Influence of Streptozotocin-induced Diabetes on Adenylate Cyclase Activity in Cultured Type II Pneumocytes Lou Ann S. Brown Department of Pediatrics, Emory University, Atlanta, Georgia

Previous studies with cultured type II pneumocytes from streptozotocin-induced diabetic rats demonstrated altered surfactant synthesis and secretion. The effects of the diabetic state were reversed by in vivo but not in vitro insulin treatment. In the current study, cultured type II pneumocytes from control and streptozotocin-induced diabetic rats were demonstrated to possess approximately 17,500 and 8,500 receptors per cell, respectively. High-affinity binding sites were determined to have a dissociation constant of 0.429 nM and 0.203 nM for control and diabetic cells, respectively. Functional capacity of the insulin receptors was determined by the initial rates of z-deoxy-o-glucose uptake. Uptake was stimulated by insulin in a dose-dependent manner and was not significantly altered by the diabetic state. This would suggest that the insulin receptor was present and functioning in cells isolated from diabetic rats. Basal adenylate cyclase activity of type II cell homogenates from diabetic rats was shown to be 16% of that for controls. In addition, isoproterenol, guanosine 5'-triphosphate (GTP), and NaF were unable to stimulate adenylate cyclase activity. However, forskolin, which directly activates the catalytic subunit of adenylate cyclase, was able to increase the cellular content of cyclic adenosine monophosphate (cAMP) in this model. This would suggest that some step prior to adenylate cyclase but not the catalytic subunit was altered by the diabetic state. But forskolin was unable to restore surfactant secretion, suggesting that in addition to adenylate cyclase, other processes are affected by the diabetic state. The effectsof the diabetic state on adenylate cyclase and surfactant secretion were reversed by in vivo but not in vitro insulin treatment.

Surfactant, a substance that prevents alveoli collapse at endexpiration, is synthesized and secreted by alveolar epithelial type II pneumocytes. Plopper and Morishige (1, 2) observed that experimentally induced diabetes selectively affected Clara cells and alveolar type II cells. In isolated perfused lung preparations of experimentally induced diabetic rats, glucose incorporation into surfactant phosphatidylcholine was greatly reduced (3, 4). In cultured type II pneumocytes of streptozotocin-induced diabetic animals, surfactant phospholipids and the rate of glucose incorporation into those phospholipids was greatly reduced, although glycerol incorporation was increased (5). In both models (3-5), in vivo but not in vitro insulin treatment reversed the effects of experimentally induced diabetes to within control values. I3-Adrenergic agents have been shown to stimulate surfactant release from adult type II cells in culture (6-8). In type

(Received in original form September 26, 1989 and in revised form August 22, 1990) Address correspondence to: Lou Ann S. Brown, Ph.D., Department ofPediatrics, Emory University, 2040 Ridgewood Drive, Atlanta, GA 30322. Abbreviations: adenosine 5'-triphosphate, ATP; bovine serum albumin, BSA; cyclic adenosine monophosphate, cAMP; Dulbecco's modified Eagle's medium, DMEM; guanosine 5'-triphosphate, GTP; isoproterenol, ISO, Am. J. Respir. Cell Mol. BioI. Vol. 4. pp. 108-114, 1991

II pneumocytes from adult streptozotocin-induced diabetic rats, the rate of surfactant secretion was biphasic, reaching a minimum at 1.5 h (9). When challenged by the l3-adrenergic agonist isoproterenol (ISO), secretion was stimulated only 30% and remained biphasic. This is in contrast to a linear increase of 100% in control cells. The effects were reversed by in vivo but not in vitro insulin treatment. These results suggest that the capacity of l3-adrenergic agents to stimulate cellular cyclic AMP (cAMP) content and consequently surfactant secretion was altered. Altered adenylate cyclase and cAMP phosphodiesterase activities in whole rat lung tissue from streptozotocin-induced diabetic rats (10) support this hypothesis. Glucose transport was reduced also in the isolated perfused lung of the streptozotocin-induced diabetic animals. However, these effects were reversed by the addition of insulin to the perfusion media (11). Although there are over 40 different cell types in the lung, these results suggested that some cellular processes altered by the diabetic state are able to respond to insulin under in vitro conditions. In cultured type II pneumocytes from control rats, insulin receptors and an insulin dose-dependent increase in glucose transport were demonstrated (12, 13). However, the presence of functional insulin receptors on type II pneumocytes isolated from diabetic rats has not been reported. The purpose of the present investigation was to determine if the al-

Brown: Cyclic AMP and Insulin Effects in Diabetic Type II Cells

tered secretory pattern observed in cells from diabetic animals was due to nonfunctional insulin receptors or to an altered adenylate cyclase cascade.

Materials and Methods Materials Materials for type II pneumocyte isolation were obtained from Sigma Chemical Co. (St. Louis, MO). Fluorocarbon was purchased from SCM Chemicals (Gainesville, FL). 2-Deoxy-o-[l-'H]glucose, insulin, [ '251]porcine, rnonoiodinated receptor grade, and [8-1 4C]adenosine 5'-triphosphate ([8- '4C]ATP) were purchased from New England Nuclear (Boston, MA). Crystalline porcine insulin, ISO, sodium fluoride, guanosine 5'-triphosphate (GTP) , forskolin, and 2-deoxy-o-glucose were obtained from Sigma. Streptozotocin was purchased from ICN Biochemicals (Irvine, CA). The cAMP radioimmunoassay kit was purchased from BTl, Inc. (Stoughton, MA). Animals and Cell Isolation Sprague-Dawley rats (175 to 200 g; Harlan Co., Madison, WI) starved for 16 h were anesthetized with ether and injected via the tail vein with 65 mg streptozotocin (citrate buffer, 0.1 M; pH 4.5) per kg body weight. After 5 d, animals with blood glucose greater than 350 mg/dl were used for cell isolation. For in vivo insulin treatment, injections of Iletin insulin (4 IV) were administered via the tail vein 30 min before surgery to rats under sodium pentobarbital anesthesia. Type II pneumocytes were isolated by trypsinization and discontinuous bovine serum albumin (BSA) density-gradient centrifugation as previously described (14). The cells were suspended at 106 cells/ml in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum, 100 U/ml penicillin, and 100 1J.g/ml streptomycin and were cultured for 20 h. After adherence, the monolayers were rinsed with medium to remove nonadherent cells. The preparation from diabetic rats yielded an average of25 ± 1.5 x 106 cells/lung with 93 % type II pneumocyte purity. The preparation from control rats yielded 35 ± 0.9 x 106 cells/lung with 96% cell purity. The lower yield of type II pneumocytes from streptozotocin-induced diabetic rats was because of decreased plating efficiency (28 %) as opposed to control cells (40%). Insulin Binding Insulin binding by type II pneumocytes cultured for 20 h in 24-well plates was measured by the procedure described by Prince and coworkers (15). The monolayers of type II pneumocytes were washed twice with binding buffer (Kreb's Ringer phosphate buffer containing 3% [wt/vol] BSA, 100 U/ml penicillin, and 100 1J.g/ml streptomycin; pH 7.4). Binding buffer containing [1251]insulin (0.5 1J.Ci/ml; 0.25 ng/ml) with or without increasing concentrations of unlabeled porcine insulin (1 to 100 ng/ml) was added to each culture well. At 15° C, the specific binding of [' 25I]insulin to cells from control and diabetic animals reached steady state at 150 min and remained constant for an additional hour (data not shown). Therefore, the cells were incubated with the binding buffer for 3 h at 15° C in a reciprocating water bath (100 oscillations/min). The incubation was terminated by remov-

109

al of the buffer and placement of the plates on ice. The monolayers were then washed 6 times with ice-cold buffer. Following the addition of 1 N NaOH to each well, the cells were incubated for 30 min at 22° C at 100 oscillations/min. The solubilized cell material was transferred to scintillation vials, each well washed with 1 N NaOH, and the radioactivity measured in a Beckman Gamma-4000 counter. Nonspecific binding was defined as the cell-associated radioactivity measured in the presence of 100 1J.g/ml nonradioactive porcine insulin (15). Under the conditions of the assay, nonspecific binding of [1251]insulin was approximately 9 % of the total radioactivity bound for cells from diabetic and control rats. Nonspecific binding was subtracted from the total radioactivity bound at each concentration of insulin and reported as specific insulin binding and normalized to 100 cells. The computer program Ligand was used to derive the best fit for receptor affinity. Insulin degradation was measured by the trichloroacetic acid precipitation method (16) and was less than 10% of the added [1251]insulin. The kinetics of insulin degradation were not significantly altered as a result of streptozotocin-induced diabetes. 2-Deoxy-o-glucose Transport Measurement of cellular uptake of 2-deoxy-o-[l-'H]glucose uptake was determined in triplicate on cell mono layers maintained in 24-well plastic culture plates for 20 h. The monolayers were washed twice with Kreb's Ringer phosphate bicarbonate buffer containing 3 % (wt/vol) BSA, 100 Ulml penicillin, and 100 1J.g/ml streptomycin. The cells were preincubated in this buffer for 1 h at 37° C without or with increasing concentrations of nonradioactive porcine insulin (0.25 nM to 50 1J.M). Following this preincubation, the transport reaction was begun by the addition of 2-deoxy-o[PH]glucose (0.4 1J.Ci; 0.1 mM) to each well. The reaction was terminated after 2 min by rapid aspiration of the buffer and by washing 4 times with ice-cold buffer containing 0.3 mM phloretin (pH 7.4). To each well, 500 1J.l of Triton X-IOO (0.5 %) was added, the wells were shaken at 100 oscillations/min for 30 min at 37° C, and the solubilized cell material was transferred to scintillation vials. The wells were rinsed with another 500 1J.l of Triton X-lOO, and the combined radioactivity was determined by liquid scintillation techniques. Simple diffusion and extracellular trapping of radioactivity were corrected for by subtracting the value for glucose-nj'Cfl.Ij] uptake. Results were expressed as nanomoles of 2-deoxy-o-glucose transported per 105 cells. Adenylate Cyclase Assay Type II pneumocytes from control and streptozotocin-induced diabetic animals were cultured for 20 h in 60-mm tissue culture plastic dishes. After the attachment period, the monolayers were washed 3 times with DMEM to remove nonadherent cells. To each dish,S mM EDTA/IO mM Hepes was added for 5 min. The dishes were gently scraped, and the cell suspension was transferred to a plastic centrifuge tube. The dishes were gently scraped again in the presence of Kreb's Ringer phosphate buffer. The pooled cell suspensions were centrifuged at 140 X g for 8 min. The cell pellet was resuspended in 1.0 mM EDTA/IO mM Hepes, and the cells were allowed to swell in this hypotonic solution for 30 s. Hepes (40 mM) was added and the cell suspension quickly

110

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 41991

mixed. Using a 7-ml Dounce homogenizer, the cells were homogenized by delivering 20 firm strokes using the tight pestle. The cell homogenate was frozen in a dry ice/acetone bath and stored at -70° C. Adenylate cyclase activity was determined within 2 to 3 d after freezing. Adenylate cyclase activity was determined in triplicate at 37° C according to the procedure of Drummond and colleagues using [8-1"C]ATP as substrate (17). Cellular protein (2 to 20 p.g) was incubated for 3 min in an assay medium containing 40 mM Tris-HCI buffer, pH 7.4; 8 mM theophylline; 2 mM cAMP; 5.5 mM KCI; 15 mM MgCI 2 ; 20 mM phosphoenolpyruvate, 130 p.glml pyruvate kinase, and added hormone. The reaction was initiated by the addition of [8- 14C]ATP (10 p.Ci/mol ATP), continued for 10 min, and terminated by heating the mixture in a boiling-water bath for 3 min. cAMP was separated on Dowex columns and the radioactivity of the cAMP fraction determined by liquid scintillation techniques (18). Column recoveries of (3H]cAMP during the purification of [,4C]cAMP were 70 to 80% during these studies. Protein was estimated by the method of Bradford (19) using BSA as the standard. Results were expressed as nanomoles of ATP converted per minute per microgram of protein. In some experiments using cells from diabetic rats, the assay mixture contained 10 nglml of crystalline porcine insulin. In other experiments, the diabetic rats received injections of insulin via the tail vein prior to surgery. Cyclic Nucleotide Determination Type II pneumocytes from control and diabetic animals were cultured for 20 h in medium containing no radioactivity. After washing to remove nonadherent cells, Kreb's Ringer phosphate bicarbonate buffer was added to all wells. Isobutylmethylxanthine (100 p.M) was added to inhibit cAMP degradation. After 2 h of equilibration at 37° C, forskolin was added and the incubation continued for 2 min. To terminate the incubation, the media was aspirated and the cellular cAMP content was determined by radioimmunoassay after succinylation. The cyclic nucleotide content was expressed as picomoles of cAMP per lQ6 cells. In some experiments, 1 p.Mcrystalline porcine insulin was added to the incubation buffer during the 2-h equilibration period. Studies of Surfactant Secretion Isolated cells were distributed onto 24-well culture plates and incubated with 0.75 p.Ci/ml of [methyl-3H]choline chloride for 20 h. After the adherence period, the wells were washed as described above and the cells were allowed to equilibrate in medium without radioactivity for 30 min at 37° C. Agonists were added according to experimental design. After 3 h of incubation, the medium was removed and saved. The adherent cells were extracted with 70% methanol. Phospholipids from both the media and the cells were extracted with chloroform:methanol (1:1, vol/vol), and the radioactivity of the organic phase was determined. The percentage of cellular phosphatidylcholine secreted was calculated as follows: [cpmmed,um/ (cpm.a,

+

Cpmmedium)]

X

100 %

where cpm is counts per minute. The percentage of surfac-

tant secreted during the 30-min equilibration period was subtracted from all samples. Other Procedures Cell viability was screened by exclusion of 0.2 % trypan blue (94 ± 5 %) and was unaltered by the diabetic state. Cell number was determined by counting the cells through an ocular grid on a Bausch and Lomb light microscope. Tests of statistical significance included analysis of variance for multiple comparisons and the Dunnett's test for difference.

Results Insulin Binding Insulin-binding experiments performed with cultured alveolar type II pneumocytes from control animals and streptozotocin-induced diabetic animals are shown in Figure 1. The absolute percentage of [125I]insulin specifically bound was 1.0 ± 0.4% per lQ6 cells and 0.49 ± 0.03% per 106 cells for control and diabetic models, respectively. Nonradioactive insulin was effective in competing with ['25I]insulin for binding sites on cells from control and diabetic animals. The competition was dose-dependent, and half-maximal displacement occurred at 6 nglml and 7 nglml of nonradioactive insulin for cells from control and diabetic animals, respectively. The value of 6 nglml for control cells agreed with the value of 4 ng/ml reported by Sugahara and associates (12) for control type II pneumocytes. Scatchard analysis of the insulin-binding data presented in Figure 1 for type II cells from control and diabetic animals is shown in Figure 2. As demonstrated previously for type II pneumocytes (12, 13), a curvilinear shape characteristic of negative cooperativity in binding by insulin receptors was observed. This characteristic binding curve was demonstrated in cells from diabetic

o CONTROL • DIABETIC

INSUUN CONCENTRATION (ng/ml)

Figure 1. Ability of porcine insulin to compete with [125I]insulin binding to alveolar type II cells from control and streptozotocininduced diabetic rats. Type II cells were incubated with 0.25 ng/ml ['25I]insulin for 3 h at 15° C as described under MATERIALS AND METHODS. Increasing concentrations of unlabeled porcine insulin (1 to 100 ng/ml) were added to each monolayer of cells. Each experiment was assayed in triplicate, and the mean ± .SEis prese~t~d for four experiments. For some points, the SE are included within the symbol.

Brown: Cyclic AMP and Insulin Effects in Diabetic Type II Cells

111

50 0



1.00

W

CONTROL DIABETIC

~

0.75

au

30

"'I Ul

20

o~ I 0

0.50

Z

>- E

a

8--a

::>

Xc

o::l

0.25

10

I

N

"""'"

0 0.00 0.00

0.05

0.10

0.15

~::::::::"~

8

Figure 2. Scatchard analysis of the insulin-binding data presented in Figure 1. For some points, the SE are included within the symbol.

and control animals. Although cells from diabetic animals showed decreased receptor binding per lQ6 cells, the binding affinity profiles for the control and diabetic state were similar. Assuming that the ratio of insulin bound to insulin receptor was 1:1, extrapolation of the abscissa gave approximately 17,500 receptor sites per control cell, of which 20 to 30% were high-affinity sites with a K, of 0.429 nM. This agreed well with the values of 17,000 receptor sites per cell with 20 to 30% representing high-affinity sites and a range of s, from 0.15 nM (13) to 0.34 nM (12). In the diabetic state, the receptor sites per cell were significantly reduced to 8,500 (P ~ 0.05). As in control cells, 20 to 30% of the sites were high-affinity sites but the K, was 0.203 nM (P ~ 0.05 when compared to control cells). 2-DeoxY-D-glucose Transport To determine if the insulin receptors were functional, the effect of insulin concentration on initial 2-deoxY-D-glucose uptake was determined (Figure 3). Insulin stimulated 2-deoxy-n-glucose uptake by type II pneumocytes in a dosedependent manner and was not significantly altered by the diabetic state. Insulin (20 nM) stimulated 2-deoxy-D-glucose uptake from 1.43 ± 0.34 to 37.4 ± 2.9 nmol/lfl' cells in control cells. In the diabetic state, insulin (20 nM) stimulated the uptake from 0.9 ± 0.13 to 43.8 ± 3.1 nmol/IO' cells.

.J

/'

10- 0

10-8 INSULIN [M]

0.20

INSUUN BOUND (ng/10 CELLS)

[1

0 - 0 CONTROL . - . DIABETIC

...J~

~

0

40

UU1 ::::>0

""""0::

Vl ...J W...J VlW

~

'"... 0

T

T/6•

~

Figure 3. Effect of insulin concentration on 2-deoxy-D-glucose uptake by alveolar type II pneumocytes from control and streptozotocin-induced diabetic rats. After a preincubation without or with increasing concentrations of insulin for 1 h at 37° C, the transport of radioactive 2-deoxY-D-glucose was assayed as described under MATERIALS AND METHODS. Each point represents a mean of triplicate determinations from eight different experiments for controls and eight different experiments for diabetics.

basal and NaP-activated adenylate cyclase activities of the cultured type II pneumocytes (Figure 5). They were no longer significantly different from control cells. Cyclic AMP While adenylate cyclase activity of type II pneumocytes from streptozotocin-induced diabetic animals was not stimulated by GTP, NaF, or ISO, a 2-min addition of 1 J-tM forskolin to these cells stimulated the cellular content of cAMP 13.6fold from a basal value of 0.65 pmol cAMP/1Q6 cells (Figure 6). The cellular cAMP content of forskolin-stimulated diabetic cells was not statistically different from forskolinstimulated control cells (0.886 ± 0.032 versus 0.738 ± 0.080 250

z

W

I-

aa:::

o 200

_

*

CONTROL DIABETIC

a, 01

~ c

150

*

'E

Influence of streptozotocin-induced diabetes on adenylate cyclase activity in cultured type II pneumocytes.

Previous studies with cultured type II pneumocytes from streptozotocin-induced diabetic rats demonstrated altered surfactant synthesis and secretion. ...
786KB Sizes 0 Downloads 0 Views