EXPEFUMENTAL
CELL RESEARCH
194,111-117
(1991)
Self-Differentiation of Human Fetal Lung Organ Culture: The Role of Prostaglandins PGE:, and PGF,, ROBERT HuME,**’ RODNEYKELLY,~ DAVID~OSSAR,$ MEXETEGILES,*ANNEHALLAS,* MARGARETGOURLAY,~AND JEANNEBELLS *Department of Child Life and Health, University of Edinburgh; tM.R.C. Centre for Reproductive Biology, Edinburgh; and *Department of Pathology, University of Edinburgh
Addition of PGEs , but not PGFB,, to fetal lung organ cultures accelerates the process of self-differentiation with increased dilatation of terminal airsacs and differentiation of the epithelial lining. Indomethacin reduces the endogenous production by organ cultures of PGE, , and 13,14-dihyPGFs, , 13,14-dihydro-15-keto-PGE,, dro-15-keto-PGF, and retards the process of self-differentiation. Prolonged exposure of cultures to indomethacin results in cell necrosis. Indomethacin inhibition of self-differentiation can be reversed and accelerated by the addition of PGE,. Addition of PGFa, in the presence of indomethacin prevents indomethatin-associated cell necrosis but does not accelerate dilatation or differentiation beyond that of cultures in serafree media without additions. We propose that the endogenous production of PGE, is a key process in the mechanism of self-differentiation of human fetal lung 0 1991 Academic Press, Inc. in organ culture.
INTRODUCIION Lung primordium is a self-differentiating system in vitro [l-4] and lung cultures from both avian and mammalian embryos can, under suitable conditions, differentiate essentially all the cellular elements of mature lung [5,6]. For example, lung explants from early mouse embryos results in further branching of the bronchial tree and development of lobules but no further cellular differentiation [7]. In contrast, lung organ cultures from mid trimester human abortuses develop type II pneumatocytes and the capacity for surfactant synthesis after a few days in sera-free culture [8, 91. In addition, terminal airsacs dilate and much of the lining epithehum flattens and differentiates to type I pneumatocytes with the morphological appearance of a 27-30 week gestation lung [lo]. The control mechanism for this precocious differentiation is unknown. What is known is
that the accelerated maturation can occur in a chemically defined medium in the absence of sera, exogenous growth factors, or hormones, which suggests the operation of endogenous paracrine or autocrine regulation. Metabolites of arachidonic acid are established autocoids and in addition act as transmembrane signals for a variety of hormones and growth factors, modulating not only cell proliferation [ll] but also cell differentiation [ 121. Differentiated lung epithelial cells produce a variety of arachidonic acid products including prostaglandins, leukotrienes, and thromboxanes [13-151 with established roles in inflammation [16], ion and water secretion [ 171,control of pulmonary vasculature [ 181, and surfactant secretion [19]. Prostaglandins are produced in developing fetal lung [20] and are clearly important in the control of the pulmonary vasculature, in utero [ 181.Other functions for prostaglandins in the undifferentiated fetal distal airway have not been defined. We have recently shown that PGEz and PGF, are released by human fetal lung in organ culture, in decreasing amounts as cultures progressed. Immunohistochemical studies revealed that as progenitor cells of distal airways differentiated into type I and II pneumatocytes, they became PGF, negative but retained PGE, reactivity for up to 10 days [21]. However, these studies did not clarify whether endogenous prostaglandins had a role in maturation of lung in organ culture or whether their presence was merely coincidental with, but noncontributory to, phenotypic differentiation. In addition, if prostaglandins did modify the process of differentiation was their action an inhibitory or stimulatory one? The aims of our studies were to answer these questions and delineate the role of PGE2 and PGFzpin the development and differentiation of human lung in organ culture. MATERIAL AND METHODS Tissue samples. Both entire lungs were obtained under sterile conditions from 15 fetuses (15-18 weeks gestation) following termination of pregnancy using Gemeprost vaginal pessaries (May and Baker). A careful estimate of developmental age was made on each fetal subject based on size including crown-heel, crown-rump, and heel-toe measurements [ZZ]; menstrual history; and ultrasound dat-
1To whom reprint requestsshould be addressed at University of Edinburgh, Simpson Memorial Maternity Pavilion, Royal Infirmary, Lauriston Place, Edinburgh, Scotland, U.K. EH3 9EF. FAX: 031668 2368. 111
0014~4827/W
$3.00
Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
112
HUME
ing of pregnancy. Normality of the fetuses was confirmed by autopsy. The study was approved by the Paediatric Reproductive Ethics and Medical Research Sub-Committee of Lothian Health Board. Lung organ culture. Lung tissue was dissected free from major airways and vessels. Cubes of lung tissue, l-2 mm, were cultured at the air-medium interface supported on Millicell CM membranes (Millipore), in serum-free Glasgow modified Eagles media (MEM) and incubated at 37°C in 5% CO,:95% air. Media contained penicillin and streptomycin (both 100 III/ml). In most instances, apart from one or two randomly chosen tissue samples processed as control uncultured material, the entire lungs were established as multiple organ cultures. Cultures were maintained for up to 14 days and media were changed daily and retained for PGE, and PGF, estimations. All cultures were observed daily by phase contrast and their progress in culture was recorded by photography. In those instances where media were supplemented with prostaglandins (Sigma Chemical Co.), PGE, or PGF, were added to concentrations of 50 rig/ml. Sodium trihydrate indomethacin (Thomas Morson Pharmaceuticals) was freshly prepared daily and diluted, in selected media, to a final concentration of 5.4 X 10m6M. Tissue processing. Tissue was fixed in 10% buffered formalin and processed routinely to paraffin wax. Sections of paraffin-embedded lung were cut at 3 gm and stained with hemotoxylin and eosin. For electron microscopy small tissue fragments were fixed in 3% gluteraldehyde in cacodylate buffer and processed to araldite. Semi-thin sections were stained with toluidine blue, and thin sections stained with lead citrate and uranyl acetate were visualized in a Jeol electron microscope and assessed independently by two observers. PGE, was methyl oxiAntisera to prostaglandins and metabolites. mated, PGEz-MOX, then coupled to bovine serum albumin, and the conjugate injected intradermally into rabbits. PGE,-MOX antiserum selected for specificity and selectivity was group specific with minimal cross-reactivity between E and F series prostaglandins [23], PGF, < 0.02%, 13,14-dihydro-15-keto-PGE, < 0.02%, and 13,14dihydro-15-keto-PGF, < 0.02%. PGE,-MOX antiserum recognizes PGE, (36% cross-reactivity) and PGE, (13% cross-reactivity) and in view of this cross-reactivity would be more accurately described as a “PGE” antiserum. Antisera were raised in rabbits to 13,14-dihydro-15-keto-PGE,MOX coupled to bovine serum albumin and in a similar manner to 13,14-dihydro-15-keto-PGF,-MOX as previously described [23] with cross-reactivities between these antisera for the 15-keto metabolites of PGE, and PGF, < 0.2%. The 13,14-dihydro-15-keto-PGEz-MOX antiserum has only a 0.05% cross-reactivity with PGE, and the PGFM-MOX antiserum similarly cross-reacts < 0.1% with PGF,. PGF, was conjugated to bovine serum albumin and then injected intradermally into rabbits. PGF,-antiserum was highly group selective with minimal cross-reactivity between F and E series prostaglandins [24]. Cross reactivity of gemeprost to PGEz-MOX antiserum was 1% and to PGF,-antiserum 0.05%. Radioimmunoassay. PGE,, 13,14-dihydro-15-keto-PGE,, and 13,14-dihydro-15-keto-PGF, in culture media were derivatized to the methyl oximes and amounts measured by radioimmunoassay using corresponding PG-MOX antiserum [25]. Amounts of PGF, in culture fluids were measured by radioimmunoassay using PGF,, antiserum [24]. RESULTS
Morphology of Uncultured Lung Phase contrast microscopy of uncultured fetal lung (15-18 weeks gestation) showed nondilated terminal airways (Fig. la) confirmed by light microscopy of lung sections stained with hemotoxylin and eosin (Fig. 2a).
ET AL.
In the absence of dilated airsacs, mesenchyme was prominent occupying a substantial volume of this immature lung (Fig. 2a). Much of the mesenchyme had a loose cellular network with no obvious structural pattern except around terminal airways where tight condensations of concentrically arranged mesenchymal cells were prominent (Fig. 2a). Light and electron microscopy confirmed that the terminal airway epithelial cells were virtually monomorphic in appearance and columnar in shape with extensive glycogen deposits and an apical condensation of mitochondria and endoplasmic reticulum; results have been reported previously [21].
Effects of PGEz or PGF,, on Lung Organ Culture Lung organ cultures in MEM showed progressive uniform dilatation of the terminal airspaces (Fig. lb) with a reduction in mesenchymal volume (Fig. lb). The epithelial lining cells flattened, initially through cuboida1 forms, to give the characteristic appearance of type I pneumatocytes (Fig. 2b). In addition other epithelial cells differentiated into type II pneumatocytes containing lamellar bodies of surfactant. The progressive dilatation and maturation occurred over the first 5 days in culture with little further development thereafter. In MEM supplemented with PGE, the terminal airsacs showed accelerated dilatation (Fig. lc), reduction in mesenchymal volume, and epithelial cell maturation, compared to those organ cultures in MEM or with added PGF,, (Fig. 2~). These differences were obvious within 24-48 h of culture. Extended cultures, up to 14 days, showed good survival in PGE,-supplemented media, and progressive dilatation of terminal airsacs continued particularly at the periphery of the culture. Small blood vessels survived well in this medium and became entrapped within septae (Fig. 2~). Cultures in MEM supplemented with PGF,, (Figs. ld and 2d) were similar in appearance to those in MEM.
Effects of Indomethacin Preliminary culture experiments using a range of indomethacin concentrations, 1.3-10.8 X lo-’ M, showed maximal inhibition of prostaglandin production at 5.4 x 10e5 M, a concentration used in subsequent investigations. Indomethacin reduced not only the amounts of active prostaglandins PGE, and PGF,, released into the media from lung organ cultures but also the corresponding metabolites 13,14-dihydro-Sketo-PGE, (PGEM) and 13,14-dihydro-Sketo-PGF,, (PGFM) (Table 1). In individual cultures indomethacin reduced total measured prostanoid released (PGE, + PGF, + PGEM + PGFM) by 88 to 96%. Lung cultured in MEM with indomethacin (Fig. 3c) showed, at 24 h, that dilatation of terminal airways was markedly reduced compared to cultures in MEM alone
PGE,
AND
PGF,
IN HUMAN
FETAL
LUNG
113
FIG. 1. Contrasting appearance of human fetal lung tissue (16 weeks gestation) examined by phase contrast microscopy: (a) before culture, and after 6 days culture in (b) serum free medium; (c) medium supplemented with PGE,; and (d) medium supplemented with PGF,,. The increased clarity of Fig. lc reflects the greatly increased tissue dilatation resulting from PGE, supplementation. Magnification, X32 in all figures.
(Fig. 3b). In addition, the epithelial lining cells retained a columnar morphology in contrast to the low cuboidal epithelium of MEM cultures. Lung cultured in MEM showed good survival up to 10 days whereas after 3 days in the presence of indomethatin most cultures had some obvious areas of necrosis.
PGEz or PGF,, Supplementation Indomethacin
in the Presence of
Inhibition of dilatation and maturation of terminal airsacs by indomethacin was reversed by the addition of PGEz (Figs. 3c and3d) andwithin 24-48 h these developmental processes were accelerated when compared to sera-free cultures. Tissue survival in these conditions was good up to 10 days. Lung cultures at 24 h, in the presence of indomethatin and PGF,, had a degree of dilatation and epithelial maturation intermediate between those in sera-free cultures and those supplemented with indomethacin (Figs. 3a and 3~). By Day 5 cultures with PGF,, plus indomethacin were similar in appearance to those in MEM but never showed the massive dilatation and epithelial
thinning seen with PGE, reversal of indomethacin inhibition. Tissue survival in the presence of PGF,, and indomethacin was generally good. The amounts of prostaglandins PGE, , PGFSa, and their metabolites 24 h after addition of 50 rig/ml PGE, to cultures in MEM in the presence of indomethacin are shown in Fig. 4. PGE, wasthe predominant prostaglandin remaining and at levels 50- to 170-fold higher than those expected from endogenous production of PGE, in the absence of indomethacin. PGE, was metabolized to 13,14-dihydro-15-keto-PGE, but no significant conversion to PGF,, or its metabolite was observed. After PGF, additions, under similar conditions, PGF,, was the main prostaglandin remaining, 5- to 120-fold in excess of levels expected from endogenous production of PGF, in the absence of indomethacin (Fig. 4). PGF, was metabolized by lung organ cultures to 13,14dihydro-15-keto-PGF,,. Analysis of the media of PGF,,supplemented cultures, in the presence of indomethatin, showed increased immunoreactivity for PGEz (P -C 0.5) and PGEM (P -C 0.01) when compared to those cultures with indomethacin alone but these changes
FIG. 2. Sections of 16 week gestation human fetal lung, comparing: (a) uncultured tissue with 5 day cultures maintained in (b) serum free medium; (c) medium supplemented with PGE,; and (d) medium supplemented with PGF,. Dilatation of aiMtays takes place in all three cultures but is maximal in PGE,-supplemented culture. Haematoxylin and eosin. All sections reproduced X125. 114
PGE,
TABLE Control (ng/mg protein/24 PGE, PGF, PGEM PGFM
1.26 9.44 3.46 23.99
+ * + f
AND
PGF,
IN HUMAN
1
h)
0.71 3.31 1.82 9.70
Plus indomethacin (ng/mg protein124 h) 0.23 1.15 0.55 0.93
+ 0.10 + 0.86 + 0.33 +- 0.55
Note. Indomethacin inhibition (5.4 X lo-’ M) of PGE,, PGF,, PGEM (13,14-dihydro-15-keto-PGE,), and PGFM (13,14-dihydro15-keto-PGF,) release from fetal lung organ cultures, time 24-48 h. Results are means * SD from cultures of three 17 week gestation fetuses (each in triplicate) and expressed as amounts of prostanoid released per total protein of lung organ culture.
were insignificant when compared to levels in MEM culture media. Amounts of exogenous PGE, or PGF,, were added to allow for this expected degradation to metabolites and levels achieved of the active prostaglandins, PGE, and PGFzo, after 24 hours were in excess and higher than those expected from endogenous production in the absence of indomethacin. DISCUSSION
We have shown that when fetal lung is established in organ culture, prostaglandins PGE, and PGF,, are released in substantial amounts into the media but as cultures become older, prostaglandin levels rapidly decrease [21]. The ontogeny of cycloxygenase and other enzymes involved in prostaglandin synthesis and metabolism in human lung epithelium is unknown. These results cannot distinguish between alterations in genetic expression of enzymes involved in arachidonic acid metabolism on the one hand and, on the other, changes in activation of preexisting enzymes. However, if exogenous prostaglandins were added to maintain initial media levels, then addition of PGE, accelerated dilatation and differentiation and this was maintained for up to 14 days, in contrast to cultures in MEM alone which showed only minimal further changes after 5 days by which time endogenous PGE, levels were low. However, additions of PGF, did not accelerate, not indeed inhibit, dilatation or differentiation of terminal airways over that found in MEM cultures. Indomethacin inhibition of cycloxygenase activity was not complete even with concentrations about 50fold higher than the Ki of indomethacin with purified cycloxygenase [26]. This may in part be explained by limitations imposed on drug diffusion by the nature of organ culture, particularly as airsacs dilate. In addition, indomethacin inhibits not only cycloxygenase but other enzymes such as prostaglandin dehydrogenase [27] which may in part explain the marked inhibition of PGEM and PGFM formation. Whereas lung cultured in
FETAL
LUNG
115
MEM showed good survival for up to 14 days, most cultures in indomethacin had some obvious areas of necrosis and little dilatation took place after only 3 days. Studies of indomethacin-supplemented cultures were therefore restricted to a 24-h period, where all the tissue remained viable and retained mitotic activity, as assessed by autoradiography and by [3H]thymidine incorporation into DNA, and was similar to cultures in MEM (Hume and Bell, unpublished). Indomethacin-associated cell death was prevented by supplementation of media with either prostaglandins PGE, or PGF,, suggesting a role for prostaglandins in maintaining basic cell function in this system. Indomethacin also lowered endogenous production of PGE, and PGF,, and markedly inhibited dilatation of terminal airsacs and differentiation of epithelial lining cells, further underlining the critical role of endogenous prostaglandins in the self-differentiation process. That indomethacin-induced inhibition can not only be reversed but dilatation and differentiation accelerated by added PGE, confirms the key role of PGE, in these precocious maturational changes. PGF,, supplementation, in the presence of indomethacin, not only prevented cell necrosis in organ cultures but allowed a slow self-differentiation process which by 5 days was comparable to cultures in MEM. It is possible that PGF,, has some weak interactions with the secondary messenger systems normally modulated by PGE, and which presumably initiate the differentiation process. To what extent may the present results be explained by crossover between PGE, and PGF,, metabolic pathways? One possible route for metabolism of PGE, is conversion to PGF,, by way of a PGE, 9-keto-reductase [28]. However, our results show that PGF,, products formed on addition of PGE, to indomethacin-inhibited cultures are low and at the limits of immunoassay crossreactivity. One possible metabolic fate for PGF,, compounds may, by dehydrogenation of the Sa-hydroxyl, give rise to E-type metabolites [29]. This reaction has been demonstrated in rat and other species, however, the human does not appear to utilize this pathway [28, 301. Increased immunoreactivity for PGE, and PGEM was noted in the media of PGF, plus indomethacinsupplemented cultures. In part, this can be accounted for by some cross-reactivity but it appears that PGF,, modified in a minor way the dynamics of PGE turnover. We cannot exclude the possibility that these small increments in PGE, are the means by which slow self-differentiation occurs with PGF,, supplementation in the presence of indomethacin. Prostaglandins produced in fetal lung are important in the control of the pulmonary vasculature in utero [ 181. Vasoconstrictors, such as PGF,, , seem to be abundant early in gestation while dilators such as prostacyclin and PGE, become predominant as term approaches [ll]. PGE, and PGF,, have important and markedly
FIG. 3. Sections of 16 week gestation human fetal lung, comparing: (a) uncultured tissue with 1 day cultures maintained in (b) sexurn free medium; (c) mtedium supplemented with indomethacin; (d) medium supplemented with indomethacin and PGE,; and (e) medium supplemented with ixrdomethacin and PGF, . Indomethacin inhibits dilatation compared to that seen in serum-free culture but this inhil oition is reversed when PGE, is also added. PGF,, has a less noticeable effect in reversing indomethacin effects. Haematoxylin and ejosin. All Isections reproduced Xl 25. 116
PGE,
PGEM
PGF2aPGFM
PGF,
IN HUMAN
PGE2
PGEM
Rudnick, D. (1933) J. Ezp. Zool. 66,125. de Jong, B. J., and de Haan, J. (1943) Acta Neerl. Morphol. 6,26. Waddell, W. R. (1949) Arch. Pathol. 47,227. Chen, J. M. (1954) Exp. Cell Res. 7,518. Sorokin, S. P. (1965) in Organogenesis (Dellaan, R., and Ursprung, H., Eds.), p. 467, Holt, New York.
6. 7. 8.
Taderera, J. B. (1967) Dev. Biol. 16,489. Spooner, B. S., and Wessels, N. K. (1970) J. Exp. 2001. 175,445. Ekelund, L., Arvidson, G., and Astedt, B. (1975) J. Clin. Lab. Invest. 35, 419. Snyder, J. M., Johnston, J. M., and Mendelson, C. R. (1981) Cell Tissue Res. 220,17. Cossar, D., Bell, J., Strange, R., Jones, M., Sandison, A., and Hume, R. (1990) Biochim. Biophys. Acta 1037,221. Wickremasinghe, R. G. (1988) Prostaglundins, Leuhotrienes Essential Fatty Acids 31,171. Zalin, R. J. (1987) Exp. Cell Res. 172, 265. Smith, P. L., Welsh, M. J., Stoff, J. S., and Frizzell, R. A. (1982) J. Membr. Biol. 70,217. Xu, G. L., Sivarajah, K., Wu, R., Netteshein, P., and Eling, T. (1986) Erp. Lung Res. 10,101. Eling, T. E., Danilowicz, R. M., Henke, D. C., Sivarajak, K., Yankaskas, J. R., and Boucher, R. C. (1986) J. Biol. Chem. 261, 12841. Samuelson, B., Dahlen, S. E., Lingreen, J. A., Rouzen, C. A., and Serhan, C. N. (1987) Science 237,117l. Welsh, M. J. (1987) Phys. Rev. 67,1143. Heymann, M. A. (1987) Clin. Perinntol. 14, 857. Rooney, S. A., and Gobran, L. I. (1988) Biochim. Biophys. Actu 960,98. Pace-Asciak, C. R. (1980) Semin. Perinutol. 4,15. Hume, R. (1990) J. Endocrinol. 124,8. Scammon, R. E., and Clakins, L. A. (1929) The Development and Growth of the External Dimensions of the Human Body in the Fetal Period, Univ. of Minnesota Press, Minneapolis. Kelly, R. W., Dean, S., Cameron, M. J., and Seamark, R. F. (1986) Prostagkutdins Leuhotrienes Med. 24, 1. Kelly, R. W., Healy, D. L., Cameron, M. J., Cameron, I. T., and Baird, D. T. (1986) J. Clin. Endocrinol. Metab. 62, 1116. Kelly, R. W., Graham, B. J. M., and O’Sullivan, M. J. (1989) Prostaglandins Leuhotrienes Essential Fatty Acids 37, 187. Kulmacz, R. J., and Lands, W. E. M. (1985) J. Biol. Chem. 260, 12572.
9.
different effects on developing lung in organ culture. Ontogenic changes in the relative levels of production of these two prostaglandins may be fundamental in triggering differentiation in human fetal lung in hero. It is of interest that in human fetal myoblasts, prostaglandins act as key regulators of development and PGF,, stimulates myoblast proliferation while PGE, initiates precocious differentiation [ 12 1. Prostaglandin E, binding to membrane receptors can modify adenyl cyclase activity and cyclic AMP levels in either direction, depending on the nature of the regulatory proteins coupling receptor and cyclase [31]. Prostaglandins of the E-type stimulate lung epithelial cells to increase intracellular levels of cyclic AMP in mature [32] and fetal lung [33]. In fetal lung, B-adrenergic agonists activate adenyl cyclase and increase cyclic AMP levels which in turn stimulates phosphatidylcholine secretion by type II pneumatocytes [34]. Cyclic AMP analogues induce the synthesis of the major surfactant apoprotein in human fetal lung in vitro and in addition accelerate the dilatation of terminal airsacs and the appearance of type II pneumatocytes [35]. We propose that, in mid trimester human fetal lung in organ culture, endogenous production of prostaglandin E, may activate the production of cyclic AMP which then leads to the process of self-differentiation of the terminal airway epithelium. We thank Birthright-Royal College of Obstetricians for financial support, Martin Hooper for advice and encouragement, and Audrey Boner for typing the manuscript. D.C. thanks the Institute of Medical Laboratory Science for permission to publish data from his thesis. Received October 19,199O Revised version received December
27,199O
117
LUNG
1. 2. 3. 4. 5.
PGFza PGFM
FIG. 4. Ratio of prostanoids in culture media with added PGE, or PGF, compared to those in MEM. In the example, lung from a 17 week gestation fetus was established at culture time 24 h in (a) MEM, (b) MEM + indomethacin (I) 5.4 X lo-‘M, (c) MEM + indomethacin + PGE, 50 rig/ml, and (d) MEM + indomethacin + PGF, 50 rig/ml. After 24 h, the media were assayed for PGE,, PGF,,, 13,14dihydro15-keto-PGE, (PGEM), and 13,14-dihydro-15-keto-PGF, (PGFM). Ratios for individual prostanoids were calculated relative to those in MEM: q indomethacin alone; n indomethacin plus PGE, or PGF,. The ratio value of unity for MEM is represented by a broken line. A similar pattern of prostanoid ratios was seen in cultures from three other fetuses.
FETAL
REFERENCES
Plus PGF2a
Plus PGE;,
PGE2
AND
10. 11. 12. 13. 14. 15.
16. 17. 18. 19. 20. 21. 22.
23. 24. 25. 26. 27. 28. 29. 30.
Hansen, H. S. (1974) Prostuglandins 8,95. Samuelsson, B., Granstrom, E., Green, K., Hamberg, M., and Hammarstrom, S. (1975) Annu. Rev. B&hem. 44,669. Pace-Asciak, C., and Miller, D. (1974) Experientia 30,590. Granstrom, E., and Samuelsson, B. (1971) J. Biol. &em. 243, 5254.
31.
Kirtland, S. J. (1988) Prostagkzndins, Fatty Acids 32,165.
32. 33. 34. 35.
Butcher, R. W., and Baird, C. E. (1968) J. Biol. Chem. 243,1713. Powell, W. S., and Solomon, S. (1980) Endocrinology 107,1469. Dobbs, L. G., and Mason, R. J. (1979) J. Clin. Invest. 63,378. Sodom, M. J., Snyder, J. M., and Mendelson, C. R. (1987) Endocrinology 121,1155.
Leuhotrienes
Essential
CELL RESEARCH
194,111-117
(1991)
Self-Differentiation of Human Fetal Lung Organ Culture: The Role of Prostaglandins PGE:, and PGF,, ROBERT HuME,**’ RODNEYKELLY,~ DAVID~OSSAR,$ MEXETEGILES,*ANNEHALLAS,* MARGARETGOURLAY,~AND JEANNEBELLS *Department of Child Life and Health, University of Edinburgh; tM.R.C. Centre for Reproductive Biology, Edinburgh; and *Department of Pathology, University of Edinburgh
Addition of PGEs , but not PGFB,, to fetal lung organ cultures accelerates the process of self-differentiation with increased dilatation of terminal airsacs and differentiation of the epithelial lining. Indomethacin reduces the endogenous production by organ cultures of PGE, , and 13,14-dihyPGFs, , 13,14-dihydro-15-keto-PGE,, dro-15-keto-PGF, and retards the process of self-differentiation. Prolonged exposure of cultures to indomethacin results in cell necrosis. Indomethacin inhibition of self-differentiation can be reversed and accelerated by the addition of PGE,. Addition of PGFa, in the presence of indomethacin prevents indomethatin-associated cell necrosis but does not accelerate dilatation or differentiation beyond that of cultures in serafree media without additions. We propose that the endogenous production of PGE, is a key process in the mechanism of self-differentiation of human fetal lung 0 1991 Academic Press, Inc. in organ culture.
INTRODUCIION Lung primordium is a self-differentiating system in vitro [l-4] and lung cultures from both avian and mammalian embryos can, under suitable conditions, differentiate essentially all the cellular elements of mature lung [5,6]. For example, lung explants from early mouse embryos results in further branching of the bronchial tree and development of lobules but no further cellular differentiation [7]. In contrast, lung organ cultures from mid trimester human abortuses develop type II pneumatocytes and the capacity for surfactant synthesis after a few days in sera-free culture [8, 91. In addition, terminal airsacs dilate and much of the lining epithehum flattens and differentiates to type I pneumatocytes with the morphological appearance of a 27-30 week gestation lung [lo]. The control mechanism for this precocious differentiation is unknown. What is known is
that the accelerated maturation can occur in a chemically defined medium in the absence of sera, exogenous growth factors, or hormones, which suggests the operation of endogenous paracrine or autocrine regulation. Metabolites of arachidonic acid are established autocoids and in addition act as transmembrane signals for a variety of hormones and growth factors, modulating not only cell proliferation [ll] but also cell differentiation [ 121. Differentiated lung epithelial cells produce a variety of arachidonic acid products including prostaglandins, leukotrienes, and thromboxanes [13-151 with established roles in inflammation [16], ion and water secretion [ 171,control of pulmonary vasculature [ 181, and surfactant secretion [19]. Prostaglandins are produced in developing fetal lung [20] and are clearly important in the control of the pulmonary vasculature, in utero [ 181.Other functions for prostaglandins in the undifferentiated fetal distal airway have not been defined. We have recently shown that PGEz and PGF, are released by human fetal lung in organ culture, in decreasing amounts as cultures progressed. Immunohistochemical studies revealed that as progenitor cells of distal airways differentiated into type I and II pneumatocytes, they became PGF, negative but retained PGE, reactivity for up to 10 days [21]. However, these studies did not clarify whether endogenous prostaglandins had a role in maturation of lung in organ culture or whether their presence was merely coincidental with, but noncontributory to, phenotypic differentiation. In addition, if prostaglandins did modify the process of differentiation was their action an inhibitory or stimulatory one? The aims of our studies were to answer these questions and delineate the role of PGE2 and PGFzpin the development and differentiation of human lung in organ culture. MATERIAL AND METHODS Tissue samples. Both entire lungs were obtained under sterile conditions from 15 fetuses (15-18 weeks gestation) following termination of pregnancy using Gemeprost vaginal pessaries (May and Baker). A careful estimate of developmental age was made on each fetal subject based on size including crown-heel, crown-rump, and heel-toe measurements [ZZ]; menstrual history; and ultrasound dat-
1To whom reprint requestsshould be addressed at University of Edinburgh, Simpson Memorial Maternity Pavilion, Royal Infirmary, Lauriston Place, Edinburgh, Scotland, U.K. EH3 9EF. FAX: 031668 2368. 111
0014~4827/W
$3.00
Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
112
HUME
ing of pregnancy. Normality of the fetuses was confirmed by autopsy. The study was approved by the Paediatric Reproductive Ethics and Medical Research Sub-Committee of Lothian Health Board. Lung organ culture. Lung tissue was dissected free from major airways and vessels. Cubes of lung tissue, l-2 mm, were cultured at the air-medium interface supported on Millicell CM membranes (Millipore), in serum-free Glasgow modified Eagles media (MEM) and incubated at 37°C in 5% CO,:95% air. Media contained penicillin and streptomycin (both 100 III/ml). In most instances, apart from one or two randomly chosen tissue samples processed as control uncultured material, the entire lungs were established as multiple organ cultures. Cultures were maintained for up to 14 days and media were changed daily and retained for PGE, and PGF, estimations. All cultures were observed daily by phase contrast and their progress in culture was recorded by photography. In those instances where media were supplemented with prostaglandins (Sigma Chemical Co.), PGE, or PGF, were added to concentrations of 50 rig/ml. Sodium trihydrate indomethacin (Thomas Morson Pharmaceuticals) was freshly prepared daily and diluted, in selected media, to a final concentration of 5.4 X 10m6M. Tissue processing. Tissue was fixed in 10% buffered formalin and processed routinely to paraffin wax. Sections of paraffin-embedded lung were cut at 3 gm and stained with hemotoxylin and eosin. For electron microscopy small tissue fragments were fixed in 3% gluteraldehyde in cacodylate buffer and processed to araldite. Semi-thin sections were stained with toluidine blue, and thin sections stained with lead citrate and uranyl acetate were visualized in a Jeol electron microscope and assessed independently by two observers. PGE, was methyl oxiAntisera to prostaglandins and metabolites. mated, PGEz-MOX, then coupled to bovine serum albumin, and the conjugate injected intradermally into rabbits. PGE,-MOX antiserum selected for specificity and selectivity was group specific with minimal cross-reactivity between E and F series prostaglandins [23], PGF, < 0.02%, 13,14-dihydro-15-keto-PGE, < 0.02%, and 13,14dihydro-15-keto-PGF, < 0.02%. PGE,-MOX antiserum recognizes PGE, (36% cross-reactivity) and PGE, (13% cross-reactivity) and in view of this cross-reactivity would be more accurately described as a “PGE” antiserum. Antisera were raised in rabbits to 13,14-dihydro-15-keto-PGE,MOX coupled to bovine serum albumin and in a similar manner to 13,14-dihydro-15-keto-PGF,-MOX as previously described [23] with cross-reactivities between these antisera for the 15-keto metabolites of PGE, and PGF, < 0.2%. The 13,14-dihydro-15-keto-PGEz-MOX antiserum has only a 0.05% cross-reactivity with PGE, and the PGFM-MOX antiserum similarly cross-reacts < 0.1% with PGF,. PGF, was conjugated to bovine serum albumin and then injected intradermally into rabbits. PGF,-antiserum was highly group selective with minimal cross-reactivity between F and E series prostaglandins [24]. Cross reactivity of gemeprost to PGEz-MOX antiserum was 1% and to PGF,-antiserum 0.05%. Radioimmunoassay. PGE,, 13,14-dihydro-15-keto-PGE,, and 13,14-dihydro-15-keto-PGF, in culture media were derivatized to the methyl oximes and amounts measured by radioimmunoassay using corresponding PG-MOX antiserum [25]. Amounts of PGF, in culture fluids were measured by radioimmunoassay using PGF,, antiserum [24]. RESULTS
Morphology of Uncultured Lung Phase contrast microscopy of uncultured fetal lung (15-18 weeks gestation) showed nondilated terminal airways (Fig. la) confirmed by light microscopy of lung sections stained with hemotoxylin and eosin (Fig. 2a).
ET AL.
In the absence of dilated airsacs, mesenchyme was prominent occupying a substantial volume of this immature lung (Fig. 2a). Much of the mesenchyme had a loose cellular network with no obvious structural pattern except around terminal airways where tight condensations of concentrically arranged mesenchymal cells were prominent (Fig. 2a). Light and electron microscopy confirmed that the terminal airway epithelial cells were virtually monomorphic in appearance and columnar in shape with extensive glycogen deposits and an apical condensation of mitochondria and endoplasmic reticulum; results have been reported previously [21].
Effects of PGEz or PGF,, on Lung Organ Culture Lung organ cultures in MEM showed progressive uniform dilatation of the terminal airspaces (Fig. lb) with a reduction in mesenchymal volume (Fig. lb). The epithelial lining cells flattened, initially through cuboida1 forms, to give the characteristic appearance of type I pneumatocytes (Fig. 2b). In addition other epithelial cells differentiated into type II pneumatocytes containing lamellar bodies of surfactant. The progressive dilatation and maturation occurred over the first 5 days in culture with little further development thereafter. In MEM supplemented with PGE, the terminal airsacs showed accelerated dilatation (Fig. lc), reduction in mesenchymal volume, and epithelial cell maturation, compared to those organ cultures in MEM or with added PGF,, (Fig. 2~). These differences were obvious within 24-48 h of culture. Extended cultures, up to 14 days, showed good survival in PGE,-supplemented media, and progressive dilatation of terminal airsacs continued particularly at the periphery of the culture. Small blood vessels survived well in this medium and became entrapped within septae (Fig. 2~). Cultures in MEM supplemented with PGF,, (Figs. ld and 2d) were similar in appearance to those in MEM.
Effects of Indomethacin Preliminary culture experiments using a range of indomethacin concentrations, 1.3-10.8 X lo-’ M, showed maximal inhibition of prostaglandin production at 5.4 x 10e5 M, a concentration used in subsequent investigations. Indomethacin reduced not only the amounts of active prostaglandins PGE, and PGF,, released into the media from lung organ cultures but also the corresponding metabolites 13,14-dihydro-Sketo-PGE, (PGEM) and 13,14-dihydro-Sketo-PGF,, (PGFM) (Table 1). In individual cultures indomethacin reduced total measured prostanoid released (PGE, + PGF, + PGEM + PGFM) by 88 to 96%. Lung cultured in MEM with indomethacin (Fig. 3c) showed, at 24 h, that dilatation of terminal airways was markedly reduced compared to cultures in MEM alone
PGE,
AND
PGF,
IN HUMAN
FETAL
LUNG
113
FIG. 1. Contrasting appearance of human fetal lung tissue (16 weeks gestation) examined by phase contrast microscopy: (a) before culture, and after 6 days culture in (b) serum free medium; (c) medium supplemented with PGE,; and (d) medium supplemented with PGF,,. The increased clarity of Fig. lc reflects the greatly increased tissue dilatation resulting from PGE, supplementation. Magnification, X32 in all figures.
(Fig. 3b). In addition, the epithelial lining cells retained a columnar morphology in contrast to the low cuboidal epithelium of MEM cultures. Lung cultured in MEM showed good survival up to 10 days whereas after 3 days in the presence of indomethatin most cultures had some obvious areas of necrosis.
PGEz or PGF,, Supplementation Indomethacin
in the Presence of
Inhibition of dilatation and maturation of terminal airsacs by indomethacin was reversed by the addition of PGEz (Figs. 3c and3d) andwithin 24-48 h these developmental processes were accelerated when compared to sera-free cultures. Tissue survival in these conditions was good up to 10 days. Lung cultures at 24 h, in the presence of indomethatin and PGF,, had a degree of dilatation and epithelial maturation intermediate between those in sera-free cultures and those supplemented with indomethacin (Figs. 3a and 3~). By Day 5 cultures with PGF,, plus indomethacin were similar in appearance to those in MEM but never showed the massive dilatation and epithelial
thinning seen with PGE, reversal of indomethacin inhibition. Tissue survival in the presence of PGF,, and indomethacin was generally good. The amounts of prostaglandins PGE, , PGFSa, and their metabolites 24 h after addition of 50 rig/ml PGE, to cultures in MEM in the presence of indomethacin are shown in Fig. 4. PGE, wasthe predominant prostaglandin remaining and at levels 50- to 170-fold higher than those expected from endogenous production of PGE, in the absence of indomethacin. PGE, was metabolized to 13,14-dihydro-15-keto-PGE, but no significant conversion to PGF,, or its metabolite was observed. After PGF, additions, under similar conditions, PGF,, was the main prostaglandin remaining, 5- to 120-fold in excess of levels expected from endogenous production of PGF, in the absence of indomethacin (Fig. 4). PGF, was metabolized by lung organ cultures to 13,14dihydro-15-keto-PGF,,. Analysis of the media of PGF,,supplemented cultures, in the presence of indomethatin, showed increased immunoreactivity for PGEz (P -C 0.5) and PGEM (P -C 0.01) when compared to those cultures with indomethacin alone but these changes
FIG. 2. Sections of 16 week gestation human fetal lung, comparing: (a) uncultured tissue with 5 day cultures maintained in (b) serum free medium; (c) medium supplemented with PGE,; and (d) medium supplemented with PGF,. Dilatation of aiMtays takes place in all three cultures but is maximal in PGE,-supplemented culture. Haematoxylin and eosin. All sections reproduced X125. 114
PGE,
TABLE Control (ng/mg protein/24 PGE, PGF, PGEM PGFM
1.26 9.44 3.46 23.99
+ * + f
AND
PGF,
IN HUMAN
1
h)
0.71 3.31 1.82 9.70
Plus indomethacin (ng/mg protein124 h) 0.23 1.15 0.55 0.93
+ 0.10 + 0.86 + 0.33 +- 0.55
Note. Indomethacin inhibition (5.4 X lo-’ M) of PGE,, PGF,, PGEM (13,14-dihydro-15-keto-PGE,), and PGFM (13,14-dihydro15-keto-PGF,) release from fetal lung organ cultures, time 24-48 h. Results are means * SD from cultures of three 17 week gestation fetuses (each in triplicate) and expressed as amounts of prostanoid released per total protein of lung organ culture.
were insignificant when compared to levels in MEM culture media. Amounts of exogenous PGE, or PGF,, were added to allow for this expected degradation to metabolites and levels achieved of the active prostaglandins, PGE, and PGFzo, after 24 hours were in excess and higher than those expected from endogenous production in the absence of indomethacin. DISCUSSION
We have shown that when fetal lung is established in organ culture, prostaglandins PGE, and PGF,, are released in substantial amounts into the media but as cultures become older, prostaglandin levels rapidly decrease [21]. The ontogeny of cycloxygenase and other enzymes involved in prostaglandin synthesis and metabolism in human lung epithelium is unknown. These results cannot distinguish between alterations in genetic expression of enzymes involved in arachidonic acid metabolism on the one hand and, on the other, changes in activation of preexisting enzymes. However, if exogenous prostaglandins were added to maintain initial media levels, then addition of PGE, accelerated dilatation and differentiation and this was maintained for up to 14 days, in contrast to cultures in MEM alone which showed only minimal further changes after 5 days by which time endogenous PGE, levels were low. However, additions of PGF, did not accelerate, not indeed inhibit, dilatation or differentiation of terminal airways over that found in MEM cultures. Indomethacin inhibition of cycloxygenase activity was not complete even with concentrations about 50fold higher than the Ki of indomethacin with purified cycloxygenase [26]. This may in part be explained by limitations imposed on drug diffusion by the nature of organ culture, particularly as airsacs dilate. In addition, indomethacin inhibits not only cycloxygenase but other enzymes such as prostaglandin dehydrogenase [27] which may in part explain the marked inhibition of PGEM and PGFM formation. Whereas lung cultured in
FETAL
LUNG
115
MEM showed good survival for up to 14 days, most cultures in indomethacin had some obvious areas of necrosis and little dilatation took place after only 3 days. Studies of indomethacin-supplemented cultures were therefore restricted to a 24-h period, where all the tissue remained viable and retained mitotic activity, as assessed by autoradiography and by [3H]thymidine incorporation into DNA, and was similar to cultures in MEM (Hume and Bell, unpublished). Indomethacin-associated cell death was prevented by supplementation of media with either prostaglandins PGE, or PGF,, suggesting a role for prostaglandins in maintaining basic cell function in this system. Indomethacin also lowered endogenous production of PGE, and PGF,, and markedly inhibited dilatation of terminal airsacs and differentiation of epithelial lining cells, further underlining the critical role of endogenous prostaglandins in the self-differentiation process. That indomethacin-induced inhibition can not only be reversed but dilatation and differentiation accelerated by added PGE, confirms the key role of PGE, in these precocious maturational changes. PGF,, supplementation, in the presence of indomethacin, not only prevented cell necrosis in organ cultures but allowed a slow self-differentiation process which by 5 days was comparable to cultures in MEM. It is possible that PGF,, has some weak interactions with the secondary messenger systems normally modulated by PGE, and which presumably initiate the differentiation process. To what extent may the present results be explained by crossover between PGE, and PGF,, metabolic pathways? One possible route for metabolism of PGE, is conversion to PGF,, by way of a PGE, 9-keto-reductase [28]. However, our results show that PGF,, products formed on addition of PGE, to indomethacin-inhibited cultures are low and at the limits of immunoassay crossreactivity. One possible metabolic fate for PGF,, compounds may, by dehydrogenation of the Sa-hydroxyl, give rise to E-type metabolites [29]. This reaction has been demonstrated in rat and other species, however, the human does not appear to utilize this pathway [28, 301. Increased immunoreactivity for PGE, and PGEM was noted in the media of PGF, plus indomethacinsupplemented cultures. In part, this can be accounted for by some cross-reactivity but it appears that PGF,, modified in a minor way the dynamics of PGE turnover. We cannot exclude the possibility that these small increments in PGE, are the means by which slow self-differentiation occurs with PGF,, supplementation in the presence of indomethacin. Prostaglandins produced in fetal lung are important in the control of the pulmonary vasculature in utero [ 181. Vasoconstrictors, such as PGF,, , seem to be abundant early in gestation while dilators such as prostacyclin and PGE, become predominant as term approaches [ll]. PGE, and PGF,, have important and markedly
FIG. 3. Sections of 16 week gestation human fetal lung, comparing: (a) uncultured tissue with 1 day cultures maintained in (b) sexurn free medium; (c) mtedium supplemented with indomethacin; (d) medium supplemented with indomethacin and PGE,; and (e) medium supplemented with ixrdomethacin and PGF, . Indomethacin inhibits dilatation compared to that seen in serum-free culture but this inhil oition is reversed when PGE, is also added. PGF,, has a less noticeable effect in reversing indomethacin effects. Haematoxylin and ejosin. All Isections reproduced Xl 25. 116
PGE,
PGEM
PGF2aPGFM
PGF,
IN HUMAN
PGE2
PGEM
Rudnick, D. (1933) J. Ezp. Zool. 66,125. de Jong, B. J., and de Haan, J. (1943) Acta Neerl. Morphol. 6,26. Waddell, W. R. (1949) Arch. Pathol. 47,227. Chen, J. M. (1954) Exp. Cell Res. 7,518. Sorokin, S. P. (1965) in Organogenesis (Dellaan, R., and Ursprung, H., Eds.), p. 467, Holt, New York.
6. 7. 8.
Taderera, J. B. (1967) Dev. Biol. 16,489. Spooner, B. S., and Wessels, N. K. (1970) J. Exp. 2001. 175,445. Ekelund, L., Arvidson, G., and Astedt, B. (1975) J. Clin. Lab. Invest. 35, 419. Snyder, J. M., Johnston, J. M., and Mendelson, C. R. (1981) Cell Tissue Res. 220,17. Cossar, D., Bell, J., Strange, R., Jones, M., Sandison, A., and Hume, R. (1990) Biochim. Biophys. Acta 1037,221. Wickremasinghe, R. G. (1988) Prostaglundins, Leuhotrienes Essential Fatty Acids 31,171. Zalin, R. J. (1987) Exp. Cell Res. 172, 265. Smith, P. L., Welsh, M. J., Stoff, J. S., and Frizzell, R. A. (1982) J. Membr. Biol. 70,217. Xu, G. L., Sivarajah, K., Wu, R., Netteshein, P., and Eling, T. (1986) Erp. Lung Res. 10,101. Eling, T. E., Danilowicz, R. M., Henke, D. C., Sivarajak, K., Yankaskas, J. R., and Boucher, R. C. (1986) J. Biol. Chem. 261, 12841. Samuelson, B., Dahlen, S. E., Lingreen, J. A., Rouzen, C. A., and Serhan, C. N. (1987) Science 237,117l. Welsh, M. J. (1987) Phys. Rev. 67,1143. Heymann, M. A. (1987) Clin. Perinntol. 14, 857. Rooney, S. A., and Gobran, L. I. (1988) Biochim. Biophys. Actu 960,98. Pace-Asciak, C. R. (1980) Semin. Perinutol. 4,15. Hume, R. (1990) J. Endocrinol. 124,8. Scammon, R. E., and Clakins, L. A. (1929) The Development and Growth of the External Dimensions of the Human Body in the Fetal Period, Univ. of Minnesota Press, Minneapolis. Kelly, R. W., Dean, S., Cameron, M. J., and Seamark, R. F. (1986) Prostagkutdins Leuhotrienes Med. 24, 1. Kelly, R. W., Healy, D. L., Cameron, M. J., Cameron, I. T., and Baird, D. T. (1986) J. Clin. Endocrinol. Metab. 62, 1116. Kelly, R. W., Graham, B. J. M., and O’Sullivan, M. J. (1989) Prostaglandins Leuhotrienes Essential Fatty Acids 37, 187. Kulmacz, R. J., and Lands, W. E. M. (1985) J. Biol. Chem. 260, 12572.
9.
different effects on developing lung in organ culture. Ontogenic changes in the relative levels of production of these two prostaglandins may be fundamental in triggering differentiation in human fetal lung in hero. It is of interest that in human fetal myoblasts, prostaglandins act as key regulators of development and PGF,, stimulates myoblast proliferation while PGE, initiates precocious differentiation [ 12 1. Prostaglandin E, binding to membrane receptors can modify adenyl cyclase activity and cyclic AMP levels in either direction, depending on the nature of the regulatory proteins coupling receptor and cyclase [31]. Prostaglandins of the E-type stimulate lung epithelial cells to increase intracellular levels of cyclic AMP in mature [32] and fetal lung [33]. In fetal lung, B-adrenergic agonists activate adenyl cyclase and increase cyclic AMP levels which in turn stimulates phosphatidylcholine secretion by type II pneumatocytes [34]. Cyclic AMP analogues induce the synthesis of the major surfactant apoprotein in human fetal lung in vitro and in addition accelerate the dilatation of terminal airsacs and the appearance of type II pneumatocytes [35]. We propose that, in mid trimester human fetal lung in organ culture, endogenous production of prostaglandin E, may activate the production of cyclic AMP which then leads to the process of self-differentiation of the terminal airway epithelium. We thank Birthright-Royal College of Obstetricians for financial support, Martin Hooper for advice and encouragement, and Audrey Boner for typing the manuscript. D.C. thanks the Institute of Medical Laboratory Science for permission to publish data from his thesis. Received October 19,199O Revised version received December
27,199O
117
LUNG
1. 2. 3. 4. 5.
PGFza PGFM
FIG. 4. Ratio of prostanoids in culture media with added PGE, or PGF, compared to those in MEM. In the example, lung from a 17 week gestation fetus was established at culture time 24 h in (a) MEM, (b) MEM + indomethacin (I) 5.4 X lo-‘M, (c) MEM + indomethacin + PGE, 50 rig/ml, and (d) MEM + indomethacin + PGF, 50 rig/ml. After 24 h, the media were assayed for PGE,, PGF,,, 13,14dihydro15-keto-PGE, (PGEM), and 13,14-dihydro-15-keto-PGF, (PGFM). Ratios for individual prostanoids were calculated relative to those in MEM: q indomethacin alone; n indomethacin plus PGE, or PGF,. The ratio value of unity for MEM is represented by a broken line. A similar pattern of prostanoid ratios was seen in cultures from three other fetuses.
FETAL
REFERENCES
Plus PGF2a
Plus PGE;,
PGE2
AND
10. 11. 12. 13. 14. 15.
16. 17. 18. 19. 20. 21. 22.
23. 24. 25. 26. 27. 28. 29. 30.
Hansen, H. S. (1974) Prostuglandins 8,95. Samuelsson, B., Granstrom, E., Green, K., Hamberg, M., and Hammarstrom, S. (1975) Annu. Rev. B&hem. 44,669. Pace-Asciak, C., and Miller, D. (1974) Experientia 30,590. Granstrom, E., and Samuelsson, B. (1971) J. Biol. &em. 243, 5254.
31.
Kirtland, S. J. (1988) Prostagkzndins, Fatty Acids 32,165.
32. 33. 34. 35.
Butcher, R. W., and Baird, C. E. (1968) J. Biol. Chem. 243,1713. Powell, W. S., and Solomon, S. (1980) Endocrinology 107,1469. Dobbs, L. G., and Mason, R. J. (1979) J. Clin. Invest. 63,378. Sodom, M. J., Snyder, J. M., and Mendelson, C. R. (1987) Endocrinology 121,1155.
Leuhotrienes
Essential
