- Email: [email protected]
The importance of the liver in normal and silicotic lung-lipid homeostasis
ENVIRONMENTAL
RESEARCH
19, 432-441 (1979)
The Importance of the Liver in Normal and Silicotic Lung-Lipid Homeostasis’ 2. Cholesterol
C.D.
ESKELSON, VIRGINIA~TIFFEL, MILOS
Division
of Surgical
J.A. OWEN,
AND
CHVAPIL
Biology, Department of Surgery, University of Arizona, Health Sciences Center, Tucson, Arizona 85724
Arizona
Received October 26. 1978 We studied the total content, concentration, and rate of synthesis of cholesterol in the lung, liver, and serum of rats intratracheally injected with quartz or saline. Intact rats served as controls. The animals were sacrificed at four time intervals ranging from 6 to 144 hr after the treatment. The presented data are based on analyses of 87 rat lungs. Instillation of silica into rat lung results in a prompt and significant increase in cholesterol content in the lung amounting to a three- to fourfold increase over the control values within 3 to 6 days after the instillation. These changes, when correlated with cholesterol changes in the serum and liver, indicate that the lung injury initiates synthesis of choiesterol in the liver. This lipid species (as well as other lipids) is then transported into the serum and deposited in the lung. The results indicate that some of the cholesterol in silica-treated rats is synthesized directly in the injured lung. The dynamics of the changes in the above indicators of cholesterol synthesis and deposition in the three tissue standards indicate that the early changes are a part of the nonspecific stress reaction (surgical trauma) and the later changes are specific to the silica related injury in the lung. Thus, the injured lung mediates the liver cholesterolgenesis and cholesterol transport into the serum and accumulation in the lung.
INTRODUCTION
A rapid increase in lung lipid content was documented soon after instillation of quartz into the lung of experimental animals (Babushkina, 1977). She found a significant increase of triglycerides and free fatty acids only 10 days after intratracheal instillation of quartz. Three and six months later all lipid species were strikingly elevated. Katsnelson et al. (1964) found a close relation between the degree of fibrogenicity of the tested dust and the magnitude of lipid accumulation in the lung. In fact, it appears that the outburst of collagen synthesis in various models of fibroproliferative inflammation is preceded by definite changes in lipid in the affected tissue (Pelliniemi, 1973; Popper and Schaffner, 1970; Chvapil and Peng, 1975; Chvapil et al., 1976). The significance of early changes in lipid composition in inflammatory lesions is unknown. There has been speculation about direct stimulation of tibroblasts by some lipids or their degradation products, change of fluidity of cell membranes following changes of cell activity, or increases in lipid peroxidation in the case of accumulated polyunsaturated fatty 1 Supported by Public Health Service Grant HL 19633. 432 0013-9351/79/040432-10$02.00/O Copyright All rights
@ 1979 by Academic Press, Inc. of reproduction in any form reserved.
LIVER
CHOLESTEROL
IN
433
SILICOSIS
acids. A close correlation between the amount of cholesterol and collagen deposited in granulation tissue was reported (Pelliniemi, 1973). The sources for the increased lipids following pulmonary injury are not well understood but some studies suggest that lung has the capacity for lipidgenesis (Tombropoulos and Hadley, 1976; Smith and Kikkawa, 1978). Our recent studies suggest, however, that phospholipids synthesized in the liver are transported into the silica injured lung. We further find that acutely damaged silicotic lung stimulates liver phospholipidgenesis and phospholipid mobilization from the liver to the serum (Eskelson et al., unpublished observations). A phospholipid, dipalmitoyl lecithin, is the major lipid component of lung surfactant; however, other lipids such as cholesterol and triacylglycerols are also components of surfactant (Ramirez et al., 1971; Ramirez and Harlon, 1968). Several histochemical data indicate that, in silicotic lung, the lipids accumulate in the interstitial tissue as well as in the periphery of the silicotic nodules and are mainly associated with macrophages (Katsnelson et al., 1976). It is the purpose of this study to determine if the acutely injured silicotic lung has the ability to stimulate liver cholesterolgenesis and cholesterol mobilization from the liver to the serum as already indicated for liver phospholipid metabolism (Eskelson et al., unpublished observations). METHODS Animal model. Sprague-Dawley male rats, 220-275 g body wt, were anesthetized by ether. A l.O-cm long midline incision was made on the anterior neck surface directly over the submaxillary gland. By blunt dissection, the trachea was exposed and 0.75 ml of either saline or silica DQ 122 (100 mg/ml of 0.9% NaCl) was instilled into the trachea through a 20-gauge needle. After a 3- to 15-set apneic pause, all animals recovered. Wound closure was made with 4-O (Ethicon, Inc.) suture. A control group of nine rats was not surgically manipulated nor was the group exposed to ether or saline and served as a basic control rat group. Sixty minutes before sacrifice the rats were given 20 FCi of sodium [2‘*C]acetate and sacrificed at 6, 24, 72, and 144 hr after silica exposure. Six to twenty-four animals were used in each group. The blood was collected by cardiac puncture, serum obtained and stored at -20°C. The animals were sacrificed by exsanguination. The carotids were cut and liver perfused through the inferior vena cava with 50 ml of 4°C isotonic saline. The liver and lungs were removed, blotted on filter paper, and weighed. Samples of liver and lung were taken for histological and chemical studies. The remaining liver and lung samples were frozen in sealed tubes and maintained at -70°C until analyzed. Lipid extraction fiorn tissue and serum. To I- 1.5 g of lung or liver, 10 ml of a chloroform-methanol (2:1), Folch reagent was added (Folch et ul., 1951). The tissues were homogenized for 30 set using a Polytron homogenizer. The homogenizer was rinsed twice for 15-25 set with 10 ml of the chloroform-methanol mixture. The washings were added to the original homogenate in a 35ml glass stoppered centrifuge tube and the homogenate and wash were mixed ‘The crystalline quartz was kindly given to us by Professor Pathology, University of Newcastle upon Tyne, England.
A. G.
Heppleston,
Department
of
434
ESKELSON
ET
AL.
thoroughly and allowed to stand for 16-20 hr. The supernatant, obtained by centrifugation for 15 min at 2OOg, was dried by adding l-2 g of anhydrous sodium sulfate and allowing it to stand for more than 1 hr. One milliliter of serum was extracted with 20 ml of Folch reagent (Folch et al., 1951), mixed thoroughly, and treated similarly to the tissue homogenate. All of the lipid analyses were done on aliquots of these extracts. Cholesterol method. Cholesterol was determined by a method described by Zak (Zak, 1965). A 0.3-ml dried Folch homogenate extract was used for tissue and 1 ml of extract for serum. Thin-layer chromatographic (tic) separation of lipid classes. A 5-ml aliquot of the Folch tissue and serum extracts was dried at 60-70°C under nitrogen. Carrier cholesterol was added to the extract along with a small amount of chloroform-methanol (2:l) and the sample was streaked on a silica gel G thin-layer plate. The thin-layer chromatographic method was developed by Okabe (Okabe et al., 1974). The cholesterol and cholesterol ester spots on the tic plates were visualized by 0.2% 2-7-dichlorofluorescein-ethanol spray reagent. Specific activity (SA) determination of cholesterol. The cholesterol and cholesterol ester spots on the tic plates were scraped off separately into scintillation vials. Two milliliters of acetic acid were added to each vial and 10 ml of POPOP, PPO solution (3 g/liter PPO) (2,5-diphenyloxazole) and 150 mg/liter of POPOP (1,4-bis-2-(methyl-5-phenyloxazolyl)-benzene). The vials were mixed and the amount of radioactivity in each vial determined on a Beckman IS 250 liquid sciutillation counting system. The amount of quench for each sample was determined using [C”]toluene as an internal standard. The total amount of cholesterol found in each Folch extract was related to the amount of radioactivity in the cholesterol and cholesterol ester spots from the tic. The specific activity of total cholesterol was determined by dividing the dpm found in the two cholesterol (cholesterol and cholesterol esters) spots from the tic by the total amount of cholesterol in the extract. Statistical analysis. A Duncan’s new multiple range statistical analysis (Duncan, 1955) was used to evaluate the data at all time periods. A 95% confidence level was chosen for the significance level. RESULTS
Changes in Liver Cholesterol following Administration of Intratracheal Silica There was no change in the concentration of cholesterol in the liver of rats sacrificed 6 hr after intratracheal injection of silica or saline. At this time interval, however, both total liver cholesterol content and its rate of synthesis were significantly increased in silica-treated rats as compared with controls (Figs. lA, B, and C). In later sampling periods total cholesterol in the liver of silica-treated rats was significantly decreased when compared to both saline-treated or untreated control rats. While at 24 hr the concentrations of cholesterol in the liver of both saline- and silica-treated rats were significantly decreased, both groups returned to the values of the control group at later times (Figs. 1A and B). The rate of liver cholesterol synthesis showed striking fluctuations after instillation of saline and silica (Fig. IC). It is worth stressing that at 6 days after the
LIVER
ii z’ qo.>e: f
a ‘0 ; ‘; 2
5
CHOLESTEROL
. Control ‘::,g ,./‘.&/‘~”
IN
435
SILICOSIS
Sallno c e-.-.-‘-‘-‘-‘* _,,,, o.....““~iiiCS”““”
43-
i2-r
A
I,
0
20
11 40
60
I 60
100
I1 120
140
HOW8
E"
20 1
10 -
0,
0
I I ,B , 20
40
60
I I ,
60
100
120
140
100
120
140
Hours 301
0
20
40
60
60
FIG. 1. Cholesterol changes in the liver following intrapulmonary injection of saline and silica. Each point in the graphs represents the mean value on a wet weight basis of 7- 14 samples at various times:
Hours 6 24 72 144 Control (n
Saline
Silica
n
n
7 7
7 8
14 13
14 13
= 9)
Graph A: Cholesteral concentration in the liver (mg/g liver) F = 5.24. Graph B: Cholesterol content of liver (mgiliver) F = 7.24. Graph C: Specific activity of cholesterol in liver (dpm/mg liver cholesterol) F = 3.44. *Significantly different from control group P s 0.05. **Significantly different from both control and saline groups, P c 0.05. +Significantly different from saline group, P s 0.05.
436
treatment, compared
ESKELSON
the synthesis was significantly to the saline-treated rat group.
ET AL.
stimulated
in the silica-treated
group
Lung Cholesterol Changes The increase in total cholesterol (Fig. 2B) was paralleled with a striking increase in the wet weight (Table 1) as well as dry weight of the lung and by the increase in the specific activity (SA) of cholesterol (Fig. 2C). Because of the change in the lung weight, the concentration of cholesterol in silica-treated rats significantly decreased at all sampling periods. After saline intratracheal injection, cholesterol concentration in the lung significantly increased at only 6 hr and 6 days (Fig. 2A). Saline administration had no effect on the rate of cholesterol synthesis in lung and liver tissue (Figs. 1C and 2C). Serum Cholesterol Changes following Intrapulmonary Silica Treatment Serum cholesterol levels rose approximately three times above normal 6 hr following saline or silica instillation into rat lung. The cholesterol levels then returned to normal values (Fig. 3A). In silica-treated rats, the serum cholesterol SA decreased at 6 hr and reached a maximum decrease to 35% of control values by 24 hr. It increased to the control values in saline-treated rats at 3 days, then decreased from the control and silica values at 6 days (Fig. 3B). The shape of this curve is similar to changes in specific activity of cholesterol in the liver of silica-treated rats (Fig. 3B). DISCUSSION
We studied early changes in the chemical composition of the lung, mainly in the lipids, after silica-induced lung injury. We have found that instillation of silica into the lung stimulates synthesis and transport of liver phospholipids which are then deposited within the inflamed lung tissue (Eskelson et al., unpublished observations). The most striking change after a large, single intratracheal dose of silica is the quadrupling of the lung weight 6 days after the insult (Table 1). This increase reflects an increase in both water content (edema) and dry mass of the lung tissue. Various species of lipids obviously participate in this increase in lung mass. Six days after silica instillation, the phospholipids increased from an original 37.8 to 149 mg/lung with a net increase of 111.2 mg. Under the same conditions 6 days after SiOs, lung cholesterol increased from 7.5 to 27 mg/lung. These results as well as our previous study indicate that silica injections into the lung .induce acute lung damage which activates lipid metabolism in the lung and liver. The possible activation of other organs cannot be ruled out. The mechanism of the liver stimulation by possible mediators released from the injured lung is unknown, Our results point to two sets of stimuli promoting the lung-liver reactivity. The first type of stimulus is represented by the surgical manipulation used, i.e., ether, surgical operations, and intratracheal injections of the saline or silica suspension in saline. This induces a stress reaction, thus increasing serum cholesterol at early times after the surgery (6 hr, Fig. 3A). At this early time, the lung is also probably under biochemical stress from the loss of surfactant due to intrapulmonary saline injection (Ramirez et al., 1971) and possible ether exposure. Surfactant is a protein-lipid complex containing principally phospholipids (Ramirez et al., 1971) but
LIVER
CHOLESTEROL
431
IN SILICOSIS
Q*+ 2r s-k., ;‘l. o
‘Y
6-.&
Control '\
ii5 \ t 5- ‘,; 54
,-
‘N.‘Q.-.‘.-’ # Saline,./'
#:
Sil~~c~,,..:.~.'"
-
'...v '...v
_...... _......
0
20
( 40
,,,,..~.~-~~"
C : E
32,,
A
, 60
, 80
, 100
( 120
, 140
HOW8 300 5 A i
2520-
. . . . . . ..'
Silica . . . . .. . .
..";:
y.."
i f
IS-
r I P::
~o-:~:~:I:~:_.--.-._~~!t~~-.ti
,.....
5
Control
E o0
1 20
I 40
6 I 60
I 60
I 100
1 120
1 140
nourr 12
Silica xx ,* . . . . . . . .._.........
IO
"" $f
,:' I
b 0 0
I 20
1 40
1 60
I 80
I 100
1 120
1 140
FIG. 2. Cholesterol changes in rat lung following intrapulmonary injection of saline and silica. Each point in the graph represents the mean value on a wet weight basis of 7- 14 samples at various times: Hours 7 24 72 144 Control (n = 9)
Saline n
Silica II
7 7 14 13
7 8 14 13
Graph A: Cholesterol concentration in the lung (mpig lung) F = 30.91. Graph B: Cholesterol content in lung (mgkng) F = 111.5. Graph C: Specific activity of cholesterol in lung (dpm/mg lung cholesterol) F = 10.47. *Significantly different from control group value, P c 0.05. **Significantly different from control and saline group values, P = 0.05.
ESKELSON
438
TABLE LUNG
Treatment Control
0 hr
ET AL. 1
WEIGHTSO
6 hr
24 hr
72 hr
144 hr
1.17 -+ 0.10 (7) 1.65 2 0.16’ (7)
1.33 f 0.12 (7) 2.60 * 0.21c
1.42 k 0.19* (14)
1.54 2 0.29b
4.06 k 0.75’ (14)
4.97 f 0.92’
1.14 ‘- 0.06
(8) Saline Silica
(8)
(20) (20)
n Lung weights are expressed in grams wet weight; the variability is given by SD. The number in parenthesis represents the number of animals in each group. * Significantly different from control lung weights at P G 0.05. c All silica-treated lung weights are significantly different from the saline and control lung weights
(F = 108.6,P c 0.05).
also containing cholesterol and other neutral lipids (Ramirez ef al., 1971). It is saline, that a strong lung expected, due to the saline “wash”, i.e., intrapulmonary lipid response will occur early after the manipulative procedures. The actual silica response is somewhat slower since it is mediated, in part, by macrophage invasion and silica phagocytosis which eventually results in cell lysis and lysosomal enzyme release resulting in lung damage and fibrosis (Heppleston, 1%9; Gross, 1%8; Heppleston er al., 1974). Serum cholesterol levels are transient since they reflect a dynamic equilibrium in which both a cholesterol input into the serum and cholesterol removal from the serum occurs. Thus, when serum cholesterol levels increase or decrease, there is either an increased or decreased removal of cholesterol from the serum or an increased or decreased input of cholesterol into the serum. Serum cholesterol SA reflects: (a) either an increased or decreased cholesterolgenesis which occurs in some tissue and is transported by the serum, (b) a decreased or increased release of de nova-formed cholesterol into the serum, (c) an increased or decreased input of nonlabeled cholesterol (depot cholesterol) into the serum, and (d) a specific uptake or decreased uptake of some newly formed radioactive cholesterol complex, i.e., specific lipoprotein cholesterol complex, cholesterol ester versus free cholesterol, from the serum and in which a normal serum cholesterol pool does not participate. In light of the above concepts, both serum cholesterol levels, serum cholesterol SA, and cholesterolgenesis in tissues are important for interpreting cholesterol metabolism occurring in lung tissues. In some mammalian species, approximately 80-85% of the cholesterol in the body is synthesized in the liver (Dietschy and Wilson, 1968) and only 0.1% in the lung. Thus, cholesterol buildup in the lung is probably caused by cholesterol being transported from the liver and other depots to the lung via the serum. Serum cholesterol SA decreased in both saline- and silica-treated rats at 6 and 24 hr, but serum cholesterol levels increased at 6 hr suggesting that nonlabeled cholesterol is being mobilized to the serum (Figs. 3A and B). The saline- and silica-treated rats
LIVER
CHOLESTEROL
IN
439
SILICOSIS
Control ; ‘; 5 c u
so-
,
I
r
AI
I
I
I
20
40
80
80
100
120
140
: oi 0
Hours
5 t 3 5 Y= 00 ra
ME;
20' 15-h 10-t: & ! '..
,:.
,:I
a..
".....
1Oo, 0 20J. 40
control
'...., ..\yled'~...gX
:' ,. A. :.',,.
8- &q.&’
. . ..
&b...
B
1 80I1100
80
-.8 1 140I
120
Hours
FIG. 3. Cholesterol changes in rat serum following intrapulmonary injection of saline and silica. Each point in the graph represents the mean value of 7- 14 samples at various times: Hours 6
Saline n
Silica n
7
7
24 72
7 14
8 14
144 Control (n = 9)
13
13
Graph A: Serum cholesterol level (mg/IOO ml serum) F = 18.911. Graph B: Serum cholesterol specific activity (dpm/mg cholesterol) F = 7.06. *Significantly different from control group value, P c 0.05 level. **Significantly different from both control and saline group values, P s 0.05 level.
responded by mobilizing cholesterol from cholesterol depots, but only the silicatreated rats responded by increasing liver cholesterolgenesis. At 6 hr the silicatreated rats responded with an increased cholesterol content in the liver and increased liver cholesterolgenesis. The finding that cholesterol mobilization is probably occurring from lipid depots other than the liver is suggested by the rise in liver cholesterol content at 6 hr without a corresponding increase of liver cholesterol SA in saline-treated rats.
440
ESKELSON
ET AL.
A crystalline quartz insult to the lung resulted in decreasing the cholesterol concentration (mg/g) in the lung but increasing its cholesterol content due to increased lung weight. The specific activity of lung cholesterol began to increase as early as 6 hr with a more rapid increase at 72 and 144 hr after the silica insult to the lung. This increase in cholesterol SA indicates that cholesterolgenesis and de nova cholesterol buildup are occurring in the lung (Fig. 2). We statistically correlated lung cholesterol levels with serum cholesterol levels in saline-treated rats, which indicated that a relationship exists between the two systems at a significance level greater than 99.9% (correlation between Figs. 2 and 3). This finding and the loss of liver cholesterol content indicates that the liver is losing cholesterol to the lung by mobilizing cholesterol into the serum. Six-hours postsurgery the liver responded by increasing cholesterol content and cholesterol SA which is associated with a highly significant increase in serum cholesterol levels. It seems that the liver over responded at this time because there was an increased liver cholesterol content associated with increased liver cholesterol SA. The increased liver cholesterol may have activated the cholesterolgenic control systems by tuming off liver cholesterolgenesis (Higgins and Rudney, 1973; Frantz 41 al., 1954; Block, 1965). This is indicated by the low liver cholesterol SA at 24-hr postsurgical manipulation. The turning off of liver cholesterolgenesis in the silica-treated rats resulted in a significant decrease in liver cholesterol content at 24 hr compared to the saline-treated and control rat liver cholesterol levels (Figs. IA and B). This decrease in liver cholesterol levels then turned on liver cholesterolgenesis which remained elevated above the 24-hr cholesterolgenic level as suggested by the increased liver cholesterol SA in the silica-treated rats. These data suggest that liver cholesterolgenesis is stimulated by silicotic rat lung and that the de nova-formed cholesterol of the liver is transported more rapidly to the serum. This serum cholesterol is then removed by the silicotic lung more rapidly so that cholesterol input from the liver equals its uptake by the lung. This rate is not only sufficient to keep pace with the cholesterol buildup in silicotic lungs (0.145 mglhr), but is sufficient to maintain constant serum cholesterol levels between 24 and 144 hr after silica treatment (Figs. 2B and 3A). At no time does intrapulmonary saline treatment seem to induce increased liver cholesterolgenesis, but does induce the removal of the de nova-formed liver cholesterol to maintain lung and other cholesterol homeostasis. ACKNOWLEDGMENT The authors wish to express deep gratitude to Dr. Ronald L. Misiorowski for his interest and advice in our project. We are particularly grateful to him for developing and providing us with computer programs to facilitate our statistical analysis.
REFERENCES 1. Babushkina, L. G. (1977). Role of lipids in the pathogenesis of sihcosis. In “Industrial Diseases Due to Exposure to Dust” (S. G. Domin and B. A. Katsnelson, Eds.), pp. 104- 114. Ministry of Health, Moscow, Russia. 2. Block, K. (1965). The biological synthesis of cholesterol. Science 150, 19-28. 3. Chvapil, M., and Peng, Y. M. (1975). Oxygen and lung fibrosis. Arch. Environ. Health 30, 528-537. 4. Chvapil, M., Stankova, L., and Malshet, V. (1976). Lipid peroxidation as one of the mechanisms of silica fibrogenicity. 1. Study with erythrocytes. Environ. Res. 11, 78-88.
LIVER
CHOLESTEROL
IN
SILICOSIS
441
5. Dietschy, J. M., and Wilson, J. D. (1968). Cholesterol synthesis in the squirrel monkey: Relative rates of synthesis in various tissue and mechanisms of control. J. C/in. Invest. 47, 166-174. 6. Duncan, D. B. (1955). Multiple range and multiple F test. Biometrics 11, l-42. 7. Eskelson, C. D., Stiffel, V., Owen, J. A., and Chvapil, M. The importance of liver in normal and silicotic lung-lipid homeostasis. 1. Phospholipids (Unpublished observations.) 8. Folch, J., Ascoli, I., Lees, M., Meath, J. A., and LeBaron, F. N. (1951). Preparation of lipid extracts from brain tissue. J. Biol. Chem. 191, 833-841. 9. Frantz, I. D., Schneider, H. S., and Kinkelman, B. T. (1954). Suppression of hepatic cholesterol synthesis in the rat by cholesterol feeding. J. Biol. Chem. 206, 465-469. Health 17, 720-725. 10. Gross, P. (1968). Experimental “acute” silicosis. Arch. Environ. 11. Heppleston, A. G. (1969). The flbrogenic action of silica. Brit. Med. Bull. 25, 282-287. 12. Heppleston, A. G., Fletcher, K., and Wyatt, I. (1974). Changes in the composition of lung lipids and the “turnover” of dipalmitoyl lecithin in experimental alveolar Iipoproteinosis induced by inhaled quartz. Brit. .I. Exp. Pathol. 55, 384-395. 13. Higgins, M., and Rudney, H. (1973). Regulation of rat liver P-hydroxy-P-methylglutaryl-CoA reductase activity by cholesterol. Nature New Biol. 246, 60-61. 14. Katsnelson, B. A., Babushkina, L. G., and Velixchdovskii, B. T. (1964). Changes in the total lipid content of the lungs of rats with experimental silicosis. Bull. Exp. Biol. Med. 57, 699-702. 15. Katsnelson, B. A., Babushkina, B. T., and Velixchdovskii, B. T. (1976). Pathogenesis, pathogenetic therapy and prophylaxis of silicones. Toxicology 7, 24-73 (in Russian). 16. Okabe, A., Katayama, T., and Kanemasa, Y. (1974). The micromethod for determination of cholesterol, cholesterol esters and phospholipids. Acta Med. Okayama 28, 403-410. 17. Pelliniemi, T. (1973). “Lipids of Connective Tissue with Special Reference to the Effect of Dietary Hyperlipidemia and Oxygen Deprivation on Experimental Granulation Tissue in the Rat.” Academic dissertation, Polytypos Turku. 18. Popper, H., and Schaffner, F. (1970). Alcohol cirrhosis and other toxic hepatopathias. In “Nordiska Bokhandelns Fortag” (A. Engel and T. Larrson, Eds.), pp. 15-46, Stockholm. 19. Ramirez, R. J., Durham, N. C., Schwartz, B., Dowell, A. R., and Lee, S. D. (1971). Biochemical composition of human pulmonary washings. Arch. Intern Med. 127, 395-400. 20. Ramirez, R. J., and Harlan, W. R. (1968). Pulmonary alveolar proteinosis. Amer. J. Med. 45, 502-512. 21. Smith, F. B., and Kikkawa, Y. (1978). The type II epithelial cells of the lung III lecithin synthesis: A comparison with pulmonary macrophages. Lab. Invest. 38, 45-51. 22. Tombropoulos, E. G., and Hadley, J. G. (1976). Lipid synthesis by perfused lung. Lipids 11, 491-495.
B. (1965). Total and free cholesterol. in “Standard Methods of Clinical Chemistry” Meites, Ed.), Vol. 5, pp. 79-89. Academic Press, New York.
23. Zak,
(S.
RESEARCH
19, 432-441 (1979)
The Importance of the Liver in Normal and Silicotic Lung-Lipid Homeostasis’ 2. Cholesterol
C.D.
ESKELSON, VIRGINIA~TIFFEL, MILOS
Division
of Surgical
J.A. OWEN,
AND
CHVAPIL
Biology, Department of Surgery, University of Arizona, Health Sciences Center, Tucson, Arizona 85724
Arizona
Received October 26. 1978 We studied the total content, concentration, and rate of synthesis of cholesterol in the lung, liver, and serum of rats intratracheally injected with quartz or saline. Intact rats served as controls. The animals were sacrificed at four time intervals ranging from 6 to 144 hr after the treatment. The presented data are based on analyses of 87 rat lungs. Instillation of silica into rat lung results in a prompt and significant increase in cholesterol content in the lung amounting to a three- to fourfold increase over the control values within 3 to 6 days after the instillation. These changes, when correlated with cholesterol changes in the serum and liver, indicate that the lung injury initiates synthesis of choiesterol in the liver. This lipid species (as well as other lipids) is then transported into the serum and deposited in the lung. The results indicate that some of the cholesterol in silica-treated rats is synthesized directly in the injured lung. The dynamics of the changes in the above indicators of cholesterol synthesis and deposition in the three tissue standards indicate that the early changes are a part of the nonspecific stress reaction (surgical trauma) and the later changes are specific to the silica related injury in the lung. Thus, the injured lung mediates the liver cholesterolgenesis and cholesterol transport into the serum and accumulation in the lung.
INTRODUCTION
A rapid increase in lung lipid content was documented soon after instillation of quartz into the lung of experimental animals (Babushkina, 1977). She found a significant increase of triglycerides and free fatty acids only 10 days after intratracheal instillation of quartz. Three and six months later all lipid species were strikingly elevated. Katsnelson et al. (1964) found a close relation between the degree of fibrogenicity of the tested dust and the magnitude of lipid accumulation in the lung. In fact, it appears that the outburst of collagen synthesis in various models of fibroproliferative inflammation is preceded by definite changes in lipid in the affected tissue (Pelliniemi, 1973; Popper and Schaffner, 1970; Chvapil and Peng, 1975; Chvapil et al., 1976). The significance of early changes in lipid composition in inflammatory lesions is unknown. There has been speculation about direct stimulation of tibroblasts by some lipids or their degradation products, change of fluidity of cell membranes following changes of cell activity, or increases in lipid peroxidation in the case of accumulated polyunsaturated fatty 1 Supported by Public Health Service Grant HL 19633. 432 0013-9351/79/040432-10$02.00/O Copyright All rights
@ 1979 by Academic Press, Inc. of reproduction in any form reserved.
LIVER
CHOLESTEROL
IN
433
SILICOSIS
acids. A close correlation between the amount of cholesterol and collagen deposited in granulation tissue was reported (Pelliniemi, 1973). The sources for the increased lipids following pulmonary injury are not well understood but some studies suggest that lung has the capacity for lipidgenesis (Tombropoulos and Hadley, 1976; Smith and Kikkawa, 1978). Our recent studies suggest, however, that phospholipids synthesized in the liver are transported into the silica injured lung. We further find that acutely damaged silicotic lung stimulates liver phospholipidgenesis and phospholipid mobilization from the liver to the serum (Eskelson et al., unpublished observations). A phospholipid, dipalmitoyl lecithin, is the major lipid component of lung surfactant; however, other lipids such as cholesterol and triacylglycerols are also components of surfactant (Ramirez et al., 1971; Ramirez and Harlon, 1968). Several histochemical data indicate that, in silicotic lung, the lipids accumulate in the interstitial tissue as well as in the periphery of the silicotic nodules and are mainly associated with macrophages (Katsnelson et al., 1976). It is the purpose of this study to determine if the acutely injured silicotic lung has the ability to stimulate liver cholesterolgenesis and cholesterol mobilization from the liver to the serum as already indicated for liver phospholipid metabolism (Eskelson et al., unpublished observations). METHODS Animal model. Sprague-Dawley male rats, 220-275 g body wt, were anesthetized by ether. A l.O-cm long midline incision was made on the anterior neck surface directly over the submaxillary gland. By blunt dissection, the trachea was exposed and 0.75 ml of either saline or silica DQ 122 (100 mg/ml of 0.9% NaCl) was instilled into the trachea through a 20-gauge needle. After a 3- to 15-set apneic pause, all animals recovered. Wound closure was made with 4-O (Ethicon, Inc.) suture. A control group of nine rats was not surgically manipulated nor was the group exposed to ether or saline and served as a basic control rat group. Sixty minutes before sacrifice the rats were given 20 FCi of sodium [2‘*C]acetate and sacrificed at 6, 24, 72, and 144 hr after silica exposure. Six to twenty-four animals were used in each group. The blood was collected by cardiac puncture, serum obtained and stored at -20°C. The animals were sacrificed by exsanguination. The carotids were cut and liver perfused through the inferior vena cava with 50 ml of 4°C isotonic saline. The liver and lungs were removed, blotted on filter paper, and weighed. Samples of liver and lung were taken for histological and chemical studies. The remaining liver and lung samples were frozen in sealed tubes and maintained at -70°C until analyzed. Lipid extraction fiorn tissue and serum. To I- 1.5 g of lung or liver, 10 ml of a chloroform-methanol (2:1), Folch reagent was added (Folch et ul., 1951). The tissues were homogenized for 30 set using a Polytron homogenizer. The homogenizer was rinsed twice for 15-25 set with 10 ml of the chloroform-methanol mixture. The washings were added to the original homogenate in a 35ml glass stoppered centrifuge tube and the homogenate and wash were mixed ‘The crystalline quartz was kindly given to us by Professor Pathology, University of Newcastle upon Tyne, England.
A. G.
Heppleston,
Department
of
434
ESKELSON
ET
AL.
thoroughly and allowed to stand for 16-20 hr. The supernatant, obtained by centrifugation for 15 min at 2OOg, was dried by adding l-2 g of anhydrous sodium sulfate and allowing it to stand for more than 1 hr. One milliliter of serum was extracted with 20 ml of Folch reagent (Folch et al., 1951), mixed thoroughly, and treated similarly to the tissue homogenate. All of the lipid analyses were done on aliquots of these extracts. Cholesterol method. Cholesterol was determined by a method described by Zak (Zak, 1965). A 0.3-ml dried Folch homogenate extract was used for tissue and 1 ml of extract for serum. Thin-layer chromatographic (tic) separation of lipid classes. A 5-ml aliquot of the Folch tissue and serum extracts was dried at 60-70°C under nitrogen. Carrier cholesterol was added to the extract along with a small amount of chloroform-methanol (2:l) and the sample was streaked on a silica gel G thin-layer plate. The thin-layer chromatographic method was developed by Okabe (Okabe et al., 1974). The cholesterol and cholesterol ester spots on the tic plates were visualized by 0.2% 2-7-dichlorofluorescein-ethanol spray reagent. Specific activity (SA) determination of cholesterol. The cholesterol and cholesterol ester spots on the tic plates were scraped off separately into scintillation vials. Two milliliters of acetic acid were added to each vial and 10 ml of POPOP, PPO solution (3 g/liter PPO) (2,5-diphenyloxazole) and 150 mg/liter of POPOP (1,4-bis-2-(methyl-5-phenyloxazolyl)-benzene). The vials were mixed and the amount of radioactivity in each vial determined on a Beckman IS 250 liquid sciutillation counting system. The amount of quench for each sample was determined using [C”]toluene as an internal standard. The total amount of cholesterol found in each Folch extract was related to the amount of radioactivity in the cholesterol and cholesterol ester spots from the tic. The specific activity of total cholesterol was determined by dividing the dpm found in the two cholesterol (cholesterol and cholesterol esters) spots from the tic by the total amount of cholesterol in the extract. Statistical analysis. A Duncan’s new multiple range statistical analysis (Duncan, 1955) was used to evaluate the data at all time periods. A 95% confidence level was chosen for the significance level. RESULTS
Changes in Liver Cholesterol following Administration of Intratracheal Silica There was no change in the concentration of cholesterol in the liver of rats sacrificed 6 hr after intratracheal injection of silica or saline. At this time interval, however, both total liver cholesterol content and its rate of synthesis were significantly increased in silica-treated rats as compared with controls (Figs. lA, B, and C). In later sampling periods total cholesterol in the liver of silica-treated rats was significantly decreased when compared to both saline-treated or untreated control rats. While at 24 hr the concentrations of cholesterol in the liver of both saline- and silica-treated rats were significantly decreased, both groups returned to the values of the control group at later times (Figs. 1A and B). The rate of liver cholesterol synthesis showed striking fluctuations after instillation of saline and silica (Fig. IC). It is worth stressing that at 6 days after the
LIVER
ii z’ qo.>e: f
a ‘0 ; ‘; 2
5
CHOLESTEROL
. Control ‘::,g ,./‘.&/‘~”
IN
435
SILICOSIS
Sallno c e-.-.-‘-‘-‘-‘* _,,,, o.....““~iiiCS”““”
43-
i2-r
A
I,
0
20
11 40
60
I 60
100
I1 120
140
HOW8
E"
20 1
10 -
0,
0
I I ,B , 20
40
60
I I ,
60
100
120
140
100
120
140
Hours 301
0
20
40
60
60
FIG. 1. Cholesterol changes in the liver following intrapulmonary injection of saline and silica. Each point in the graphs represents the mean value on a wet weight basis of 7- 14 samples at various times:
Hours 6 24 72 144 Control (n
Saline
Silica
n
n
7 7
7 8
14 13
14 13
= 9)
Graph A: Cholesteral concentration in the liver (mg/g liver) F = 5.24. Graph B: Cholesterol content of liver (mgiliver) F = 7.24. Graph C: Specific activity of cholesterol in liver (dpm/mg liver cholesterol) F = 3.44. *Significantly different from control group P s 0.05. **Significantly different from both control and saline groups, P c 0.05. +Significantly different from saline group, P s 0.05.
436
treatment, compared
ESKELSON
the synthesis was significantly to the saline-treated rat group.
ET AL.
stimulated
in the silica-treated
group
Lung Cholesterol Changes The increase in total cholesterol (Fig. 2B) was paralleled with a striking increase in the wet weight (Table 1) as well as dry weight of the lung and by the increase in the specific activity (SA) of cholesterol (Fig. 2C). Because of the change in the lung weight, the concentration of cholesterol in silica-treated rats significantly decreased at all sampling periods. After saline intratracheal injection, cholesterol concentration in the lung significantly increased at only 6 hr and 6 days (Fig. 2A). Saline administration had no effect on the rate of cholesterol synthesis in lung and liver tissue (Figs. 1C and 2C). Serum Cholesterol Changes following Intrapulmonary Silica Treatment Serum cholesterol levels rose approximately three times above normal 6 hr following saline or silica instillation into rat lung. The cholesterol levels then returned to normal values (Fig. 3A). In silica-treated rats, the serum cholesterol SA decreased at 6 hr and reached a maximum decrease to 35% of control values by 24 hr. It increased to the control values in saline-treated rats at 3 days, then decreased from the control and silica values at 6 days (Fig. 3B). The shape of this curve is similar to changes in specific activity of cholesterol in the liver of silica-treated rats (Fig. 3B). DISCUSSION
We studied early changes in the chemical composition of the lung, mainly in the lipids, after silica-induced lung injury. We have found that instillation of silica into the lung stimulates synthesis and transport of liver phospholipids which are then deposited within the inflamed lung tissue (Eskelson et al., unpublished observations). The most striking change after a large, single intratracheal dose of silica is the quadrupling of the lung weight 6 days after the insult (Table 1). This increase reflects an increase in both water content (edema) and dry mass of the lung tissue. Various species of lipids obviously participate in this increase in lung mass. Six days after silica instillation, the phospholipids increased from an original 37.8 to 149 mg/lung with a net increase of 111.2 mg. Under the same conditions 6 days after SiOs, lung cholesterol increased from 7.5 to 27 mg/lung. These results as well as our previous study indicate that silica injections into the lung .induce acute lung damage which activates lipid metabolism in the lung and liver. The possible activation of other organs cannot be ruled out. The mechanism of the liver stimulation by possible mediators released from the injured lung is unknown, Our results point to two sets of stimuli promoting the lung-liver reactivity. The first type of stimulus is represented by the surgical manipulation used, i.e., ether, surgical operations, and intratracheal injections of the saline or silica suspension in saline. This induces a stress reaction, thus increasing serum cholesterol at early times after the surgery (6 hr, Fig. 3A). At this early time, the lung is also probably under biochemical stress from the loss of surfactant due to intrapulmonary saline injection (Ramirez et al., 1971) and possible ether exposure. Surfactant is a protein-lipid complex containing principally phospholipids (Ramirez et al., 1971) but
LIVER
CHOLESTEROL
431
IN SILICOSIS
Q*+ 2r s-k., ;‘l. o
‘Y
6-.&
Control '\
ii5 \ t 5- ‘,; 54
,-
‘N.‘Q.-.‘.-’ # Saline,./'
#:
Sil~~c~,,..:.~.'"
-
'...v '...v
_...... _......
0
20
( 40
,,,,..~.~-~~"
C : E
32,,
A
, 60
, 80
, 100
( 120
, 140
HOW8 300 5 A i
2520-
. . . . . . ..'
Silica . . . . .. . .
..";:
y.."
i f
IS-
r I P::
~o-:~:~:I:~:_.--.-._~~!t~~-.ti
,.....
5
Control
E o0
1 20
I 40
6 I 60
I 60
I 100
1 120
1 140
nourr 12
Silica xx ,* . . . . . . . .._.........
IO
"" $f
,:' I
b 0 0
I 20
1 40
1 60
I 80
I 100
1 120
1 140
FIG. 2. Cholesterol changes in rat lung following intrapulmonary injection of saline and silica. Each point in the graph represents the mean value on a wet weight basis of 7- 14 samples at various times: Hours 7 24 72 144 Control (n = 9)
Saline n
Silica II
7 7 14 13
7 8 14 13
Graph A: Cholesterol concentration in the lung (mpig lung) F = 30.91. Graph B: Cholesterol content in lung (mgkng) F = 111.5. Graph C: Specific activity of cholesterol in lung (dpm/mg lung cholesterol) F = 10.47. *Significantly different from control group value, P c 0.05. **Significantly different from control and saline group values, P = 0.05.
ESKELSON
438
TABLE LUNG
Treatment Control
0 hr
ET AL. 1
WEIGHTSO
6 hr
24 hr
72 hr
144 hr
1.17 -+ 0.10 (7) 1.65 2 0.16’ (7)
1.33 f 0.12 (7) 2.60 * 0.21c
1.42 k 0.19* (14)
1.54 2 0.29b
4.06 k 0.75’ (14)
4.97 f 0.92’
1.14 ‘- 0.06
(8) Saline Silica
(8)
(20) (20)
n Lung weights are expressed in grams wet weight; the variability is given by SD. The number in parenthesis represents the number of animals in each group. * Significantly different from control lung weights at P G 0.05. c All silica-treated lung weights are significantly different from the saline and control lung weights
(F = 108.6,P c 0.05).
also containing cholesterol and other neutral lipids (Ramirez ef al., 1971). It is saline, that a strong lung expected, due to the saline “wash”, i.e., intrapulmonary lipid response will occur early after the manipulative procedures. The actual silica response is somewhat slower since it is mediated, in part, by macrophage invasion and silica phagocytosis which eventually results in cell lysis and lysosomal enzyme release resulting in lung damage and fibrosis (Heppleston, 1%9; Gross, 1%8; Heppleston er al., 1974). Serum cholesterol levels are transient since they reflect a dynamic equilibrium in which both a cholesterol input into the serum and cholesterol removal from the serum occurs. Thus, when serum cholesterol levels increase or decrease, there is either an increased or decreased removal of cholesterol from the serum or an increased or decreased input of cholesterol into the serum. Serum cholesterol SA reflects: (a) either an increased or decreased cholesterolgenesis which occurs in some tissue and is transported by the serum, (b) a decreased or increased release of de nova-formed cholesterol into the serum, (c) an increased or decreased input of nonlabeled cholesterol (depot cholesterol) into the serum, and (d) a specific uptake or decreased uptake of some newly formed radioactive cholesterol complex, i.e., specific lipoprotein cholesterol complex, cholesterol ester versus free cholesterol, from the serum and in which a normal serum cholesterol pool does not participate. In light of the above concepts, both serum cholesterol levels, serum cholesterol SA, and cholesterolgenesis in tissues are important for interpreting cholesterol metabolism occurring in lung tissues. In some mammalian species, approximately 80-85% of the cholesterol in the body is synthesized in the liver (Dietschy and Wilson, 1968) and only 0.1% in the lung. Thus, cholesterol buildup in the lung is probably caused by cholesterol being transported from the liver and other depots to the lung via the serum. Serum cholesterol SA decreased in both saline- and silica-treated rats at 6 and 24 hr, but serum cholesterol levels increased at 6 hr suggesting that nonlabeled cholesterol is being mobilized to the serum (Figs. 3A and B). The saline- and silica-treated rats
LIVER
CHOLESTEROL
IN
439
SILICOSIS
Control ; ‘; 5 c u
so-
,
I
r
AI
I
I
I
20
40
80
80
100
120
140
: oi 0
Hours
5 t 3 5 Y= 00 ra
ME;
20' 15-h 10-t: & ! '..
,:.
,:I
a..
".....
1Oo, 0 20J. 40
control
'...., ..\yled'~...gX
:' ,. A. :.',,.
8- &q.&’
. . ..
&b...
B
1 80I1100
80
-.8 1 140I
120
Hours
FIG. 3. Cholesterol changes in rat serum following intrapulmonary injection of saline and silica. Each point in the graph represents the mean value of 7- 14 samples at various times: Hours 6
Saline n
Silica n
7
7
24 72
7 14
8 14
144 Control (n = 9)
13
13
Graph A: Serum cholesterol level (mg/IOO ml serum) F = 18.911. Graph B: Serum cholesterol specific activity (dpm/mg cholesterol) F = 7.06. *Significantly different from control group value, P c 0.05 level. **Significantly different from both control and saline group values, P s 0.05 level.
responded by mobilizing cholesterol from cholesterol depots, but only the silicatreated rats responded by increasing liver cholesterolgenesis. At 6 hr the silicatreated rats responded with an increased cholesterol content in the liver and increased liver cholesterolgenesis. The finding that cholesterol mobilization is probably occurring from lipid depots other than the liver is suggested by the rise in liver cholesterol content at 6 hr without a corresponding increase of liver cholesterol SA in saline-treated rats.
440
ESKELSON
ET AL.
A crystalline quartz insult to the lung resulted in decreasing the cholesterol concentration (mg/g) in the lung but increasing its cholesterol content due to increased lung weight. The specific activity of lung cholesterol began to increase as early as 6 hr with a more rapid increase at 72 and 144 hr after the silica insult to the lung. This increase in cholesterol SA indicates that cholesterolgenesis and de nova cholesterol buildup are occurring in the lung (Fig. 2). We statistically correlated lung cholesterol levels with serum cholesterol levels in saline-treated rats, which indicated that a relationship exists between the two systems at a significance level greater than 99.9% (correlation between Figs. 2 and 3). This finding and the loss of liver cholesterol content indicates that the liver is losing cholesterol to the lung by mobilizing cholesterol into the serum. Six-hours postsurgery the liver responded by increasing cholesterol content and cholesterol SA which is associated with a highly significant increase in serum cholesterol levels. It seems that the liver over responded at this time because there was an increased liver cholesterol content associated with increased liver cholesterol SA. The increased liver cholesterol may have activated the cholesterolgenic control systems by tuming off liver cholesterolgenesis (Higgins and Rudney, 1973; Frantz 41 al., 1954; Block, 1965). This is indicated by the low liver cholesterol SA at 24-hr postsurgical manipulation. The turning off of liver cholesterolgenesis in the silica-treated rats resulted in a significant decrease in liver cholesterol content at 24 hr compared to the saline-treated and control rat liver cholesterol levels (Figs. IA and B). This decrease in liver cholesterol levels then turned on liver cholesterolgenesis which remained elevated above the 24-hr cholesterolgenic level as suggested by the increased liver cholesterol SA in the silica-treated rats. These data suggest that liver cholesterolgenesis is stimulated by silicotic rat lung and that the de nova-formed cholesterol of the liver is transported more rapidly to the serum. This serum cholesterol is then removed by the silicotic lung more rapidly so that cholesterol input from the liver equals its uptake by the lung. This rate is not only sufficient to keep pace with the cholesterol buildup in silicotic lungs (0.145 mglhr), but is sufficient to maintain constant serum cholesterol levels between 24 and 144 hr after silica treatment (Figs. 2B and 3A). At no time does intrapulmonary saline treatment seem to induce increased liver cholesterolgenesis, but does induce the removal of the de nova-formed liver cholesterol to maintain lung and other cholesterol homeostasis. ACKNOWLEDGMENT The authors wish to express deep gratitude to Dr. Ronald L. Misiorowski for his interest and advice in our project. We are particularly grateful to him for developing and providing us with computer programs to facilitate our statistical analysis.
REFERENCES 1. Babushkina, L. G. (1977). Role of lipids in the pathogenesis of sihcosis. In “Industrial Diseases Due to Exposure to Dust” (S. G. Domin and B. A. Katsnelson, Eds.), pp. 104- 114. Ministry of Health, Moscow, Russia. 2. Block, K. (1965). The biological synthesis of cholesterol. Science 150, 19-28. 3. Chvapil, M., and Peng, Y. M. (1975). Oxygen and lung fibrosis. Arch. Environ. Health 30, 528-537. 4. Chvapil, M., Stankova, L., and Malshet, V. (1976). Lipid peroxidation as one of the mechanisms of silica fibrogenicity. 1. Study with erythrocytes. Environ. Res. 11, 78-88.
LIVER
CHOLESTEROL
IN
SILICOSIS
441
5. Dietschy, J. M., and Wilson, J. D. (1968). Cholesterol synthesis in the squirrel monkey: Relative rates of synthesis in various tissue and mechanisms of control. J. C/in. Invest. 47, 166-174. 6. Duncan, D. B. (1955). Multiple range and multiple F test. Biometrics 11, l-42. 7. Eskelson, C. D., Stiffel, V., Owen, J. A., and Chvapil, M. The importance of liver in normal and silicotic lung-lipid homeostasis. 1. Phospholipids (Unpublished observations.) 8. Folch, J., Ascoli, I., Lees, M., Meath, J. A., and LeBaron, F. N. (1951). Preparation of lipid extracts from brain tissue. J. Biol. Chem. 191, 833-841. 9. Frantz, I. D., Schneider, H. S., and Kinkelman, B. T. (1954). Suppression of hepatic cholesterol synthesis in the rat by cholesterol feeding. J. Biol. Chem. 206, 465-469. Health 17, 720-725. 10. Gross, P. (1968). Experimental “acute” silicosis. Arch. Environ. 11. Heppleston, A. G. (1969). The flbrogenic action of silica. Brit. Med. Bull. 25, 282-287. 12. Heppleston, A. G., Fletcher, K., and Wyatt, I. (1974). Changes in the composition of lung lipids and the “turnover” of dipalmitoyl lecithin in experimental alveolar Iipoproteinosis induced by inhaled quartz. Brit. .I. Exp. Pathol. 55, 384-395. 13. Higgins, M., and Rudney, H. (1973). Regulation of rat liver P-hydroxy-P-methylglutaryl-CoA reductase activity by cholesterol. Nature New Biol. 246, 60-61. 14. Katsnelson, B. A., Babushkina, L. G., and Velixchdovskii, B. T. (1964). Changes in the total lipid content of the lungs of rats with experimental silicosis. Bull. Exp. Biol. Med. 57, 699-702. 15. Katsnelson, B. A., Babushkina, B. T., and Velixchdovskii, B. T. (1976). Pathogenesis, pathogenetic therapy and prophylaxis of silicones. Toxicology 7, 24-73 (in Russian). 16. Okabe, A., Katayama, T., and Kanemasa, Y. (1974). The micromethod for determination of cholesterol, cholesterol esters and phospholipids. Acta Med. Okayama 28, 403-410. 17. Pelliniemi, T. (1973). “Lipids of Connective Tissue with Special Reference to the Effect of Dietary Hyperlipidemia and Oxygen Deprivation on Experimental Granulation Tissue in the Rat.” Academic dissertation, Polytypos Turku. 18. Popper, H., and Schaffner, F. (1970). Alcohol cirrhosis and other toxic hepatopathias. In “Nordiska Bokhandelns Fortag” (A. Engel and T. Larrson, Eds.), pp. 15-46, Stockholm. 19. Ramirez, R. J., Durham, N. C., Schwartz, B., Dowell, A. R., and Lee, S. D. (1971). Biochemical composition of human pulmonary washings. Arch. Intern Med. 127, 395-400. 20. Ramirez, R. J., and Harlan, W. R. (1968). Pulmonary alveolar proteinosis. Amer. J. Med. 45, 502-512. 21. Smith, F. B., and Kikkawa, Y. (1978). The type II epithelial cells of the lung III lecithin synthesis: A comparison with pulmonary macrophages. Lab. Invest. 38, 45-51. 22. Tombropoulos, E. G., and Hadley, J. G. (1976). Lipid synthesis by perfused lung. Lipids 11, 491-495.
B. (1965). Total and free cholesterol. in “Standard Methods of Clinical Chemistry” Meites, Ed.), Vol. 5, pp. 79-89. Academic Press, New York.
23. Zak,
(S.