Mechanisms of Phosphatidylcholine Acyl Remodeling by Human Fetal Lung Petula A. Caesar, Mary C. McElroy, Frank J. Kelly, I. Colin S. Normand, and Anthony D. Postle Departments of Child Health and Human Nutrition, Faculty of Medicine, Southampton General Hospital, Southampton, United Kingdom
The molecular specificity of phosphatidylcholine (PC) synthesis by the de novo pathway in postmortem samples of human fetal lung (15 to 20 wk of gestation) was determined from the incorporation pattern in isolated microsomal preparations of CDP:[I4C]choline into individual molecular species of Pc. These analyses are based on the assumption that the molecular species composition of the pool of endogenous diacylglycerol used for PC synthesis by isolated microsomes reflects that of the authentic pool of diacylglycerol converted to PC by intact cells. Comparison of this microsomal incorporation pattern of radiolabel into PC with tissue PC composition suggested that even at this early stage of gestation 50% oflung dipalmitoyl PC was derived from synthesis de novo, with the remainder coming from acyl remodeling mechanisms. Analysis of PC synthesis de novo by organ cultures of human fetal lung showed that these acyl remodeling mechanisms were lost in culture. Despite evidence for differentiation of type II alveolar epithelial cells in culture, equilibrium labeling of PC with [I4C]choline over 18 h resulted in a progressive decline in fractional incorporation into dipalmitoyl PC with time in culture. By 4 days in culture, this value was no different from the fractional incorporation of CDP: [I4C]choline into microsomal PC in vitro over 3 h. The pattern of PC synthesized was not altered when total PC synthesis was stimulated by exposure of cultures to dexamethasone and tri-iodothyronine but was readily manipulated by exposure to exogenous fatty acids. These results demonstrate for the first time the activity of PC acyl remodeling mechanisms in human fetal lung, well before the initiation of surfactant production. They support the conclusion that conditions that permit accelerated differentiation of type II epithelial cells from human fetal lung in organ culture do not necessarily support the spectrum of PC species synthesis characteristic of the differentiated state.
The spectrum of phosphatidylcholine (PC) molecular species synthesized by lung tissue is determined largely by the balance between the rate of synthesis de novo and the activities of a variety of acyI remodeling mechanisms. The pattern of PC species synthesized de novo by the microsomal enzyme cholinephosphotransferase reflects the molecular species composition of the substrate pool of diacylglycerol (1, 2); this enzyme in lung appears to possess little inherent sub(Received in original form December 7, 1990 and in revised form March 29, 1991) Address correspondence to: Dr. A. D. Postle, Child Health, Level G, Centre Block, Southampton General Hospital, Tremona Road, Southampton S09 4XY, United Kingdom. Abbreviations: cytidine 5!diphospho-[methyJ-l4C]choline, CDP: [14C]choline; [methyl-r'Cjcholine chloride, [14C]choline; disaturated phosphatidylcholine, DSPC; Hanks' balanced salt solution, HBSS; high performance liquid chromatography, HPLC; lyso-l-palmitoyl-sn-glycero-3-phosphatidylcholine, lysoPC; phosphatidylcholine, PC; sn-l-myristoyl sn-2-palmitoyl phosphatidylcholine, PCI4:0/16:0; dipaimitoylphosphatidylcholine, PCI6:0/16:0; sn-l-palmitoyl sn-2-palmitoleoyl phosphatidylcholine, PCI6:0/ 16:1; sn-l-palmitoylsn-2-oleoylphosphatidylcholine, PCI6:0/18:1; sn-l-palmitoyl sn-2-linoleoyl phosphatidyIchoIine, PCI6:0/18:2; sn-l-stearoyl sn-2linoleoyl phosphatidylcholine, PC18:0118:2; dioleoyl phosphatidylcholine, PCI8:1I18:1; ultraviolet, Uv, Am. J. Respir. Cell Mol. BioI. Vol. S. pp. 363-370, 1991
strate specificity (3, 4). The composition of newly formed PC is subsequently modified by acyl remodeling, involving sequential actions of phospholipase and acyltransferase enzymes (5). In addition to these synthetic mechanisms, which are common in principle to)llost cell types, many cells possess specialized selectiop-tnechanisms for the incorporation of specific PC species/into different subcellular fractions. One example of such 'selection is the assembly of surfactant phospholipids into lamellar body stores within type II pneumocytes (6). Recognition of the central role of dipalmitoyl PC (PCI6: 0/16:0) in lung surfactant function (7) has been the major stimulus for most studies of lung phospholipid metabolism. Consequently, studies of acyl remodeling in lung have been directed toward understanding mechanisms for enrichment of PCI6:0/I6:0 in lung parenchyma in general and more specifically within the surfactant PC pool. Analyses of the fractional contents of PCI6:0/I6:0 in lung tissue and microsomal PC with those synthesized de novo by microsomes isolated from adult rat lung (1) or rat lung type II pneumocytes (3) have concluded that up to 50% of total PCI6:0/I6:0 is provided directly and that the remainder is formed by remodeling mechanisms. An important central role for phospholipase A2 in this remodeling, converting
364
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 5 1991
I-palmitoyl-2-enoyl PC to I-palmitoyl-sn-glycero-3-phosphorylcholine (lysoPC), was suggest~d by Post ~d ~o-~~rk ers, who incubated rat type II cells with the specific inhibitor 4-bromophenacyl bromide, causing a significant decrease in the saturation of the PC formed (3). Proposed donor molecules to supply the palmitoyl groups subsequently reesterified into the sn-2 position oflysoPC are palmitoyl CoA (8) or a second molecule of lysoPC (9), catalyzed respectively by lysoPC acyltransferase and lysoPC:lysoPC acyltransferase. Comparisons of specific activities in adult rat type II pneumocytes (10) and in neonatal mouse lung (11) suggest that lysoPC acyltransferase may be the more important enzyme for PCI6:0/16:0 synthesis. One central assumption underlying much of this work is that PC acyl remodeling mechanisms in the lung are active principally to supply PCI6:0/16:0 for s~rfactant synthesis. For instance, the developmental expressions of acyltransferase enzymes in late gestation in rat and mouse fetal lung h~ve been compared with parameters of surfactant maturation (11). However, in adult rat lung, only 30% of total PCI6:0/ 16:0 is associated directly with surfactant PC pools (12, 13), and fetal human lung PC at 15 wk of gestation already contains 25 to 30% PCI6:0/16:0 well before the initiation of surfactant production (14, 15). Moreover, disaturated PC concentration in postmortem human fetal and neonatal lung samples throughout gestation and the neonatal period remained relatively constant except for an increase and subsequent decrease in the perinatal period (16). These observations imply both that PCI6:0/16:0 synthesis by lung is not solely directed toward surfactant synthesis and that mechanisms for PCI6:0/16:0 synthesis exist very early in human lung development. However, no analyses of PC acyl remodeling mechanisms comparable to studies in adult rat lung have yet been reported for human tissue, either adult or fetal. Consequently, the initial objective in this study was to characterize such mechanisms in postmortem samples of human fetal lung. The second part of this study addressed further the changes in the specificity of PC synthesis that accompany the differentiation in vitro of human fetal lung maintained in organ culture. This widely used model system for the study of lung maturation is characterized by the development of alveolar-like structures, the acquisition of intracellular lamellar bodies, and the expression of surfactant-specific apoproteins (17-20). Additionally, PC synthesis by organ cultures can be stimulated in a synergistic fashion by exposure to dexamethasone and tri-iodothyronine (17-20). By contrast, our previous results demonstrated clearly that these changes in organ culture are not necessarily accompanied by any increased fractional synthesis of PC16:0/16:0 (15). Indeed, the percentage incorporation of [I4C]choline into PCI6:0/16:0 decreased rather than increased with time in culture. In the present study, we have investigated remodeling mechanisms of PC synthesis in organ cultures of human fetal lung in response to a variety of hormonal and nutritional modifications.
Materials and Methods Chemicals Tissue culture reagents were supplied by Flow Laboratories
(Rickmansworth, UK), chemicals from Sigma Chemical Co. (Poole, UK), high performance liquid chromatography (HPLC) solvents by Rathburn Ltd. (Walkerburn, Scotland, UK), collagen sponge (Colgen Compres~) by ,!,hames Laboratories Ltd. (Middlesex, UK), and radiochemicals by Amersham International (Amersham, UK). Human Fetal Lung Microsomal Preparation and Cell Culture Lung tissue was obtained aseptically from 13 prostaglan~in induced human fetal abortions at 15 to 20 wk of gestation, in accordance with local ethical committee guidelines. A sample of lung (approximately 150 ~g) was re~o~ed and frozen in liquid nitrogen for PC analysis. The remammg lung tissue was used either for preparation of a microsomal fraction or was maintained in organ culture for up to 5 days. For isolation of microsomes, lung was homogenized in 4 vol of 0.145 M NaCI containing 50 mM Tris HCl (pH 7.4), 5 mM EDTA.Na2' and 0.2 mM phenylmethyl sulfonylfluoride at 4 0 C using a Teflon' and glass homogenizer. Mitochondrial, nuclear, and plasma membrane components were removed by an initial centrifugation at 10,000 X g for 10 min, and the microsomal fraction was then pelleted by centrifugation at 100,000 X g for 60 min in a 10 X 10 ml angle rotor using an SS65 centrifuge (MSE, Loughborough, UK). The pellet was washed once by resuspension in the same volume of buffer, followed by recentrifugation.The washed pellet was resuspended in buffer with gentle hand homogenization and the final volume adjusted to a final concentration of 1.4 mg/ml protein. Microsomal fractions were isolated from pooled (n = 8 to 10) replicate organ cultures by a similar procedure. Because of the small amounts of tissue available for these studies, microsomal enzyme markers were not routinely measured. Consequently, a degree of mitochondrial contamination of these preparations cannot be discounted. . Lung tissue for organ culture (15) was first ?iced int~ l-rnm' pieces and washed free of blood cells WIth Hanks balanced salt solution (HBSS). Organ cultures were established by placing approximately 20 of these lung pieces on a collagen sponge (1 X 1 X 0.7 cm) in a 60-n.un-diame~er tissue culture dish. Serum-free RPMI 1640 medium containing antibiotics (100 J-Lg/ml kanamycin, 100 J-Lg/ml fungizone, 50 U/ml penicillin, 50 nglml streptomycin, and 5 U/ml nystatin) was added and subsequently replaced every 2 days. Cultures were maintained at 3r C under 5 % CO 2 for up to 5 days, and were placed on a rocking platform ~t. 6 oscillations/min. The hormones dexamethasone and tri-iodothyronine were added to indicated cultures at concentrations of 10-6 and 10-7 M, respectively. Hormones were added at time of culture and at each change of medium. This combination of high doses of hormones was chosen simply as a tool to manipulate the rate of PC synthesis in culture (19,20), and not to investigate the mechanism of hormone action. Parallel cultures were supplemented throughout the culture period with exogenous fatty acids. Palmitate and oleate, as the Na salts, were complexed to bovine serum albumin (essentially fatty acid free) to give final concentrations in the cUl~re medium of 200 J-LM fatty acid and 3 % (wt/vol) albumin. Control cultures received either no additions (hormones) or albumin alone (fatty acids).
Caesar, McElroy, Kelly et al.: Phosphatidylcholine Remodeling by Human Fetal Lung
PC Synthesis in Isolated Microsomes Microsomal preparations (100 ~l) were incubated at 37° C for up to 3 h in homogenization buffer with additions of 20 mM MgCb and 0.15 ~Ci cytidine 5'-diphospho-[methyl14C]choline (CDP:[I4C]choline) (50 Ci/mol). Incubations were terminated by the addition of 2 ml methanol, followed by PC extraction as described below. PC Synthesis in Organ Cultures of Human Fetal Lung PC synthesis by organ cultures of human fetal lung was measured by the incorporation of [methyl-t'Clcholine chloride ([I4C]choline). Fresh medium (5 ml) was added to the cultures, together with 100 ~l of (14C]choline (58 Ci/mol) to give final concentrations and specific radioactivities in the medium of 400 nCi/ml and 16.3 Ci/mol, respectively. Cultures were incubated with label at 37° C in 5 % CO 2 for either 3 h to measure rates of synthesis or 18 h to determine the equilibrium pattern of synthesis of PC molecular species. After incubation with radiolabel, cultures were extensively washed, suspended in HBSS, and extracted for phospholipid. Analysis of PC Molecular Species Fresh lung tissue homogenates were prepared (100 mg in 1 ml HBSS) using an IKA Ultra-Thrrax homogenizer (Janke & Kunkel, Stanfen, Germany) while organ culture homogenates were prepared by sonication using an MSE Soniprep 150 (MSE). Internal standards added were PCI4:0/14:0 (40 nmol) and (3H]PCI6:0/16:0 (77 mCi/~mol, 50 nCi) dissolved in 25 ~l trifluoroethanol. Total lipids were extracted with chloroform and methanol, dried under N2 , and dissolved in chloroform. The PC fraction from this extract was purified using disposable aminopropyl-Bondelut columns as described previously (15), dried under Nz, and dissolved in trifluoroethanol. PC was resolved by HPLC on a 25 em X 4.6 rom Apex II ODS column (Jones Chromatography, Hengoed, Mid Glamorgan, Wales, UK) thermostatically maintained at 50° C (21). The mobile phase of 40 mM choline chloride in methanol/water (37:3, vol/vol) was delivered at 1 ml/min by a Waters 6000A pump (Waters, Milford, MA). PC mass content of eluted peaks was determined by postcolumn fluorescent-derivatization using 1,6-diphenyl-l,3,5hexatriene (DPH) (21). This technique provided a linear measure of PC concentration relative to the internal standard PCI4:0/14:0 over the range of 0.5 to 100 nmol, irrespective of the degree of acyl unsaturation. The ultraviolet (UV) absorbance of eluted PC species, due to the number of double carbon:carbon bonds, was simultaneously measured using an HPLC UV detector (21). A comparison of UV and fluorescence responses was used to construct an unsaturation index, which was then used for the routine identification of individual PC species. The radioactivity content of eluted PC peaks was determined as previously described, using a dualchannel HPLC radioactivity monitor after postcolumn admixture with scintillant (15). Isolation of Surfactant and Residual Fractions Subcellular fractions were prepared from organ cultures by a density gradient centrifugation procedure designed to isolate surfactant-associated (primarily lamellar body) and residual membrane fractions of culture explants (22). Explants were homogenized in 5 ml HBSS using an IKA Ultra-Turrax
365
homogenizer (Janke & Kunkel), layered over 6 ml of 0.75 M sucrose, 0.145 M NaCI, and centrifuged at 160,000 X g for 30 min in a 6 X 14 ml swingout rotor. The surfactant fraction was recovered from the interface between sucrose and saline layers and the residual fraction from the pelleted material. DNA and Protein Assay DNA concentration was measured by an automated fluorescence binding technique using the fluorochrome Hoechst 33258 (23). Protein was measured using the phenol:Ciocalteau reagent (24). Statistics Significances between experimental groups were analyzed by an unpaired Student's t test. Results are presented as mean ± SEM.
Results Analysis of PC Molecular Species Analyses of PC composition and synthesis are presented in terms of individual molecular species, resolved by HPLC on the basis of their combinations offatty acyl groups at the sn-l and sn-2 positions. A typical chromatogram of intact PC species isolated from human fetal lung (15 wk of gestation) is illustrated in Figure 1. The fluorescence trace provides a quantitative measure of concentration, whereas the UV absorption response is determined both by concentration and degree of acyl unsaturation. Comparison of fluorescence and UV absorption responses provided a convenient means to confirm PC species identities on a routine basis. For analyses of the incorporations of radiolabeled precursors, the same
6 7
Q)
2
III C
5
8
o
Fluorescence 9
a.
III
~
o c
'" Vi
o
10
20
30
40
50
60
Retention time (min)
Figure 1. Reverse-phase high performance liquid chromatography of individual molecular species of phosphatidylcholine (PC) from human fetal lung. PC from human fetal lung (15 wk of gestation) was resolved by high performance liquid chromatography and detected by ultraviolet absorbance at 205 nm (lower trace) and by quantitative postcolumn fluorescence derivatization using 1,6-diphenyl-l,3,5-hexatriene (DPH) (upper trace). The ratio of absorbance to fluorescence responses gavea routine indication of species composition. Identified eluted peaks are: 1 = PC14:0/14:0 (internal standard); 2 = sphingomyelin; 3 = PC14:0/16:0; 4 = PC16:0/16:1; 5 = PC16:0/18:2; 6 = PC16:0/16:0; 7 = PC16:0/18:1; 8 = PC18:0/18:2 + PC18:1/18:1; 9 = PC16:0/18:0.
366
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 5 1991
60
'"
c
~n
50
o
e-~ U
x
U
U -e-
40
c c .-
",,,,
c ....
0
0
U
20
20
uc> ~" u,
20
~a.
10
~o
0
g
0 +-'----"'=--+-l-0 14:0/16:0 16:0/16:1 16:0/18:2 16:0/16:0 16:0/18:1 18:0/18:2
~
Phosphatidylcholine molecular species
;!:E .• -0
2S~
u
a.
10
10
2
2.5
3
Incubation time (h)
Figure 2. Incorporation of cytidine 5'-diphospho-[methyl-14C]choline (CDP:[I4C]choline) into PC by microsomes isolated from human fetal lung.
HPLC resolution was combined with a flow-through radioactivity monitor. Characterization of Microsomal Incorporation of CDP:(14C]Choline Previous investigators have established optimal conditions to measure the activity of the microsomal enzyme cholinephosphotransferase (1,2). In these studies, we have used similar conditions to convert endogenous microsomal diacylglycerol to PC and monitored this process by the incorporation of CDP:[I4C]choline. The time course of this incorporation (Figure 2) was not linear, reaching a maximum at approximately 3 h. The patterns of CDP:[I4C]choline incorporations after 1- and 3-h incubation were, however, essentially identical. In addition, this incorporation pattern did not change using a protocol of a 1-h pulse of CDP:[J4C]choline followed by a chase incubation of as long as 10 h with excess cold CDP:choline (results not shown). Consequently, results presented below are for 3-h incubations with CDP:[J4C] choline as these provided greater radiolabel incorporations for analysis. Analysis of Human Fetal Lung PC Synthesis The pattern of CDP:[J4C]choline incorporation into PC by microsomes isolated from human fetal lung is shown in Figure 3. The major species synthesized were PC16:0/18:2 and PC16:0/18:1, together with appreciable amounts of PC16: 0/16:0, PC14:0/16:0, PC16:0/16:1, and PC18:0/18:2. In all cases, these six species represented more than 90 % of the total radioactivity incorporated. Comparison of this incorporation pattern in vitro with the PC species distribution in whole lung (Figure 3) gave an indication of which molecular species were modified by remodeling mechanisms between their synthesis de novo on the microsomes and their final distribution within the lung. De novo synthesis, when compared with mass distribution, was characterized by greater microsomal incorporations of CDP:[J4C]choline into the linoleate-containing species PC16:0/18:2 and PC18:0/18:2. Conversely, the fractional contents of PC16:0/16:0 and PC16:0/16:1 masses were greater than their fractional rates of de novo synthesis. This result suggests that linoleatecontaining PC species synthesized initially are subsequently remodeled to more saturated PC species. Comparison of the
(;
:3
eo
1.5
In
s:
o
0.5
s: a. 0
c
0
>-
:;:;
o
0
U
-0
0-
30
m .,
.s s:
30
';'30
.- 0
00. s:
40
.s::4O
'-I 0
50
050
c;
0
~
Figure 3. PC remodeling by human fetal lung. The pattern of incorporation of CDP:[14C]choline into PC molecular species by microsomes isolated from human fetal lung (open bars) was compared with the species distribution of PC content from the same lungs (hatched bars).
fractional synthesis de novo (14.4 %) and concentration in vivo (27.8%) suggests that in immature human fetal lung approximately 50% of lung PC16:0/16:0 was synthesized directly on the microsomes, with the remainder being provided by acyl remodeling mechanisms. Differentiation of Human Fetal Lung in Organ Culture The changes in epithelial cell shape and acquisitions of apical microvilli and intracellular lamellar bodies characteristic of the spontaneous maturation of immature lung explants maintained under our organ culture conditions have been described previously (15). Additionally, some changes of PC metabolism by fetal lung in organ culture are also characteristic of a maturation process. PC synthesis by human fetal lung increases with time in organ culture, and this is further stimulated in response to glucocorticoid and thyroid hormones (17-20). Consequently, we analyzed the effect of continuous exposure to 10-6 M dexamethasone and 10-7 M triiodothyronine for 4 days on the rate of PC synthesis by cultures of human fetal lung. The incorporation rate of [I4C]choline over 3 h into total PC was significantly stimulated (P < 0.05) by this hormonal exposure from 3 to 6 days in culture, the response being maximal at 4 days. The mean hormonal stimulation of PC synthesis at this time point for six different fetal lung preparations was 346 ± 81%. In contrast to some previous reports, however (20), the morphologic maturation of human fetal lung maintained in our culture conditions was not hormone dependent. Morphometric analysis demonstrated that following 4 days in culture 40 ± 2 % (n = 5) and 41 ± 3 % (n = 5) of identifiable cells contained lamellar bodies for control and hormone-treated cultures, respectively. Equivalent numbers of lamellar bodies in sections of each of these presumptive type II pneumocytes were 21.5 ± 3.0% and 16.4 ± O. %. Patterns of PC Synthesis by Explants of Human Fetal Lung The distribution of [t4C]choline incorporated into PC in surfactant and residual fractions was determined after isolation of the relevant fractions by sucrose density gradient centrifugation. Expressed as a percentage of the total radioactivity recovered in PC in both fractions, [I4C]choline
Caesar, McElroy, Kelly et al.: Phosphatidylcholine Remodeling by Human Fetal Lung
367
TABLE 1
The effect of 10-6 M dexamethasone and 10- 7 M tri-iodothyronine on the incorporation of p4C]choline into phosphatidylcholine (PC) isolated from organ cultures of human fetal lung: incorporation of [14C]choline into PC molecular species (% total) * Control Cultures Molecular Species
PCI4:0/16:0 PCI6:0/16: 1 PCI6:0/18:2 PCI6:0/16:0 PCI6:0/18: 1 PCI8:0/18:2 } PCI8: 1/18: 1
Whole Culture
6.4 11.7 16.6 16.6 36.1 12.2
Hormone-treated Cultures
Surfactant Fraction
± 0.7 ± 0.6 ± 1.4 ± 1.7 ± 3.2 ± 0.4
8.8 15.0 15.4 22.8 27.0 10.8
Residual Fraction
± 0.7 ± 0.6 ± 0.9 ± 1.1 ± 0.8 ± 0.6
7.7 12.4 16.8 20.2 30.6 12.2
Whole Culture
± 0.3 ± 0.6 ± 1.0 ± 1.1 ± 1.0 ± 0.6
9.8 13.6 14.8 18.6 31.3 11.8
Surfactant Fraction
± 0.8 ± 1.5 ± 1.7 ± 0.7 ± 1.2 ± 1.4
9.2 13.6 16.4 21.0 27.6 13.1
± 0.4 ± 1.5 ± 1.0 ± 0.5 ± 0.3 ± 0.7
Residual Fraction
8.5 10.7 17.7 20.3 29.0 13.7
± 0.4 ± 0.5 ± 1.0 ± 0.7 ± 0.3 ± 0.5
* Cultures were incubated after 4 days in culture with 400 nCi/m1 [l4C)choline for 18 h. Surfactant and residual fractions were recovered by discontinuous gradient centrifugation. Results are mean ± SEM for six fetal preparations. incorporation into surfactant was 16.6 ± 3.5% (n = 6) for control and 28.1 ± 9.9% (n = 5) for hormone-treated cultures. This difference, although marked, was not statistically significant. In contrast to the stimulation of total PC synthesis, there was no comparable hormonal stimulation of the fractional synthesis of PCI6:0/16:0 (Table 1). There were no differences in the synthetic patterns of PC molecular species in whole cultures or in surfactant or residual fractions between the control and hormone-treated groups. In addition, the effects of exposure to a concentration range of dexamethasone from 10-9 to 10-6 M on the synthesis of molecular species of PC by human fetal lung explants was determined. However, no effect of dexamethasone was found at any concentration tested on the pattern of incorporation of [14C]choline into PC species over 18 h (results not shown). We believed that, in the absence of any hormonal effect, it was important to demonstrate that the pattern of synthesis of PC molecular species could be altered by an appropriate manipulation of the culture conditions. The results in Table 2 clearly demonstrate that treatment with fatty acid-albumin mixtures could readily alter the pattern of synthesis of PC molecular species by cultured human fetal lung. The addition of albumin alone caused no significant change in this
pattern when compared with the control group in Table 1. For all three tissue fractions, PCI6:0/18:1 remained the predominant species synthesized. Incubation with 200 ~M palmitate-albumin caused a small but significant stimulation of the fractional incorporation of (14C]choline into PCI6:0/ 16:0 at the expense of PCI6:0/18:1 and PCI8:0/18:2. There was no selective increase ofPCI6:0/16:0 synthesis in the surfactant fraction; addition of palmitate caused proportional increases in the fractional incorporations into PCI6:0/16:0 in all three tissue fractions. Incubation with 200 ~M oleate-albumin resulted in the preferential incorporation of (14C]choline into PCI8:1I18:1 at the expense of all other PC species not containing oleate. PC18:1/18:1 co-eluted with PCI8:0/18:2 but was a minor component in cultures not supplemented with oleate. Incorporation of CDP:[I4C]Choline into PC by Microsomes from Organ Cultures of Human Fetal Lung In addition to measurements of PC synthesis by whole cultures, parallel analyses of microsomal synthesis were performed de novo. Table 3 details the patterns of incorporation of CDP:[14C]choline over 3 h into PC by microsomal fractions from control, hormone-treated, and palmitate-treated
TABLE 2
The effects of exogenous fatty acid additions on the synthesis of PC molecular species by organ cultures of human fetal lung: incorporation of [14C]choline into PC molecular species (% total)* Palmitate
Albumin Molecular Species
PCI4:0/16:0 PCI6:0/16: 1 PCI6:0/18:2 PCI6:0/16:0
PCI6:0/18:1 PCI8:0/18:2 }
PCI8:1/18:1
Whole Culture
5.3 10.7 17.1 19.0 35.4 12.0
± 0.3 ± 0.7 ± 0.7 ± 0.5 ± 0.6 ± 0.7
Surfactant Fraction
6.2 9.3 16.0 26.8 31.5 10.1
± 0.3 ± 1.4 ± 0.6 ± 1.9 ± 1.1 ± 1.2
Residual Fraction
5.2 9.9 17.5 21.4 34.3 11.4
± 0.3 ± 0.8 ± 0.5 ± 0.5 ± 0.4 ± 0.3
Whole Culture
4.3 9.6 19.5 28.4 31.4 6.2
± 0.1 ± 0.5 ± 1.1 ± O.4t ± 0.7 ± 0.3t
Surfactant Fraction
4.2 10.6 14.4 38.8 27.6 4.3
± 0.5 ± 0.3 ± 1.7 ± 2.5t ± 0.7 ± 0.7t
Oleate Residual Fraction
4.6 10.0 18.0 29.4 31.2 6.7
± 0.3 ± 0.5 ± 0.7 ± 0.3t ± 0.7 ± 0.6t
Whole Culture
3.6 7.7 11.7 12.1 32.2 32.4
* Cultures were incubated for 4 days with albumin (3% wt/vol) or fatty acid (200 IlM)-albumin, and then with [14C]choline for 18 h. Results are mean for six fetal preparations. t p < 0.05 versus control cultures. p < 0.01 versus control cultures.
*
± 1.2 ± 1.4 ± 0.2 ± 0.6* ± 1.1 ± 1.4* ± SEM
368
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 5 1991
TABLE 3
50
PC synthesis by lung microsomes: incorporation of CDP:[l4C]choline into PC molecular species (% total)* Molecular Species
Hormone-treated Cultures
Control Cultures
Palmitate-treated Cultures
Cll •
The molecular specificity of phosphatidylcholine (PC) synthesis by the de novo pathway in postmortem samples of human fetal lung (15 to 20 wk of gestation) was determined from the incorporation pattern in isolated microsomal preparations of CDP:[I4C]choline into individual molecular species of Pc. These analyses are based on the assumption that the molecular species composition of the pool of endogenous diacylglycerol used for PC synthesis by isolated microsomes reflects that of the authentic pool of diacylglycerol converted to PC by intact cells. Comparison of this microsomal incorporation pattern of radiolabel into PC with tissue PC composition suggested that even at this early stage of gestation 50% oflung dipalmitoyl PC was derived from synthesis de novo, with the remainder coming from acyl remodeling mechanisms. Analysis of PC synthesis de novo by organ cultures of human fetal lung showed that these acyl remodeling mechanisms were lost in culture. Despite evidence for differentiation of type II alveolar epithelial cells in culture, equilibrium labeling of PC with [I4C]choline over 18 h resulted in a progressive decline in fractional incorporation into dipalmitoyl PC with time in culture. By 4 days in culture, this value was no different from the fractional incorporation of CDP: [I4C]choline into microsomal PC in vitro over 3 h. The pattern of PC synthesized was not altered when total PC synthesis was stimulated by exposure of cultures to dexamethasone and tri-iodothyronine but was readily manipulated by exposure to exogenous fatty acids. These results demonstrate for the first time the activity of PC acyl remodeling mechanisms in human fetal lung, well before the initiation of surfactant production. They support the conclusion that conditions that permit accelerated differentiation of type II epithelial cells from human fetal lung in organ culture do not necessarily support the spectrum of PC species synthesis characteristic of the differentiated state.
The spectrum of phosphatidylcholine (PC) molecular species synthesized by lung tissue is determined largely by the balance between the rate of synthesis de novo and the activities of a variety of acyI remodeling mechanisms. The pattern of PC species synthesized de novo by the microsomal enzyme cholinephosphotransferase reflects the molecular species composition of the substrate pool of diacylglycerol (1, 2); this enzyme in lung appears to possess little inherent sub(Received in original form December 7, 1990 and in revised form March 29, 1991) Address correspondence to: Dr. A. D. Postle, Child Health, Level G, Centre Block, Southampton General Hospital, Tremona Road, Southampton S09 4XY, United Kingdom. Abbreviations: cytidine 5!diphospho-[methyJ-l4C]choline, CDP: [14C]choline; [methyl-r'Cjcholine chloride, [14C]choline; disaturated phosphatidylcholine, DSPC; Hanks' balanced salt solution, HBSS; high performance liquid chromatography, HPLC; lyso-l-palmitoyl-sn-glycero-3-phosphatidylcholine, lysoPC; phosphatidylcholine, PC; sn-l-myristoyl sn-2-palmitoyl phosphatidylcholine, PCI4:0/16:0; dipaimitoylphosphatidylcholine, PCI6:0/16:0; sn-l-palmitoyl sn-2-palmitoleoyl phosphatidylcholine, PCI6:0/ 16:1; sn-l-palmitoylsn-2-oleoylphosphatidylcholine, PCI6:0/18:1; sn-l-palmitoyl sn-2-linoleoyl phosphatidyIchoIine, PCI6:0/18:2; sn-l-stearoyl sn-2linoleoyl phosphatidylcholine, PC18:0118:2; dioleoyl phosphatidylcholine, PCI8:1I18:1; ultraviolet, Uv, Am. J. Respir. Cell Mol. BioI. Vol. S. pp. 363-370, 1991
strate specificity (3, 4). The composition of newly formed PC is subsequently modified by acyl remodeling, involving sequential actions of phospholipase and acyltransferase enzymes (5). In addition to these synthetic mechanisms, which are common in principle to)llost cell types, many cells possess specialized selectiop-tnechanisms for the incorporation of specific PC species/into different subcellular fractions. One example of such 'selection is the assembly of surfactant phospholipids into lamellar body stores within type II pneumocytes (6). Recognition of the central role of dipalmitoyl PC (PCI6: 0/16:0) in lung surfactant function (7) has been the major stimulus for most studies of lung phospholipid metabolism. Consequently, studies of acyl remodeling in lung have been directed toward understanding mechanisms for enrichment of PCI6:0/I6:0 in lung parenchyma in general and more specifically within the surfactant PC pool. Analyses of the fractional contents of PCI6:0/I6:0 in lung tissue and microsomal PC with those synthesized de novo by microsomes isolated from adult rat lung (1) or rat lung type II pneumocytes (3) have concluded that up to 50% of total PCI6:0/I6:0 is provided directly and that the remainder is formed by remodeling mechanisms. An important central role for phospholipase A2 in this remodeling, converting
364
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 5 1991
I-palmitoyl-2-enoyl PC to I-palmitoyl-sn-glycero-3-phosphorylcholine (lysoPC), was suggest~d by Post ~d ~o-~~rk ers, who incubated rat type II cells with the specific inhibitor 4-bromophenacyl bromide, causing a significant decrease in the saturation of the PC formed (3). Proposed donor molecules to supply the palmitoyl groups subsequently reesterified into the sn-2 position oflysoPC are palmitoyl CoA (8) or a second molecule of lysoPC (9), catalyzed respectively by lysoPC acyltransferase and lysoPC:lysoPC acyltransferase. Comparisons of specific activities in adult rat type II pneumocytes (10) and in neonatal mouse lung (11) suggest that lysoPC acyltransferase may be the more important enzyme for PCI6:0/16:0 synthesis. One central assumption underlying much of this work is that PC acyl remodeling mechanisms in the lung are active principally to supply PCI6:0/16:0 for s~rfactant synthesis. For instance, the developmental expressions of acyltransferase enzymes in late gestation in rat and mouse fetal lung h~ve been compared with parameters of surfactant maturation (11). However, in adult rat lung, only 30% of total PCI6:0/ 16:0 is associated directly with surfactant PC pools (12, 13), and fetal human lung PC at 15 wk of gestation already contains 25 to 30% PCI6:0/16:0 well before the initiation of surfactant production (14, 15). Moreover, disaturated PC concentration in postmortem human fetal and neonatal lung samples throughout gestation and the neonatal period remained relatively constant except for an increase and subsequent decrease in the perinatal period (16). These observations imply both that PCI6:0/16:0 synthesis by lung is not solely directed toward surfactant synthesis and that mechanisms for PCI6:0/16:0 synthesis exist very early in human lung development. However, no analyses of PC acyl remodeling mechanisms comparable to studies in adult rat lung have yet been reported for human tissue, either adult or fetal. Consequently, the initial objective in this study was to characterize such mechanisms in postmortem samples of human fetal lung. The second part of this study addressed further the changes in the specificity of PC synthesis that accompany the differentiation in vitro of human fetal lung maintained in organ culture. This widely used model system for the study of lung maturation is characterized by the development of alveolar-like structures, the acquisition of intracellular lamellar bodies, and the expression of surfactant-specific apoproteins (17-20). Additionally, PC synthesis by organ cultures can be stimulated in a synergistic fashion by exposure to dexamethasone and tri-iodothyronine (17-20). By contrast, our previous results demonstrated clearly that these changes in organ culture are not necessarily accompanied by any increased fractional synthesis of PC16:0/16:0 (15). Indeed, the percentage incorporation of [I4C]choline into PCI6:0/16:0 decreased rather than increased with time in culture. In the present study, we have investigated remodeling mechanisms of PC synthesis in organ cultures of human fetal lung in response to a variety of hormonal and nutritional modifications.
Materials and Methods Chemicals Tissue culture reagents were supplied by Flow Laboratories
(Rickmansworth, UK), chemicals from Sigma Chemical Co. (Poole, UK), high performance liquid chromatography (HPLC) solvents by Rathburn Ltd. (Walkerburn, Scotland, UK), collagen sponge (Colgen Compres~) by ,!,hames Laboratories Ltd. (Middlesex, UK), and radiochemicals by Amersham International (Amersham, UK). Human Fetal Lung Microsomal Preparation and Cell Culture Lung tissue was obtained aseptically from 13 prostaglan~in induced human fetal abortions at 15 to 20 wk of gestation, in accordance with local ethical committee guidelines. A sample of lung (approximately 150 ~g) was re~o~ed and frozen in liquid nitrogen for PC analysis. The remammg lung tissue was used either for preparation of a microsomal fraction or was maintained in organ culture for up to 5 days. For isolation of microsomes, lung was homogenized in 4 vol of 0.145 M NaCI containing 50 mM Tris HCl (pH 7.4), 5 mM EDTA.Na2' and 0.2 mM phenylmethyl sulfonylfluoride at 4 0 C using a Teflon' and glass homogenizer. Mitochondrial, nuclear, and plasma membrane components were removed by an initial centrifugation at 10,000 X g for 10 min, and the microsomal fraction was then pelleted by centrifugation at 100,000 X g for 60 min in a 10 X 10 ml angle rotor using an SS65 centrifuge (MSE, Loughborough, UK). The pellet was washed once by resuspension in the same volume of buffer, followed by recentrifugation.The washed pellet was resuspended in buffer with gentle hand homogenization and the final volume adjusted to a final concentration of 1.4 mg/ml protein. Microsomal fractions were isolated from pooled (n = 8 to 10) replicate organ cultures by a similar procedure. Because of the small amounts of tissue available for these studies, microsomal enzyme markers were not routinely measured. Consequently, a degree of mitochondrial contamination of these preparations cannot be discounted. . Lung tissue for organ culture (15) was first ?iced int~ l-rnm' pieces and washed free of blood cells WIth Hanks balanced salt solution (HBSS). Organ cultures were established by placing approximately 20 of these lung pieces on a collagen sponge (1 X 1 X 0.7 cm) in a 60-n.un-diame~er tissue culture dish. Serum-free RPMI 1640 medium containing antibiotics (100 J-Lg/ml kanamycin, 100 J-Lg/ml fungizone, 50 U/ml penicillin, 50 nglml streptomycin, and 5 U/ml nystatin) was added and subsequently replaced every 2 days. Cultures were maintained at 3r C under 5 % CO 2 for up to 5 days, and were placed on a rocking platform ~t. 6 oscillations/min. The hormones dexamethasone and tri-iodothyronine were added to indicated cultures at concentrations of 10-6 and 10-7 M, respectively. Hormones were added at time of culture and at each change of medium. This combination of high doses of hormones was chosen simply as a tool to manipulate the rate of PC synthesis in culture (19,20), and not to investigate the mechanism of hormone action. Parallel cultures were supplemented throughout the culture period with exogenous fatty acids. Palmitate and oleate, as the Na salts, were complexed to bovine serum albumin (essentially fatty acid free) to give final concentrations in the cUl~re medium of 200 J-LM fatty acid and 3 % (wt/vol) albumin. Control cultures received either no additions (hormones) or albumin alone (fatty acids).
Caesar, McElroy, Kelly et al.: Phosphatidylcholine Remodeling by Human Fetal Lung
PC Synthesis in Isolated Microsomes Microsomal preparations (100 ~l) were incubated at 37° C for up to 3 h in homogenization buffer with additions of 20 mM MgCb and 0.15 ~Ci cytidine 5'-diphospho-[methyl14C]choline (CDP:[I4C]choline) (50 Ci/mol). Incubations were terminated by the addition of 2 ml methanol, followed by PC extraction as described below. PC Synthesis in Organ Cultures of Human Fetal Lung PC synthesis by organ cultures of human fetal lung was measured by the incorporation of [methyl-t'Clcholine chloride ([I4C]choline). Fresh medium (5 ml) was added to the cultures, together with 100 ~l of (14C]choline (58 Ci/mol) to give final concentrations and specific radioactivities in the medium of 400 nCi/ml and 16.3 Ci/mol, respectively. Cultures were incubated with label at 37° C in 5 % CO 2 for either 3 h to measure rates of synthesis or 18 h to determine the equilibrium pattern of synthesis of PC molecular species. After incubation with radiolabel, cultures were extensively washed, suspended in HBSS, and extracted for phospholipid. Analysis of PC Molecular Species Fresh lung tissue homogenates were prepared (100 mg in 1 ml HBSS) using an IKA Ultra-Thrrax homogenizer (Janke & Kunkel, Stanfen, Germany) while organ culture homogenates were prepared by sonication using an MSE Soniprep 150 (MSE). Internal standards added were PCI4:0/14:0 (40 nmol) and (3H]PCI6:0/16:0 (77 mCi/~mol, 50 nCi) dissolved in 25 ~l trifluoroethanol. Total lipids were extracted with chloroform and methanol, dried under N2 , and dissolved in chloroform. The PC fraction from this extract was purified using disposable aminopropyl-Bondelut columns as described previously (15), dried under Nz, and dissolved in trifluoroethanol. PC was resolved by HPLC on a 25 em X 4.6 rom Apex II ODS column (Jones Chromatography, Hengoed, Mid Glamorgan, Wales, UK) thermostatically maintained at 50° C (21). The mobile phase of 40 mM choline chloride in methanol/water (37:3, vol/vol) was delivered at 1 ml/min by a Waters 6000A pump (Waters, Milford, MA). PC mass content of eluted peaks was determined by postcolumn fluorescent-derivatization using 1,6-diphenyl-l,3,5hexatriene (DPH) (21). This technique provided a linear measure of PC concentration relative to the internal standard PCI4:0/14:0 over the range of 0.5 to 100 nmol, irrespective of the degree of acyl unsaturation. The ultraviolet (UV) absorbance of eluted PC species, due to the number of double carbon:carbon bonds, was simultaneously measured using an HPLC UV detector (21). A comparison of UV and fluorescence responses was used to construct an unsaturation index, which was then used for the routine identification of individual PC species. The radioactivity content of eluted PC peaks was determined as previously described, using a dualchannel HPLC radioactivity monitor after postcolumn admixture with scintillant (15). Isolation of Surfactant and Residual Fractions Subcellular fractions were prepared from organ cultures by a density gradient centrifugation procedure designed to isolate surfactant-associated (primarily lamellar body) and residual membrane fractions of culture explants (22). Explants were homogenized in 5 ml HBSS using an IKA Ultra-Turrax
365
homogenizer (Janke & Kunkel), layered over 6 ml of 0.75 M sucrose, 0.145 M NaCI, and centrifuged at 160,000 X g for 30 min in a 6 X 14 ml swingout rotor. The surfactant fraction was recovered from the interface between sucrose and saline layers and the residual fraction from the pelleted material. DNA and Protein Assay DNA concentration was measured by an automated fluorescence binding technique using the fluorochrome Hoechst 33258 (23). Protein was measured using the phenol:Ciocalteau reagent (24). Statistics Significances between experimental groups were analyzed by an unpaired Student's t test. Results are presented as mean ± SEM.
Results Analysis of PC Molecular Species Analyses of PC composition and synthesis are presented in terms of individual molecular species, resolved by HPLC on the basis of their combinations offatty acyl groups at the sn-l and sn-2 positions. A typical chromatogram of intact PC species isolated from human fetal lung (15 wk of gestation) is illustrated in Figure 1. The fluorescence trace provides a quantitative measure of concentration, whereas the UV absorption response is determined both by concentration and degree of acyl unsaturation. Comparison of fluorescence and UV absorption responses provided a convenient means to confirm PC species identities on a routine basis. For analyses of the incorporations of radiolabeled precursors, the same
6 7
Q)
2
III C
5
8
o
Fluorescence 9
a.
III
~
o c
'" Vi
o
10
20
30
40
50
60
Retention time (min)
Figure 1. Reverse-phase high performance liquid chromatography of individual molecular species of phosphatidylcholine (PC) from human fetal lung. PC from human fetal lung (15 wk of gestation) was resolved by high performance liquid chromatography and detected by ultraviolet absorbance at 205 nm (lower trace) and by quantitative postcolumn fluorescence derivatization using 1,6-diphenyl-l,3,5-hexatriene (DPH) (upper trace). The ratio of absorbance to fluorescence responses gavea routine indication of species composition. Identified eluted peaks are: 1 = PC14:0/14:0 (internal standard); 2 = sphingomyelin; 3 = PC14:0/16:0; 4 = PC16:0/16:1; 5 = PC16:0/18:2; 6 = PC16:0/16:0; 7 = PC16:0/18:1; 8 = PC18:0/18:2 + PC18:1/18:1; 9 = PC16:0/18:0.
366
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 5 1991
60
'"
c
~n
50
o
e-~ U
x
U
U -e-
40
c c .-
",,,,
c ....
0
0
U
20
20
uc> ~" u,
20
~a.
10
~o
0
g
0 +-'----"'=--+-l-0 14:0/16:0 16:0/16:1 16:0/18:2 16:0/16:0 16:0/18:1 18:0/18:2
~
Phosphatidylcholine molecular species
;!:E .• -0
2S~
u
a.
10
10
2
2.5
3
Incubation time (h)
Figure 2. Incorporation of cytidine 5'-diphospho-[methyl-14C]choline (CDP:[I4C]choline) into PC by microsomes isolated from human fetal lung.
HPLC resolution was combined with a flow-through radioactivity monitor. Characterization of Microsomal Incorporation of CDP:(14C]Choline Previous investigators have established optimal conditions to measure the activity of the microsomal enzyme cholinephosphotransferase (1,2). In these studies, we have used similar conditions to convert endogenous microsomal diacylglycerol to PC and monitored this process by the incorporation of CDP:[I4C]choline. The time course of this incorporation (Figure 2) was not linear, reaching a maximum at approximately 3 h. The patterns of CDP:[I4C]choline incorporations after 1- and 3-h incubation were, however, essentially identical. In addition, this incorporation pattern did not change using a protocol of a 1-h pulse of CDP:[J4C]choline followed by a chase incubation of as long as 10 h with excess cold CDP:choline (results not shown). Consequently, results presented below are for 3-h incubations with CDP:[J4C] choline as these provided greater radiolabel incorporations for analysis. Analysis of Human Fetal Lung PC Synthesis The pattern of CDP:[J4C]choline incorporation into PC by microsomes isolated from human fetal lung is shown in Figure 3. The major species synthesized were PC16:0/18:2 and PC16:0/18:1, together with appreciable amounts of PC16: 0/16:0, PC14:0/16:0, PC16:0/16:1, and PC18:0/18:2. In all cases, these six species represented more than 90 % of the total radioactivity incorporated. Comparison of this incorporation pattern in vitro with the PC species distribution in whole lung (Figure 3) gave an indication of which molecular species were modified by remodeling mechanisms between their synthesis de novo on the microsomes and their final distribution within the lung. De novo synthesis, when compared with mass distribution, was characterized by greater microsomal incorporations of CDP:[J4C]choline into the linoleate-containing species PC16:0/18:2 and PC18:0/18:2. Conversely, the fractional contents of PC16:0/16:0 and PC16:0/16:1 masses were greater than their fractional rates of de novo synthesis. This result suggests that linoleatecontaining PC species synthesized initially are subsequently remodeled to more saturated PC species. Comparison of the
(;
:3
eo
1.5
In
s:
o
0.5
s: a. 0
c
0
>-
:;:;
o
0
U
-0
0-
30
m .,
.s s:
30
';'30
.- 0
00. s:
40
.s::4O
'-I 0
50
050
c;
0
~
Figure 3. PC remodeling by human fetal lung. The pattern of incorporation of CDP:[14C]choline into PC molecular species by microsomes isolated from human fetal lung (open bars) was compared with the species distribution of PC content from the same lungs (hatched bars).
fractional synthesis de novo (14.4 %) and concentration in vivo (27.8%) suggests that in immature human fetal lung approximately 50% of lung PC16:0/16:0 was synthesized directly on the microsomes, with the remainder being provided by acyl remodeling mechanisms. Differentiation of Human Fetal Lung in Organ Culture The changes in epithelial cell shape and acquisitions of apical microvilli and intracellular lamellar bodies characteristic of the spontaneous maturation of immature lung explants maintained under our organ culture conditions have been described previously (15). Additionally, some changes of PC metabolism by fetal lung in organ culture are also characteristic of a maturation process. PC synthesis by human fetal lung increases with time in organ culture, and this is further stimulated in response to glucocorticoid and thyroid hormones (17-20). Consequently, we analyzed the effect of continuous exposure to 10-6 M dexamethasone and 10-7 M triiodothyronine for 4 days on the rate of PC synthesis by cultures of human fetal lung. The incorporation rate of [I4C]choline over 3 h into total PC was significantly stimulated (P < 0.05) by this hormonal exposure from 3 to 6 days in culture, the response being maximal at 4 days. The mean hormonal stimulation of PC synthesis at this time point for six different fetal lung preparations was 346 ± 81%. In contrast to some previous reports, however (20), the morphologic maturation of human fetal lung maintained in our culture conditions was not hormone dependent. Morphometric analysis demonstrated that following 4 days in culture 40 ± 2 % (n = 5) and 41 ± 3 % (n = 5) of identifiable cells contained lamellar bodies for control and hormone-treated cultures, respectively. Equivalent numbers of lamellar bodies in sections of each of these presumptive type II pneumocytes were 21.5 ± 3.0% and 16.4 ± O. %. Patterns of PC Synthesis by Explants of Human Fetal Lung The distribution of [t4C]choline incorporated into PC in surfactant and residual fractions was determined after isolation of the relevant fractions by sucrose density gradient centrifugation. Expressed as a percentage of the total radioactivity recovered in PC in both fractions, [I4C]choline
Caesar, McElroy, Kelly et al.: Phosphatidylcholine Remodeling by Human Fetal Lung
367
TABLE 1
The effect of 10-6 M dexamethasone and 10- 7 M tri-iodothyronine on the incorporation of p4C]choline into phosphatidylcholine (PC) isolated from organ cultures of human fetal lung: incorporation of [14C]choline into PC molecular species (% total) * Control Cultures Molecular Species
PCI4:0/16:0 PCI6:0/16: 1 PCI6:0/18:2 PCI6:0/16:0 PCI6:0/18: 1 PCI8:0/18:2 } PCI8: 1/18: 1
Whole Culture
6.4 11.7 16.6 16.6 36.1 12.2
Hormone-treated Cultures
Surfactant Fraction
± 0.7 ± 0.6 ± 1.4 ± 1.7 ± 3.2 ± 0.4
8.8 15.0 15.4 22.8 27.0 10.8
Residual Fraction
± 0.7 ± 0.6 ± 0.9 ± 1.1 ± 0.8 ± 0.6
7.7 12.4 16.8 20.2 30.6 12.2
Whole Culture
± 0.3 ± 0.6 ± 1.0 ± 1.1 ± 1.0 ± 0.6
9.8 13.6 14.8 18.6 31.3 11.8
Surfactant Fraction
± 0.8 ± 1.5 ± 1.7 ± 0.7 ± 1.2 ± 1.4
9.2 13.6 16.4 21.0 27.6 13.1
± 0.4 ± 1.5 ± 1.0 ± 0.5 ± 0.3 ± 0.7
Residual Fraction
8.5 10.7 17.7 20.3 29.0 13.7
± 0.4 ± 0.5 ± 1.0 ± 0.7 ± 0.3 ± 0.5
* Cultures were incubated after 4 days in culture with 400 nCi/m1 [l4C)choline for 18 h. Surfactant and residual fractions were recovered by discontinuous gradient centrifugation. Results are mean ± SEM for six fetal preparations. incorporation into surfactant was 16.6 ± 3.5% (n = 6) for control and 28.1 ± 9.9% (n = 5) for hormone-treated cultures. This difference, although marked, was not statistically significant. In contrast to the stimulation of total PC synthesis, there was no comparable hormonal stimulation of the fractional synthesis of PCI6:0/16:0 (Table 1). There were no differences in the synthetic patterns of PC molecular species in whole cultures or in surfactant or residual fractions between the control and hormone-treated groups. In addition, the effects of exposure to a concentration range of dexamethasone from 10-9 to 10-6 M on the synthesis of molecular species of PC by human fetal lung explants was determined. However, no effect of dexamethasone was found at any concentration tested on the pattern of incorporation of [14C]choline into PC species over 18 h (results not shown). We believed that, in the absence of any hormonal effect, it was important to demonstrate that the pattern of synthesis of PC molecular species could be altered by an appropriate manipulation of the culture conditions. The results in Table 2 clearly demonstrate that treatment with fatty acid-albumin mixtures could readily alter the pattern of synthesis of PC molecular species by cultured human fetal lung. The addition of albumin alone caused no significant change in this
pattern when compared with the control group in Table 1. For all three tissue fractions, PCI6:0/18:1 remained the predominant species synthesized. Incubation with 200 ~M palmitate-albumin caused a small but significant stimulation of the fractional incorporation of (14C]choline into PCI6:0/ 16:0 at the expense of PCI6:0/18:1 and PCI8:0/18:2. There was no selective increase ofPCI6:0/16:0 synthesis in the surfactant fraction; addition of palmitate caused proportional increases in the fractional incorporations into PCI6:0/16:0 in all three tissue fractions. Incubation with 200 ~M oleate-albumin resulted in the preferential incorporation of (14C]choline into PCI8:1I18:1 at the expense of all other PC species not containing oleate. PC18:1/18:1 co-eluted with PCI8:0/18:2 but was a minor component in cultures not supplemented with oleate. Incorporation of CDP:[I4C]Choline into PC by Microsomes from Organ Cultures of Human Fetal Lung In addition to measurements of PC synthesis by whole cultures, parallel analyses of microsomal synthesis were performed de novo. Table 3 details the patterns of incorporation of CDP:[14C]choline over 3 h into PC by microsomal fractions from control, hormone-treated, and palmitate-treated
TABLE 2
The effects of exogenous fatty acid additions on the synthesis of PC molecular species by organ cultures of human fetal lung: incorporation of [14C]choline into PC molecular species (% total)* Palmitate
Albumin Molecular Species
PCI4:0/16:0 PCI6:0/16: 1 PCI6:0/18:2 PCI6:0/16:0
PCI6:0/18:1 PCI8:0/18:2 }
PCI8:1/18:1
Whole Culture
5.3 10.7 17.1 19.0 35.4 12.0
± 0.3 ± 0.7 ± 0.7 ± 0.5 ± 0.6 ± 0.7
Surfactant Fraction
6.2 9.3 16.0 26.8 31.5 10.1
± 0.3 ± 1.4 ± 0.6 ± 1.9 ± 1.1 ± 1.2
Residual Fraction
5.2 9.9 17.5 21.4 34.3 11.4
± 0.3 ± 0.8 ± 0.5 ± 0.5 ± 0.4 ± 0.3
Whole Culture
4.3 9.6 19.5 28.4 31.4 6.2
± 0.1 ± 0.5 ± 1.1 ± O.4t ± 0.7 ± 0.3t
Surfactant Fraction
4.2 10.6 14.4 38.8 27.6 4.3
± 0.5 ± 0.3 ± 1.7 ± 2.5t ± 0.7 ± 0.7t
Oleate Residual Fraction
4.6 10.0 18.0 29.4 31.2 6.7
± 0.3 ± 0.5 ± 0.7 ± 0.3t ± 0.7 ± 0.6t
Whole Culture
3.6 7.7 11.7 12.1 32.2 32.4
* Cultures were incubated for 4 days with albumin (3% wt/vol) or fatty acid (200 IlM)-albumin, and then with [14C]choline for 18 h. Results are mean for six fetal preparations. t p < 0.05 versus control cultures. p < 0.01 versus control cultures.
*
± 1.2 ± 1.4 ± 0.2 ± 0.6* ± 1.1 ± 1.4* ± SEM
368
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 5 1991
TABLE 3
50
PC synthesis by lung microsomes: incorporation of CDP:[l4C]choline into PC molecular species (% total)* Molecular Species
Hormone-treated Cultures
Control Cultures
Palmitate-treated Cultures
Cll •
