In Vivo Regulation of Surfactant Proteins by ... - ATS Journals

27 downloads 45 Views 1MB Size Report
James H. Fisher, Frank McCormack, Sung Soo Park, Tom Stelzner, John M. ... Roosevelt Institute for Cancer Research; Webb-Waring Lung Institute; and ...
In Vivo Regulation of Surfactant Proteins by Glucocorticoids James H. Fisher, Frank McCormack, Sung Soo Park, Tom Stelzner, John M. Shannon, and Talia Hofmann Department of Medicine, University of Colorado Health Sciences Center, Division of Pulmonary Sciences; Eleanor Roosevelt Institute for Cancer Research; Webb-Waring Lung Institute; and National Jewish Center for Immunology and Respiratory Medicine, Department of Medicine, Denver, Colorado

Surfactant proteins have key roles in regulating surfactant secretion, in recycling, and in the assembly of the surfactant monolayer but little is known about their regulation in vivo. Surfactant proteins SP-A, SP-B, and SP-C have been shown to be upregulated by glucocorticoids in vitro, but the role of glucocorticoids in the physiologic regulation of surfactant protein synthesis remains unknown. We have studied the effects of exogenously administered glucocorticoids on the regulation of steady-state surfactant protein mRNA accumulation. We have also studied the effects of adrenalectomy on the accumulation of the surfactant protein mRNAs. Surfactant protein genes appear to have quantitatively different responses to exogenously administered glucocorticoids, with SP-C mRNA increasing at the lowest dose, SP-Aand SP-BmRNA increasing in response to similar glucocorticoids doses but with SP-B yielding the highest maximum response. Adrenalectomy, however, does not alter surfactant protein mRNA levels. These observations support a minor role for glucocorticoids in maintaining the steady-state accumulation of surfactant protein mRNA. Adrenalectomy decreases total pulmonary SP-A when compared to sham-operated animals in the absence of changes in its mRNA. Therefore, glucocorticoids may have translational or post-translational effectsthat regulate total pulmonary SP-A accumulation, but the effects appear to be minor. These findings support a potential role for the adrenal in the pulmonary response to stress and demonstrate for the first time differential accumulation of the surfactant protein mRNAs to glucocorticoids in vivo.

The importance of lung surfactant in maintaining normal lung function is well documented (1), but the mechanisms that regulate its synthesis, secretion, and recyclingare incompletely understood. Surfactant is composed primarily of phospholipid but contains, approximately2 to 4% surfactantspecific proteins by weight (1, 2). It is believed that the surfactant proteins are important in the assembly of the surfactant monolayer and in regulating surfactant homeostasis. Supporting evidence for this hypothesis includes the observation that surfactant protein A (SP-A) inhibits the stimulated secretion of phosphatidylcholine by cultured alveolar type II cells (3, 4) by binding a high-affinity saturable cell membrane receptor (5,6). SP-A can also facilitate the internationalization of synthetic mixtures of phospholipids by cultured alveolar type II cells (7).

(Received in original form August 28, 1990 and in revised form November 27, 1990) Address correspondence to: James H. Fisher, M.D., Department of Medicine, University of Colorado Health Sciences Center, Division of Pulmonary Sciences, 4200 East Ninth Avenue, Denver, Colorado 80262. Abbreviations: counts per minute, cpm; enzyme-linked immunosorbent assay, ELISA; 4 M guanidine isothiocyanate, 0.025 M Na citrate (pH 7.0), 05% sarcosyl, 0.7% 2-mercaptoethanol, GITe solution; phosphate-buffered saline, PBS; sodium dodecyl sulfate, SDS; surfactant proteins A, B, and C, SP-A, SP-B, and SP-C, respectively; trichloroacetic acid, TeA. Am. J. Respir. Cell Mol. BioI. Vol. S. pp. 63-70, 1991

The hydrophobic surfactant proteins SP-B and SP-C are believed to have primary roles in accelerating surface spreading of surfactant phospholipid (8, 9), but the ratio of SP-A to SP-B and phospholipid may be important in determining the morphologic structure of surfactant, with specific ratios of SP-A to SP-B favoringthe formation of tubular myelin (10). Tubular myelin is concentrated in the rapidly sedimenting fractions of surfactant that tend to have more surface activity than buoyant fractions (11). These observations suggest that the ratio of surfactant proteins to phospholipid and each other in the alveolus may impact surfactant activity and recycling in vivo. It is apparent that there are multiple levels of regulation of surfactant secretion, morphology, spreading, and recycling in which surfaetantprotein concentrations have potential importance. Therefore, understanding the regulation of the expression of each surfactant protein gene may be of fundamental importance in understanding surfactant function and homeostasis. Examination of DNA sequences upstream from the coding region for each surfactant protein gene reveals consensus recognition elements for both cyclic/adenosine monophosphate and glucocorticoids (12-14), suggesting that humoral agents might be important in regulating the production of surfactant proteins. Fetal lung explants exposed to glucocorticoids demonstrate increased mRNA for SP-B and SP-C as well as increased SP-B and SP-C protein (16, 17). The effects of glucocorticoids on SP-A are more complex, with high doses

64

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 51991

inhibiting SP-A gene expression but lower doses stimulating the accumulation of both mRNA and protein (18-21), only in part by transcriptional mechanisms (22). Thus, glucocorticoids appear to increase the expression of all surfactant protein genes at some dose in vitro. Glucocorticoids have been shown to increase SP-A mRNA and protein in fetal and adult rats in vivo (15, 24). The role of cell differentiation and local factors in the regulation of surfactant proteins is poorly understood, but the accelerated epithelial maturation observed in fetal explants makes it likely that both factors could be altered in vitro. Therefore, it is important to study the potential role of humoral factors in vivo as well as in vitro. It is also uncertain to what extent glucocorticoids have a physiologic role in the regulation of surfactant proteins in adult animals. We have studied the effects of various doses of glucocorticoids and adrenalectomy on steady-state accumulation of mRNA for each surfactant protein, both in whole lung and isolated alveolar type II cells. We have also measured the effects of glucocorticoids on total lung SP-A protein content.

Materials and Methods Sprague-Dawley rats weighing 250 to 300 g were purchased from Sasco Inc. (Grand Island, NE) and allowed to acclimatize for at least 2 wk. Animals were allowed food and water ad libitum and given a 12-h light/dark cycle. For assessment of diurnal variation, animals were removed from the animal care facility, kept in a quiet room with no change in the preexisting light/dark cycles, and at specific times were euthanized by lethal pentobarbital injection. Animals were given various doses of dexamethasone (Ivenex Laboratories, Rosemont, IL) diluted to 1 mg/ml in bacteriostatic sterile water containing 0.9% benzoyl alcohol preservative by subcutaneous injection. Control animals were given identical volumes of bacteriostatic sterile water by subcutaneous injection. For adrenalectomy, animals were anesthetized with ketamine/xylazine given by intramuscular injection. Bilateral adrenalectomies were performed through posterior incisions. Adrenalectomy was confirmed by postmortem examination and by measurement of serum corticosterone levels 1 wk after surgery (Hazelton Laboratories). Sham-operated animals were subjected to an identical procedure, with the adrenal glands visualized but not manipulated. Lung tissue was harvested by giving the animals a lethal intraperitoneal dose of pentobarbital, transecting the abdominal aorta, then removing the lungs and heart en bloc. Major vessels and airways were dissected away from the lung, and a portion of each lung was flash-frozen in dry ice and ethanol for SP-A and DNA determination. The remaining lung tissue was homogenized in 4 M guanidine isothiocyanate, 0.025 M Na citrate (pH 7.0),0.5% sarcosyl, 0.7% 2-mercaptoethanol (GITe solution) (25) 10 ml/g tissue and frozen at -70° C. SP-A Quantification Lung SP-A was determined by double-sandwich enzymelinked imrnunosorbent assay (ELISA) using a polyclonal antibody raised in New Zealand white (Hazelton) rabbits against purified rat SP-A, which has been described previously (26). Briefly, weighed lung fragments were homogen-

ized in 4 ml 1% Triton X-IOO/phosphate-buffered saline (PBS) using a polytron and then sonicated briefly. Untreated microtiter plates (Dynatech Laboratories, Alexandria, VA) were incubated with a 0.1 ml/well rabbit anti-rat SP-A IgG fraction (0.1 #-,g/ml and 0.1 M NaHC03 [pH 9.3]) overnight at 22° C. The antibody solution was removed, and the wells were incubated with PBS without calcium or magnesium (pH 7.4) containing 1% Triton X-IOO and 3 % bovine serum albumin for 30 min, then washed twice with the same buffer. Then 0.1 ml of various dilutions of purified rat SP-A standard (0 to 20 ng/ml) or dilution of sample lung homogenate (1:250 to 1:1,000) were added to wells and incubated at 37° C for 90 min. The wells were then washed 3 times with 3% albumin/PBS/I % Triton X-100, and 0.1 ml of rabbit anti-rat SP-A IgG fraction conjugated with horseradish peroxidase (Boehringer Mannheim, Indianapolis, IN) was added to each well and incubated for 90 min at 37° C. Wells were washed 4 times with 1% Triton X-IOO/PBS, and 0.1 ml of substrate solution (0.1 % [wt/vol] a-phenylenediamine (Sigma Chemical Co., St. Louis, MO), and 0.03% (vol/vol) hydrogen peroxide in 0.1 M citrate buffer at pH 4.6 was added to each well. Plates were incubated in the dark at 22° C for 10 min. The reaction was stopped by the addition of 0.1 ml of 2 M H2S04 per well. Adsorption at 490 nm was determined with a microtiter autoreader EL307 (Biotech Instruments). DNA Determination DNA was precipitated from 0.1- to 0.2-ml aliquots of 1% Triton X-IOO/PBS lung homogenate by the addition of trichloroacetic acid (TeA) to 10% final concentration. The precipitate was collected by centrifugation and washed with methanol followedby absolute alcohol. DNA precipitate was resuspended in TE buffer (10 mM Tris-HCl [pH 8.0], 1 mM Na2 EDTA), and DNA was quantified by the fluorometric method of Setaro and Morley (27). Preparation of RNA RNA was purified by the acid phenol method of Chomczynski and Sacchi (25). Briefly, 500-#-,1 aliquots of lung homogenized in GITe solution 10 ml/g lung were acidified by the addition of 50 #-,1 of 2 M sodium acetate (pH 4.0) followed by the addition of 500 #-,1 of phenol buffered to pH 7.5 with 0.1 M Tris (pH 7.5), 10 mM EDTA. The mixture was vortexed, and 100 #-,1 of chloroform/isoamyl alcohol (49:1) was added followed by vortexing. The mixture was incubated on ice for 15 min and the phases separated by roomtemperature centrifugation for 5 min at 10,000 x g in a microfuge. The aqueous phase was removed with a micropipet, and the RNA precipitated by the addition of an equal volume of isopropanol. After incubating at -20° C for at least 2 h, the precipitate was collected by centrifugation for 5 min at 10,000 x g. Supernatant was decanted and precipitate redissolved in a 150 #-,1 of GITe solution followed by precipitation with an equal volume of isoproterenol overnight at -20° C. After centrifugation, the precipitate was collected at 10,000 x g for 5 min in a microfuge and washed once with 70% ethanol. The precipitate was dried and resuspended in 0.1% diethyl pyrocarbonate-treated 1 mM EDTA(pH 8). RNA was quantified by adsorption at 260 nm, and the quality of RNA was assessed by visual inspection of an ethidium bromide-stained formaldehyde/agarose dena-

Fisher, McCormack, Park et al.: In Vivo Regulation of Surfactant Proteins by Glucocorticoids

turing gel. If there was visible degradation, the sample was discarded. RNA Hybridization Assays The abundance of each mRNA was estimated as a fraction of total RNA and in relation to beta cytoskeletal actin mRNA by a filter hybridization assay developed in this laboratory. The complete coding regions for each rat surfactant protein cDNA (SP-A, SP-B, and SP-C) (28-30) were subcloned individually into pGem 3Z or 4Z such that either antisense or sense transcripts were obtained by using SP6 RNA polymerase. Yields from transcription reactions averaged 20 to 30 p,g of full-length transcript/ug of-linearized vector. That transcripts were full length was verified by analysis of ethidium bromide-stained formaldehyde/agarose denaturing gels. Known amounts of sense transcript (0, 0.1, 0.5, 1.0, 2.5, and 5 ng) or whole lung RNA (l p,g) were applied in triplicate to 13-mm nitrocellulose circles (pore size, 0.45 p,m, BA 85; Schleicher and Schuell) in lOx SSC/50% formarnide after denaturing at 65° C for 10 min. RNA was applied in a 20-p,1 vol/filter followed by air drying and heating at 80° C for 2 h. Filters were prehybridized at 56° C in 0.2 to 0.5 ml/filter of prehybridization solution consisting of 0.8 M NaCl, 10% dextran sulfate, 50% formamide, and 1% sodium dodecyl sulfate (SDS) for 6 to 12 h with vigorous shaking. Hybridizations were conducted in 0.2 to 0.5 ml/filter of hybridization mix consisting of Sx SSC, ix Denhardt's solution, 1 mg/ml sonicated denatured salmon sperm DNA, 10% dextran sulfate, and 0.5 % SDS using 5 X 105 to 1 X lQ6 cpm/ml probe at 57° C with vigorous shaking. cDNA inserts for SP-A (28), SP-B (29), SP-C (30), or the untranslated 3' region of human beta actin (31) were labeled to 1 to 5 X 109 cpm/p,g by random oligonucleotide primed secondstrand DNA synthesis using a kit from BRL. After 24 h of hybridization at 56° C for SP-A, SP-B, and SP-C or 37° C for beta actin, the filters were washed at room temperature 3 times in zx SSC, 0.2 % SDS and then washed 3 times at 65° C in O.1x SSC, 0.2 % SDS. The filters were air-dried and counted individually in scintillation vials using a commercially available scintillation cocktail. Hybridization conditions were the same as those yielding a specific low background banding pattern on RNA blot analysis. That nonspecific hybridization was not occurring was determined by including filters loaded with RNA from tissues that do not express surfactant proteins such as liver, kidney, and heart, and also filters containing no RNA. Counts per minute (cpm) deposited on these controls and filters without RNA were not significantly different from one another. Content of a specific mRNA species was determined by applying a regression equation derived from the standard curve to the cpm deposited on triplicate filters containing sample RNA. Statistics For RNA analysis, comparisons between groups were made using one-way ANOVA with correction for multiple comparisons. The presence of a linear dose response for single dexa'methasone injections was derived using an orthogonal polynomial contrast on the group means (34). Todetermine if the dose-re-sponse relationships were significantly different for SP-A, SP-B, and SP-C, an analysis of covariance was used

65

(40). Regression equations were derived using the Epistat statistical package. For statistical analyses, the SAS package was used. For all analyses, statistical significance was defined by a P value of Jess than or equal to 0.05.

Results Hybridization Assay Documenting significant differences in RNA content less than 2- to 3-fold is difficult. We sought to improve on the standard analyses such as slot blot and Northern blot by using a standardized filter hybridization assay. To assess the accuracy of filter hybridization, its linearity, and its specificity, the following experiments were performed. (1) To determine that our method of applying RNA to filters resulted in quantitative and reproducible binding, 10,000 cpm of 35S-labeled sense orientation SP-A RNA transcript was applied to 13-mm-diameter nitrocellulose filters alone and in the presence of 1 to 20 p,gof total cellular RNA. The filters were incubated in prehybridization buffer at 56° C for 24 h and then with hybridization buffer at 56° C for 24 h, washed according to the hybridization protocol, and counted in scintillation vials. Ninety-four percent of the input leA precipitable counts bound to the filters. Binding of 35S RNA was identical at all cellular RNA inputs (data not shown), indicating that the RNA binding capacity of the filters was not saturated under these conditions. (2) Various amounts of lung RNA were applied to filters and hybridized with labeled cDNA inserts for SP-A, SP-B, SP-C, or actin (Figure lA). There was a linear relationship between cpm deposited/filter and RNA input over a wide range, with an acceptable degree of variation of less than 10% among triplicate filters. (3) To determine if a standard curve would be linear under the same conditions, varying amounts of SP-A, SP-B, and SP-C RNA transcript were loaded on filters and hybridized. As Figure IB demonstrates, the reaction is linear over approximately a 50-fold input of sense orientation transcript. (4) To determine the effect of nonhomologous RNA content on hybridization, various amounts of total cellular RNA from spleen, which contains no SP-A transcript, were applied to filters simultaneously with various amounts of sense transcript for SP-A (Figure lC). At all spleen RNA inputs, deposited cpm/filter were nearly identical and the slope of the curves was not significantly different. Thus, nonspecific binding was not significant, nor was displacement of transcript by an excess of cellular RNA evident. Therefore, it is apparent that this method yields a linear increase in hybridized cpm as a function of sense orientation RNA input over a range of approximately 50-fold. Linearity and reproducibility of the assay is not altered by a concomitant IO-folddifference in input of nonspecific RNA. Finally, identical estimates of SP-A, SP-B, and SP-C mRNA content were obtained with both standardized filter hybridization and solution hybridization (32) using labeled antisense RNA transcript as a probe (data not shown). One significant advantage of this method is that statistical analysis can easily be applied to either cpm/filter or total content of message as derived from regression analysis of a standard curve. Additionally, by loading various numbers of identical filters, mRNA copy number for different mRNA species can be assayed simultaneously.

66

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 51991

A

FILTER HYBRIDIZATION ASSAY FOR SP-A

2000

o

1800



1600 o:

,

1400

c;:

1200

ld

-'