- Email: [email protected]
Lysophosphatidylcholine uptake and metabolism in the adult rabbit lung
328
Bwchimico
et Blophysm
Actu, 961 (1988) 328-336 Elsevier
BBA 52883
Lysophosphatidylcholine
uptake and metabolism in the adult rabbit lung
Steven R. Seidner, Alan H. Jobe, Machiko Ikegami, Andrea Anathalie Priestley and Lynda Ruffini Department
of Pedintrics, Harbor- UCLA Medical Center, UCLA School of Medune. (Received
Key words:
Lysophosphatidylcholine;
1 February
Pettenazzo,
Torrance, CA (U.S.A.)
1988)
Phospholipid
metabolism;
(Lung);
(Rabbit)
Tracer quantities of 3H-labeled IysoPC and 32P-labeled natural rabbit surfactant were given intratracbeally via a broncboscope and [ 14C]palmitate was given intravenously to 25 rabbits with labeled PC and IysoPC measured in the alveolar wash, lung homogenate, lamellar bodies and microsomes at five times from 10 min to 6 h after tracheal injection. Surprisingly, only 31% of the administered IysoPC remained in its original form in the total lungs (alveolar wash + lung homogenate) by 10 min, of which 77% was in the alveolar wash. Meanwhile, by 10 min an additional 37% was already converted to PC, of which more than 98% was in the lung homogenate. LysoPC continued to be rapidly and efficiently converted to PC, with 62% conversion measured at 3 h. The converted IysoPC initially appeared with high specific activity in microsomes, then in lamellar bodies, and finally in the alveolar wash. The intravascular palmitate labeled lung PC had similar specific activity-time profiles in the s&cellular fractions, while intratracheally administered natural rabbit surfactant had a constantly low specific activity in microsomes and much higher specific activities in lamellar bodies and alveolar wash. Another 25 rabbits received intratracheal IysoPC labeled in both the choline and palmitate moieties and then were studied from 1 to 24 h after tracheal injection. The ratio of the palmitate to choline labels indicated uptake and conversion to PC primarily by direct acylation rather than transacylation and by intact reuptake and conversion rather than breakdown and resynthesis. LysoPC is an attractive ‘metabolic probe’ of surfactant metabolism which undergoes very rapid and efficient intracellular conversion to PC via a subcellular pathway that parallels the remodeling and de novo synthetic pathways.
Introduction Reutilization of the various phospholipids of pulmonary surfactant by the type II cell occurs to some degree in the mature and to a greater degree in the developing lung [l-4]. The histochemical studies of Williams with cationic and MPA ferri-
Abbreviations:
PC, phosphatidylcholine;
DPPC,
dipalmitoyl-
phosphatidylcholine. Correspondence: Harbor-UCLA rance,
CA 90509,
0005-2760/88/$03.50
S.R.
Seidner,
Medical
Center,
Department 1000
of
W. Carson
Pediatrics, Street,
U.S.A. c 1988 Elsevier
Science
Publishers
Tor-
tins indicated that there was a somewhat nonspecific pathway for endocytosis. These large molecules were endocytosed into pinocytotic vesicles which then fused into larger endosomes or multivesicular bodies. These then fused with lysosomes and/or became incorporated into lamellar bodies prior to resecretion [5,6]. The pathways for phospholipid movement during recycling have not been identified, but probably follow the multivesicular body pathway identified with the ferritins. Jacobs et al. [7] found a lack of specificity in the reutilization of various phosphatidylcholine analogues in 3-day-old rabbits with apparent bulk uptake and resecretion. However, in the case of
B.V. (Biomedical
Division)
329
L-cu-lysophosphatidylcholine, 1-palmitoyl (lysoPC), a relatively efficient conversion to phosphatidylcholine (PC) occurred during the process of reutilization. This most likely occurred by remodeling of 1ysoPC via the action of the lysoacyltransferase and/or lyso : lyso transacylase enzymes [g-11]. Since the major activities of these enzymes are in the rough endoplasmic reticulum and cytosol, respectively [12,13], this remodeling would require a diversion of reutilized 1ysoPC from the pathways suggested by the histochemical studies. The synthesis of phosphatidylcholine from labeled 1ysoPC could be extremely useful as a probe of alveolar uptake of phospholipids as well as recycling, since the radiolabel will be converted to a new compound after reuptake and prior to resecretion. The precursor and the product can easily be separated to allow a clear distinction between alveolar phospholipids and intracellular phospholipids, avoiding some of the problems with fraction contamination previously experienced with radiolabeled phosphatidylcholine [14]. The fate of alveolar 1ysoPC has not been well characterized in the adult rabbit.
Materials and Methods Intratracheal injection solutions. 32P radiolabeled natural rabbit surfactant was made by injecting 3-day-old rabbits intratracheally with [ 32P]orthophosphate purchased from ICN, and then recovering the synthesized surfactant 20 h later by alveolar wash with 0.9% saline as previously described [15]. The large surface active surfactant aggregates were isolated by centrifugation at 8000 X g for 20 min at 4 o C over a 0.7 M sucrose cushion to remove cellular debris. The large aggregates recovered at the interface were then pelleted at 40000 X g and resuspended in 1.9% saline for later use. Unlabeled natural rabbit surfactant was obtained from the pooled lung lavages of healthy adult rabbits also in large aggregate form as previously described [16]. L-a-Lysophosphatidylcholine radiolabeled with tritium in the choline moiety (800 Ci/mol) was made from L-cY-dipalmitoylphosphatidylcholine (DPPC) by specific cleavage of the 2-palmitoyl residue using phospholipase A, from Crotalus
adamenteus venom (Sigma). The 3H-labeled DPPC was originally synthe&ed.from [ 3H]methyl iodide and unlabeled phosphatidyldimethylethanolamine [17] and repurified by treatment with osmium tetroxide and elution through an alumina column as described by Mason et al. [18]. Both repurified DPPC and the enzymatically prepared 1ysoPC were over 99% pure by thin-layer chromatography on silica-gel H plates using chloroform/methanol/ acetic acid/water (65 : 25 : 8 : 4, v/v) as the solvent. L-a-Lyso-3-phosphatidylcholine, 1-[1-i4C]palmitoy1 (58.5 Ci/mol) was purchased from Amersham. For the 6 h protocol, approx. 0.4 mCi of [3H]choline-labeled 1ysoPC (0.5 pmol) was dried under nitrogen after adding 15 pmol of DPPC. 6 ml of 0.9% saline were added and the mixture was sonicated into solution using a Fisher Sonic dismembrator at 80% energy output for 60 s followed by 30 s of cooling with ice for a total of four sonication cycles [19]. Then approx. 15 PCi in 30 mg lipid of 32P-labeled natural rabbit surfactant and 22 mg lipid of unlabeled natural rabbit surfactant were added to the sonicated solution, gently mixed and centrifuged at 40000 x g for 15 min. Following this, an additional 22 mg of rabbit surfactant was added to the supernatant and again centrifuged at 40000 X g for 15 min, after which the combined pellet of large aggregate surfactant and associated 1ysoPC and DPPC liposomes were resuspended in 0.45% saline for intratracheal injection in 2 ml aliquots each containing 0.5 PCi 32P, 10 PCi 3H, 2 mg surfactant lipid, and 7 pg 1ysoPC. Approx. 90% of the 3H and i4C radiolabels originally added were associated with the large aggregates in this manner [19]. For the 24 h protocol the intratracheal injection solution was prepared in a similar manner but 0.05 mCi of [14C]palmitate-labeled 1ysoPC was added to 0.8 mCi of [ 3H]choline-labeled 1ysoPC prior to sonication with DPPC. Therefore, each 2 ml aliquot injected contained 0.5 PCi 32P, 20 PCi 3H, 1 PCi r4C, 2 mg surfactant lipid and 25 pg 1ysoPC. Intravascular injection solution. [ l-l4 Clpalmitic acid (56 Ci/mol) was purchased from ICN and converted to the sodium salt by boiling for 5 min in alkaline saline (pH 10) followed by dilution with an equal volume of 6% bovine serum albumin in 0.9% saline and pH adjustment to 8.0. The final
330
solution contained 30 pCi/ml [‘4C]palmitate and 1 ml was injected into each of the rabbits in the 6 h protocol. 6 h protocol. 25 New Zealand White female rabbits weighing 1 kg were lightly anesthetized with diethyl ether and given 2 ml of the appropriate injection solution intratracheally under direct visualization via a 2 mm catheter passed through an 8 mm Olympus flexible bronchoscope as previously described [20]. The chest of each rabbit was compressed by hand during the injection and immediately released with no reflux of fluid seen from the trachea by direct visualization. The entire procedure took 2-3 min per rabbit, after which each rabbit was resuscitated with oxygen and had visibly recovered by 5 min following the injection. While still recovering from the ether anesthesia, each rabbit was injected via a marginal ear vein with 30 PCi of the [i4C]palmitate solution. The rabbits were killed with intravenous pentobarbital via a marginal ear vein in groups of five animals at 10 min, 40 min, 1.5 h, 3 h and 6 h after isotope injection, after which the abdominal aortas were severed to drain the blood from the lungs prior to lung lavage. 24 h protocol. 25 rabbits were individually anesthetized and injected intratracheally as above with a 2 ml solution that contained [‘4C]palmitoyllysoPC in addition to [3H]choline 1ysoPC and “P-labeled rabbit surfactant. They were killed in groups of five at 1, 4, 8, 16 and 24 h following isotope injection. Processing of alveolar wash and lung homogenates. After killing and exsanguination the chest of each rabbit was opened, the trachea was cannulated, the lungs were filled to total lung capacity with 0.9% saline at room temperature, and the saline was rinsed in and out three times. The cycle of three rinses was repeated four more times with volume measurement of the pooled alveolar washes prior to storage at - 20” C. The pooled washes should contain more than 90% of the surfactant recoverable by a wash procedure [21]. The lungs were then removed, weighed, chopped with scissors, and degassed under vacuum. After adding 20 ml of 0.25 M sucrose buffer (0.15 M NaCI/O.Ol M Tris-HCl/ 0.001 M CaCl,/ 0.001 M MgSO,/ 0.0001 M EDTA (pH 7.4)) we briefly homogenized the lungs for 15-20 s with a Tekmar homo-
genizer at 35% speed and then homogenized them with two passes of a Teflon pestel in a glass homogenizer [22]. After removal of aliquots for total lung homogenate phospholipid analysis, the remainder of the homogenate was used for a series of differential and sucrose density gradient centrifugations in order to isolate lamellar body and microsome fractions for phospholipid analysis as previously described [22]. The lamellar body fraction was that fraction isolated at the interface between 0.45 and 0.6 M sucrose. All samples and aliquots were kept on crushed ice during processing prior to storage at -20°C. Phospholipid analysis. Lipids in the alveolar wash, lung homogenate, lamellar body, and microsome fractions, as well as duplicate samples from the intratracheal injection solutions were extracted with chloroform/methanol (2 : 1) and dried under N, at 50” C [23]. All samples had additional unlabeled 1ysoPC added prior to extraction in order to maximize the recovery and permit visualization of 1ysoPC by thin-layer chromatography in chloroform/methanol/acetic acid/water (65 : 25 : 8 : 4, v/v). Duplicate spots of lipid were plated for each sample, from which two spots of PC and a single spot of 1ysoPC were isolated following one-dimensional thin-layer chromatography with one of the two subsequently isolated phosphatidylcholine spots used for phosphate assay [24]. The resulting 1ysoPC spot and the other phosphatidylcholine spot were used for radioactivity measurements by liquid scintillation counting with Aquasol II (New England Nuclear). All ‘H, 14C and 32P cpm were corrected for quenching using internal standards and cross-channel corrections. 32P was then allowed to decay for 6 weeks and the radioactivity in the other labels was again verified. Saturated PC cpm and total phosphate were measured following treatment of the lipid extracts with osmium tetroxide and elution through activated alumina columns as per Mason et al. [18]. The eluted samples were then separated by one-dimensional thin-layer chromatography to remove any amount of contaminating 1ysoPC in the saturated PC sample. Data analysis. All values unless otherwise specified are given as group means -t S.E. Specific activities are expressed as cpm per pmol saturated PC and then normalized as described in Results.
331
RMlIt.9 Description of animals The rabbits for the 24 h protocol were significantly larger (mean 1.2 kg, range 0.9-1.7 kg) than those in the 6 h protocol (mean 0.8 kg, range 0.7-1.0 kg) with significantly larger alveolar pool sizes (10.4 pmol vs. 8.2 pmol saturated PC). This led to small quantitative differences in specific activities for similar time points but had no effect on the general relationships to be described. 6 h protocol Conversion of lysoPC to PC The 1ysoPC given intratracheally was rapidly and efficiently taken up by the lungs and converted to PC with 60-658 of the administered 1ysoPC eventually being converted to PC in the total lungs (alveolar wash + lung homogenate) (Fig. 1). Surprisingly, only 31% of the labeled 1ysoPC remained in the total lungs as 1ysoPC by 10 min (77% of which was recovered in the alveolar wash), while an additional 37% of the labeled 1ysoPC was already converted to PC (with more than 98% in the lung homogenate). This indicated rapid conversion to PC once 1ysoPC was taken up by the lung. While some of the 32% which was unaccounted for at 10 min may have been lost due to adherence to the inside of the
10 ml”
40
I5hr
3
6
TIME
Fig. 1. 6 h recovery curves for [3H]lysoPC given intratracheally. Recoveries are expressed as percentages of the cpm of [3H]lysoPC administered. Each point represents the group mean + S.E. for rabbits hilled at the indicated times. Error bars not shown fall within the data point in this and subsequent figures. The percentages recovered in the form of [’ H]lysoPC are shown for the total lung and alveolar wash (AW) as well as the percentages converted to [ ‘H]PC in the total lung.
injection catheter and the larger airways, there appeared to be an initial rapid loss of some of the 1ysoPC from the lungs. This radioactivity remained unaccounted for over the remainder of the time-course of the experiment. However, between 10 min and 6 h there was essentially complete uptake and conversion of the remaining 31% unconverted 1ysoPC. Interestingly the PC resulting from 1ysoPC conversion was 55560% saturated in the lung homogenate at 10 min with a slight increase to 65-70% saturation by 6 h. The converted PC secreted in the alveolar wash at 6 h was also 65570% saturated. Subcellular processing The normalized specific activities for saturated PC in the subcellular fractions of the lung are shown in Fig. 2. The results are expressed as specific activities rather than cpm because of the variability in the efficiency of recovering the subcellular fractions. The specific activities were calculated by dividing the cpm of radioactivity by the pmol of saturated PC and then normalized by dividing by the specific activity of saturated PC in the total lung at the time point of maximal value. Therefore, a totally homogeneous distribution of labeled saturated PC would have resulted in all subcellular fractions having a specific activity of l.O-times the fraction of peak saturated PC remaining at that time point. For the 1ysoPC converted to saturated PC there were high specific activities in the microsome fractions by 10 min, which persisted for 3 h (Fig. 2A). The lamellar bodies had very low specific activities initially but equaled those of the microsomes by 3 h and surpassed them by 6 h. The specific activities of the alveolar washes (Fig. 2B) lagged behind those of the lamellar bodies, consistent with resecretion. Since the overwhelming majority of the 1ysoPC converted to PC in the total lung was located in the intracellular compartment, the normalized specific activity in the lung homogenate was about 1.0 following 1ysoPC conversion. However, while one might expect increased concentrations of the converted PC in microsomes and lamellar bodies, the normalized specific activities in the microsomes and lamellar bodies were actually less than those in the lung homogenates at all times.
332
A
l.d
#.A
1.2
1.2 1.0 Q8 a6 0.4 a2 I-
-10 ml”
D
1.4 1.2
i ,,“----___
;
1.0
_*.,
&,
LH
#’
---------‘em__“_-*
f
0.8 0.6
6
3
I5 hr
40
~
-
-+-
0,4
Aw
0.2
>
1,
irz: 7.0
/
10 40 Cl?,”
> 5 Q
i
l
_I-
IShi
3
/
6
/
10
40
IShr
6
i
m,n
E
2.8 2.4 2.0 1.6 1.2
Gas z g ad
LH
0.8 g--I_.,
i
-
-~*____-----I------,-__________~
MIC
T--Em_,
_,_..____-.-g
_____________*----_
5
a4
~I
10 *en
Fig. 2. Normalized washes (AW). 6 h shown in (A) and normalized specific
a0
1 Shr
h
specific actj~ty-tome curves of mierosomes (MIC), lamellar bodies (LB), lung homogenates (LH), and alveofar specific activity-time curves for intratrachealfy administered [3H]IysoPC converted to saturated PC @PC) are (B), while corresponding curves for intravenously administered [‘4C]paknitate are shown in (C) and (D). The activity profiles for intratracheaily administered “P natural rabbit surfactant are shown in (E) and (F). Note the double scale on the y axis in (E) and (F).
For the IV palmitate converted to saturated PC the same sequence of rises in specific activities in the fractions occurred as with IysoPC (Figs. 2C and ZD). The relative specific activities in the alveolar washes compared to the lamellar bodies were also similar to those seen with saturated PC converted from 1ysoPC. Again, neither the microsomes nor famelfar bodies had specific activities greater than 1.0 except for an early high value in the micfosomes at IO min.
The normalized specific activities for the labeled rabbit surfactant remained very high in the lamellar bodies and alveolar wash (l.O-2.2), with almost no activity in the microsomes at early times (Figs. 2E and 2F). The activity in the lamellar bodies was high by 10 min after the intratracheal injection. By comparing the ratios of the specific activities of the microsome to lamellar body fractions one can appreciate that the lung-tissue-associated
333
tam recovered by alveolar wash. By 24 h 8.8% of the initial label given as 1ysoPC was recovered by alveolar wash compared to only 6.5% of the natural rabbit surfactant.
~$----&&~MIT~c ;
AClD
%“RP4CT*N: -5 --__ -.--_.____-.-
&-
40
10
mtn
3
15hi
-
-
-. 6
TIME
Fig. 3. Ratio of microsome to lamellar body specific activities of saturated PC (SPC) during 6 h protocol. Each point represents the group mean + S.E. of the individual ratios of microsome to lamellar body specific activities of saturated PC for each of the three isotopes given. The ratios of the specific activities of saturated PC are shown for ‘H-labeled saturated PC (0) converted from intratracheally administered [ ‘H]lysoPC, “C-labeled saturated PC (0) synthesized from intravenously administered [ 14C]palmitate, and 32P-labeled saturated PC (A) from 32P-labeled rabbit surfactant given intratracheally.
rabbit surfactant saturated PC was rapidly localized to lamellar bodies following uptake (Fig. 3). Meanwhile saturated PC formed from 1ysoPC was initially more specifically localized to microsomes followed by transfer to lamellar bodies with specific activity ratios similar to those seen following administration of labeled IV palmitate.
g
10 z -I 1 e
Q
1R
‘,
o*.....“’
AW ,..O.........................~........~.........,~~....~
~._‘...”
I
A
8
16
24
24 h protocol ~aturared PC clearance and resecret~o~ In this protocol the intratracheally administered 1ysoPC was labeled with 14C in the lpalmitoyl residue as well as 3H in the choline moiety and no labeled precursor was given by intravascular injection. Total radioactivity recovered from the alveolar wash, lung homogenate, and total lung over 24 h is shown in Fig. 4. The clearance of the 32P-labeled natural rabbit surfactant from the total lung (Fig. 4A) was approximately 3% per h, a value similar to the clearance rate of natural rabbit surfactant measured previously [20]. The resecretion into the alveolar space of saturated PC derived from 1ysoPC reached a peak of 12% of the original intratracheal tritium label at 8 h (Fig. 4B). At that time there was 21% of the intratrach~lly administered natural rabbit surfac-
Fig. 4. Recovery curves for radiolabels used in 24 h protocol. (A) Percentage of 32P-labeled saturated PC administered intratracheally in rabbit surfactant recovered in total lungs, alveolar washes (AW), and lung homogenates (LH) over 24 h. Each point represents the group mean f SE. for rabbits killed at the indicated times. (B) Percentage of 13H~ho~e-la~l~ 1ysoPC recovered as 3H-labeled saturated PC following conversion in total lungs, AW and LH over 24 h. The percentages recovered as 3H-labeled total PC were 30-401 higher. (C) Ratio of 3H/‘4C in saturated PC recovered in the total lungs following conversion from [ 3H]choline-labeIed 1ysoPC and l[l-‘4C]pal~toyi-ly~~ over 24 h.
334
”
__A---
;h--c__A__--
/
J2P-RABBlT
I6
.4
T”lME
SURFACTAN, 74
(HOURS)
Fig. 5. Ratio of microsome to lamellar body specific activities of saturated PC during 24 h protocol. Each point represents the group mean + S.E. of the individual ratios of microsome to lamellar body specific activities of saturated PC for each of the three isotopes Siven intratracheally. The ratios of the specific activities of saturated PC are shown for 3H-labeled saturated PC (0) converted from administered [3H]choline IysoPC, 14Clabeled saturated PC (0) converted from administered l-[l“C]palmitoyllysoPC, and “P-labeled saturated PC (A) from 32P-labeled rabbit surfactant given intratracheally.
Ratio of labels from IysoPC in PC As in the 6 h protocol the uptake and conversion of 1ysoPC occurred rapidly and efficiently with peak efficiencies of 75% and 80% at 8 h for the [ 3HIcholine-labeled and [ 14C]palmitoyl-labeled 1ysoPC. The ratio of 3H/‘4C in the form of 1ysoPC converted to PC versus the ratio in the injection solution was 0.8-1.0 at 1-16 h (Fig. 4C). At 24 h the ratio was significantly increased to 1.2 (P < 0.01 by ANOVA). Subcellular processing of 1ysoPC and converted PC The intratracheal 1ysoPC was preferentially targeted to the microsomal fraction during the time of maximal conversion to saturated PC as seen by again comparing the ratios of the specific activities of the microsomal to lamellar body fractions (Fig. 5). The 3H and i4C radiolabels showed similar relationships, as would be expected. After 8 h all three radiolabels were relatively more apparent in the microsome fractions, a finding that is consistent with breakdown and resynthesis following PC catabolism. Discussion Studies of surfactant metabolism have been complicated by technical difficulties. In whole animal studies, it has not been possible to tell reliably how much extracellular surfactant PC
contaminated the various subcellular fractions. It has also not been possible to differentiate extracellular surfactant components that have been endocytosed and resecreted from those that remained in the alveolar pool. Mathematical models such as those of Zilversrnit et al. [25] have been used to estimate turnover times and percent reutilization and were probably fairly accurate in the healthy animal with stable pool sizes [1,4]. However, animals with immature lungs or lung injury may have unstable surfactant pools. In addition, the mathematical models assume a strict precursor-product relationship, which may not always be the case. In vitro studies have the disadvantage of altering the normal physiologic environment of the type II cell. Histochemical studies have utilized lectins which were specific for receptors on type II cells, but such lectins may not follow the same pathways as recycled surfactant phospholipids [5,6]. An ideal probe of surfactant phospholipid reutilization is one that can be used without toxic effects in animals and will permit the differentiation of extracellular surfactant PC contaminants from PC localized during subcellular fractionation. It will allow for separation of resecreted alveolar PC from phospholipids that were never endocytosed. Finally, the ideal probe will have similar kinetics as it follows the same pathways of uptake, subcellular processing, resecretion, and clearance as the phospholipids of interest. A ‘metabolic probe’ such as 1ysoPC which is converted to PC rapidly after uptake is attractive. since unconverted extracellular contaminants can easily be separated from labeled subcellular components by thin-layer chromatography. Resecreted labeled PC is similarly separated from labeled 1ysoPC that was never endocytosed and converted to PC by the lung. One potential problem with 1ysoPC is its cytotoxicity with lytic effects on cell membranes [26]. Increased permeability of the alveolar epithelium to both sucrose and dextran 70 (M, 70000) was seen in perfused hamster lungs with intratracheal administration of 1ysoPC in concentrations as low as 8 pg/ml [27]. While our intratracheal injections contained 3.5 pg/ml and 12.5 pg/ml of 1ysoPC in the 6 and 24 h protocols, respectively, the total dose administered per kg body weight was only
335
10% of the lowest dose administered in the perfused lung model. In addition, our 1ysoPC was son&ted into liposomes with DPPC and then associated with rabbit surfactant prior to administration [19]. Labeled DPPC sonicated into liposomes and associated with rabbit surfactant in this manner was previously shown to function metabolically like endogenously labeled rabbit surfactant PC 119,281. The clinical status of the animals and the gross appearance of the lungs did not indicate any toxicity and there was no significant increase in the total proteins recovered from the alveolar washes compared to untreated 1 kg rabbits (40 mg protein/alveolar wash in the rabbits receiving 25 pg 1ysoPC vs. 55 mg protein alveolar wash in untreated 1 kg rabbits [20]). There was very rapid uptake in the lysoPC, with 43% of the total counts given being lung associated by 10 min after administration. This finding was consistent with the rate of uptake seen in the perfused hamster lung and the increases in liposome fusion to mammalian cells in culture when 1ysoPC was added to mixtures of dipalmitoylphosphatidylglycerol and DPPC [26]. The uptake of 1ysoPC was more rapid than that of rabbit surfactant PC, which was 33% lung associated by 10 min. There was a similar rapid and efficient conversion of 1ysoPC to PC seen at much higher doses in the perfused hamster lung [27]. At the subcellular level, 1ysoPC clearly deviated from the pathway of endocytosed rabbit surfactant PC during uptake and intra~llul~ conversion to PC. It appeared to follow a pathway of remodeling that paralleled that of labeled IV palmitate which probably entered pathways of both remodeling and de novo synthesis. In addition, the normalized specific activities for saturated PC derived from IV palmitate and intratracheal 1ysoPC were lower than one might expect in the lamellar body and microsome fractions. In contrast, the intratracheal natural rabbit surfactant had much higher specific activities in the lamellar body fractions than the lung homogenate and was therefore probably much more specifically localized to lamellar bodies and type II cells. Therefore, synthesis of lung saturated PC utilizing 1ysoPC or intravascular palmitate could have occurred largely in cell types other than type II cells or lung saturated PC may have been
transported to other cell types following synthesis in type II cells. However, the high percent saturation of the phosphatidylcholine derived from 1ysoPC suggests targeting of the type II cell. In a recent review, Wright and Clements estimated that only 21% of lung saturated PC was in the type II cell [29]. While any lack of efficiency in recovering type II cells and loss of saturated PC during isolation would make their estimate low, any localization of saturated PC outside the type II cells could help explain why the normalized specific activities are lower in the lamellar bodies following synthesis or remodeling when compared to endocytosed saturated PC. An alternative explanation is that our fractionation procedure missed an important subcellular site of 1ysoPC metabolism. In the perfused hamster lung given higher doses of intratracheal lysoPC, the major toxicity by electron microscopy was seen in type I cells 1271. The mechanism of the deviation of 1ysoPC from the recycling pathway of endocytosed saturated PC-is unknown, but could be related to specific transfer proteins [30]. Prior to conversion to saturated PC the clearance of 1ysoPC could not be used as a measure for surfactant clearance. However, once 1ysoPC was converted to saturated PC, it appeared to be cleared from the lung at the same rate as labeled rabbit surfactant PC. For the 24 h protocol we utilized 1ysoPC labeled both with 3H in the choline moiety of the backbone and 14C in the I-palmitoyl side-chain in order to evaluate the relative cont~butions of lysoacyltransferase and lyso : lyso acyltransferase, since any conversion by the latter enzyme would result in a decrease in the isotope ratios if two labeled 1ysoPC molecules participated in the reaction. The 3H/‘4C ratio of 0.8-1.0 for radioactivity in PC converted from 1ysoPC in the total lung from 1 to 16 h compared to the ‘H/14C ratio in the solution used to inject the rabbits indicated that the 1ysoPC was predominantly converted to PC by lysoacyltransferase with the addition of an unlabeled fatty acid via an acyl-CoA to the 2-position of the glycerol backbone. The combination of the high efficiency and rapid conversion of 1ysoPC to PC and the 3H/14C ratios close to 1 indicated intact uptake and conversion of the 1ysoPC to PC rather than breakdown and resynthesis from labeled precursors at early times. However, at 24 h
336
the ratio of [ 3H]choline/[‘4C]palmitate increased significantly in both the total lung and the various subcellular fractions accompanied by an increase in the relative proportion of activity in the microsomes. These findings were consistent with a shorter turnover time for palmitate than choline and/or the occurrence of breakdown and resynthesis with preferential conservation of labeled choline as found by Chander et al. [31] and Fisher et al. [32] in the isolated perfused lung and type II cells of the adult rat. However, we saw no significant increase in the percent of unsaturated PC or 1ysoPC with time in any of our radiolabels or subcellular fractions (data not shown) as was reported in the above in vitro models. It is possible that in those studies the PC remodeling pathway was somehow depressed by the in vitro conditions. This study showed that intratracheally administered 1ysoPC was taken up rapidly and efficiently by the adult rabbit lung where rapid conversion to PC occurred in the endoplasmic reticulum followed by movement to the lamellar bodies and resecretion into the alveolus. The metabolic conversion made it simple to differentiate resecretion from decreased uptake and to eliminate contamination of subcellular fractions by extracellular label. The lung uptake of 1ysoPC was more rapid than PC and 1ysoPC took a pathway that paralleled that of the remodeling and de novo synthetic pathways followed by labeled IV palmitate. However, while the conversion pathway of 1ysoPC to PC deviated from that of recycled surfactant PC, it appeared to parallel the clearance and resecretion pathways after conversion. Therefore, 1ysoPC is useful as a ‘metabolic probe’ of phospholipid reutilization, especially in the immature and/or injured lung, when it is given intratracheally in combination with a differentially labeled surfactant PC used to evaluate initial uptake. Perhaps other phospholipid probes exist which would be structurally modified in multivesicular or in lamellar bodies and could be used to further evaluate the specific pathways of initial uptake and preferential lamellar body localization of recycled saturated PC in the lung. Acknowledgement This work was supported by Grant HD-12714 from the Department of Health and Human Services.
References 1 Jacobs,
H.C., Ikegami, M., Jobe, A.H., Berry, D.D. and Jones, S. (1985) Biochim. Biophys. Acta 837, 77-84. 2 Hallman, M., Epstein, B.L. and Gluck, L. (1981) J. Clin. Invest. 68, 742-751. 3 Glatz, T., Ikegami, M. and Jobe. A. (1982) Pediatr. Res. 16. 711-715. 4 Jacobs, H., Jobe, A., Ikegami, M. and Jones, S. (1982) J. Biol. Chem. 257. 1805-1810. 5 Williams, M.C. (1984) Proc. Natl. Acad. Sci. USA 81. 6054-6058. 6 Williams, M.C. (1984) Proc. Natl. Acad. Sci. USA 81. 6383-6387. 7 Jacobs, H., Jobe, A., Ikegami, M., Miller, D. and Jones. S. (1984) Biochim. Biophys. Acta 793, 300-309. 8 Hallman, M. and Raivio. K. (1974) Pediatr. Res. 8. 874-879. 9 Lands, W.E.M. (1960) J. Biol. Chem. 235, 2233-2237. 10 Van den Bosch, H., Bonte, H.A. and Van Deenen, L.L.M. (1965) Biochim. Biophys. Acta 98. 648-651. 11 Van Heusden, G.P.H. and Van den Bosh, H. (1981) Biochim. Biophys. Acta 666, 508-511. 12 Oldenborg, V. and Van Golde, L.M.G. (1976) 441.433-442. 13 Tsao, F.H.C. and Zachman, R.D. (1977) Pediatr. Res. 11. 849-857. 14 Oyarzun, M.J., Clements. J.A. and Baritussio. A. (1980) Am. Rev. Respir. Dis. 121, 709-721. 15 Jacobs. H.C., Jobe, A.H., Ikegami, M. and Jones, S. (1983) Biochim. Biophys. Acta 752, 178-181. 16 Jobe, A., Ikegami, M., Glatz, T., Yoshida, Y., Diakomanolis, E. and Padbury, J. (1981) J. Clin. Invest. 67, 370-375. 17 Stoffel, W., Lekim, D. and Tschung. T.S. (1971) Z. Physiol. Chem. 352, 1058-1064. 18 Mason, R.J., Nellenbogen, J. and Clement% J.A. (1976) J. Lipid Res. 17, 281-284. 19 Ikegami, M., Jobe, A. and Duane. G. (1985) Biochim. Biophys. Acta 835, 352-359. 20 Pettenazzo, A., Ikegami, M., Seidner, S. and Jobe. A. (1988) J. Appl. Physiol. 64, 120-I 27. 21 Toshima, N. and Akino. T. (1972) Tohoku. J. Exp. Med. 108, 253-263. 22 Jobe, A. (1977) B&him. Biophys. Acta 489, 440-453. 23 Bligh, E.G. and Dyer. W.J. (1959) Can. J. Biochem. Physiol. 37, 911-917. 24 Bartlett, G.R. (1959) J. Biol. Chem. 234, 466-468. 25 Zilversmit. D.B., Entenman, C. and Fishler, M.C. (1943) J. Gen. Physiol. 26, 325-331. 26 Papahadjopoulos, D.. Poste, G. and Schaeffer, B.E. (1973) Biochim. Biophys. Acta 323, 23-42. 27 Niewoehner, D.E., Rice, K., Sinha, A.A. and Wangensteen, D. (1987) J. Appl. Physiol. 63, 1979-1986. 28 Jacobs, H., Jobe, A., Ikegami, M. and Jones, S. (1982) J. Biol. Chem. 257, 1805-1810. 29 Wright, J.R. and Clement& J.A. (1987) Am. Rev. Respir. Dis. 135, 426-444. 30 Spalding, J.W. and Hook, G.E.R. (1979) Lipids 14, 606-613. 31 Chander, A., Reicherter, J. and Fisher, A.B. (1987) J. Clin. Invest. 79, 1133-1138. 32 Fisher, A.B.. Dodia, C. and Chander. A. (1987) J. Appl. Physiol. 62, 2295-2299.
Bwchimico
et Blophysm
Actu, 961 (1988) 328-336 Elsevier
BBA 52883
Lysophosphatidylcholine
uptake and metabolism in the adult rabbit lung
Steven R. Seidner, Alan H. Jobe, Machiko Ikegami, Andrea Anathalie Priestley and Lynda Ruffini Department
of Pedintrics, Harbor- UCLA Medical Center, UCLA School of Medune. (Received
Key words:
Lysophosphatidylcholine;
1 February
Pettenazzo,
Torrance, CA (U.S.A.)
1988)
Phospholipid
metabolism;
(Lung);
(Rabbit)
Tracer quantities of 3H-labeled IysoPC and 32P-labeled natural rabbit surfactant were given intratracbeally via a broncboscope and [ 14C]palmitate was given intravenously to 25 rabbits with labeled PC and IysoPC measured in the alveolar wash, lung homogenate, lamellar bodies and microsomes at five times from 10 min to 6 h after tracheal injection. Surprisingly, only 31% of the administered IysoPC remained in its original form in the total lungs (alveolar wash + lung homogenate) by 10 min, of which 77% was in the alveolar wash. Meanwhile, by 10 min an additional 37% was already converted to PC, of which more than 98% was in the lung homogenate. LysoPC continued to be rapidly and efficiently converted to PC, with 62% conversion measured at 3 h. The converted IysoPC initially appeared with high specific activity in microsomes, then in lamellar bodies, and finally in the alveolar wash. The intravascular palmitate labeled lung PC had similar specific activity-time profiles in the s&cellular fractions, while intratracheally administered natural rabbit surfactant had a constantly low specific activity in microsomes and much higher specific activities in lamellar bodies and alveolar wash. Another 25 rabbits received intratracheal IysoPC labeled in both the choline and palmitate moieties and then were studied from 1 to 24 h after tracheal injection. The ratio of the palmitate to choline labels indicated uptake and conversion to PC primarily by direct acylation rather than transacylation and by intact reuptake and conversion rather than breakdown and resynthesis. LysoPC is an attractive ‘metabolic probe’ of surfactant metabolism which undergoes very rapid and efficient intracellular conversion to PC via a subcellular pathway that parallels the remodeling and de novo synthetic pathways.
Introduction Reutilization of the various phospholipids of pulmonary surfactant by the type II cell occurs to some degree in the mature and to a greater degree in the developing lung [l-4]. The histochemical studies of Williams with cationic and MPA ferri-
Abbreviations:
PC, phosphatidylcholine;
DPPC,
dipalmitoyl-
phosphatidylcholine. Correspondence: Harbor-UCLA rance,
CA 90509,
0005-2760/88/$03.50
S.R.
Seidner,
Medical
Center,
Department 1000
of
W. Carson
Pediatrics, Street,
U.S.A. c 1988 Elsevier
Science
Publishers
Tor-
tins indicated that there was a somewhat nonspecific pathway for endocytosis. These large molecules were endocytosed into pinocytotic vesicles which then fused into larger endosomes or multivesicular bodies. These then fused with lysosomes and/or became incorporated into lamellar bodies prior to resecretion [5,6]. The pathways for phospholipid movement during recycling have not been identified, but probably follow the multivesicular body pathway identified with the ferritins. Jacobs et al. [7] found a lack of specificity in the reutilization of various phosphatidylcholine analogues in 3-day-old rabbits with apparent bulk uptake and resecretion. However, in the case of
B.V. (Biomedical
Division)
329
L-cu-lysophosphatidylcholine, 1-palmitoyl (lysoPC), a relatively efficient conversion to phosphatidylcholine (PC) occurred during the process of reutilization. This most likely occurred by remodeling of 1ysoPC via the action of the lysoacyltransferase and/or lyso : lyso transacylase enzymes [g-11]. Since the major activities of these enzymes are in the rough endoplasmic reticulum and cytosol, respectively [12,13], this remodeling would require a diversion of reutilized 1ysoPC from the pathways suggested by the histochemical studies. The synthesis of phosphatidylcholine from labeled 1ysoPC could be extremely useful as a probe of alveolar uptake of phospholipids as well as recycling, since the radiolabel will be converted to a new compound after reuptake and prior to resecretion. The precursor and the product can easily be separated to allow a clear distinction between alveolar phospholipids and intracellular phospholipids, avoiding some of the problems with fraction contamination previously experienced with radiolabeled phosphatidylcholine [14]. The fate of alveolar 1ysoPC has not been well characterized in the adult rabbit.
Materials and Methods Intratracheal injection solutions. 32P radiolabeled natural rabbit surfactant was made by injecting 3-day-old rabbits intratracheally with [ 32P]orthophosphate purchased from ICN, and then recovering the synthesized surfactant 20 h later by alveolar wash with 0.9% saline as previously described [15]. The large surface active surfactant aggregates were isolated by centrifugation at 8000 X g for 20 min at 4 o C over a 0.7 M sucrose cushion to remove cellular debris. The large aggregates recovered at the interface were then pelleted at 40000 X g and resuspended in 1.9% saline for later use. Unlabeled natural rabbit surfactant was obtained from the pooled lung lavages of healthy adult rabbits also in large aggregate form as previously described [16]. L-a-Lysophosphatidylcholine radiolabeled with tritium in the choline moiety (800 Ci/mol) was made from L-cY-dipalmitoylphosphatidylcholine (DPPC) by specific cleavage of the 2-palmitoyl residue using phospholipase A, from Crotalus
adamenteus venom (Sigma). The 3H-labeled DPPC was originally synthe&ed.from [ 3H]methyl iodide and unlabeled phosphatidyldimethylethanolamine [17] and repurified by treatment with osmium tetroxide and elution through an alumina column as described by Mason et al. [18]. Both repurified DPPC and the enzymatically prepared 1ysoPC were over 99% pure by thin-layer chromatography on silica-gel H plates using chloroform/methanol/ acetic acid/water (65 : 25 : 8 : 4, v/v) as the solvent. L-a-Lyso-3-phosphatidylcholine, 1-[1-i4C]palmitoy1 (58.5 Ci/mol) was purchased from Amersham. For the 6 h protocol, approx. 0.4 mCi of [3H]choline-labeled 1ysoPC (0.5 pmol) was dried under nitrogen after adding 15 pmol of DPPC. 6 ml of 0.9% saline were added and the mixture was sonicated into solution using a Fisher Sonic dismembrator at 80% energy output for 60 s followed by 30 s of cooling with ice for a total of four sonication cycles [19]. Then approx. 15 PCi in 30 mg lipid of 32P-labeled natural rabbit surfactant and 22 mg lipid of unlabeled natural rabbit surfactant were added to the sonicated solution, gently mixed and centrifuged at 40000 x g for 15 min. Following this, an additional 22 mg of rabbit surfactant was added to the supernatant and again centrifuged at 40000 X g for 15 min, after which the combined pellet of large aggregate surfactant and associated 1ysoPC and DPPC liposomes were resuspended in 0.45% saline for intratracheal injection in 2 ml aliquots each containing 0.5 PCi 32P, 10 PCi 3H, 2 mg surfactant lipid, and 7 pg 1ysoPC. Approx. 90% of the 3H and i4C radiolabels originally added were associated with the large aggregates in this manner [19]. For the 24 h protocol the intratracheal injection solution was prepared in a similar manner but 0.05 mCi of [14C]palmitate-labeled 1ysoPC was added to 0.8 mCi of [ 3H]choline-labeled 1ysoPC prior to sonication with DPPC. Therefore, each 2 ml aliquot injected contained 0.5 PCi 32P, 20 PCi 3H, 1 PCi r4C, 2 mg surfactant lipid and 25 pg 1ysoPC. Intravascular injection solution. [ l-l4 Clpalmitic acid (56 Ci/mol) was purchased from ICN and converted to the sodium salt by boiling for 5 min in alkaline saline (pH 10) followed by dilution with an equal volume of 6% bovine serum albumin in 0.9% saline and pH adjustment to 8.0. The final
330
solution contained 30 pCi/ml [‘4C]palmitate and 1 ml was injected into each of the rabbits in the 6 h protocol. 6 h protocol. 25 New Zealand White female rabbits weighing 1 kg were lightly anesthetized with diethyl ether and given 2 ml of the appropriate injection solution intratracheally under direct visualization via a 2 mm catheter passed through an 8 mm Olympus flexible bronchoscope as previously described [20]. The chest of each rabbit was compressed by hand during the injection and immediately released with no reflux of fluid seen from the trachea by direct visualization. The entire procedure took 2-3 min per rabbit, after which each rabbit was resuscitated with oxygen and had visibly recovered by 5 min following the injection. While still recovering from the ether anesthesia, each rabbit was injected via a marginal ear vein with 30 PCi of the [i4C]palmitate solution. The rabbits were killed with intravenous pentobarbital via a marginal ear vein in groups of five animals at 10 min, 40 min, 1.5 h, 3 h and 6 h after isotope injection, after which the abdominal aortas were severed to drain the blood from the lungs prior to lung lavage. 24 h protocol. 25 rabbits were individually anesthetized and injected intratracheally as above with a 2 ml solution that contained [‘4C]palmitoyllysoPC in addition to [3H]choline 1ysoPC and “P-labeled rabbit surfactant. They were killed in groups of five at 1, 4, 8, 16 and 24 h following isotope injection. Processing of alveolar wash and lung homogenates. After killing and exsanguination the chest of each rabbit was opened, the trachea was cannulated, the lungs were filled to total lung capacity with 0.9% saline at room temperature, and the saline was rinsed in and out three times. The cycle of three rinses was repeated four more times with volume measurement of the pooled alveolar washes prior to storage at - 20” C. The pooled washes should contain more than 90% of the surfactant recoverable by a wash procedure [21]. The lungs were then removed, weighed, chopped with scissors, and degassed under vacuum. After adding 20 ml of 0.25 M sucrose buffer (0.15 M NaCI/O.Ol M Tris-HCl/ 0.001 M CaCl,/ 0.001 M MgSO,/ 0.0001 M EDTA (pH 7.4)) we briefly homogenized the lungs for 15-20 s with a Tekmar homo-
genizer at 35% speed and then homogenized them with two passes of a Teflon pestel in a glass homogenizer [22]. After removal of aliquots for total lung homogenate phospholipid analysis, the remainder of the homogenate was used for a series of differential and sucrose density gradient centrifugations in order to isolate lamellar body and microsome fractions for phospholipid analysis as previously described [22]. The lamellar body fraction was that fraction isolated at the interface between 0.45 and 0.6 M sucrose. All samples and aliquots were kept on crushed ice during processing prior to storage at -20°C. Phospholipid analysis. Lipids in the alveolar wash, lung homogenate, lamellar body, and microsome fractions, as well as duplicate samples from the intratracheal injection solutions were extracted with chloroform/methanol (2 : 1) and dried under N, at 50” C [23]. All samples had additional unlabeled 1ysoPC added prior to extraction in order to maximize the recovery and permit visualization of 1ysoPC by thin-layer chromatography in chloroform/methanol/acetic acid/water (65 : 25 : 8 : 4, v/v). Duplicate spots of lipid were plated for each sample, from which two spots of PC and a single spot of 1ysoPC were isolated following one-dimensional thin-layer chromatography with one of the two subsequently isolated phosphatidylcholine spots used for phosphate assay [24]. The resulting 1ysoPC spot and the other phosphatidylcholine spot were used for radioactivity measurements by liquid scintillation counting with Aquasol II (New England Nuclear). All ‘H, 14C and 32P cpm were corrected for quenching using internal standards and cross-channel corrections. 32P was then allowed to decay for 6 weeks and the radioactivity in the other labels was again verified. Saturated PC cpm and total phosphate were measured following treatment of the lipid extracts with osmium tetroxide and elution through activated alumina columns as per Mason et al. [18]. The eluted samples were then separated by one-dimensional thin-layer chromatography to remove any amount of contaminating 1ysoPC in the saturated PC sample. Data analysis. All values unless otherwise specified are given as group means -t S.E. Specific activities are expressed as cpm per pmol saturated PC and then normalized as described in Results.
331
RMlIt.9 Description of animals The rabbits for the 24 h protocol were significantly larger (mean 1.2 kg, range 0.9-1.7 kg) than those in the 6 h protocol (mean 0.8 kg, range 0.7-1.0 kg) with significantly larger alveolar pool sizes (10.4 pmol vs. 8.2 pmol saturated PC). This led to small quantitative differences in specific activities for similar time points but had no effect on the general relationships to be described. 6 h protocol Conversion of lysoPC to PC The 1ysoPC given intratracheally was rapidly and efficiently taken up by the lungs and converted to PC with 60-658 of the administered 1ysoPC eventually being converted to PC in the total lungs (alveolar wash + lung homogenate) (Fig. 1). Surprisingly, only 31% of the labeled 1ysoPC remained in the total lungs as 1ysoPC by 10 min (77% of which was recovered in the alveolar wash), while an additional 37% of the labeled 1ysoPC was already converted to PC (with more than 98% in the lung homogenate). This indicated rapid conversion to PC once 1ysoPC was taken up by the lung. While some of the 32% which was unaccounted for at 10 min may have been lost due to adherence to the inside of the
10 ml”
40
I5hr
3
6
TIME
Fig. 1. 6 h recovery curves for [3H]lysoPC given intratracheally. Recoveries are expressed as percentages of the cpm of [3H]lysoPC administered. Each point represents the group mean + S.E. for rabbits hilled at the indicated times. Error bars not shown fall within the data point in this and subsequent figures. The percentages recovered in the form of [’ H]lysoPC are shown for the total lung and alveolar wash (AW) as well as the percentages converted to [ ‘H]PC in the total lung.
injection catheter and the larger airways, there appeared to be an initial rapid loss of some of the 1ysoPC from the lungs. This radioactivity remained unaccounted for over the remainder of the time-course of the experiment. However, between 10 min and 6 h there was essentially complete uptake and conversion of the remaining 31% unconverted 1ysoPC. Interestingly the PC resulting from 1ysoPC conversion was 55560% saturated in the lung homogenate at 10 min with a slight increase to 65-70% saturation by 6 h. The converted PC secreted in the alveolar wash at 6 h was also 65570% saturated. Subcellular processing The normalized specific activities for saturated PC in the subcellular fractions of the lung are shown in Fig. 2. The results are expressed as specific activities rather than cpm because of the variability in the efficiency of recovering the subcellular fractions. The specific activities were calculated by dividing the cpm of radioactivity by the pmol of saturated PC and then normalized by dividing by the specific activity of saturated PC in the total lung at the time point of maximal value. Therefore, a totally homogeneous distribution of labeled saturated PC would have resulted in all subcellular fractions having a specific activity of l.O-times the fraction of peak saturated PC remaining at that time point. For the 1ysoPC converted to saturated PC there were high specific activities in the microsome fractions by 10 min, which persisted for 3 h (Fig. 2A). The lamellar bodies had very low specific activities initially but equaled those of the microsomes by 3 h and surpassed them by 6 h. The specific activities of the alveolar washes (Fig. 2B) lagged behind those of the lamellar bodies, consistent with resecretion. Since the overwhelming majority of the 1ysoPC converted to PC in the total lung was located in the intracellular compartment, the normalized specific activity in the lung homogenate was about 1.0 following 1ysoPC conversion. However, while one might expect increased concentrations of the converted PC in microsomes and lamellar bodies, the normalized specific activities in the microsomes and lamellar bodies were actually less than those in the lung homogenates at all times.
332
A
l.d
#.A
1.2
1.2 1.0 Q8 a6 0.4 a2 I-
-10 ml”
D
1.4 1.2
i ,,“----___
;
1.0
_*.,
&,
LH
#’
---------‘em__“_-*
f
0.8 0.6
6
3
I5 hr
40
~
-
-+-
0,4
Aw
0.2
>
1,
irz: 7.0
/
10 40 Cl?,”
> 5 Q
i
l
_I-
IShi
3
/
6
/
10
40
IShr
6
i
m,n
E
2.8 2.4 2.0 1.6 1.2
Gas z g ad
LH
0.8 g--I_.,
i
-
-~*____-----I------,-__________~
MIC
T--Em_,
_,_..____-.-g
_____________*----_
5
a4
~I
10 *en
Fig. 2. Normalized washes (AW). 6 h shown in (A) and normalized specific
a0
1 Shr
h
specific actj~ty-tome curves of mierosomes (MIC), lamellar bodies (LB), lung homogenates (LH), and alveofar specific activity-time curves for intratrachealfy administered [3H]IysoPC converted to saturated PC @PC) are (B), while corresponding curves for intravenously administered [‘4C]paknitate are shown in (C) and (D). The activity profiles for intratracheaily administered “P natural rabbit surfactant are shown in (E) and (F). Note the double scale on the y axis in (E) and (F).
For the IV palmitate converted to saturated PC the same sequence of rises in specific activities in the fractions occurred as with IysoPC (Figs. 2C and ZD). The relative specific activities in the alveolar washes compared to the lamellar bodies were also similar to those seen with saturated PC converted from 1ysoPC. Again, neither the microsomes nor famelfar bodies had specific activities greater than 1.0 except for an early high value in the micfosomes at IO min.
The normalized specific activities for the labeled rabbit surfactant remained very high in the lamellar bodies and alveolar wash (l.O-2.2), with almost no activity in the microsomes at early times (Figs. 2E and 2F). The activity in the lamellar bodies was high by 10 min after the intratracheal injection. By comparing the ratios of the specific activities of the microsome to lamellar body fractions one can appreciate that the lung-tissue-associated
333
tam recovered by alveolar wash. By 24 h 8.8% of the initial label given as 1ysoPC was recovered by alveolar wash compared to only 6.5% of the natural rabbit surfactant.
~$----&&~MIT~c ;
AClD
%“RP4CT*N: -5 --__ -.--_.____-.-
&-
40
10
mtn
3
15hi
-
-
-. 6
TIME
Fig. 3. Ratio of microsome to lamellar body specific activities of saturated PC (SPC) during 6 h protocol. Each point represents the group mean + S.E. of the individual ratios of microsome to lamellar body specific activities of saturated PC for each of the three isotopes given. The ratios of the specific activities of saturated PC are shown for ‘H-labeled saturated PC (0) converted from intratracheally administered [ ‘H]lysoPC, “C-labeled saturated PC (0) synthesized from intravenously administered [ 14C]palmitate, and 32P-labeled saturated PC (A) from 32P-labeled rabbit surfactant given intratracheally.
rabbit surfactant saturated PC was rapidly localized to lamellar bodies following uptake (Fig. 3). Meanwhile saturated PC formed from 1ysoPC was initially more specifically localized to microsomes followed by transfer to lamellar bodies with specific activity ratios similar to those seen following administration of labeled IV palmitate.
g
10 z -I 1 e
Q
1R
‘,
o*.....“’
AW ,..O.........................~........~.........,~~....~
~._‘...”
I
A
8
16
24
24 h protocol ~aturared PC clearance and resecret~o~ In this protocol the intratracheally administered 1ysoPC was labeled with 14C in the lpalmitoyl residue as well as 3H in the choline moiety and no labeled precursor was given by intravascular injection. Total radioactivity recovered from the alveolar wash, lung homogenate, and total lung over 24 h is shown in Fig. 4. The clearance of the 32P-labeled natural rabbit surfactant from the total lung (Fig. 4A) was approximately 3% per h, a value similar to the clearance rate of natural rabbit surfactant measured previously [20]. The resecretion into the alveolar space of saturated PC derived from 1ysoPC reached a peak of 12% of the original intratracheal tritium label at 8 h (Fig. 4B). At that time there was 21% of the intratrach~lly administered natural rabbit surfac-
Fig. 4. Recovery curves for radiolabels used in 24 h protocol. (A) Percentage of 32P-labeled saturated PC administered intratracheally in rabbit surfactant recovered in total lungs, alveolar washes (AW), and lung homogenates (LH) over 24 h. Each point represents the group mean f SE. for rabbits killed at the indicated times. (B) Percentage of 13H~ho~e-la~l~ 1ysoPC recovered as 3H-labeled saturated PC following conversion in total lungs, AW and LH over 24 h. The percentages recovered as 3H-labeled total PC were 30-401 higher. (C) Ratio of 3H/‘4C in saturated PC recovered in the total lungs following conversion from [ 3H]choline-labeIed 1ysoPC and l[l-‘4C]pal~toyi-ly~~ over 24 h.
334
”
__A---
;h--c__A__--
/
J2P-RABBlT
I6
.4
T”lME
SURFACTAN, 74
(HOURS)
Fig. 5. Ratio of microsome to lamellar body specific activities of saturated PC during 24 h protocol. Each point represents the group mean + S.E. of the individual ratios of microsome to lamellar body specific activities of saturated PC for each of the three isotopes Siven intratracheally. The ratios of the specific activities of saturated PC are shown for 3H-labeled saturated PC (0) converted from administered [3H]choline IysoPC, 14Clabeled saturated PC (0) converted from administered l-[l“C]palmitoyllysoPC, and “P-labeled saturated PC (A) from 32P-labeled rabbit surfactant given intratracheally.
Ratio of labels from IysoPC in PC As in the 6 h protocol the uptake and conversion of 1ysoPC occurred rapidly and efficiently with peak efficiencies of 75% and 80% at 8 h for the [ 3HIcholine-labeled and [ 14C]palmitoyl-labeled 1ysoPC. The ratio of 3H/‘4C in the form of 1ysoPC converted to PC versus the ratio in the injection solution was 0.8-1.0 at 1-16 h (Fig. 4C). At 24 h the ratio was significantly increased to 1.2 (P < 0.01 by ANOVA). Subcellular processing of 1ysoPC and converted PC The intratracheal 1ysoPC was preferentially targeted to the microsomal fraction during the time of maximal conversion to saturated PC as seen by again comparing the ratios of the specific activities of the microsomal to lamellar body fractions (Fig. 5). The 3H and i4C radiolabels showed similar relationships, as would be expected. After 8 h all three radiolabels were relatively more apparent in the microsome fractions, a finding that is consistent with breakdown and resynthesis following PC catabolism. Discussion Studies of surfactant metabolism have been complicated by technical difficulties. In whole animal studies, it has not been possible to tell reliably how much extracellular surfactant PC
contaminated the various subcellular fractions. It has also not been possible to differentiate extracellular surfactant components that have been endocytosed and resecreted from those that remained in the alveolar pool. Mathematical models such as those of Zilversrnit et al. [25] have been used to estimate turnover times and percent reutilization and were probably fairly accurate in the healthy animal with stable pool sizes [1,4]. However, animals with immature lungs or lung injury may have unstable surfactant pools. In addition, the mathematical models assume a strict precursor-product relationship, which may not always be the case. In vitro studies have the disadvantage of altering the normal physiologic environment of the type II cell. Histochemical studies have utilized lectins which were specific for receptors on type II cells, but such lectins may not follow the same pathways as recycled surfactant phospholipids [5,6]. An ideal probe of surfactant phospholipid reutilization is one that can be used without toxic effects in animals and will permit the differentiation of extracellular surfactant PC contaminants from PC localized during subcellular fractionation. It will allow for separation of resecreted alveolar PC from phospholipids that were never endocytosed. Finally, the ideal probe will have similar kinetics as it follows the same pathways of uptake, subcellular processing, resecretion, and clearance as the phospholipids of interest. A ‘metabolic probe’ such as 1ysoPC which is converted to PC rapidly after uptake is attractive. since unconverted extracellular contaminants can easily be separated from labeled subcellular components by thin-layer chromatography. Resecreted labeled PC is similarly separated from labeled 1ysoPC that was never endocytosed and converted to PC by the lung. One potential problem with 1ysoPC is its cytotoxicity with lytic effects on cell membranes [26]. Increased permeability of the alveolar epithelium to both sucrose and dextran 70 (M, 70000) was seen in perfused hamster lungs with intratracheal administration of 1ysoPC in concentrations as low as 8 pg/ml [27]. While our intratracheal injections contained 3.5 pg/ml and 12.5 pg/ml of 1ysoPC in the 6 and 24 h protocols, respectively, the total dose administered per kg body weight was only
335
10% of the lowest dose administered in the perfused lung model. In addition, our 1ysoPC was son&ted into liposomes with DPPC and then associated with rabbit surfactant prior to administration [19]. Labeled DPPC sonicated into liposomes and associated with rabbit surfactant in this manner was previously shown to function metabolically like endogenously labeled rabbit surfactant PC 119,281. The clinical status of the animals and the gross appearance of the lungs did not indicate any toxicity and there was no significant increase in the total proteins recovered from the alveolar washes compared to untreated 1 kg rabbits (40 mg protein/alveolar wash in the rabbits receiving 25 pg 1ysoPC vs. 55 mg protein alveolar wash in untreated 1 kg rabbits [20]). There was very rapid uptake in the lysoPC, with 43% of the total counts given being lung associated by 10 min after administration. This finding was consistent with the rate of uptake seen in the perfused hamster lung and the increases in liposome fusion to mammalian cells in culture when 1ysoPC was added to mixtures of dipalmitoylphosphatidylglycerol and DPPC [26]. The uptake of 1ysoPC was more rapid than that of rabbit surfactant PC, which was 33% lung associated by 10 min. There was a similar rapid and efficient conversion of 1ysoPC to PC seen at much higher doses in the perfused hamster lung [27]. At the subcellular level, 1ysoPC clearly deviated from the pathway of endocytosed rabbit surfactant PC during uptake and intra~llul~ conversion to PC. It appeared to follow a pathway of remodeling that paralleled that of labeled IV palmitate which probably entered pathways of both remodeling and de novo synthesis. In addition, the normalized specific activities for saturated PC derived from IV palmitate and intratracheal 1ysoPC were lower than one might expect in the lamellar body and microsome fractions. In contrast, the intratracheal natural rabbit surfactant had much higher specific activities in the lamellar body fractions than the lung homogenate and was therefore probably much more specifically localized to lamellar bodies and type II cells. Therefore, synthesis of lung saturated PC utilizing 1ysoPC or intravascular palmitate could have occurred largely in cell types other than type II cells or lung saturated PC may have been
transported to other cell types following synthesis in type II cells. However, the high percent saturation of the phosphatidylcholine derived from 1ysoPC suggests targeting of the type II cell. In a recent review, Wright and Clements estimated that only 21% of lung saturated PC was in the type II cell [29]. While any lack of efficiency in recovering type II cells and loss of saturated PC during isolation would make their estimate low, any localization of saturated PC outside the type II cells could help explain why the normalized specific activities are lower in the lamellar bodies following synthesis or remodeling when compared to endocytosed saturated PC. An alternative explanation is that our fractionation procedure missed an important subcellular site of 1ysoPC metabolism. In the perfused hamster lung given higher doses of intratracheal lysoPC, the major toxicity by electron microscopy was seen in type I cells 1271. The mechanism of the deviation of 1ysoPC from the recycling pathway of endocytosed saturated PC-is unknown, but could be related to specific transfer proteins [30]. Prior to conversion to saturated PC the clearance of 1ysoPC could not be used as a measure for surfactant clearance. However, once 1ysoPC was converted to saturated PC, it appeared to be cleared from the lung at the same rate as labeled rabbit surfactant PC. For the 24 h protocol we utilized 1ysoPC labeled both with 3H in the choline moiety of the backbone and 14C in the I-palmitoyl side-chain in order to evaluate the relative cont~butions of lysoacyltransferase and lyso : lyso acyltransferase, since any conversion by the latter enzyme would result in a decrease in the isotope ratios if two labeled 1ysoPC molecules participated in the reaction. The 3H/‘4C ratio of 0.8-1.0 for radioactivity in PC converted from 1ysoPC in the total lung from 1 to 16 h compared to the ‘H/14C ratio in the solution used to inject the rabbits indicated that the 1ysoPC was predominantly converted to PC by lysoacyltransferase with the addition of an unlabeled fatty acid via an acyl-CoA to the 2-position of the glycerol backbone. The combination of the high efficiency and rapid conversion of 1ysoPC to PC and the 3H/14C ratios close to 1 indicated intact uptake and conversion of the 1ysoPC to PC rather than breakdown and resynthesis from labeled precursors at early times. However, at 24 h
336
the ratio of [ 3H]choline/[‘4C]palmitate increased significantly in both the total lung and the various subcellular fractions accompanied by an increase in the relative proportion of activity in the microsomes. These findings were consistent with a shorter turnover time for palmitate than choline and/or the occurrence of breakdown and resynthesis with preferential conservation of labeled choline as found by Chander et al. [31] and Fisher et al. [32] in the isolated perfused lung and type II cells of the adult rat. However, we saw no significant increase in the percent of unsaturated PC or 1ysoPC with time in any of our radiolabels or subcellular fractions (data not shown) as was reported in the above in vitro models. It is possible that in those studies the PC remodeling pathway was somehow depressed by the in vitro conditions. This study showed that intratracheally administered 1ysoPC was taken up rapidly and efficiently by the adult rabbit lung where rapid conversion to PC occurred in the endoplasmic reticulum followed by movement to the lamellar bodies and resecretion into the alveolus. The metabolic conversion made it simple to differentiate resecretion from decreased uptake and to eliminate contamination of subcellular fractions by extracellular label. The lung uptake of 1ysoPC was more rapid than PC and 1ysoPC took a pathway that paralleled that of the remodeling and de novo synthetic pathways followed by labeled IV palmitate. However, while the conversion pathway of 1ysoPC to PC deviated from that of recycled surfactant PC, it appeared to parallel the clearance and resecretion pathways after conversion. Therefore, 1ysoPC is useful as a ‘metabolic probe’ of phospholipid reutilization, especially in the immature and/or injured lung, when it is given intratracheally in combination with a differentially labeled surfactant PC used to evaluate initial uptake. Perhaps other phospholipid probes exist which would be structurally modified in multivesicular or in lamellar bodies and could be used to further evaluate the specific pathways of initial uptake and preferential lamellar body localization of recycled saturated PC in the lung. Acknowledgement This work was supported by Grant HD-12714 from the Department of Health and Human Services.
References 1 Jacobs,
H.C., Ikegami, M., Jobe, A.H., Berry, D.D. and Jones, S. (1985) Biochim. Biophys. Acta 837, 77-84. 2 Hallman, M., Epstein, B.L. and Gluck, L. (1981) J. Clin. Invest. 68, 742-751. 3 Glatz, T., Ikegami, M. and Jobe. A. (1982) Pediatr. Res. 16. 711-715. 4 Jacobs, H., Jobe, A., Ikegami, M. and Jones, S. (1982) J. Biol. Chem. 257. 1805-1810. 5 Williams, M.C. (1984) Proc. Natl. Acad. Sci. USA 81. 6054-6058. 6 Williams, M.C. (1984) Proc. Natl. Acad. Sci. USA 81. 6383-6387. 7 Jacobs, H., Jobe, A., Ikegami, M., Miller, D. and Jones. S. (1984) Biochim. Biophys. Acta 793, 300-309. 8 Hallman, M. and Raivio. K. (1974) Pediatr. Res. 8. 874-879. 9 Lands, W.E.M. (1960) J. Biol. Chem. 235, 2233-2237. 10 Van den Bosch, H., Bonte, H.A. and Van Deenen, L.L.M. (1965) Biochim. Biophys. Acta 98. 648-651. 11 Van Heusden, G.P.H. and Van den Bosh, H. (1981) Biochim. Biophys. Acta 666, 508-511. 12 Oldenborg, V. and Van Golde, L.M.G. (1976) 441.433-442. 13 Tsao, F.H.C. and Zachman, R.D. (1977) Pediatr. Res. 11. 849-857. 14 Oyarzun, M.J., Clements. J.A. and Baritussio. A. (1980) Am. Rev. Respir. Dis. 121, 709-721. 15 Jacobs. H.C., Jobe, A.H., Ikegami, M. and Jones, S. (1983) Biochim. Biophys. Acta 752, 178-181. 16 Jobe, A., Ikegami, M., Glatz, T., Yoshida, Y., Diakomanolis, E. and Padbury, J. (1981) J. Clin. Invest. 67, 370-375. 17 Stoffel, W., Lekim, D. and Tschung. T.S. (1971) Z. Physiol. Chem. 352, 1058-1064. 18 Mason, R.J., Nellenbogen, J. and Clement% J.A. (1976) J. Lipid Res. 17, 281-284. 19 Ikegami, M., Jobe, A. and Duane. G. (1985) Biochim. Biophys. Acta 835, 352-359. 20 Pettenazzo, A., Ikegami, M., Seidner, S. and Jobe. A. (1988) J. Appl. Physiol. 64, 120-I 27. 21 Toshima, N. and Akino. T. (1972) Tohoku. J. Exp. Med. 108, 253-263. 22 Jobe, A. (1977) B&him. Biophys. Acta 489, 440-453. 23 Bligh, E.G. and Dyer. W.J. (1959) Can. J. Biochem. Physiol. 37, 911-917. 24 Bartlett, G.R. (1959) J. Biol. Chem. 234, 466-468. 25 Zilversmit. D.B., Entenman, C. and Fishler, M.C. (1943) J. Gen. Physiol. 26, 325-331. 26 Papahadjopoulos, D.. Poste, G. and Schaeffer, B.E. (1973) Biochim. Biophys. Acta 323, 23-42. 27 Niewoehner, D.E., Rice, K., Sinha, A.A. and Wangensteen, D. (1987) J. Appl. Physiol. 63, 1979-1986. 28 Jacobs, H., Jobe, A., Ikegami, M. and Jones, S. (1982) J. Biol. Chem. 257, 1805-1810. 29 Wright, J.R. and Clement& J.A. (1987) Am. Rev. Respir. Dis. 135, 426-444. 30 Spalding, J.W. and Hook, G.E.R. (1979) Lipids 14, 606-613. 31 Chander, A., Reicherter, J. and Fisher, A.B. (1987) J. Clin. Invest. 79, 1133-1138. 32 Fisher, A.B.. Dodia, C. and Chander. A. (1987) J. Appl. Physiol. 62, 2295-2299.