESLINSKI, L. B., AND ENG, W. K. 1991. Expression of human recombinant changes in endothelial cell PDE activity, CAMP phosphodiesterase isozyme IV reverses and the possible consequences of these growth arrest phenotypes in phosphodiesterasedeficient yeast. Molecular Pharmacology 39, 109changes on endothelial cell participation in 113. microvascular perfusion noted previously, are entirely consistent with our longstand- MORRIS, S. A., BILEZIKIAN, J. P., HATCHER, V., WEISS, L. M., TANOWITZ, H. B., AND WITTNER, ing hypothesis regarding the critical role of M. 1989. Trypanosoma cruzi: Infection of cultured the microvasculature in the pathogenesis of human endothelial cells alters inositol phosphate Chagas’ cardiomyopathy. synthesis. Experimental Parasitology 69, 330-339. ACKNOWLEDGMENTS Grants
HL20859,
HL28958,
AI 12770, HL37025,
DK39880, and HL35882 were from the National Institutes of Health. REFERENCES
BILEZIKIAN, J. P., DORNFELD,A. M., AND GAMMON, D. E. 1978. Structure activity binding analysis of beta adrenergic amines. Biochemistry and Pharmacology 21, 1445-1454.
MORRIS, S. A., BILEZIKIAN, J. P., TANOWITZ, H., AND WITTNER, M. 1987. Infection of L6E9 Myoblasts with Trypanosoma cruzi alters adenylate cyclase activity and guanine nucleotide binding proteins. Journal of Cellular Physiology 133, W71. MORRIS,S. A., HATCHER,V., TANOWITZ, H. B., AND WITTNER, M. 1988. Alterations in intracellular calcium following infection of human endothelial cells with Trypanosoma cruzi. Molecular Biochemistry and Parasitology
29, 213-221.
MORRIS,S. A., AND MARMAN, M. H. 1976.Adenylate cyclase activity in cultured rat astrocytoma cells. Molecular
Pharmacology
12, 362-372.
MORRIS ET AL.
76
MORRIS, S. A., TANOWITZ, H. B., FACTOR, S. A., BILEZIKIAN, J. P., AND WITTNER, M. 1988. Myocardial beta adrenergic adenylate cyclase activity in acute murine Chagas’ disease. Circulation Research 62, 800-810. MORRIS, S. A., TANOWITZ, H. B., ROWIN, K. S., WITTNER, M., AND BILEZIKIAN, J. P. 1984. Alterations in the pattern of B-adrenergic desensitization in cultured L6E9 muscle cells infected with Trypanosoma cruzi. Molecular tology 13, 227-234.
Biochemistry
and Parasi-
MORRIS, S. A., TANOWITZ, H. B., WITTNER, M., AND BILEZIKIAN, J. P. 1990. Pathophysiological insights into the cardiomyopathy of Chagas’ disease. Circulation 82, 1900-1909. MORRIS, S. A., WEISS, L. M., FACTOR, S., BILEZIKIAN, J. P., TANOWITZ, H. B., AND WITTNER, M. 1989. Verapamil ameliorates clinical, pathological and biochemical manifestations of experimental chagasic cardiomyopathy in mice. Journal of the American College of Cardiology 14, 782789. MORRIS,S. A., WITTNER, M., WEISS, L., HATCHER, V. B., TANOWITZ, H. B., BILEZIKIAN, J. P. B., AND GORDON,P. B. 1990. Extracellular Matrix derived from Trypanosoma cruzi infected Endothelial cells directs phenotypic expression. Journal of Cellular Physiology 145, 340-346. MUNSON, P. J., AND RODBARD,D. 1980. LIGAND A versatile computerized approach for characterization of ligand binding systems. Analytical Biochemistry 107, 220-239. RITTENHOUSE~IMMONS, S. 1979. Production of diglyceride from phosphatidylinositol in activated human platelets. Journal of Clinical Investigation 63, 580-587.
ROSSI, M. A., GONCALVES, S., AND RIBEIRO-DOS SANTOS,R. 1984. Experimental Trypanosoma cruzi cardiomyopathy in BALBlc mice. American Journal of Pathology 114, 209-216. ROWIN, K. S., TANOWITZ, H. B., WITTNER, M., NYGUEN, H. T., AND NADAL-GINARD, B. 1983. Inhibition of muscle differentiation by Trypanosoma cruzi. Proceedings of the National ences USA 80, 63906394.
Academy of Sci-
SALOMON, Y. C., LONDOS, C., AND RODBELL, M. 1974. A highly sensitive adenylate cyclase assay. Analytical Biochemistry 58, 541-548. SCHAFER, A. I., GIMBRONE, M. A., AND HANDIN, R. I. 1984.Regulation of endothelial cell function by cyclic nucleotides. In “Biology of Endothelial Cells” (E. A. Jaffe, Ed.), pp. 248-158. Nijhoff, Boston. STELZNER,T. J., WEIL, J. V., AND O’BRIEN, R. F. 1989. Role of cyclic adenosine monophosphate in the induction of endothelial barrier properties. Journal of Cellular Physiology 139, 26-35. TANOWITZ, H. B., BURNS,E., SINHA, A., KAHN, N., MORRIS, S. A., FACTOR, S., HATCHER, V. B., BILEZIKIAN, J. P., AND WITTNER, M. 1990. Pathogenesis of cardiomyopathy: Microvascular alterations and platelet reactivity in Chagas’ disease. American Journal of Tropical Diseases 43, 274-281.
TRISCHMANN,T., TANOWITZ, H. B., WITTNER, M., AND BLOOM, B. 1978. Trypanosoma cruzi: The role of the immune response in the natural resistance of inbred strains of mice. Experimental Parasitology 45. 160-166. Received 26 April 1991;accepted with revision 12 September 1991
PARASITOLOGY
T~panosoma
74,69-76 (1992)
cruzi: Alteration of CAMP Metabolism Infection of Human Endothelial Cells
STEPHEN A. MORRIS,*Y'HERBERTTANOWITZ,* VICTOR B. HATCHER,* JOHN P. BILEZIKIAN,~AND
following
MAYNARDMAKMAN, MURRAY WITTNERt
*Departments of Pathology and Medicine and Biochemistry, Albert Einstein College of Medicine, Bronx, New York, New York 10461 USA; and tDepartments of Medicine and Pharmacology, College of Physicians and Surgeons, Columbia University, New York, New York 10032 USA
S. A., TANOWITZ, H., MAKMAN, M., HATCHER, V. B., BILEZIKIAN, J. P., AND (1992). Trypanosoma cruzi: Alteration of CAMP metabolism following infection of human endothelial cells. Experimental Parasitology 74, 69-76. We have previously reported that Trypanosoma cruzi infection of endothelial cells results in alterations in the metabolism of Cazf , inositol triphosphate (IP,), and prostacycline (PC&). In this report, we demonstrate that infection also alters the metabolism of CAMP. Infection of endothelial cells does not significantly alter P-adrenergic receptor density or affinity, adenylate cyclase activity, and whole-cell CAMP levels. However, incubation of infected endothelial cells with the phosphodiesterase inhibitor isobutylmethylxanthine (IBMX) resulted in less than a 60% increase in cell CAMP in contrast to the greater than a 100% increase observed in uninfected endothelial cells under otherwise identical reaction conditions. Infected endothelial cells demonstrated a twofold increase in phosphodiesterase activity when measured directly. Moreover, homogenates prepared from infected endothelial cells previously incubated with isoproterenol for 20 min showed little or no change in PDE activity. In contrast, homogenates prepared from uninfected endothelial cells treated under otherwise identical reaction conditions showed a 5.7-fold increase in PDE activity. In the presence of IBMX, isoproterenol-dependent stimulation of CAMP levels in infected endothelial cells reached a maximum level at 5 min of incubation, and thereafter rapidly declined. In contrast, CAMP levels in uninfected endothelial cells reached a maximum at 2 min of incubation, and thereafter remained elevated throughout the duration of the incubation. Infection-associated changes in isoproterenol dependent stimulation of CAMP accumulation appear to relate, in part, to changes in PDE activity. 8 19% Academic press. IK. Hemoflagellates; Trypanosoma cruzi; EndoINDEX DESCRIPTORS AND ABBREVIATIONS: thelial cells: Cvclic AMP (CAMP); Isobutylmethylxanthine (IBMX); Phosphodiesterase activity (PDE). MORRIS,
WITTNER,
M.
and in vitro are consistent with abnormalities in the microvasculature, resulting in foThe pathophysiological basis for cardiac cal pathology. In support of this hypothesis involvement in Chagas’ disease remains un- we have demonstrated that direct infection clear. We (Morris et al. 1989,199O)and oth- of endothelial cells in vitro results in the ers (Rossi et al. 1984) have hypothesized a perturbation of a number of biochemical central role for the microvasculature in processes thought to be critical to the maincontributing to the cardiac pathology ob- tenance of microvasculature perfusion served in both the chronic and acute stages (Morris et al. 1988). For example, infection of the disease. Pathological studies in vivo of endothelial cells has been shown to interfere with Ca2’ homeostasis (Morris et ’ To whom correspondence and reprint requests al. 1988), inositol triphosphate (IP3) genershould be addressed at Department of Medicine, College of Physicians and Surgeons, 630 West 168th ation (Morris et al. 1989) and the synthesis of prostaglandin I, (Tanowitz et al. 1990). Street, New York, NY 10032. 69 0014-4894/92$3.00 Copyright Q 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
70
MORRIS ET AL.
More recent investigations by others (Luckhoff et al. 1990; Schafer et al. 1984; Stelzner et al. 1989) into the biochemical basis by which endothelial cells participate in the regulation of microvascular perfusion have focused on the role of CAMP metabolism. In this regard, a number of observations have demonstrated areas in which CAMP may play a critical role. Cyclic AMP was shown to influence the concentration of intracellular Ca2+ in response to administration of ATP (Luckhoff et al. 1990) and play a protective role in defending endothelial cells from direct or indirect damage caused by adhesion and aggregation of circulating platelets. In addition, CAMP has been assigned an important role in the regulation of angiotensin converting enzyme synthesis, an enzyme involved in the formation of the vasoconstrictor angiotensin II (Lloyd et al. 1987). In this report, we show that infection of endothelial cells with Trypanosoma cruzi results in an alteration of CAMP metabolism. MATERIALS
AND METHODS
"'1 , [w~~P]ATP, and [3H]cAMP were Materials. obtained from New England Nuclear-duPont (Boston, MA). S’Guanylylimidodiphosphate (Gpp(NH)p) was obtained from Sigma Corporation (St. Louis, MO). All other reagents were of the highest purity commercially available. Culture and maintenance of endothelial cells. Human umbilical vein endothelial cells were isolated according to the method of Jaffe (Jaffe 1980)and grown as described by Gordon et al. (1985). Parasites. The Tulahuen strain of T. cruzi was passaged by syringe in A/J mice (Jackson Laboratories, Bar Harbor, ME; Trischmann et al. 1978;Morris et al. 1988). Trypomastigotes used for infection of endothelial cells were maintained in L,E, myoblasts as previously described (Rowin et al. 1983;Morris et al. 1984). Infection of endothelial cells was performed as described (Morris et al. 1984). Cells were harvested after 2 to 3 days of infection when the percentage of parasitism ranged from 5 to 50% as determined by May Grunwald-Giemsa staining (Tanowitz et al. 1990). Determination of P-receptor sites. The binding assay used [‘*51]iodocyanopindolol (ICYP). Cyanopindolol was iodinated and purified to a specific activity of 2200 Ci/mmol according to methods previously pub-
lished (Morris et al. 1988). Aliquots of endothelial cell membranes (100-300 up) were distributed to tubes containing [“‘IIICYP in a final volume of 1 ml whole cell buffer (0.15 M NaCl, 0.01 M Tris at pH 7.5), 0.01 M KCl, and 1 mg/ml bovine serum albumin with 2 mgiml dextrose. The binding assay was carried out in a shaking water bath for 20 min at 37°C. By 20 min of incubation under these conditions, binding of [“‘I]ICYP had reached equilibrium (results not shown) and values obtained from longer incubation periods (30-60 min) were identical to those reported here. The reaction was terminated by filtering the suspension over Gelman A/E glass fiber filters. The filters were washed with 10 ml 0.01 M Tris at room temperature, and radioactivity was determined in a Packard Auto-Gamma scintillation counter. Specific binding was defined as the difference between total binding and binding inhibited by (-)propranolol (0.1 PM). Maximal binding capacity was determined by Scatchard analysis (Morris et al. 1988) using the LIGAND binding analysis program (Munson and Rodbard 1980). Adenylate cyclase assay. The reaction mixture contained 50 mM Tris-HCI (pH 7.4), 2.5 mM magnesium unless otherwise noted, 0.143 mM ATP, an ATP regenerating system (creatine phosphate/creatine phosphokinase), ATP ([w~*P]ATP, 1-2 X lo6 cpm/assay tube) and 6 mM theophylhne in a final volume of 75 ~1 (Bilezikian et al. 1978). Agents dissolved in 50 mM Tris-HCl (pH 7.5) were added in concentrations noted in individual experiments. The adenylate cyclase reaction was terminated by the addition of an ATP-cyclic AMP stopping solution. Isolation of [32P]cyclic AMP was accomplished by sequential Dowex and alumina chromatography using [3H]cyclic AMP as a recovery marker (Salomon et al. 1974). Measurement of CAMP levels. For the measurement of CAMP levels in intact cells, the medium was removed rapidly and an aqueous acetic acid solution (1: I v/v) was added. The resulting suspensions were heated for 3 min at 90°C and centrifuged to remove cellular debris. Aliquots of the supernatant fluid were dried at 80°C in small glass tubes and assayed for CAMP by the modified binding assay of Gilman (Brown et a[. 1975). Determination of protein levels. Protein concentration was determined using the Bio-Rad method (BioRad Laboratories; Richmond, CA). Phosphodiesterase activity. Cyclic nucleotide phosphodiesterase activities were determined in a two-step incubation method previously reported (Duttagupta et al. 1974). Plates of endothelial cells prepared as described were washed two times in ice-cold 0.9% saline and scraped into a buffer containing 0.25 M sucrose and 50 mM Tris. The suspension was homogenized using a Polytron homogenizer. For the first incubation, aliquots (100-300 wg protein) of this homogenate were distributed to tubes containing [‘HIcAMP (25 Ci/ mmole, approximately 25,000 CPM@, MgCl, [5 mM],
T. cruzi: INFECTION
AND CELL PHOSPHODIESTERASE
2 mercaptoethanol [3.75 mM], and Tris buffer [40 mM], pH 8; incubated at 30°C for 10 min in a shaking water bath; and terminated by boiling for 3 min. In the second incubation, snake venom nucleotidase was added to the suspensions, which were then incubated for 10 min at 30°C. The second reaction was stopped by the addition of a 1:3 slurry of Bio-Rad resin AGlX2, followed by centrifugation and counting of [3H]adenosine in the supematant in a Packard liquid scintillation counter. RESULTS
The possibility that infection of endothelial cells may alter the density or affinity of the p-adrenergic receptors responsible for the activation of adenylate cyclase was first examined. As shown in Fig. 1, analysis of endothelial cell membranes prepared from uninfected or endothelial cells infected with T. cruzi revealed no significant differences. In membranes prepared from uninfected or infected endothelial cells, P-adrenergic receptor density was found to be 81 * 6 fmol/ mg protein with a l3-adrenergic affinity of 50 +6pM. As reported previously, 6 days after infection of mice we were able to demonstrate organisms in the endothelium of the coronary microvasculature (Morris et al.
71
ACTIVITY
1988). At this same time, hormone-sensitive myocardial adenylate cyclase activity was depressed. Accordingly, we examined whether infection of endothelial cells in vitro may result in a similar depression of adenylate cyclase activity despite the absence of a change in the density of j3-adrenergic receptors. As shown in Fig. 2, infection of cultured endothelial cells did not substantially alter adenylate cyclase activity whether measured alone or in response to isoproterenol, forskolin, Gpp(NH)p, or the combinations of isoprotereno1 plus Gpp(NH)p or forskolin plus Gpp(NH)p. Hence, under the reaction conditions employed here, infection did not substantially alter the capacity of endothe-
20 B/'Wk
IS0 FORSK Gpp
ISO FORSK GYP
FIG. 1. P-Adrenergic receptors in endothelial cells: Influence of Trypanosoma cruzi infection. Homogenates of endothelial cells were prepared as described under Materials and Methods from uninfected (solid circles) and infected (open circles) cells, and l3-adrenergic receptor density and affinity determined as described using [‘*‘I]ICYP. Individual points are the means of three separate determinations. The standard deviations are less than 10% of the means and, for purposes of clarity, are not included.
+ Gbp
FIG. 2. Endothelial cell adenylate cyclase activity: Influence of infection. Homogenates were prepared from uninfected (open bars) and infected (hatched bars) endothelial cells, and adenylate cyclase activity was determined in response to no agent (Basal), isoproterenol (1 ~JW), forskolin (1 pm, Gpp(NH)p (1 FM), or the combination of isoproterenol and Gpp(NH)p or forskolin and Gpp(NH)p in an adenylate cyclase assay for 15 min performed as described. There are no significant differences between the activities measured in uninfected or infected endothelial cells. Results shown are means and standard deviations of six separate experiments.
72
MORRIS
lial cells to generate CAMP. We therefore ascertained whether infection altered the levels of whole cell CAMP, since cellular steady-state levels of CAMP reflect not only CAMP synthesis, but also its metabolism by phosphodiesterase (PDE). Despite interexperimental variation, in a given preparation of endothelial cells whole cell basal levels of CAMP were not substantially altered by infection over a range of parasitism (252 + 32 pmol cAMP/mg total protein in uninfected endothelial cells vs 246 -+ 25 pmol cAMP/mg total protein in infected endothelial cells; results not shown). However, the addition of isobutylmethylxanthine (IBMX), an inhibitor of phosphodiesterase activity, revealed a specific infection associated change. When infected endothelial cells were incubated with IBMX (2 rniV) for 45 min, the increase in whole-cell CAMP levels (354 + 12 pmol cAMP/mg total protein) reflected a 1.4 * 0.2-fold increase in CAMP levels due to the presence of the inhibitor. In contrast, under otherwise identical reaction conditions, IBMX incubation increased cell CAMP in uninfected endothelial cells to 554 5 38 pmol cAMP/mg total protein, or an increase of 2.2 ? 0.2-fold. The difference in response to IBMX between uninfected and infected endothelial cells was significant (P < .Ol; results not shown). These results suggested that infection may have altered the metabolic disposition of CAMP by virtue of an influence on PDE activity. To explore this possibility, we determined PDE activity directly. When PDE activity was determined in whole endothelial cell homogenates, infection was observed to induce several important changes. First, as shown in Fig. 3, basal phosphodiesterase activity was markedly increased (>2.5-fold increase). Second, infection influenced PDE activity in response to prolonged hormone-dependent elevation of cell CAMP. An increase in PDE activity has been consistently observed following prolonged accumulation of cell
ET AL.
lE4 lE4 lE4 9000 z
6000
g
7000
E”
6000
2i i 6
4000
5000 3000 2000 1000 0 Unmfected
Infected
3. Phosphodiesterase activity in endothelial cells: Influence of infection. Whole-cell homogenates were prepared for uninfected and infected endothelial cells previously incubated without (open bar) or with (hatched bar) isoproterenol (10 pA4) for 20 min, and phosphodiesterase activity was determined as described under Materials and Methods. Results are expressed as CPMs adenosine per milligram total protein and represent the means and standard deviations determined from five separate experiments. *Significantly different from corresponding value in uninfected cells, P < .Ol. FIG.
CAMP levels, in response to either hormone stimulation or exposure to CAMP analogues (Duttagupta et al. 1974). When we determined PDE activity in uninfected endothelial cells previously incubated with isoproterenol for 20 min, we observed the expected marked increase in PDE activity. In marked contrast, however, there was no significant change in the PDE activity measured in infected endothelial cells previously incubated with isoproterenol. Thus, infection resulted not only in an increase in basal PDE activity but also a reduction in sensitivity to increases in cellular CAMP following accumulation of cellular CAMP. To ascertain the consequences of the infection-associated changes in phosphodiesterase activity on hormone-stimulated CAMP accumulation in intact cells, we determined the time course of CAMP accumulation in response to isoproterenol in uninfected and infected endothelial cells. In many cell culture systems known to possess a B-adrenergic adenylate cyclase complex, the addition of isoproterenol may not result in a large or significant increase in
T. cruzi: INFECTION AND CELL PHOSPHODIESTERASE ACTIVITY CAMP by virtue of an inhibitory influence of serum on hormone-sensitive adenylate cyclase activation (Morris and Makman, 1976). Consistent with this observation, (results not shown) we determined that addition of isoproterenol to either uninfected or infected endothelial cells grown in complete medium with serum did not substantially alter the levels of whole-cell CAMP. Nonetheless, as shown in Fig. 4A, despite the removal of serum from uninfected endothelial cells, a small increase in CAMP was observed after 10 min exposure to isoproterenol (from 307 + 28 pmol CAMP/ whole cell protein at the start of the incubation to 484 + 36 pmol CAMP/whole-cell protein after 10 min of incubation, P < 0.05; Fig. 4A). This increase in CAMP in response to isoproterenol returned to baseline by 20 min incubation with the agonist, suggesting a high turnover of CAMP by these cells as reported by others (Luckhoff ef al. 1990). In contrast, there was no significant change in CAMP levels in infected endothelial cells exposed to isoproterenol under otherwise identical reaction conditions (Fig. 4B). When uninfected endothelial cells are exposed to isoproterenol in the presence of IBMX (Fig. 4A), a threefold increase in cell CAMP occurs at 2 min of incubation; with longer time periods of isoproterenol exposure, whole-cell CAMP declined slightly to a level about twofold above the unstimulated CAMP levels. Infection altered several aspects of this response to isoproterenol. In the presence of IBMX, isoproterenol stimulation of infected endothelial cells resulted in a rise in CAMP that reached a maximal level similar to that observed in uninfected endothelial cells (Fig. 4A) but only after 5 min of incubation. Thereafter, in contrast to uninfected cells, whole-cell levels of CAMP in infected endothelial cells rapidly returned to basal unstimulated levels within 10 min of incubation despite the continued presence of isoproterenol and IBMX. These differences in CAMP generation in infected
1400
B
73
+
B 1200 s 1000 E ii s
A
TIME (minutes)
4. Isoproterenol-dependent stimulation of CAMP levels in endothelial cells: Influence of infection. Uninfected (A) or infected (B) endothelial cells were washed free of their growth medium and fed PBS without (open circles) or with IBMX (closed circles; 2 rnhf) and incubated for 45 min, at the end of which isoproterenol (1 pkf) was added. At the indicated times after the addition of the hormone, total cell CAMP levels were determined. In (C), the ratio of CAMP determined in the presence of isoproterenol and IBMX at time f over the level of CAMP in the presence of IBMX but in the absence of isoproterenol is plotted as a function of time of incubation with isoproterenol for uninfected (open circles) and infected (closed circles) endothelial cells. Results shown are means and standard deviations of three separate determinations. *Significantly different from corresponding values obtained in uninfected endothelial cells, P < .Ol. FIG.
and uninfected endothelial cells in response to the presence of isoproterenol are more clearly revealed when plotted as the ratio of CAMP in the presence of isoproterenol at time t over the amount of CAMP in the whole cell at time zero (Fig. 4C). Clearly, the peak in CAMP accumulation in re-
74
MORRIS
sponse to isoproterenol in infected endothelial cells is not only delayed in onset, but rapidly attenuated despite the continued presence of both isoproterenol and IBMX.
ET AL.
myocardial dysfunction (Morris et al. 1990). Infection of endothelial cells resulted in an increase in PDE activity. While the physiological role of CAMP-dependent phosphodiesterase remains to be defined, DISCUSSION the metabolism of CAMP may well be an In this report we have documented the integral part of the total CAMP accumulated consequences of T. cruzi infection on the in response to hormone-dependent stimulasynthesis and metabolism of whole cell tion of adenylate cyclase. In this regard, it CAMP. In contrast to the influence of in- has been reported that expression of human fection in the animal, direct infection of cul- recombinant CAMP-dependent PDE isotured human endothelial cells did not ap- form IV in PDE-deficient yeast reversed parently influence CAMP generation in the growth arrest phenotype in this organresponse to isoproterenol in whole homoge- ism, notably without a significant change in nates. Thus, infection altered neither p-ad- basal levels of CAMP (McHale et al. 1991). renergic receptor density and affinity nor Hence, the physiological consequences of adenylate cyclase activity per se. However, the infection-associated changes in PDE acinfection altered the metabolism of endo- tivity may impact on the modulation of thelial cell CAMP. Infection increased basal CAMP accumulation in response to physioPDE activity and abolished inducible PDE logical stimulation. Of direct relevance to activity observed in uninfected endothelial our study there has been a report that cells incubated with isoproterenol. The CAMP levels positively correlate with the consequences of these infection-assoexpression of thrombomodulin receptors in ciated changes in PDE activity may impact cultured human umbilical vein endothelial on the infection-associated change ob- cells (Ishii et al. 1990). Hence, the marked served in the accumulation of CAMP in re- attenuation in CAMP accumulation in response to isoproterenol. sponse to isoproterenol in infected endoThe observation that direct infection of thelial cells may negatively influence the endothelial cells in vitro did not alter B-ad- expression of thrombomodulin receptors renergic adenylate cyclase activity con- and, consistent with our hypothesis of a trasts with in viva studies. Six days after role of the microvasculature in the pathoinfection in the mouse, when parasites can genesis of Chagas’ cardiomyopathy (Morris be demonstrated only in the endothelial et al. 1990), foster a procoagulant state. cells of the coronary microcirculation and Infection also abolished the usual pattern levels of parasitism are extremely low, of PDE induction observed after prolonged myocardial adenylate cyclase activity in re- isoproterenol stimulation which requires elsponse to isoproterenol is decreased (Mor- evation of cellular CAMP (Duttagupta et al. ris et al. 1988). It would appear therefore 1974). However, despite the infectionthat the depression in myocardial adenylate associated increased in PDE activity, we cyclase activity may be related in part to can still demonstrate an increase in CAMP alterations in the function of infected endo- levels in infected cells exposed to isoprothelial cells with a compromise in the mi- terenol. Hence, the absence of inducible crocirculation of the myocardium. This ob- PDE activity in infected endothelial cells servation is consistent with our hypothesis reflects additional, as yet to be identified implicating an important and early role for infection-associated alterations. The physithe microvasculature, and particularly the ological role for the induction of CAMPendothelial cell, in the pathogenesis of dependent PDE activity following the in-
T. cruzi:
INFECTION
AND
CELL
PHOSPHODIESTERASE
ACTIVITY
75
crease in cytosolic CAMP remains to be es- BROWN, J. H., MISRA, R. K., AND MAKMAN, M. H. 1975. A modified method for the analysis of CAMP. tablished. Moreover, it would appear that Advances in Cyclic Nucleic Research 5, 365-372. the absence of inducible PDE activity fol- DE CASTRO, S. L., MEIRELLES, M. DE N. L., AND lowing infection renders it unlikely that OLIVEIRA, M. M. 1987. Trypanosoma cruzi: AdrenPDE activity substantially contributes to ergic modulation of cyclic AMP role in proliferation and differentiation of amastigotes in vitro. Experithe rapid decline in CAMP levels following mental Parasitology 64, 369-375. isoproterenol stimulation per se. In this reDUTTAGUPTA, C., RIFAS, L., AND MAKMAN, M. H. gard, we reported that infection greatly ac1974. Phosphodiesterase activity in cultured glial celerated the rate of P-adrenergic desensicells. In Vitro 10, 347-351. tization (loss of responsiveness to isopro- GORDON, P. V., CONN G., AND HATHCER, V. B. 1985. Glycosaminoglycan synthesis in cultures of terenol-dependent stimulation of adenylate early and late passage human endothelial cells: The cyclase activity) in L,E, myoblasts (Morris influence of an anionic endothelial cell growth factor et al. 1984). Hence, infection-associated acand the extracellular matrix. Journal of Cellular celeration of p-adrenergic desensitization Physiology 133, 671. may be an additional contribution to the ISHII, H., KIZAKI, K., UCHIYAMA, H., HORIE, S., AND KAZAMA, M. 1990. Cyclic AMP increases precipitous decline in whole-cell CAMP levthrombomodulin expression on membrane surface els following prolonged incubation with isoof cultured human umbilical vein endothelial cells. proterenol (Morris et al. 1984). Thrombosis Research 59, 841-850. We have not determined as yet the spec- JAFFE, E. 1980. Culture of Human endothelial cells. ificity of the infection associated changes in Transplant. Proceedings 12, 49-53. PDE activity, i.e., whether such changes LLOYD, C. J., CARY, D. A., AND MENDELSOHN, F. A. 0. 1987. Angiotensin converting enzyme incan be demonstrated in other tissues. Deduction by cyclic AMP and analogues in cultured spite the fact that during acute infection, a endothelial cells. Molecular and Cellular Endocrinumber of diverse tissues are involved, it nology 52, 219-225. may turn out that the specificity of the con- LUCKHOFF, A., MULSCH, A., AND BUSSE, R. 1990. sequences of infection with regard to PDE CAMP attenuates autocoid release from endothelial cells: relation to internal calcium. American Journal activity relates more to the unique imporof Physiology 258, H960-H966. tance of this particular pathway in endothelial cell physiology. Nonetheless, the MCHALE, M. M., CIESLINSKI, L. B., AND ENG, W. K. 1991. Expression of human recombinant changes in endothelial cell PDE activity, CAMP phosphodiesterase isozyme IV reverses and the possible consequences of these growth arrest phenotypes in phosphodiesterasedeficient yeast. Molecular Pharmacology 39, 109changes on endothelial cell participation in 113. microvascular perfusion noted previously, are entirely consistent with our longstand- MORRIS, S. A., BILEZIKIAN, J. P., HATCHER, V., WEISS, L. M., TANOWITZ, H. B., AND WITTNER, ing hypothesis regarding the critical role of M. 1989. Trypanosoma cruzi: Infection of cultured the microvasculature in the pathogenesis of human endothelial cells alters inositol phosphate Chagas’ cardiomyopathy. synthesis. Experimental Parasitology 69, 330-339. ACKNOWLEDGMENTS Grants
HL20859,
HL28958,
AI 12770, HL37025,
DK39880, and HL35882 were from the National Institutes of Health. REFERENCES
BILEZIKIAN, J. P., DORNFELD,A. M., AND GAMMON, D. E. 1978. Structure activity binding analysis of beta adrenergic amines. Biochemistry and Pharmacology 21, 1445-1454.
MORRIS, S. A., BILEZIKIAN, J. P., TANOWITZ, H., AND WITTNER, M. 1987. Infection of L6E9 Myoblasts with Trypanosoma cruzi alters adenylate cyclase activity and guanine nucleotide binding proteins. Journal of Cellular Physiology 133, W71. MORRIS,S. A., HATCHER,V., TANOWITZ, H. B., AND WITTNER, M. 1988. Alterations in intracellular calcium following infection of human endothelial cells with Trypanosoma cruzi. Molecular Biochemistry and Parasitology
29, 213-221.
MORRIS,S. A., AND MARMAN, M. H. 1976.Adenylate cyclase activity in cultured rat astrocytoma cells. Molecular
Pharmacology
12, 362-372.
MORRIS ET AL.
76
MORRIS, S. A., TANOWITZ, H. B., FACTOR, S. A., BILEZIKIAN, J. P., AND WITTNER, M. 1988. Myocardial beta adrenergic adenylate cyclase activity in acute murine Chagas’ disease. Circulation Research 62, 800-810. MORRIS, S. A., TANOWITZ, H. B., ROWIN, K. S., WITTNER, M., AND BILEZIKIAN, J. P. 1984. Alterations in the pattern of B-adrenergic desensitization in cultured L6E9 muscle cells infected with Trypanosoma cruzi. Molecular tology 13, 227-234.
Biochemistry
and Parasi-
MORRIS, S. A., TANOWITZ, H. B., WITTNER, M., AND BILEZIKIAN, J. P. 1990. Pathophysiological insights into the cardiomyopathy of Chagas’ disease. Circulation 82, 1900-1909. MORRIS, S. A., WEISS, L. M., FACTOR, S., BILEZIKIAN, J. P., TANOWITZ, H. B., AND WITTNER, M. 1989. Verapamil ameliorates clinical, pathological and biochemical manifestations of experimental chagasic cardiomyopathy in mice. Journal of the American College of Cardiology 14, 782789. MORRIS,S. A., WITTNER, M., WEISS, L., HATCHER, V. B., TANOWITZ, H. B., BILEZIKIAN, J. P. B., AND GORDON,P. B. 1990. Extracellular Matrix derived from Trypanosoma cruzi infected Endothelial cells directs phenotypic expression. Journal of Cellular Physiology 145, 340-346. MUNSON, P. J., AND RODBARD,D. 1980. LIGAND A versatile computerized approach for characterization of ligand binding systems. Analytical Biochemistry 107, 220-239. RITTENHOUSE~IMMONS, S. 1979. Production of diglyceride from phosphatidylinositol in activated human platelets. Journal of Clinical Investigation 63, 580-587.
ROSSI, M. A., GONCALVES, S., AND RIBEIRO-DOS SANTOS,R. 1984. Experimental Trypanosoma cruzi cardiomyopathy in BALBlc mice. American Journal of Pathology 114, 209-216. ROWIN, K. S., TANOWITZ, H. B., WITTNER, M., NYGUEN, H. T., AND NADAL-GINARD, B. 1983. Inhibition of muscle differentiation by Trypanosoma cruzi. Proceedings of the National ences USA 80, 63906394.
Academy of Sci-
SALOMON, Y. C., LONDOS, C., AND RODBELL, M. 1974. A highly sensitive adenylate cyclase assay. Analytical Biochemistry 58, 541-548. SCHAFER, A. I., GIMBRONE, M. A., AND HANDIN, R. I. 1984.Regulation of endothelial cell function by cyclic nucleotides. In “Biology of Endothelial Cells” (E. A. Jaffe, Ed.), pp. 248-158. Nijhoff, Boston. STELZNER,T. J., WEIL, J. V., AND O’BRIEN, R. F. 1989. Role of cyclic adenosine monophosphate in the induction of endothelial barrier properties. Journal of Cellular Physiology 139, 26-35. TANOWITZ, H. B., BURNS,E., SINHA, A., KAHN, N., MORRIS, S. A., FACTOR, S., HATCHER, V. B., BILEZIKIAN, J. P., AND WITTNER, M. 1990. Pathogenesis of cardiomyopathy: Microvascular alterations and platelet reactivity in Chagas’ disease. American Journal of Tropical Diseases 43, 274-281.
TRISCHMANN,T., TANOWITZ, H. B., WITTNER, M., AND BLOOM, B. 1978. Trypanosoma cruzi: The role of the immune response in the natural resistance of inbred strains of mice. Experimental Parasitology 45. 160-166. Received 26 April 1991;accepted with revision 12 September 1991
