0013-7227/78/1035-1725S02.00/0 Endocrinology Copyright © 1978 by The Endocrine Society

Vol. 103, No. 5 Printed in U.S.A.

Triiodothyronine Nuclear Binding in Fetal and Adult Rabbit Lung and Cultured Lung Cells* JEFFREY A. LINDENBERG.f ARLETTE BREHIER, AND PHILIP L. BALLARD Cardiovascular Research Institute and Department of Pediatrics, University of California, San Francisco, California 94143 ABSTRACT. To investigate the possible mechanism of thyroid hormone action in the lung, we examined fetal and adult rabbit lung, and cell lines derived from lung, for specific nuclear binding sites for T3. Using incubation of isolated nuclei with L-[125I]T3 at 37 C, we found approximately 2400 specific binding sites/cell in fetal lung and 1120 sites/cell in adult lung, with a similar dissociation constant (~500 pM) for both tissues. The L2 and A549 cell lines, which may have originated from pulmonary type II alveolar cells, contained 2280 and 1580 nuclear sites/cell, and the dissociation constants

I

DIOPATHIC respiratory distress syndrome of premature infants results from inadequate amounts of pulmonary surfactant, the lecithin-rich material required for stabilization of lung alveoli. Studies in animals have shown that glucocorticoids accelerate both surfactant production and morphological development in fetal lungs (1), and corticoid treatment of pregnant women reduces the incidence of respiratory distress syndrome in premature infants (2). Recent reports by Wu et al. (3) and by Smith and Torday (4) suggest that administration of thyroid hormone has similar stimulatory effects on lung development in the fetal rabbit. Many effects of thyroid hormone in target tissues involve stimulation of both RNA and protein synthesis after localization of hormone in the nucleus. Binding of thyroid hormone to specific nuclear receptors seems to be an early step in this mechanism (5-11). In the present study, we examined lung tissue and cultured Received January 20,1978. Address all correspondence and requests for reprints to: P. L. Ballard, M.D., Cardiovascular Research Institute, University of California, San Francisco, California 94143; phone, (415)666-1940. * Tins work was supported by Pulmonary Specialized Centers of Research Grant HL-19185 from the NHLBI. t James Scholar medical student at the University of Illinois, Chicago, Illinois.

were 280 and 200 pM, respectively. In fetal lung, the ability of various analogs to compete for L-T3 (100%) binding was: 3,5-diiodo-3'-isopropylthyronine, 81%; DT3, 73%; L-T4, 6.7%; 3,3'-diiodothyronine, 0.19%; 3,5-dimethyl-3'-isopropyl-L-thyronine, 0.15%; and rT3, 0.08%. These findings indicate that both fetal and adult lung, and cultured lung cells, contain specific nuclear binding sites for T3, suggesting that these tissues and their type II alveolar cells may be directly influenced by thyroid hormones. (Endocrinology 103: 1725, 1978)

L-2 and A549 lung cells for nuclear binding of T3. We found specific binding sites in both fetal and adult lung, indicating that these tissues may be directly influenced by thyroid hormones. Similar binding occurred in two cell lines which may have originated from alveolar type II cells, the site of surfactant synthesis. A preliminary report of some of these results has been published (12). Materials and Methods Materials L-[ 1 2 5 I]T 3 (58-910 juCi/jug) was purchased from New England Nuclear Co. Nonlabeled L-T 3 , T4, and D-T3 were obtained from Sigma Chemical Co. 3,5Diiodo-3'-isopropyl-L-thyronine; 3,3'-diiodo-L-thyronine; 3,5-dimethyl-3'-isopropyl-L-thyronine (DIMIT); and rT 3 were gifts from Dr. E. C. Jorgensen of the University of California, School of Pharmacy, San Francisco, CA. All other chemicals were reagent grade and were obtained from commercial suppliers. Stock solutions of thyroid hormones were prepared in 100% n-propanol and stored in the dark at 5 C until used, at which time they were diluted to desired concentrations in 0.1 N NaOH.

Methods Preparation of nuclei. Adult male and time-dated pregnant New Zealand White or New Zea-

1725 The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 November 2015. at 17:26 For personal use only. No other uses without permission. . All rights reserved.

1726

LINDENBERG, BREHIER, AND BALLARD

Endo • 1978 Vol 103 • No 5

land/California hybrid rabbits were supplied by ml more of wash solution, and centrifuging it again commercial breeders. Rabbits were killed by iv at 1200 X g for 10 min. The wash procedure was injection of 4-6 ml sodium pentobarbital (60 repeated once. The final pellet was assayed for radioactivity in mg/ml), and the fetuses were obtained by uterotomy. Fetal lungs were dissected free of trachea and a y-scintillation spectrometer with 50% efficiency. major bronchi, weighed, and frozen at —20 C. All Specific binding was determined by subtracting the results reported are from tissues stored for less than counts per min bound in the incubations with excess competing nonlabeled T 3 (nonspecific binding) 5 days. Tissue was minced while frozen and homoge- from the counts per min bound in the incubations nized in 4 vol 25 raM iV-Tris-(hydroxymethyl)- without competitor (total binding); results are exmethyl-glycine (Tricine), 0.25 M sucrose, 1 mM pressed as counts per min or femtomoles of T3 MgCl2, 1 mM dithiothreitol, pH 7.4 (TSM buffer), bound per tube or per jug DNA (13). Binding paramwith four to six strokes (2000 rpm) of a motor- eters were assessed by Scatchard analysis (14) of driven Teflon pestle (24- and 26-day gestational age dose-response experiments using duplicate detertissue required fewer strokes). This and all subse- minations. Statistical comparisons used Student's quent procedures before incubation were carried t test. out at 0-4 C. The homogenate was filtered under Binding in cultured lung cells. The A549 cell line, pressure through four layers of surgical gauze. The which was originally established from a human gauze was rinsed once with TSM buffer, and the alveolar cell carcinoma, was obtained from W. Nelhomogenate was adjusted to 10 ml TSM buffer/g son-Rees at the Naval Biomedical Research Labotissue and centifuged at 1200 X g for 10 min. The ratory in Oakland, CA, and cultured as described pellet was washed by resuspending in 10 vol wash by Lieber et al. (15). L-2 cells were provided by W. solution (TSM buffer with 0.2% Triton X-100) and H. J. Douglas of the W. Alton Jones Cell Science centrifuging as before. The pellet was then resus- Center, Lake Placid, NY, who originally established pended in wash medium and filtered under pressure the line by cloning from adult rat lung (16). Binding through four layers of surgical gauze and then by studies were carried out at passages 15-47 for the gravity through nylon mesh (75 /xm). The suspen- A549 cells and at passages 8-12 for L-2 cells. sion was again centrifuged at 1200 X g for 10 min, Intact cells (2-4 x 106) suspended in 0.5 ml and the pellet was resuspended in 5-8 ml TSM Swim's S77 medium without serum at pH 7.4 were buffer/g original tissue. This final preparation was incubated for 90 min at 37 C with various concenfiltered through a fresh piece of nylon mesh, as trations of [125I]T with or without 250 nM nonlaabove, and on the average contained 16 X 106 beled L-T . After 3incubation, the cells were lysed, 3 nuclei/ml, as determined by counting in a hemo- and the nuclear fraction was washed, as previously cytometer in the presence of trypan blue. Nuclei described (17), except that 0.2% Triton X-100 was from adult rabbit lungs were prepared in the same present in the wash solution. T binding in the fashion. In fetuses of 24 and 26 days, the first wash nuclear pellets was determined as 3described above. step was omitted. This procedure yielded whole nuclei generally free of cytoplasmic contamination Results as judged by examining trypan blue-stained preparations by light microscopy. Nuclear binding. The reaction mixture (total volume, 1 ml) contained 0.5 ml freshly prepared nuclear suspension (40-120 jig DNA), 0.45 ml TSM buffer, 0.025 ml L-[125I]T3 (2-300 nM), and 0.025 ml 0.1 N NaOH, nonlabeled L-T 3 (10~5 M), or analog. Unless stated otherwise, incubations were at 37 C for 90 min. After incubation, samples were chilled 1 min in an ice bath and centrifuged at 1200 X g for 10 min. When appropriate, 0.1 or 0.2 ml supernatant was removed and assayed for radioactivity to determine free [125I]T3 concentration. The remaining supernatant was then decanted. The pellet was washed by resuspending it in 1 ml wash solution, agitating it for 10 sec on a vortex mixer, adding 1

Since, to our knowledge, binding of T3 by nuclei of fetal mammalian tissue has not been described previously, we investigated several parameters of the nuclear assay system which has been utilized for tissues of adult animals (5, 6,10). Storage of fetal lung tissue for 1 h to 5 days at —20 C did not affect the level or properties of T 3 binding compared with fresh tissue. We found that 0.2% Triton X-100 in the wash solution provided the highest ratio of specific bound to total bound counts per min. Figure 1 demonstrates the effects of repeated nuclear washes on binding. Although a single wash removed most of the nonspecific

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 November 2015. at 17:26 For personal use only. No other uses without permission. . All rights reserved.

T 3 BINDING IN LUNG

1727 Non -Lab led T ,

0.6 -

37'

f)

0

A—

7/

0.2

f

—O—
40 CO

£

20

& CO

0

L. 10"

I0"9

lO-io

Competitor

10 " 8

Concentrotion (M)

TABLE: 2. Relative affinities of thyroid hormone analogs for binding to nuclei of 28-day-old fetal rabbit lung Compound

n

L-T 3

T2iPr"

3

D-T3 L-T4 3,3'-h-T2b

4 4 2 2 2

Relative affinity (mean and range) 100 81 73 6.7

(71-91) (62-92) (4.2-12) 0.19 (0.16-0.22) DIMIT 0.15 (0.06-0.23) rT3 0.08 (0.05-0.1) The relative affinity reflects the concentration of analog compared to that for L-T3 at which there is 50% inhibition of [125I]T3 binding, as determined in experiments such as shown in Fig. 4. 0 3,5-Diiodo-3'-isopropyl-L-thyronine. 6 3,3'-Diiodo-L-thyronine.

from at least day 24 of gestation, before both morphological maturation of lung cells and appearance of alveolar surfactant. The presence of T3-binding sites in fetal lung is consistent with a direct action of thyroid hormones in this tissue. We find no change in the affinity of T 3 binding during late fetal life and compared with adult rabbit lung. The concentration of binding sites, however, is lower in adult than in fetal life. We estimate the average number of binding sites per cell to be 2400 for fetal lung and 1120 for adult lung. This difference may reflect fewer binding sites in every adult cell type and/or loss of binding in certain cell types. In L-2 and A549 cells, there are approximately 2280 and 1580 nuclear sites/cell, respectively. The presence of T3-binding sites as well as glucocorticoid receptors (23) in these cell lines suggests that type II alveolar cells

10'

1729 FIG. 4. Competition by various T4 analogs for [125I]T3 binding in 28-day-old fetal lung nuclei. Reaction mixtures contained 500 pM L-[ 125 I]T 3 and various concentrations of unlabeled L-T3 or other unlabeled analog. Maximal specific binding was determined in the presence and absence of 250 nM unlabeled T3, as described in Materials and Methods. Data shown are combined from two representative experiments. T2iPr, 3,5-Diiodo-3'-isopropyl-Lthyronine; 3,3'-L-T2, 3,3'-diiodoL-thyronine.

are potentially responsive in vivo to both hormones. The values in lung tissue and cells for both Kd and concentration of nuclear sites are comparable to those found previously in GHi cells (8,19), adult rat liver (6, 8-10), and other T4-responsive adult rat tissues (9). Our survey of several thyroid hormone analogs indicates that binding by fetal lung has a specificity similar to that observed with adult rat liver (19-21). It was of particular interest that both rT3 and its metabolite, 3,3'diiodo-L-thyronine, have less than 0.2% the activity of T3. These compounds also have a low affinity for nuclear receptors in adult rat liver (19-21) and in cultured rat pituitary cells, where they act as weak agonists (19). rT 3 is present in relatively high levels in the fetal circulation (24, 25) and in amniotic fluid (25, 26) compared to adult plasma, raising the possibility that this compound might be a physiologically important thyroid hormone during fetal life. However, this possibility is not supported by our finding of a low binding affinity in fetal lung. Assuming that the affinity of rT 3 is similarly low in the human, we calculate that endogenous rT 3 would occupy