0013.7227/92/1316-2717$03,00/O Endocrinology Copyright 0 1992 by The Endocrine

Vol. 131, No. 6 Printed in U.S.A.

Society

In Viuo Growth Hormone Treatment Stimulates Secretion of Very Low Density Lipoprotein by the Isolated Perfused Rat Liver* MARSHALL B. ELAM, HENRY MURRAY HEIMBERG

G. WILCOX,

SOLOMON

S. SOLOMON,

AND

Veterans Administration Hospital Research Service and Departments University of Tennessee, Memphis, Tennessee 38163

of

Pharmacology

and Medicine,

ABSTRACT We have previously demonstrated in hepatocyte suspensions prepared after in uiuo GH deprivation [hypophysectomy (hypox)] that rates of esterification of [l-‘C]oleic acid into triglyceride (TG) and phospholipid (PL) were diminished, and that these esterification rates were correspondingly restored by repletion with recombinant GH. The current studies were designed to determine if GH exerts a similar effect on the secretion of very low density lipoprotein (VLDL), the primary plasma carrier of TG. We assessed rates of secretion of VLDL lipid and apoprotein by perfused livers prepared from cortisol/T3-replaced hypox female rats in the presence and absence of recombinant human (h) GH infusion. We also determined rates of synthesis and secretion of VLDL TG from infused [l-“Cloleic acid. After hypox, rates of secretion of VLDL lipid (TG, PL, and cholesterol) and apoprotein

(total) were significantly decreased. In addition, VLDL secreted under these conditions was depleted of PL, relative to the other lipid components. Secretion of newly synthesized VLDL TG from [1-“Cloleic acid was also decreased; however, neither intracellular accumulation of labeled TG nor absolute tissue levels of TG were significantly changed. Conversely, GH treatment of hypox rats effectively restored rates of secretion of VLDL TG, PL, cholesterol (C) and apoprotein to control levels. These findings support the putative role of GH in regulating VLDL secretion in viuo by demonstrating that alterations in plasma GH are accompanied by changes in VLDL secretion. The findings further suggest that GH may regulate VLDL secretion by altering the amount of PL and/or apoprotein available for formation of the VLDL particle. (Endocrinology 131: 2717-2722, 1992)

I

men of the endoplasmic reticulum to form the VLDL before its transfer to the Golgi apparatus and subsequentsecretion (5). It appears that rates of VLDL secretion by the liver are dependent upon the availability of each of its component parts, including fatty acid (TG) (6, 7), PL (8), cholesterol, and (possibly) cholesteryl ester (9). It is of interest, therefore, to ascertain whether GH, by altering rates of synthesis of the lipid components of the VLDL, influences the rate of secretion of that lipoprotein. The current study addressesthese questions by examining the effect of in viva GH deprivation (hypox cortisol-T3-replaced rat) and repletion (recombinant hGH infusion) on rates of synthesis and secretion of VLDL lipid (TG, PL, and cholesterol) and, in addition, on rates of secretion of VLDL apoprotein by isolated perfused livers of adult female Sprague-Dawley rats.

N PREVIOUS studies we observed that hepatocyte suspensionsprepared from female rats after 2 weeks of GH deprivation [hypophysectomized (hypox)] exhibited reduced ability to esterify [l-‘4C]oleic acid into triglyceride (TG), phospholipid (PL), and cholesteryl ester, and that after in viuo repletion of GH, rates of TG and PL (but not cholesteryl ester) synthesis were restored to control levels (1). Similarly, GH administration in the male increased rates of TG and PL synthesis(1). These observations are significant, in that they demonstrate the ability of GH, given in viva, to regulate the rates of synthesis of two critical components of the very low density lipoprotein (VLDL). The VLDL particle is synthesized and secretedby the liver and serves as the primary transport vehicle to deliver fatty acid, in the form of TG, to peripheral tissues. GH excess (acromegaly) has been associated with excess plasma VLDL and hypertriglyceridemia (2). Conversely, reduction of plasma GH in acromegalics either with the somatostatinanalog octreotide (3) or by selective pituitary adenomectomy (4) resulted in reductions in plasma VLDL and TG. TG and PL are assembledalong with newly synthesized apoprotein, cholesterol, and cholesteryl ester within the luReceived May 11, 1992. Address requests for reprints to: Dr. Marshall B. Elam, Department of Pharmacology, University of Tennessee Medical Center, 800 Madison Avenue, Memphis, Tennessee 38163. * This work was supported by the V.A. Research Service (to M.B.E.) and Grants HL-27850 and BRSG-09 from the NIH, U’SPHS (to M.H.). The opinions expressed in this manuscript are those of the authors and do not necessarily reflect those of the Department of Veterans Affairs.

Materials and Methods Female Sprague-Dawley rats (control and hypox), 8-9 weeks of age, were obtained from Harlan Laboratories (Indianapolis, IN) and housed under standardized conditions of light (0600-1800 h) and temperature (23 * 1 C). The animals were fed ad libitum with rat chow (no. 5001, Ralston-Purina, St. Louis, MO) until the time of perfusion. To avoid the potential confounding effects of thyroid and adrenal hormone deficiency on hepatic fatty acid metabolism (10, ll), all hypox rats were treated with hydrocortisone (1 mg/day) and T3 (2.5 pg/day) administered SC using Alzet implantable osmotic minipumps (Alza Corp., Palo Alto, CA). Pumps were implanted in the SC tissue of the dorsal surface of the rats under light ether anesthesia. Alzet pumps containing vehicle (polyethylene glycol) were implanted in the control rats. Recombinant human GH (hGH), dissolved in a small volume of 0.01 M NaOH and diluted with physiological saline, was also continuously administered SC via

2717

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 August 2016. at 16:07 For personal use only. No other uses without permission. . All rights reserved.

2718

GH STIMULATES

HEPATIC

Alzet pump at a rate of 5 Kg/h. The animals were studied after 2 weeks of hypophysectomy with or without hGH treatment. On the day of the experiments, the rats were anesthetized with an ip injection of Na pentobarbital (60 mg/kg), and the livers were removed, using a protocol approved by the University animal use committee. Livers were perfused in vitro in a recycling perfusion apparatus, as described previously (6). The perfusion medium consisted of 25% bovine erythrocytes (vol/vol), 6 g BSA/dl, and 100 mg glucose/d1 in KrebsHenseleit bicarbonate buffer (pH 7.4). The initial volume of the perfusate was 80 ml, and the flow rate was maintained at 35-50 ml/min at 20 cm hydrostatic pressure. The medium was gassed continuously with 95% 02.5% CO2. Throughout the 3-h perfusion, [l-‘4C]oleic acid, as a complex with delipidated bovine albumin (6 g/dl), was infused at a rate of 166 pmol/h (3.0 x lo4 dpm/pmol). Perfusate (20 ml) was taken at hourly intervals for isolation of VLDL by ultracentrifugation at a density of 1.006 (7). The less than 1.006 density fraction was isolated and dialyzed for 48 h against 0.15 M NaCl0.002 M EDTA, pH 7.4, and 0.01% NaN3. Total perfusate, hepatic and VLDL TG (12, 13), cholesterol (14), cholesteryl esters (15), and PL (16) were determined by calorimetric methods after separation of lipid extracts by TLC on Silica gel G plates (Analtech, Newark, DE) with petroleum ether-ethyl ether-glacial acetic acid (84:15:1). Lipids were identified by direct visualization under UV light after treatment with Rhodamine 6G (0.02%). Lipid zones were eluted for calorimetric analysis or assayed for radioactivity in a liquid scintillation cocktail (Biocount, Research Products International, Mount Prospect, IL), using a Beckman Liquid Scintillation System model LS3801 (Beckman Instruments, Inc., Fullerton, CA). Incorporation of [l-i4C]oleic acid into ketones was estimated by assaying radioactivity in the perchloric acid-soluble fraction of the perfusate (17). VLDL apoprotein (total) was measured by the method of Lowry et al. (18) modified to eliminate turbidity by the addition of sodium dodecyl sulfate (19). Perfusate albumin and apolipoprotein-Al (apoA1) were determined by RIA (20). At the end of the perfusion, samples of liver (1 g) were taken, and lipids were extracted for chemical analysis and measurement of radioactivity in cellular lipids. [1-“C]Oleic acid (SA, 40-60 mCi/mmol) was obtained from New England Nuclear (Boston, MA). Oleic acid was supplied by Nu-Chek Laboratories (Elysian, MN). BSA (Sigma) was purified as described previously (6). hGH (Protropin, Genentech, South San Francisco, CA) was obtained through a commercial supplier. Plasma concentrations of cortisol, T3, and hGH were determined by RIA. hGH was measured using reagents supplied by Immunonuclear (Stillwater, MN). Cortisol and T3 were measured by RIA using reagents supplied by Micromedic, Inc. (Horsham, PA). Statistical

VLDL SECRETION

TABLE hormone

1. Effect of hypox and hormone replacement concentrations and body/liver weights Plasma

Group

Endo. 1992 Vol131. No 6

(n)

T3 (rig/ml)

Controls (5) Hwox (3) Hypox-GH (4)

0.86 * 0.14 0.97 k 0.16 0.72 & 0.14

hormone

cont.

Cortisol

(pg/dl)

9.2 zk 2.4 7.1 zk 2.2

on plasma

Wt M Animal

Liver

216.4 + 3.7 197.3 t 5.4” 185.8 + 4.7”

9.18 + 0.26 5.93 + 0.93” 6.70 + 0.73”

Data are the mean + SEM plasma T3 and cortisol levels, total body weight, and liver weight. Hypox animals were treated with T, (2.5 fig/ day) and cortisol (1000 Kg/day). Hypox-GH animals were treated with hGH (5 pg/h) in addition to cortisol and T,. Hormone levels were determined by RIA, as described in Materials and Methods. “P < 0.05 us. control (by Student’s t test).

.-5

1.2

t,

T

Triglyceride

Phospholipid

Triglyceride

Phospholipid

Cholesterol

methods

The means of each of the two experimental groups (hypox and hypox hGH-treated) were compared to those of a common control group using Student’s t test for pairwise comparisons (21). These statistical tests were made as orthogonal contrasts (22). P < 0.05 was accepted as significant. Data are depicted as the mean + SEM.

Results The efficacy of the hormone replacement protocols in the hypox and hypox GH-treated rats was assessed by measurement of plasma TJ, cortisol, and hGH. Infusion of T3 at a rate of 2.5 pg/day resulted in plasma T3 concentrations comparable to those in sham-operated control rats in both hypox and hypox GH-treated rats (Table 1). Furthermore, in both hypox and hypox GH-treated rats, infusion of hydrocortisone at rates comparable to those previously reported (23) resulted in plasma cortisol levels consistent with adequate replacement of corticosteroid (Table 1). The plasma GH levels observed after the administration of hGH (64.9 + 14.1 &ml) were comparable to reported mean plasma GH levels in the normal female rat (70-96 &ml) (24, 25). Hypox

Cholesterol

FIG. 1. Effect of hypox and GH replacement on hepatic VLDL lipid secretion. Data are depicted as both the absolute quantity of VLDL lipid (micromoles per h/g liver) secreted into the perfusate (A) and as a percentage of the control values (B). Livers were perfused after 2 weeks of in viuo GH deprivation (hypox) and repletion (hypox-GH). Hypox and hypox-GH animals were treated with Ta (2.5 pg/day) and cortisol (1000 pg/day). hGH was administered continuously SC (5 pg/ day). Liver perfusions were carried out in the presence of oleic acid, administered as a continuous infusion (166 Fmol/h). *, P < 0.05 us. female control (by Student’s t test).

rats failed to gain weight at the same rate as the shamoperated controls, and liver weights were lower in the hypox group compared to sham-operated values. Surprisingly, despite replacement of somatotropin (hGH) at rates comparable to those previously reported to restore weight gain in hypox rats (26) and despite the ability of hGH to reverse the effect of hypox on VLDL secretion (Table 1 and Figs. 1 and 2),

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 August 2016. at 16:07 For personal use only. No other uses without permission. . All rights reserved.

GH

HEPATIC-TG E? E D > 5 P 8

q

.c 0 8 .o a, 0 5*

STIMULATES

HEPATIC

* T

KETONE BODIES

2.0

1.0

0 5. 0.0 CONTROL

2. secretion are the metabolic t test). cellular

FIG.

HYPOX

HYPOX-GH

Effect of hypox and GH replacement on the synthesis and of newly synthesized TG and ketone by perfused livers. Data mean t SEM micromoles of [l-‘%]oleic acid incorporated into products per g liver/h. *, P < 0.05 us. control (by Student’s **, Total TG = micromoles of fatty acid incorporated into and perfusate TG.

TABLE composition

2. Effect

of hypox

and GH

replacement

Treatment VLDL

on VLDL

group (II)

liDids Control

PL/TG Cholesterol/TG PL/cholesterol

(5)

0.209 3~ 0.010 0.063 5 0.006 3.358 f 0.202

&pox 0.137 0.062 2.167

(3)

+ 0.043” f 0.003 f 0.601*

Hypox-GH

0.182 0.054 3.410

(4)

+ 0.026 + 0.007 + 0.247

Data are the mean + SEM lipid ratios of VLDL secreted by perfused livers after hypophysectomy with and without hGH treatment. The molar ratios are derived from concentrations of lipids in perfusate VLDL (micromoles per ml cell free perfusate). 1; 1 ~.~W,us. controls (by Student’s t test).

continuous infusion of somatotropin did not accelerate weight gain or reverse the effect of hypophysectomy (hypox) on liver weight (Table 1). Rates of secretion of VLDL lipids (TG, PL, and cholesterol) by perfused livers of control, hypox, and hypox GH-treated rats are depicted in Fig. 1 as absolute rates (micromoles per h/g liver; Fig. 1A) and as a fraction of the control rates (Fig. 1B). GH deprivation (hypox) resulted in a significant reduction in the secretion of VLDL TG, PL, and cholesterol (Fig. 1). This decrease occurred despite T3 and cortisol replacement. This decrease in VLDL lipid secretion was effectively reversed by selective replacement with GH (5 pg/h hGH; 2 weeks). Specifically, after hGH infusion in the hypox animal, rates of secretion of VLDL TG, PL, and cholesterol were restored to levels indistinguishable from those in the shamoperated controls (Fig. 1). Hypox affected not only the quantity, but also the composition, of the secreted VLDL. After hypox, VLDL PL was reduced to a greater extent than were other VLDL lipids. This was detected as a reduced PL to cholesterol ratio in the VLDL, and a similar trend for the Pl to TG ratio (Table 2). Conversely, on a molar basis, the relative cholesterol content of the VLDL was unaltered. hGH treatment effectively corrected the observed changes in

VLDL

2719

SECRETION

VLDL PL (Table 2). Decreased secretion of VLDL lipids by livers from hypox rats was accompanied by a decreased rate of secretion of total VLDL apoprotein. This indicates that in addition to reduced VLDL lipid secretion, fewer particles were secreted. We also observed similar reductions in the secretion of apoAl and albumin (Table 3). GH treatment effectively restored VLDL apo secretion, but had no discernable effect on the secretion of either apoA-1 or albumin (Table 3). This observation suggests a specific effect of GH on the secretion of VLDL apoprotein in addition to its well known generalized anabolic effect on protein synthesis (27, 28). In addition to determinations of perfusate VLDL lipid by chemical assay, we also determined the effects of hypox and hGH infusion on the rates of synthesis and secretion of VLDL TG and ketone bodies from infused [1-r4C]oleic acid. Rates of incorporation of [l-‘4C]oleic acid into perfusate VLDL TG, ketone bodies, and cellular TG were assessed after extraction of lipid from perfusate and liver. Rates of secretion of VLDL TG were reduced in hypox livers and were restored to control levels after in viva hGH treatment of the hypox rat (Fig. 2). These data support the observations of VLDL TG mass measurements presented in Fig. 1 and, in addition, allow an assessment of the rates of TG synthesis. It is apparent that under these experimental conditions, the observed changes in VLDL secretion were not the result of altered rates of TG synthesis, but, rather, reflected an impairment of the ability of the hypox liver to secrete newly synthesized TG in the VLDL. This was manifested as a net accumulation of newly synthesized TG in the hypox liver during the 3-h perfusion (Fig. 2). Total (perfusate plus cellular) TG synthesis was, in fact, not significantly different in hypox or hypox GH-replaced livers compared with that in control livers (Fig. 2). After hypox, secretion of 14C-labeled ketones (perchloric acid-soluble counts) into the perfusate was significantly increased (P < 0.01) compared to that in control livers (Fig. 2). Addition of hGH to hypox reduced ketone body secretion to control levels. The effect of hypox and hGH treatment on hepatic lipids after perfusion was determined by measurement of tissue concentration of TG, cholesterol, and cholesteryl ester. Hepatic TG was not affected by hypox; however, cholesteryl ester was significantly reduced in livers of both hypox and hypox GH-treated rats (Table 4). There was a trend toward reduced hepatic cholesterol (P = 0.08) in the hypox livers, and hepatic cholesterol was significantly decreased in the TABLE 3. Effect of hypophysectomy and albumin secretion by the perfused protein (pg/g liver. h)

Treatment

Perfusate

VLDL (total) ApoAl Albumin

and GH replacement liver.

Controls

(5)

107.82 + 7.94 14.42 +- 2.63 2,654 f 455

on apo

group (n)

HYPO~ (3) 36.83 -C 7.54” 5.85 + 2.14” 846 + 138”

Hypox-GH

(4)

107.12 + 12.19 6.86 * 0.94” 1,015 f 43”

Data are the mean + SEM rates of secretion of hepatic albumin and apo by the perfused liver after hypox with and without hGH treatment (5 pg/h; 2 weeks). All hypox animals were treated with T3 (2.5 pg/day) and cortisol(lOO0 wg/day). ApoAl and albumin were measured by RIA. a P < 0.05 compared to controls (by Student’s t test).

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 August 2016. at 16:07 For personal use only. No other uses without permission. . All rights reserved.

GH STIMULATES TABLE hepatic

4. Effect lipids

Hepatic (pmol/g

Triglyceride Cholesterol Cholesteryl Data perfusion. ‘I’ < *P < cP

In vivo growth hormone treatment stimulates secretion of very low density lipoprotein by the isolated perfused rat liver.

We have previously demonstrated in hepatocyte suspensions prepared after in vivo GH deprivation [hypophysectomy (hypox)] that rates of esterification ...
753KB Sizes 0 Downloads 0 Views