Biocllimic¢ et BiophyslcaAcre, 1134( 1~9"2l105-I I 1
I 0~;
© 1992 ElsevierScience PublishersB.V. All righls reserved 0167-4889/92/$05.00
BBAMCR 1311S
Effect of hypoxia on endothelin-1 production by pulmonary vascular endothelial cells J e n n i f e r L. W i e b k e , C h a h r z a d M o n t r o s e - R a f i z a d e h , a n d W i l l i a m B. G u g g i n o
P a m e l a L. Z e i t l i n
The Del~rlments of Pediettric$ertd PtlyJiology, The 1ohtt~ Hopkins Medicallnstitutior~ Balttreumr.MD (USA~
(Received21]August 199D (RevisedmanuscrilRreceived 19 November 1991l Key words: Endothe~iin-I;l-lypoxia Endolhelin-I (ET-I), a pcptide product of endolhelial cons, is mRogeaie for fibmblasts and smooth muscle cells. In this study we examined the effect of hypoxia on ET-I production by bovlnc pulmonary vascular endothelial cells. Boy:me pofmona~ artery (BPAE) and microvascular endothelial (BMVE) cells were isolated, grown in tissue culture, and characterized by tl~ presence of Factor VIII related antigen and LDL uptake. Baseline production of ET-I by BPAE cells (measured by radioimmunoass~) increased over time. BMVE cells produced one tenth the amount of ET-I as produced by the pulm~aarj artery endothelial cells under the same conditions. In both cell types, hypoxia (0% 0 2) signifmandy reduced the amount of ET-! at 48 h. Restcvration of normosia in 21% O 2 for 48 h resulted in a term, of ET-I levels to baseline. Northern blot analysis showed decreased ET-I mRNA in cells eXlmSed to h~ao~aa for 48 h. These data demonstrate that pulmonaq~ vascular endotlmlial cells respond to hypoxia by reversibly decreasing ET-I production, and 1his altcnuation is likely regulated at the level of n'anseril~km.
lntrtaluctiea Chronic hypoxia leads to proliferation of pulmonary vascular smooth muscle cells and flbroblasts in the pulmonary circulation [1], In response to hypoxia the vessel wall demonstrates thickening of the media and adventitia secondary to smooth muscle cell hypertrophy and differentiation of non-contractile precursor cells. There is also accumulation of elastin and collagen fibers in the vessel wall. These changes are most prominent at the pro.capillary level in the distal p u b monary vaseulatutc [2], The signal linking Iwpo~a to this process of pulmonary vascular remodeling is unknown. A candidate for a role in this process is the endu~helial cell, once thought to have only barrier furtetion, but now rgcognized to play an integral part in many vascular responses. The endothelial cell is capable of not only metabolizing certain vasoactive substances such as angiotensin [3]. but has been found to synthesize a number of vasoactive substances as well [4,5].
Eud~theiin-I (lET-l) is a product of endotbeUal cells. It is a 21 amino acid peptide first described as vasoconstrictive in action [6]. ET-I has also been shown m be mitogonic for smooth muscle cells and fibroblasts in culture. This peptide causes an inc~,ase in protooncogene expression and cell proliferation in smooth muscle cells, as well as an inorease in [3Hlthymidine uptake in target cells ['7], contdstent with mitogenic activity. ET-I also stimulates the turnovec of phosphatidylinositol and alkalinizatioR of the cytosol [8], both of which are associated with cell proliferation. In this study we investigated cultured lung endothelial cells from the pulmonary artery and the pulanonaqw microcircelatiou to determine baseline ET-1 production by cells derived from these different sites along th~ pulmonary vascular bed. The effect o f hypoaia on ET-I production by the pulmonary vascular endothelial cells in culture was investigated by measuring ET-1 peptide and m R N A from cells exposed to hypoxia.
Materials and Metlmds Cell isolation and culture
Correspondence: W.IL GuHiao, Depa~ment of Physiology, The Johns HopkinsUniversity,Schoolof Medicine,Baltinmre,MD 21205,
USA.
Cultures of bovine pulmonary artery endothelial cells and pulmonary micr(wascular endothelial cells derived from the same animal were used in this study. In initial experiments for which large numbers of cells were
106 required, we used an established line of bovine pulmonary artery cells (CCL-209, American Type Tissue Collection, gockvillc, M D ) w h o s e endothelial origin has been confirmed by the presence of angiotensinconverting enzyme. Cell isolation. Bovine pulmonary artery endothelial (BPAE) cells were isolated as described [9]. Pulmonary arteries were trimmed of excess fat and further procussed under sterile conditions. Vessels were washed three times in Ringer's sotution containing 3 × antibiotics (I × antibiotics consists of penicillin, 100 p.I/ml and streptomycin, 100 /.tg/ml; Gilx:o, Grand Island, NY). Vessels were then cut open to expose the endothelial surface which was gently scraped with a #10 scalpel blade. Cells were centrifuged in DME (Gibco) with 3 × antibiotics at 250Xg for 10 rain at 4°C, resuspcnded in growth medium (DME with 20% FC~ and 1 × antibiotics) and placed in 100 mm culture dishes at 37°C in a 95% air,/5% CO 2 incubator. To obtain pure endothelial cell lines, the primary culture plates were evaluated daily. Colonies of 50-I00 endnthelial cells were identified by their cuboidal, cobblestone appearance using phase microscopy and then isolated on the plate using glass cloning cylinders (Belice, Viueland, N J). The select~cd ceils were trypsinized with 0.C,5% trypsin with 1% polyvinylpyrrolidone and 0.02% EGTA in Hepes-buffered saline (PET; Biofluids, Roclccille, MD). Ceils were then transferred to 24 well plates from which they were serially passaged. Bovine pulmonary microvascular endothelial (BMVE) cells were obtained using collagenase digestion of peripheral lung as previously described [10l. Peripheral lung was dissected from the pleura to avoid contamination with mesothelial cells. The lung tissue was finely minced irk a solution of Ringer's with 3 × antibiotics. After h.lsing the minced lung three times with Ringer's, the tissue was incubated with 0.1% collagenase (Type IA, Sigma, St. Louis, MO) for 20 rain at 37~C. The dissociated cells were centrifuged at 250 × g for 5 rain at ¢°C. The cell pellet was resnsponded in growth medium and placed in 100 nltm culture dishes. To obtain pure microvaseniar endothelial cell lines further isolation was performed as described above. Cell cullure. Cells were passed when confluent using 0.05% trypsin (Biofluids). Experiments were performed on cells at passages 6-10 and on the CCL-209 endothelial cell line at passages 18-22. AI[ cells were maintained in growth medium containing DME supplemented with 20% FCS (Inovar Biologics, Gaithersburg, MD) and 1 x a,:tibiotics. Cell characterizatiog The BPAE and BMVE cell lines were demonstrated to meet two criteria for eudnthelia[ cells: the presence of Factor VIII antigen detected by primary antibody to Factor VIII (Dace, Carpinteria, CA) complexed with FITC-conjugated lgG
(Dace) and the uptake of rhodamine labeled acetylated low density [ipoprot,~in (LDL) (Biomedical Technologies, Stoughton, MA) following a 4 h incubation.
Normoxic and hypoxic exposures Endothelial monolayers in 35 mm tissue culture dishes w e n rinsed three times in serum free medium. Serum free medium (1 ml) was then added to each dish. Normo~ic exposures were accomplished in a standard 5% CO 2 incubator. Hypoxie exposures were performed in an airtight chamber (Billups Rothenberg, Del Mar, CA) gassed with a mixture of 95% N2/5% CO 2 for 1,5 rain. Oxygen concentration in the chamber was monitored with an oxygen microelectrode (Microelectrodes, Londonderry, NH). The chamber was then placed in a 37°C warm room for 2 to 48 h. For hypoxic exposures of 48 h, the chamber was regas~d at 24 h. At the end of each hypexic exposure the percent O 2 in the chamber was < 0.05%. The Pc2 of the media of cells incubated in hypoxic conditions was 20-30 mmHg at all time points compared to a Pc2 of 120-130 mmHg in the media of cells incubated in normoxia. In experiments performed on BPAE and BMVE cells the expo. sures were modified by the addition of serum to the medium. ET-I was not detectable in serum-containing medium alone. In a series of experiments, the BPAE cells were restored to normoxia following the initial 48 h hypo~dc exposure and ET-1 production was measured.
Measurement of endothelin-I At the end of each KVpox[c or normoxie exposure media were aspirated from culture plates and volume of the media determined. The media ware centrifuged to remove any cell debris and supernatants were frozen at - 2WC until further analysis. Cells were assessed for viability by examination of morphology, adherent cell counts and Trypan blue exclusion. The cells were counted both by hemocytometer and by Coulter counter (Coulter Electronics, Hialeah, FL) and then frozen at - 2 W C until further analysis. ET-1 in the media was measured using a competi. tive radioimmunoassay according to the procedure outlined by the manufacturer (Peninsula, Belmont, CA). The limit of detection o f this assay is 2 pg per sample. Intracellular ET-1 was measured after cell Iysis with 0.1% Triton X-100 (Sigma). The results were expressed as pg/10 s cells.
Northern blot analysis Cells exposed to normoxia or hypoxia were rinsed with phosphate buffered saline and trypsinized from the plate. Total RNA was extracted using ~a acid guanidinium phenol-chloroform protocol [11] and assayed by measuring absorbance at 260 am. 10 ~tg of total RNA were electrophoresed on a 1% agarosc gel
I07 containing 0.66 M formaldehyde. After capillary transfer to nylon membranes (MSI, Westborough, MA) in 10 x SSC (1 × SSC =0.15 M NaCI/0.015 M sodium citrate, pH 7.4) the samples were crosslinked by ultraviolet radiation. A 38 base pair oligonucleotide synthesized by MacCumber [12] based on the sequence of ET was 3' end labeled with [azPIdATP (NEG-012H; Dupont New England Nuclear, Wilmington, DE) to a specific activity of 3' l0 s cpm/p.g using terminal transferase (BRL, Bethesda, MD). Northern blots were prehybridized for 2 h at .42°C in buffer containing 50% formamide and 4 × SSPE (1 X S S P E = 0 . 1 5 M NaCl/0.09 M NaH2PO4/1 mM EDTA, pH 7.4) and the RNA w0s then hybridized with the labelled ET probe overnight in the same buffer at 37°C. The blots were washed twice at 42°(2 in 1 × SSC + 0.5% SDS f,-" 30 rain and then exposed to X-ray film (X-OMAT Kodak, Rochester, NY) at -80"C for 72 h in the presence of one intensifying screen (Dupont). The nylon membranes were stripped by boiling in 0.1% SDS for 15 rain. The RNA was then reprobed with a po|~(dT) oligonueleotide labelled with [32P]dTTP (PB10387, Amersham, Arlington Heights, IL) using terminal transferase. The synthesis and labelling of this probe were performed as described [13].
Ana!yais of data The samples were done in duplicate and each RIA measureraent was w:rformed in duplicate. Statistical analysis included both paired and unpaired Student's t-test. Significance was defined as P < 0.05.
Results
Cell characterization The endothelial origin of the cell lines isolated and cultured from the pulmonary artery and the lung mi. crovasCulature was confirmed by phase and fluorescent microscopy (Fig. I). Cells were Bored to have a typical ¢uboidal appearance in a cobhlestone monolayer [1.4]. The BPAE and BMVE cells were also positive for the presence o f Factor VIII related antigen and incorpo. rated acetylated LDL over a .4 h incubation, two characteristics used to positively identify cells as endothelial [15,16].
Cell viability Following hypo~ic exposures for 2, 4, L2, 24 and 48 h the endothelial cells were assessed for viability by retention of a cobblestone appearance, adherent cell counts and trypan blue exclusion. No significant differences between cells exposed to h ~ , i a and normoxia were seen at any time point (Table 1).
TABLE I Viability of pulmonary aeteo endothelial celh ~ r ~ w d to ~ormoxia and Izypoxi~ "
T[me (h)
"rrypanbloc exclusion Adherentcell comus (percent) (1O¢'cells) .ormoxia hypoxia noremxia h,/po~ia
2 4 12 24 48
93+2 91 4- I 93+2 95£1 92+1
91±1 91 ± 2 90±5 90±4 88±2
0,63± 0.L2 0.43-~0.01 0..56~: 0.O')' a.50 -I-0.05 0.63+0.08
0,61-J:0,l ! 0.65 ±l)A0 0.654-0.12 0..532]:O.08 e.56± 0,0~
Cellswere exposedto nom~xia or hypmia as describedil~Materials and Methods, No significant difference at amy time point (P > 0.05). Data are expressedas mcan:l:S.E.M, n = 10 for each condition.
Endothelin- l production During normoxia the ET-I in the medium increased over 48 h, indicating a constitutive production by the CCL-209 endothelial cells. The baseline production of ET-I by the bovine pulmonary artery endothelial cell line was approfimately 75-150 pg/106 cells per h. This amount is comparable to what has been reported previously for this c~ll line [17]. The amen'it of ET-I detected from the corresponding cell lysates was measurable hut significantly less than the quantity released into the media, indicating that there is no substantial intracellular store of ET-1 (Table ll). The amount of ET-1 produced by the established cell line (CCL-209) at 4g h (3845 ± 1195 pg/106 cells) was not significontly different from the amount of ET-1 produced by the BPAE cells (5970 ~e 490 pg/106 cells). However, ET-1 production by the microvascular cells (BMVE) at 48 h, 242 ± 96 pg/10 ° cells, was significantly lower (P < 0,05) than that from the corresponding BPAE cells. These results suggested a regional difference in ET-I production along the pulmona~v vasculature rather than an effect of primary cell culture on ET-1 production. The2¢ was no difference in the ET-1 found in the cell lysates of the BMVE cells as compared to the BPAE cells (data not shown).
Effect of hypoxia At the initial time imints studied, there was no significant difference in ET-1 release into the media by CCL-2(D cells exposed to normoaia or hypoxia. How. ever, by 48 h the amount of ET-I detected in the media from ~ cells was significantly less than the amoum in the media from corresponding normoaie cells (Fig. 2). The delay in attenuation of ET-I production suggests that hypoxia affects transcription or peptide synthesis rather than the secretion of ET-1. BPAE ceils responded similarly to h!qx~xia with a decrease in production of ET-I by 50-60% rehearing hypoxic exposure for 48 h. The amounts of ET-I that remained in
108 TABLE li ET. 1 cont,! o f lyxalex ~ d conditioned media ~mm cells malntaiswd m norem~ic c,~did¢~ ~
Time (h)
Cell IDatcs (pg/10 e cells)
Media (pg/10 h cello
2 4 12 24 48
47±32 23+ 4 13± 0 27±13 19+ 9
3294 61 569± 150 985-1- 24Z 14~,'0~: 488 3845:1: ] i~l
• Cells were maiataine,d in 5% CO2/21% O2 for 48 h ia scrota free mmdia. Data are expressed as mean:t:S,E.M, n ~ 2 for cell lysatcs
at each lime polnt; t~= !0 for medium sml~lcs at each time point.
the cells after normoxic or hypoxic exposure were not significantly different (Fig. 3), again suggesting that the decrease in ET-I in the media was not due to a decrease in release or a n increase in ceLl storage of this peptide. B M V E cells responded similarly to hypoxia by decreasing p r o d u c t i o n of ET-1. A t 48 h the a m o u n t of ET-I p~)duced by the B M V E cells exposed to hypoxia was below the level detectable by the assay used (2 pg/sample). When BPAE cells w e r e restored to normoxia during a second 48 h period, ET-1 levels were c o m p a r a b l e to levels produced by normoxic controls (Fig. 4). Thus,
FiB. t. Microsrapbs of primary pulmonary mlcrovascular endothelial celLs, (a) phase mk:rojraph. Bar - 50 microns. (b) fluorescent mieroglraph of
incorporallon of acetylslad L D L
109 I 7C30~' ~Q013
75 ¢
~
5000
50
4000 ¢ 3000 200(3 10Gc
0
4
12
24
48
"1~fl e (hOurs}
FJB, ~ ~ ' ¢ e [ of ] l y ~ a on E.T-i production, The amount of ET-I released Jmo the mcdi~ by eeL-Z09 ¢~lls under Itypox~c (itullow baf~) o r normoxic (filled ba~s) ¢¢tndJt[ofis for 2 thlough 48 It is sltown. 1"he data are exl~e~ed a~; a percet'ttag¢ Of the maximum prodacti¢~ at 48 h. Dais represeat mean 4- S.E. n = ]O for each condition at each time point, Asterisk indicates P < 0.05.
hypoxic induced down regulation of ET-I was reversible. Further, reoxygenation following a period of hypoxia did not stimulate ET-I production above baseline.
0
aermmd4t bype~,
r~wjr|Gnattoa
Fig. 4. R~ersal of hypoxJc inhll~ition of ET-1 I~roduclion. The amount of ET-L released ~ t o the media by pflman, BPAJE c e l l after 48 h nf nomlnxie condilio~ (n - 2 . ~ condido~ (.,I ~ 10) aa¢l reoxygenalion followi.ng h,jpooda i n - - 5 ) is shown. Bars rcptcscn! mean-I- S.E..a,sle risk indlcales P < 0.05.
mRNA (the labelled pol~dT) oligonudeotide). Though its use has been reported in the literature [17] this probe has not served previously to standardize RNA
H
N 285
Northern blot analysis Hybridization of RNA from the BPAE cells with tke labelled ET oligonucleotide probe detected the presenoe of a transcript of the expected size, 2.4 kb [12]. ET-1 mRNA was diminished in cells exposed to hypoxia (Fig 5a). Thus, a reduction in the steady state ET-1 mRNA results in a reduction in secreted peptide. Nonspecific reduction of mRNA by hypoxia was cons[dered by the use of a probe for the polyA tail of
185
ao
H
N 28S
80
"io i 185 ~1
4
]~ 1;m, (hour~)
z4
.tit
Fill. 3, lu|racellhda¢ ET-I eo,~teat. The amount of ET.1 contained in the cell lyules of CCL-:i~J ¢e]l~. exposed IO Ether h~/p0eda (hollow bars) 0~ aormoxia (filled bars) ~ 2 tl~ro01~ 48 b is shown here. Bars mprcsen! mean 4-$.E. ~ ~ 4 I~r ©Itch ~ d i t i o n a! each dine poP,t.
Fig.5. Northern blolan~tsJsof R N A ram-acted~
cells to hyp0xia (H) or nonneeda (P0. Tite ISS and 28S rRNA b~db arc used ~ size ma~ken. (a) l-bbridiz~tio~ ~ t h a labeJed E T olisoaucleotide p l e b e de~c.B a 2.4 Irb d l n a t sko;m in the ripper panel, (b) HybridizatioB w#h [abe.k~i ~ d ' D ¢~uclcotk]c standardizes for total mP-JqA shown in the lower panel.
llO on Northern blots following a hypoxic exposure. Fig. 5b demonstrates that the observed difference in ET hybridization was not due to variations in the amounts of loaded RNA or degradation of mRNA. Dismssien Studies on the synthesis and release of ET.1 indicate that this peptide is synthesized as a preproendothelin of approx. 300 amino acids and cleaved first to pmendothelin and then to the mature, biologically active form [6,19,20]. Though ET-I production was first described in endothelial cells, other cell types including airway epithelial cells and vascular smooth muscle cells have been shown to produce ET-I [21,22|. Pulmonary vascular endothelial cells were used in this study to investigate ET-1 production by cultured cells from different anatomic regions (pulmonary artery and pulmonary microcirculation). A new line of BPAE cells and tbe established bovine pulmonary artery endothelial cell line, CCL-209, produced similar amounts of ET-I. Cells derived from the pulmonary microvasculature produced significantly less ET-1 than did cells derived from the pulmonary artery of the same animals, Dodge and coworkers have reported similar results for bovine pulmonary artery and pulmonary microvascnlar endothelial cells [23]. The basis for the difference in ET-I production by BPAE and BMVE cells was not due to differences in culture conditions as the BMVE ceils were derived from the same animals and after initial isolation were cultured in the same way as the BPAE cells. Further, two endothelial characteristics, the presence of Factor VIII related antigen and the incorporation of acetylated LDL, were not different between the BPAE and BMVE ceils. The physiologic consequences of a regional difference in ET-I production in the pulmonary vaseulature are unknown. ET-] is a known vasoactive substance and a difference in peptide production along the vascular bed may be involved in local regulation of circulatory tone. In our experiments cultured pulmonary artery and pulmonary microvascular endothelial cells exposed to lvjpoxia produced significantly less ET-1 at 48 h compared to control cells exposed to normoxia. Other investigators have found variable responses of ET-1 production to hypoxia. Similar to our findings, Yoshimote and coworkers demonstrated that porcine cerebral microvascular endothelial cells in culture produce less ET-1 under hypoxic conditions [24]. This difference was noted at 12 h in their study. Later time points were not reported. Hieda reported the production of ET-I by calf coronary artery endotbeliai cells increased following exposure to hypoxia for 24 h [25]. Out experimental model differs in that we investigated endothelial cells from the pulmonary vaseulature rather than
the systemic circulation. One interesting possibility is that the effect of hypoxia on ET-I production by cultured vascular endothelial cells may be dependent on the anatomic origin of the cells studied. When BPAE cells were restored to normoxia following a hypoxie exposure, ET-1 production returned to baseline levels. This indicates that the down regulation of ET-I production by hypoxia is reversible. Also, these data suggest that reo~genation following hypoxia, as might be seen in an ischemia-reperfusion injury, does not stimulate production of ET-1 above baseline. The decrease in ET-I peptide following hyposic exposure for 48 h was associated with a decrease in ET-1 mRNA levels in the cells, suggesting thal hypoxia attenuates ET-1 production at the transcriptional level. Furthermore, although we can not rule out the possibility of hypoxia causing enhanced degradation of the ET-1 peptide, our results suggest that this is not the sole mechanism for decreased ET-1 levels after 48 h of hypoxia. Because the steady state message is measured by Northern blot analysis, it is unknown whether the decrease in ET mRNA is secondary to decreased production or increased degradation of the transcript. There is evidence that prepmendothelin mRNA level is regulated by degradation, with the transcript having a half life of 15 rain [26]. Factors which increase the ET-I message levels such as transforming growth factor beta [6], thrombin [6], and shear stress [27] may do so by attenuation of degradation. Conversely, decreases in mRNA may be due to an upregulation of degradation. It is possllfle that hypoxia upregulates the degradation of the ET-1 transcript. in summary, this study demonstrates that the amount of ET-1 produced by both pulmonaff artery and pulmonary micmvascular endothelial cells in culture decreases in response to hypoxia for 48 h. This attenuation of peptide production is reversible and is associated with a decrease in the level of ET mRNA sacondary to an upregulation of degradation or a down regulation of production of the ET-1 transcript. In addition, our data show that ET-1 is not a stored peptide and that there is a difference in the baseline release of this peptide which is dependant on the site of origin of the cell from the pulmonary vascular bed. Our findings in this model of cultured pulmonary vascular endothelial cells do not support a role for increased ET-1 syntheseis or release as a falter regulating smooth muscle cell and fibroblast proliferation in the distal pulmonary vasculature in response to 48 i, of hypoxia. Admowledgements The authors appreciate the technical assistance of Betsy Rohland and the statisticaL analysis by Diane
111 Markakis. T h e E T - I oligonueleotidc p r o b e w a s kindly provided by M . MacCumber, The w a l k was funded by a Cystic Fibrosis Foundation R e s e a r c h Fellowship 0-o J,W.), a Kidney F o u n d a t i o n Fellowship (to C.M-R.), N I H G r a n t # K 0 8 H L - 0 2 [ 8 8 (to P.Z.~ and N I H G r a n t # H L 40178 and H L 47122 (to W.B.G,).
References I Muysiek. B. and Reid, L 11982) Lab. Invest, 46. 51L5-514, 2 Muir[ok, B. and Reid, L. 11981) Exp, Lun8 Res. 2, 2~7~271, 3 Ryan, U.S.. [~yan,/.W., Whitaker, C. and Chiu. A. (1976) Tissue Celt 8, 125-146. 4 Furchgott, R.F. and Zawadzki, 3.V. 11980) Nalure 2H8.373-376. 5 Hickey, K.A., Rubanyi, G., Paul, RJ. al~d Highsmilh, R.F. 1198S) Am, J. Physiol. 248, C.550-C556. h Yauagisawa, M., Kurihara, H.. Kimura, S., Tomobe. Y., Kobayashi, M., Mitsui. Y., Yazaki. Y., Gala. K, and Mas~ki, 1". (1988) Nature 332. 411-415. 7 Komllro. 1.. Kurihara, H., Shugiyam3,T., Takaku. F. and Ya~aki. Y+ 41q88) FEBS Lint. 238, 249-252. 8 Simoflson, M.S., Warm, S., Menu, P.. Dabyak, G.R.. Kester+ M-. Nakazalo, Y., Sedor, J.R. and Dunn, M.J. (1989}J. Cli~. invest. 83, 708-712. 9 Ryan. U.S. (1984) Euv. Health Persp. 56, 103-114. 10 M~rJck, B., Hoover, R., Jones, MR,, Ber~. L.C. and Brigham. K.L. 11989)J. Cell. Physiol. 138, 165-174. 11 Choacz3'nski, P. and Sacehi, N. (1987) Anal. B[ochem. 162, 156159. 12 MacCumber, M.W., Ross, C.A. Glaser, B.M. and Snyder. S.H. 11989) Proc. Natl. Aead. Sci. USA $6, 7285-72~,
13 Lubberl, H., Shuleh, T.P., l~tscal,N., l..~ter,H. and David~4~, N. 0.987)J. Nen~i. 7, 1157-1165. 14 Balconi, G. and Dejana, E. (l°/~6) Mud. Biol. 64, 231~245. 1.5 Jaffe, E A.. Boycr, L.W. and Nachfaan. R,L (1973) J. C'lin. Irrv~t. S2, 2757-~764. I~ Voyla, J.C..Via. D.P., Bunerfteld. C,E. ald Zgtter, B.R. 11980 J. Cell. Biol. 99, 2034-2040. 17 Ohlstein, E.H., Atleth, A . Eckiel. M., tlomhonich, S., Arm M~ Caltahiano. M M . and Sung, C. (1990) Life Sci. 46, 18|-188. 18 Jrnt~h, T J. Steint'~¢r. g. and Schwar'z,G. (1990) Mature 510-514 19 Emori. T , Hiraia. Y., Ohta+ g., Shlchiri. M.. Shifnokado, K., and Mmumo, F, I19~) ~hem. Bioplty& Rc~ Commtm. 162, 217223. ~1 Matsumma, Y, Ikegawa. R., Tsokabara, Y., Takaoka, M, andl Morirn0to. 5. (t990) FEBS Loll. 272, 111~-]1170. 2l Black, P.N. Ohatei, M.A., Taltah~hi, K., l~ethellon-Watt, D., KJau~, T., Dollery, C,T, and ]Bloom. S.R. (11~9) FIEBS LetL 24,$. 129-132. 22 Resink, T.J., Hahn, A,W.A., S~li-But'der*, T,, IN>well,J., weber, E. and Buhlcf, F.R. {l~JO) Bioehe~. Biolph,J'& Res. Common. 168. 1303-13|~. 23 Dodse, A.B.+ Hechtman, H.3. amJ sl~pm, D. (199t) Cell ] 1 L-",,jtosk. 18. 180--~& 24 Yoshiinolo, S., [shizaki, Y., Sasaki, T. and Mttrota, S. (|991} Stroke 22, 378-383. 25 Hiedo, H,$. and G0mcz.Sanchez, C-E. 11990) Life Sci. 47, ~1.725l. 26 YanaBisawa, M., Inane, A., Takuwa, Y., Mitsui, Y.. Kobayashk M. and Masukl, T. (19~) J. C.mdiovasc.Pharm. 13, $13.--SI7. 27 Yoshizumi. M., Kudhara, H., Sugiyalua, T., Tatmku, F., Yanagisawa, M.. Masaki, T. and Yazald, Y. (1989) Biochem, Bioldh~ Res. Coramun. 161,859-864.
I 0~;
© 1992 ElsevierScience PublishersB.V. All righls reserved 0167-4889/92/$05.00
BBAMCR 1311S
Effect of hypoxia on endothelin-1 production by pulmonary vascular endothelial cells J e n n i f e r L. W i e b k e , C h a h r z a d M o n t r o s e - R a f i z a d e h , a n d W i l l i a m B. G u g g i n o
P a m e l a L. Z e i t l i n
The Del~rlments of Pediettric$ertd PtlyJiology, The 1ohtt~ Hopkins Medicallnstitutior~ Balttreumr.MD (USA~
(Received21]August 199D (RevisedmanuscrilRreceived 19 November 1991l Key words: Endothe~iin-I;l-lypoxia Endolhelin-I (ET-I), a pcptide product of endolhelial cons, is mRogeaie for fibmblasts and smooth muscle cells. In this study we examined the effect of hypoxia on ET-I production by bovlnc pulmonary vascular endothelial cells. Boy:me pofmona~ artery (BPAE) and microvascular endothelial (BMVE) cells were isolated, grown in tissue culture, and characterized by tl~ presence of Factor VIII related antigen and LDL uptake. Baseline production of ET-I by BPAE cells (measured by radioimmunoass~) increased over time. BMVE cells produced one tenth the amount of ET-I as produced by the pulm~aarj artery endothelial cells under the same conditions. In both cell types, hypoxia (0% 0 2) signifmandy reduced the amount of ET-! at 48 h. Restcvration of normosia in 21% O 2 for 48 h resulted in a term, of ET-I levels to baseline. Northern blot analysis showed decreased ET-I mRNA in cells eXlmSed to h~ao~aa for 48 h. These data demonstrate that pulmonaq~ vascular endotlmlial cells respond to hypoxia by reversibly decreasing ET-I production, and 1his altcnuation is likely regulated at the level of n'anseril~km.
lntrtaluctiea Chronic hypoxia leads to proliferation of pulmonary vascular smooth muscle cells and flbroblasts in the pulmonary circulation [1], In response to hypoxia the vessel wall demonstrates thickening of the media and adventitia secondary to smooth muscle cell hypertrophy and differentiation of non-contractile precursor cells. There is also accumulation of elastin and collagen fibers in the vessel wall. These changes are most prominent at the pro.capillary level in the distal p u b monary vaseulatutc [2], The signal linking Iwpo~a to this process of pulmonary vascular remodeling is unknown. A candidate for a role in this process is the endu~helial cell, once thought to have only barrier furtetion, but now rgcognized to play an integral part in many vascular responses. The endothelial cell is capable of not only metabolizing certain vasoactive substances such as angiotensin [3]. but has been found to synthesize a number of vasoactive substances as well [4,5].
Eud~theiin-I (lET-l) is a product of endotbeUal cells. It is a 21 amino acid peptide first described as vasoconstrictive in action [6]. ET-I has also been shown m be mitogonic for smooth muscle cells and fibroblasts in culture. This peptide causes an inc~,ase in protooncogene expression and cell proliferation in smooth muscle cells, as well as an inorease in [3Hlthymidine uptake in target cells ['7], contdstent with mitogenic activity. ET-I also stimulates the turnovec of phosphatidylinositol and alkalinizatioR of the cytosol [8], both of which are associated with cell proliferation. In this study we investigated cultured lung endothelial cells from the pulmonary artery and the pulanonaqw microcircelatiou to determine baseline ET-1 production by cells derived from these different sites along th~ pulmonary vascular bed. The effect o f hypoaia on ET-I production by the pulmonary vascular endothelial cells in culture was investigated by measuring ET-1 peptide and m R N A from cells exposed to hypoxia.
Materials and Metlmds Cell isolation and culture
Correspondence: W.IL GuHiao, Depa~ment of Physiology, The Johns HopkinsUniversity,Schoolof Medicine,Baltinmre,MD 21205,
USA.
Cultures of bovine pulmonary artery endothelial cells and pulmonary micr(wascular endothelial cells derived from the same animal were used in this study. In initial experiments for which large numbers of cells were
106 required, we used an established line of bovine pulmonary artery cells (CCL-209, American Type Tissue Collection, gockvillc, M D ) w h o s e endothelial origin has been confirmed by the presence of angiotensinconverting enzyme. Cell isolation. Bovine pulmonary artery endothelial (BPAE) cells were isolated as described [9]. Pulmonary arteries were trimmed of excess fat and further procussed under sterile conditions. Vessels were washed three times in Ringer's sotution containing 3 × antibiotics (I × antibiotics consists of penicillin, 100 p.I/ml and streptomycin, 100 /.tg/ml; Gilx:o, Grand Island, NY). Vessels were then cut open to expose the endothelial surface which was gently scraped with a #10 scalpel blade. Cells were centrifuged in DME (Gibco) with 3 × antibiotics at 250Xg for 10 rain at 4°C, resuspcnded in growth medium (DME with 20% FC~ and 1 × antibiotics) and placed in 100 mm culture dishes at 37°C in a 95% air,/5% CO 2 incubator. To obtain pure endothelial cell lines, the primary culture plates were evaluated daily. Colonies of 50-I00 endnthelial cells were identified by their cuboidal, cobblestone appearance using phase microscopy and then isolated on the plate using glass cloning cylinders (Belice, Viueland, N J). The select~cd ceils were trypsinized with 0.C,5% trypsin with 1% polyvinylpyrrolidone and 0.02% EGTA in Hepes-buffered saline (PET; Biofluids, Roclccille, MD). Ceils were then transferred to 24 well plates from which they were serially passaged. Bovine pulmonary microvascular endothelial (BMVE) cells were obtained using collagenase digestion of peripheral lung as previously described [10l. Peripheral lung was dissected from the pleura to avoid contamination with mesothelial cells. The lung tissue was finely minced irk a solution of Ringer's with 3 × antibiotics. After h.lsing the minced lung three times with Ringer's, the tissue was incubated with 0.1% collagenase (Type IA, Sigma, St. Louis, MO) for 20 rain at 37~C. The dissociated cells were centrifuged at 250 × g for 5 rain at ¢°C. The cell pellet was resnsponded in growth medium and placed in 100 nltm culture dishes. To obtain pure microvaseniar endothelial cell lines further isolation was performed as described above. Cell cullure. Cells were passed when confluent using 0.05% trypsin (Biofluids). Experiments were performed on cells at passages 6-10 and on the CCL-209 endothelial cell line at passages 18-22. AI[ cells were maintained in growth medium containing DME supplemented with 20% FCS (Inovar Biologics, Gaithersburg, MD) and 1 x a,:tibiotics. Cell characterizatiog The BPAE and BMVE cell lines were demonstrated to meet two criteria for eudnthelia[ cells: the presence of Factor VIII antigen detected by primary antibody to Factor VIII (Dace, Carpinteria, CA) complexed with FITC-conjugated lgG
(Dace) and the uptake of rhodamine labeled acetylated low density [ipoprot,~in (LDL) (Biomedical Technologies, Stoughton, MA) following a 4 h incubation.
Normoxic and hypoxic exposures Endothelial monolayers in 35 mm tissue culture dishes w e n rinsed three times in serum free medium. Serum free medium (1 ml) was then added to each dish. Normo~ic exposures were accomplished in a standard 5% CO 2 incubator. Hypoxie exposures were performed in an airtight chamber (Billups Rothenberg, Del Mar, CA) gassed with a mixture of 95% N2/5% CO 2 for 1,5 rain. Oxygen concentration in the chamber was monitored with an oxygen microelectrode (Microelectrodes, Londonderry, NH). The chamber was then placed in a 37°C warm room for 2 to 48 h. For hypoxic exposures of 48 h, the chamber was regas~d at 24 h. At the end of each hypexic exposure the percent O 2 in the chamber was < 0.05%. The Pc2 of the media of cells incubated in hypoxic conditions was 20-30 mmHg at all time points compared to a Pc2 of 120-130 mmHg in the media of cells incubated in normoxia. In experiments performed on BPAE and BMVE cells the expo. sures were modified by the addition of serum to the medium. ET-I was not detectable in serum-containing medium alone. In a series of experiments, the BPAE cells were restored to normoxia following the initial 48 h hypo~dc exposure and ET-1 production was measured.
Measurement of endothelin-I At the end of each KVpox[c or normoxie exposure media were aspirated from culture plates and volume of the media determined. The media ware centrifuged to remove any cell debris and supernatants were frozen at - 2WC until further analysis. Cells were assessed for viability by examination of morphology, adherent cell counts and Trypan blue exclusion. The cells were counted both by hemocytometer and by Coulter counter (Coulter Electronics, Hialeah, FL) and then frozen at - 2 W C until further analysis. ET-1 in the media was measured using a competi. tive radioimmunoassay according to the procedure outlined by the manufacturer (Peninsula, Belmont, CA). The limit of detection o f this assay is 2 pg per sample. Intracellular ET-1 was measured after cell Iysis with 0.1% Triton X-100 (Sigma). The results were expressed as pg/10 s cells.
Northern blot analysis Cells exposed to normoxia or hypoxia were rinsed with phosphate buffered saline and trypsinized from the plate. Total RNA was extracted using ~a acid guanidinium phenol-chloroform protocol [11] and assayed by measuring absorbance at 260 am. 10 ~tg of total RNA were electrophoresed on a 1% agarosc gel
I07 containing 0.66 M formaldehyde. After capillary transfer to nylon membranes (MSI, Westborough, MA) in 10 x SSC (1 × SSC =0.15 M NaCI/0.015 M sodium citrate, pH 7.4) the samples were crosslinked by ultraviolet radiation. A 38 base pair oligonucleotide synthesized by MacCumber [12] based on the sequence of ET was 3' end labeled with [azPIdATP (NEG-012H; Dupont New England Nuclear, Wilmington, DE) to a specific activity of 3' l0 s cpm/p.g using terminal transferase (BRL, Bethesda, MD). Northern blots were prehybridized for 2 h at .42°C in buffer containing 50% formamide and 4 × SSPE (1 X S S P E = 0 . 1 5 M NaCl/0.09 M NaH2PO4/1 mM EDTA, pH 7.4) and the RNA w0s then hybridized with the labelled ET probe overnight in the same buffer at 37°C. The blots were washed twice at 42°(2 in 1 × SSC + 0.5% SDS f,-" 30 rain and then exposed to X-ray film (X-OMAT Kodak, Rochester, NY) at -80"C for 72 h in the presence of one intensifying screen (Dupont). The nylon membranes were stripped by boiling in 0.1% SDS for 15 rain. The RNA was then reprobed with a po|~(dT) oligonueleotide labelled with [32P]dTTP (PB10387, Amersham, Arlington Heights, IL) using terminal transferase. The synthesis and labelling of this probe were performed as described [13].
Ana!yais of data The samples were done in duplicate and each RIA measureraent was w:rformed in duplicate. Statistical analysis included both paired and unpaired Student's t-test. Significance was defined as P < 0.05.
Results
Cell characterization The endothelial origin of the cell lines isolated and cultured from the pulmonary artery and the lung mi. crovasCulature was confirmed by phase and fluorescent microscopy (Fig. I). Cells were Bored to have a typical ¢uboidal appearance in a cobhlestone monolayer [1.4]. The BPAE and BMVE cells were also positive for the presence o f Factor VIII related antigen and incorpo. rated acetylated LDL over a .4 h incubation, two characteristics used to positively identify cells as endothelial [15,16].
Cell viability Following hypo~ic exposures for 2, 4, L2, 24 and 48 h the endothelial cells were assessed for viability by retention of a cobblestone appearance, adherent cell counts and trypan blue exclusion. No significant differences between cells exposed to h ~ , i a and normoxia were seen at any time point (Table 1).
TABLE I Viability of pulmonary aeteo endothelial celh ~ r ~ w d to ~ormoxia and Izypoxi~ "
T[me (h)
"rrypanbloc exclusion Adherentcell comus (percent) (1O¢'cells) .ormoxia hypoxia noremxia h,/po~ia
2 4 12 24 48
93+2 91 4- I 93+2 95£1 92+1
91±1 91 ± 2 90±5 90±4 88±2
0,63± 0.L2 0.43-~0.01 0..56~: 0.O')' a.50 -I-0.05 0.63+0.08
0,61-J:0,l ! 0.65 ±l)A0 0.654-0.12 0..532]:O.08 e.56± 0,0~
Cellswere exposedto nom~xia or hypmia as describedil~Materials and Methods, No significant difference at amy time point (P > 0.05). Data are expressedas mcan:l:S.E.M, n = 10 for each condition.
Endothelin- l production During normoxia the ET-I in the medium increased over 48 h, indicating a constitutive production by the CCL-209 endothelial cells. The baseline production of ET-I by the bovine pulmonary artery endothelial cell line was approfimately 75-150 pg/106 cells per h. This amount is comparable to what has been reported previously for this c~ll line [17]. The amen'it of ET-I detected from the corresponding cell lysates was measurable hut significantly less than the quantity released into the media, indicating that there is no substantial intracellular store of ET-1 (Table ll). The amount of ET-1 produced by the established cell line (CCL-209) at 4g h (3845 ± 1195 pg/106 cells) was not significontly different from the amount of ET-1 produced by the BPAE cells (5970 ~e 490 pg/106 cells). However, ET-1 production by the microvascular cells (BMVE) at 48 h, 242 ± 96 pg/10 ° cells, was significantly lower (P < 0,05) than that from the corresponding BPAE cells. These results suggested a regional difference in ET-I production along the pulmona~v vasculature rather than an effect of primary cell culture on ET-1 production. The2¢ was no difference in the ET-1 found in the cell lysates of the BMVE cells as compared to the BPAE cells (data not shown).
Effect of hypoxia At the initial time imints studied, there was no significant difference in ET-1 release into the media by CCL-2(D cells exposed to normoaia or hypoxia. How. ever, by 48 h the amount of ET-I detected in the media from ~ cells was significantly less than the amoum in the media from corresponding normoaie cells (Fig. 2). The delay in attenuation of ET-I production suggests that hypoxia affects transcription or peptide synthesis rather than the secretion of ET-1. BPAE ceils responded similarly to h!qx~xia with a decrease in production of ET-I by 50-60% rehearing hypoxic exposure for 48 h. The amounts of ET-I that remained in
108 TABLE li ET. 1 cont,! o f lyxalex ~ d conditioned media ~mm cells malntaiswd m norem~ic c,~did¢~ ~
Time (h)
Cell IDatcs (pg/10 e cells)
Media (pg/10 h cello
2 4 12 24 48
47±32 23+ 4 13± 0 27±13 19+ 9
3294 61 569± 150 985-1- 24Z 14~,'0~: 488 3845:1: ] i~l
• Cells were maiataine,d in 5% CO2/21% O2 for 48 h ia scrota free mmdia. Data are expressed as mean:t:S,E.M, n ~ 2 for cell lysatcs
at each lime polnt; t~= !0 for medium sml~lcs at each time point.
the cells after normoxic or hypoxic exposure were not significantly different (Fig. 3), again suggesting that the decrease in ET-I in the media was not due to a decrease in release or a n increase in ceLl storage of this peptide. B M V E cells responded similarly to hypoxia by decreasing p r o d u c t i o n of ET-1. A t 48 h the a m o u n t of ET-I p~)duced by the B M V E cells exposed to hypoxia was below the level detectable by the assay used (2 pg/sample). When BPAE cells w e r e restored to normoxia during a second 48 h period, ET-1 levels were c o m p a r a b l e to levels produced by normoxic controls (Fig. 4). Thus,
FiB. t. Microsrapbs of primary pulmonary mlcrovascular endothelial celLs, (a) phase mk:rojraph. Bar - 50 microns. (b) fluorescent mieroglraph of
incorporallon of acetylslad L D L
109 I 7C30~' ~Q013
75 ¢
~
5000
50
4000 ¢ 3000 200(3 10Gc
0
4
12
24
48
"1~fl e (hOurs}
FJB, ~ ~ ' ¢ e [ of ] l y ~ a on E.T-i production, The amount of ET-I released Jmo the mcdi~ by eeL-Z09 ¢~lls under Itypox~c (itullow baf~) o r normoxic (filled ba~s) ¢¢tndJt[ofis for 2 thlough 48 It is sltown. 1"he data are exl~e~ed a~; a percet'ttag¢ Of the maximum prodacti¢~ at 48 h. Dais represeat mean 4- S.E. n = ]O for each condition at each time point, Asterisk indicates P < 0.05.
hypoxic induced down regulation of ET-I was reversible. Further, reoxygenation following a period of hypoxia did not stimulate ET-I production above baseline.
0
aermmd4t bype~,
r~wjr|Gnattoa
Fig. 4. R~ersal of hypoxJc inhll~ition of ET-1 I~roduclion. The amount of ET-L released ~ t o the media by pflman, BPAJE c e l l after 48 h nf nomlnxie condilio~ (n - 2 . ~ condido~ (.,I ~ 10) aa¢l reoxygenalion followi.ng h,jpooda i n - - 5 ) is shown. Bars rcptcscn! mean-I- S.E..a,sle risk indlcales P < 0.05.
mRNA (the labelled pol~dT) oligonudeotide). Though its use has been reported in the literature [17] this probe has not served previously to standardize RNA
H
N 285
Northern blot analysis Hybridization of RNA from the BPAE cells with tke labelled ET oligonucleotide probe detected the presenoe of a transcript of the expected size, 2.4 kb [12]. ET-1 mRNA was diminished in cells exposed to hypoxia (Fig 5a). Thus, a reduction in the steady state ET-1 mRNA results in a reduction in secreted peptide. Nonspecific reduction of mRNA by hypoxia was cons[dered by the use of a probe for the polyA tail of
185
ao
H
N 28S
80
"io i 185 ~1
4
]~ 1;m, (hour~)
z4
.tit
Fill. 3, lu|racellhda¢ ET-I eo,~teat. The amount of ET.1 contained in the cell lyules of CCL-:i~J ¢e]l~. exposed IO Ether h~/p0eda (hollow bars) 0~ aormoxia (filled bars) ~ 2 tl~ro01~ 48 b is shown here. Bars mprcsen! mean 4-$.E. ~ ~ 4 I~r ©Itch ~ d i t i o n a! each dine poP,t.
Fig.5. Northern blolan~tsJsof R N A ram-acted~
cells to hyp0xia (H) or nonneeda (P0. Tite ISS and 28S rRNA b~db arc used ~ size ma~ken. (a) l-bbridiz~tio~ ~ t h a labeJed E T olisoaucleotide p l e b e de~c.B a 2.4 Irb d l n a t sko;m in the ripper panel, (b) HybridizatioB w#h [abe.k~i ~ d ' D ¢~uclcotk]c standardizes for total mP-JqA shown in the lower panel.
llO on Northern blots following a hypoxic exposure. Fig. 5b demonstrates that the observed difference in ET hybridization was not due to variations in the amounts of loaded RNA or degradation of mRNA. Dismssien Studies on the synthesis and release of ET.1 indicate that this peptide is synthesized as a preproendothelin of approx. 300 amino acids and cleaved first to pmendothelin and then to the mature, biologically active form [6,19,20]. Though ET-I production was first described in endothelial cells, other cell types including airway epithelial cells and vascular smooth muscle cells have been shown to produce ET-I [21,22|. Pulmonary vascular endothelial cells were used in this study to investigate ET-1 production by cultured cells from different anatomic regions (pulmonary artery and pulmonary microcirculation). A new line of BPAE cells and tbe established bovine pulmonary artery endothelial cell line, CCL-209, produced similar amounts of ET-I. Cells derived from the pulmonary microvasculature produced significantly less ET-1 than did cells derived from the pulmonary artery of the same animals, Dodge and coworkers have reported similar results for bovine pulmonary artery and pulmonary microvascnlar endothelial cells [23]. The basis for the difference in ET-I production by BPAE and BMVE cells was not due to differences in culture conditions as the BMVE ceils were derived from the same animals and after initial isolation were cultured in the same way as the BPAE cells. Further, two endothelial characteristics, the presence of Factor VIII related antigen and the incorporation of acetylated LDL, were not different between the BPAE and BMVE ceils. The physiologic consequences of a regional difference in ET-I production in the pulmonary vaseulature are unknown. ET-] is a known vasoactive substance and a difference in peptide production along the vascular bed may be involved in local regulation of circulatory tone. In our experiments cultured pulmonary artery and pulmonary microvascular endothelial cells exposed to lvjpoxia produced significantly less ET-1 at 48 h compared to control cells exposed to normoxia. Other investigators have found variable responses of ET-1 production to hypoxia. Similar to our findings, Yoshimote and coworkers demonstrated that porcine cerebral microvascular endothelial cells in culture produce less ET-1 under hypoxic conditions [24]. This difference was noted at 12 h in their study. Later time points were not reported. Hieda reported the production of ET-I by calf coronary artery endotbeliai cells increased following exposure to hypoxia for 24 h [25]. Out experimental model differs in that we investigated endothelial cells from the pulmonary vaseulature rather than
the systemic circulation. One interesting possibility is that the effect of hypoxia on ET-I production by cultured vascular endothelial cells may be dependent on the anatomic origin of the cells studied. When BPAE cells were restored to normoxia following a hypoxie exposure, ET-1 production returned to baseline levels. This indicates that the down regulation of ET-I production by hypoxia is reversible. Also, these data suggest that reo~genation following hypoxia, as might be seen in an ischemia-reperfusion injury, does not stimulate production of ET-1 above baseline. The decrease in ET-I peptide following hyposic exposure for 48 h was associated with a decrease in ET-1 mRNA levels in the cells, suggesting thal hypoxia attenuates ET-1 production at the transcriptional level. Furthermore, although we can not rule out the possibility of hypoxia causing enhanced degradation of the ET-1 peptide, our results suggest that this is not the sole mechanism for decreased ET-1 levels after 48 h of hypoxia. Because the steady state message is measured by Northern blot analysis, it is unknown whether the decrease in ET mRNA is secondary to decreased production or increased degradation of the transcript. There is evidence that prepmendothelin mRNA level is regulated by degradation, with the transcript having a half life of 15 rain [26]. Factors which increase the ET-I message levels such as transforming growth factor beta [6], thrombin [6], and shear stress [27] may do so by attenuation of degradation. Conversely, decreases in mRNA may be due to an upregulation of degradation. It is possllfle that hypoxia upregulates the degradation of the ET-1 transcript. in summary, this study demonstrates that the amount of ET-1 produced by both pulmonaff artery and pulmonary micmvascular endothelial cells in culture decreases in response to hypoxia for 48 h. This attenuation of peptide production is reversible and is associated with a decrease in the level of ET mRNA sacondary to an upregulation of degradation or a down regulation of production of the ET-1 transcript. In addition, our data show that ET-1 is not a stored peptide and that there is a difference in the baseline release of this peptide which is dependant on the site of origin of the cell from the pulmonary vascular bed. Our findings in this model of cultured pulmonary vascular endothelial cells do not support a role for increased ET-1 syntheseis or release as a falter regulating smooth muscle cell and fibroblast proliferation in the distal pulmonary vasculature in response to 48 i, of hypoxia. Admowledgements The authors appreciate the technical assistance of Betsy Rohland and the statisticaL analysis by Diane
111 Markakis. T h e E T - I oligonueleotidc p r o b e w a s kindly provided by M . MacCumber, The w a l k was funded by a Cystic Fibrosis Foundation R e s e a r c h Fellowship 0-o J,W.), a Kidney F o u n d a t i o n Fellowship (to C.M-R.), N I H G r a n t # K 0 8 H L - 0 2 [ 8 8 (to P.Z.~ and N I H G r a n t # H L 40178 and H L 47122 (to W.B.G,).
References I Muysiek. B. and Reid, L 11982) Lab. Invest, 46. 51L5-514, 2 Muir[ok, B. and Reid, L. 11981) Exp, Lun8 Res. 2, 2~7~271, 3 Ryan, U.S.. [~yan,/.W., Whitaker, C. and Chiu. A. (1976) Tissue Celt 8, 125-146. 4 Furchgott, R.F. and Zawadzki, 3.V. 11980) Nalure 2H8.373-376. 5 Hickey, K.A., Rubanyi, G., Paul, RJ. al~d Highsmilh, R.F. 1198S) Am, J. Physiol. 248, C.550-C556. h Yauagisawa, M., Kurihara, H.. Kimura, S., Tomobe. Y., Kobayashi, M., Mitsui. Y., Yazaki. Y., Gala. K, and Mas~ki, 1". (1988) Nature 332. 411-415. 7 Komllro. 1.. Kurihara, H., Shugiyam3,T., Takaku. F. and Ya~aki. Y+ 41q88) FEBS Lint. 238, 249-252. 8 Simoflson, M.S., Warm, S., Menu, P.. Dabyak, G.R.. Kester+ M-. Nakazalo, Y., Sedor, J.R. and Dunn, M.J. (1989}J. Cli~. invest. 83, 708-712. 9 Ryan. U.S. (1984) Euv. Health Persp. 56, 103-114. 10 M~rJck, B., Hoover, R., Jones, MR,, Ber~. L.C. and Brigham. K.L. 11989)J. Cell. Physiol. 138, 165-174. 11 Choacz3'nski, P. and Sacehi, N. (1987) Anal. B[ochem. 162, 156159. 12 MacCumber, M.W., Ross, C.A. Glaser, B.M. and Snyder. S.H. 11989) Proc. Natl. Aead. Sci. USA $6, 7285-72~,
13 Lubberl, H., Shuleh, T.P., l~tscal,N., l..~ter,H. and David~4~, N. 0.987)J. Nen~i. 7, 1157-1165. 14 Balconi, G. and Dejana, E. (l°/~6) Mud. Biol. 64, 231~245. 1.5 Jaffe, E A.. Boycr, L.W. and Nachfaan. R,L (1973) J. C'lin. Irrv~t. S2, 2757-~764. I~ Voyla, J.C..Via. D.P., Bunerfteld. C,E. ald Zgtter, B.R. 11980 J. Cell. Biol. 99, 2034-2040. 17 Ohlstein, E.H., Atleth, A . Eckiel. M., tlomhonich, S., Arm M~ Caltahiano. M M . and Sung, C. (1990) Life Sci. 46, 18|-188. 18 Jrnt~h, T J. Steint'~¢r. g. and Schwar'z,G. (1990) Mature 510-514 19 Emori. T , Hiraia. Y., Ohta+ g., Shlchiri. M.. Shifnokado, K., and Mmumo, F, I19~) ~hem. Bioplty& Rc~ Commtm. 162, 217223. ~1 Matsumma, Y, Ikegawa. R., Tsokabara, Y., Takaoka, M, andl Morirn0to. 5. (t990) FEBS Loll. 272, 111~-]1170. 2l Black, P.N. Ohatei, M.A., Taltah~hi, K., l~ethellon-Watt, D., KJau~, T., Dollery, C,T, and ]Bloom. S.R. (11~9) FIEBS LetL 24,$. 129-132. 22 Resink, T.J., Hahn, A,W.A., S~li-But'der*, T,, IN>well,J., weber, E. and Buhlcf, F.R. {l~JO) Bioehe~. Biolph,J'& Res. Common. 168. 1303-13|~. 23 Dodse, A.B.+ Hechtman, H.3. amJ sl~pm, D. (199t) Cell ] 1 L-",,jtosk. 18. 180--~& 24 Yoshiinolo, S., [shizaki, Y., Sasaki, T. and Mttrota, S. (|991} Stroke 22, 378-383. 25 Hiedo, H,$. and G0mcz.Sanchez, C-E. 11990) Life Sci. 47, ~1.725l. 26 YanaBisawa, M., Inane, A., Takuwa, Y., Mitsui, Y.. Kobayashk M. and Masukl, T. (19~) J. C.mdiovasc.Pharm. 13, $13.--SI7. 27 Yoshizumi. M., Kudhara, H., Sugiyalua, T., Tatmku, F., Yanagisawa, M.. Masaki, T. and Yazald, Y. (1989) Biochem, Bioldh~ Res. Coramun. 161,859-864.
