Atrial natriuretic factor in experimental cirrhosis in rats

GASTROENTEROLOGY 1992;102:1356-1362

Atria1 Natriuretic Factor in Experimental Cirrhosis in Rats TIMOTHY R. MORGAN, KENGATHEVY and INTHIRANY THILLAINADARAJAH

MORGAN,

GAVIN M. JONAS,

Department of Medicine, Veterans Administration Medical Center, Long Beach, California; and University of Caiifornia, Irvine, California

Atria1 natriuretic factor (ANF) is a cardiac hormone with potent natriuretic, diuretic, and vasorelaxant properties. Although abnormalities in ANF release, plasma level, and renal receptors have been described in humans and/or animals with cirrhosis and ascites, little is known about the ANF hormonal system in cirrhosis without ascites. The aim of this study was to examine the ANF hormonal system in an animal model of cirrhosis to determine whether compensated cirrhosis is associated with changes in the ANF hormonal system. Pair-fed rats were studied 5-7 weeks after either sham surgery or bile duct ligation. Bile duct-ligated (BDL) rats had elevated portal pressure and cirrhosis but did not have ascites. ANF messenger RNA levels were increased onefold in the atria of BDL rats. Shamoperated and BDL rats had similar plasma ANF levels. Competitive binding inhibition studies of isolated glomeruli showed a single class of receptors in sham-operated and BDL rats. The equilibrium dissociation constant was similar in sham-operated (0.51 nmol/L) and BDL (0.63 nmol/L) rats. Glomerular ANF receptor density increased significantly in BDL rats. Cyclic guanosine monophosphate generation in isolated glomeruli in response to 100 nmol/L ANF decreased slightly but not significantly in BDL rats. It was concluded that the ANF hormonal system is altered in cirrhosis without ascites; atria1 messenger RNA level and glomerular ANF receptor density are increased. trial natriuretic factor (ANF) is a 28 amino acid hormone with natriuretic, diuretic, and vasoreof ANF’s potent natrilaxant properties. l Because uretic properties, investigators have closely examined the ANF hormonal system in sodium retaining

A

states to determine whether alterations in the system contribute to sodium retention. Several abnormalities in the ANF hormonal system have been reported in cirrhosis with ascites. Gines et al. reported increased cardiac release of ANF in patients with cirrhosis and ascites.’ Numerous investigators have

reported increased or normal plasma ANF levels in patients with cirrhosis and ascites.3-s The natriuretic response to intravenously administered ANF is decreased in cirrhosis with ascites.‘-” Gerbes et al. reported increased ANF clearance receptor (C-receptor) density in glomeruli from ascitic rats.l’ Renal resistance to the natriuretic effect of ANF may be caused by alterations in ANF receptors within the kidney or other systemic factors such as increased renal sympathetic tone, decreased arterial pressure, or elevated plasma aldosterone levels. In addition to its renal effects, ANF is also a potent vasodilator. Exogenous infusion of ANF may cause arterial vasodilation in patients with cirrhosis. Although the ANF hormonal system is known to be altered in cirrhosis with ascites, little is known about the time course of changes in the ANF hormonal system in cirrhosis and, specifically, whether the ANF hormonal system is altered before the presence of ascites. Several studies have reported normal plasma ANF levels in compensated cirrhosis.‘3-‘5 This finding is in keeping with other studies in cirrhosis without ascites that show normal plasma renin activity and normal plasma aldosterone levels.16 However, infusion of ANF into patients with cirrhosis without ascites results in decreased natriuresis compared with healthy controls,‘1”3 suggesting alterations in the ANF hormonal system or other natriuretic homeostatic mechanisms (e.g., aldosterone, sympathetic nerves) in these patients. In some cirrhotic patients ANF infusion produces vasodilation and hypotension, suggesting that ANF is a potent vasodilator in compensated cirrhosis. To better evaluate the ANF hormonal system in cirrhosis, we undertook a study of ANF synthesis [i.e., cardiac ANF messenger RNA (mRNA) level], plasma ANF level, and renal ANF receptor affinity, and cyclic guanosine monophosphate density, (cGMP) response to ANF in bile duct-ligated (BDL) This is a U.S. government work. There are no restrictions on its use. 0016-5085/92/$0.00

ANF IN CIRRHOSIS 1357

April 1992

rats with advanced fibrosis/cirrhosis but without ascites. Our results suggest that although the plasma level of ANF is normal in cirrhosis, cardiac synthesis of ANF is increased and glomerular ANF receptors are altered. Thus, alterations in the ANF hormonal system are present in cirrhosis before the onset of ascites and are in the same direction as those reported in cirrhosis with ascites.

Materials and Methods Materials Rat ANF,,,,, was purchased from Calbiochem (San Diego, CA); iodinated rat ANFS,,,, (“‘I-rANF, 2200 Ci/ mmol) and GeneScreen Plus nylon membranes from DuPont/New England Nuclear (N. Billerica, MA); and ANF antibody from Research and Diagnostic Antibodies (Berkeley, CA). The dCT32P was purchased from Amersham (Chicago, IL); guanidinium isothiocyanate and sodium Nsarkosyl from Fluka (Ronkonkoma, NY); and enzymes and oligonucleotides from Boehringer Mannheim (Indianapolis, IN). All other chemicals were purchased from Sigma Chemical Co. (St. Louis, MO) and were of the highest purity available.

Bile Duct Ligation This study was approved by the Animal Use Committee of the Veterans Administration Medical Center, Long Beach. Male Sprague-Dawley rats weighing 175-200 g, purchased from Charles River Inc. (Wilmington, MA), were anesthetized by methoxyflurane inhalation (Metofane; Pitman-Moore, Inc., Washington Crossing, NJ). The rats’ abdominal hair was shaved and the skin was cleaned with Betadine (The Purdue Frederick Company, Norwalk, CT). A 3-cm incision was made caudally from the xiphoid to locate the bile duct and tie a 6-O silk suture around the bile duct in two places approximately 1 cm apart. The bile duct was not severed. The abdominal wall and skin were closed separately, and the animals were allowed to recover. Control rats underwent the same procedure except that sutures were not tied around the bile duct. The rats were maintained in an American Association for Accreditation of Laboratory Animal Care-approved facility with controlled light (12 hour light/dark cycle), temperature, and humidity. Free access to water was provided. After the rats had recovered fully from surgery and were eating normally (4 days), they were pair fed for the remainder of the study (approximately 6 weeks). Pair feeding was achieved by giving each rat 20-21 g of food each day. Only rats that consumed all of their food every day were studied. On the day of study, rats were sedated with an intraperitoneal injection of 50 mg/kg sodium pentobarbital (BDL rats often required slightly less medication). Sodium pentobarbital has been shown to not affect plasma ANF levels.” The abdomen was opened and the portal vein was located. Portal pressure was measured by inserting a saline-filled, needle (no. 22)-tipped tube into the portal vein and measuring the height of the meniscus, using the inferior vena cava as the reference point. Blood was then re-

moved from the aorta and saved for subsequent measurement of ANF. Both kidneys were removed, and glomerular isolation was started immediately. The liver was removed, and a sample was placed in 10% formalin for subsequent histological study. The heart was removed, and the atria were excised and frozen immediately in liquid nitrogen. The distal half of the heart (apex) was then removed and frozen in liquid nitrogen. Cardiac tissue was stored at -70’C until extraction of RNA (storage time approximately 1 week). Preparation

of Glomeruli

Glomeruli were prepared using a variation of the graded-sieving technique described by Misra et al.‘* The kidneys from two rats were combined and used in each experiment. The kidneys were decapsulated and bisected longitudinally, then the medullae and papillae were removed. The cortex was minced with scissors and pressed through a stainless steel sieve with a 180~pm pore size. The glomeruli were collected from the underside of the sieve, mixed with approximately 10 mL of phosphate-buffered saline (PBS), pH 7.4, containing 50 mmol/L dextrose (buffer A), and then decapsulated by passing twice through 20-, 22-, and 25-gauge needles. The glomerular suspension was then poured onto stacked sieves with pore sizes of 104 and 75 pm and washed with 4 L of ice-cold buffer A. The glomeruli were collected in 10 mL of buffer A from the top surface of the 75-pm sieve and centrifuged at 1200 X g for 90 seconds at 4°C. The pellet was resuspended in 10 mL of 0.5 mol/L sodium acetate-0.9% NaCl buffer (pH 5.0) at 4”C, and immediately centrifuged at 1200 X g for 90 seconds. This buffer dissociates the remaining bound ANF from the receptors.‘g The pellet was then resuspended in a lo-ml ice-cold Hanks’ balanced salt solution (HBSS) (NaCl, 137 mmol/L; KCl, 5.4 mmol/L; KH,PO,, 0.44 mmol/L; Na,HPO,, 0.33 mmol/L; MgSO,, 0.40 mmol/L; MgCl,, 0.50 mmol/L; CaCl,, 1.25 mmol/L; and NaHCO,, 4.0 mmol/L; pH 7.4). This suspension was centrifuged at 1200 X g for 60 seconds at 4°C. The glomerular precipitate was then suspended in 10.0 mL of HBSS, and four 100~FL

aliquots were taken for protein determination using the Pierce BCA method (Pierce, Rockford, IL). Then 0.2% bovine serum albumin (BSA) (wt/vol), 5 mmol/L dextrose, and 10 mmol/L HEPES (final concentrations) were added to the glomerular suspension. Bacitracin (1 mmol/L) and phenylmethylsulfonyl fluoride (1 mmol/L) (final concentrations) were added to the glomeruli used for receptorbinding studies but not to those used for cGMP studies. Receptor-Binding Glomeruli

Studies

for Isolated

Competitive binding inhibition studies were conducted according to the method of Ballermann et a1.19 Freshly isolated glomeruli from two rats were used in each experiment. A 200~yL aliquot of glomerular suspension was added to 12 X 75-mm, round-bottom, polypropylene tubes containing 50 pL of ‘251-ANF (approximately 30,000 dpm) and appropriate concentrations of unlabeled ANF. The 250~pL

1358

GASTROENTEROLOGY Vol. 102, No. 4

MORGAN ET AL.

suspension was incubated in a shaking-ice bath (4’C) until equilibrium was reached (75 minutes). Bound radioactivity was separated from free within 15 seconds by filtration through 2.4-cm-diameter nylon filters (presoaked with 0.2% BSA in PBS) with a 0.45~pm pore size (Tetko Inc., Elmsford, NY). Filters were washed three times with 3 mL of ice-cold binding buffer. Binding was unchanged by continuous washing for 30 seconds. Nonspecific binding was defined as radioactivity bound in the presence of saturating concentrations of ANF (10m6mol/L) and represented 40% of total binding. Filters were transferred to 12 X 75mm borosilicate glass tubes and radioactivity counted in an Iso-Data 20/20 series gamma counter (70% efficiency; Iso-Data, Inc., Rolling Meadows, IL). Eight different concentrations of unlabeled ANF were used to span the 3-log spread used in this experiment (lo-” through lo-’ mol/L). Each point was determined in triplicate. ANF Radioimmunoassay The method for ANF radioimmunoassay (RIA) of rat plasma is a variation of our published method for ANF RIA of human plasma.4 Blood was withdrawn from the aorta into a syringe containing ethylenediaminetetraacetic acid (EDTA; 10 mg/mL). After centrifugation at 4”C, aprotinin (500 KIU/mL) and soybean trypsin inhibitor [500 sodium benzoyl-L’arginine ethyl ester (BAEE) U/mL] (final concentrations) were added and 1.0 mL was frozen at -7O’C for subsequent RIA. One milliliter of plasma was applied to a Sep-Pak C-18 cartridge (Waters Associates, Milford, MA) that had been equilibrated with 0.1 mol/L acetic acid. The samples were washed with 20 mL of 0.1 mol/L acetic acid, and ANF was eluted in 3 mL of 80% methanol water (vol/vol), lyophilized, and redissolved in an RIA buffer containing 0.1 mol/ L Tris-acetate (pH 7.4), 0.1% BSA, 50 KIU/mL aprotinin, 50 BAEE U/mL soybean trypsin inhibitor, 1 mmol/L EDTA, and 0.02% sodium azide. The extracted plasma samples (or ANF standards) were incubated in an RIA buffer with antiserum for 24 hours at 4°C. Iodinated rat ANF (lz51-ANF) was added, and the incubation was continued for another 24 hours. Antibodybound and free ligand were separated by centrifugation after the addition of 12.5% polyethylene glycol (final concentration, wt/vol) and 0.2% bovine y-globulin (wt/vol). Each sample was determined in triplicate.

mRNA Measurement Complementary DNA probes. A pBR322 plasmid containing a 750-base pair complementary DNA (cDNA) complementary to the rat ANF gene” was used to probe for ANF, and a 1.07-kb cDNA complementary to the 18s ribosomal RNA was used to probe for 18s ribosomal RNA.‘l The plasmids were purified using a Qiagen kit according to the manufacturer’s specifications (Qiagen Inc., Studio City, CA). The cDNAs were isolated by electrophoresis in low-melting agar (Sea Plaque; FMC, Rockland, ME) and labeled with dCT3*P by the random-primer method.22~23 RNA extraction. Total RNA was extracted from the atria and the apex of the heart using the guanidinium isothiocyanate-phenol-chloroform-extraction method of Chomczynski and SacchieZ4 RNA in 70% ethanol was stored at -20°C for later use. Before use, RNA was vacuum dried (without heat), resuspended in 0.15 mol/L sodium acetate, and quantified by measuring absorbance at 260 nm. Northern Blot. Formamide-denatured RNA, 10 pg, was electrophoresed on a 1% agarose gel with 2.2 mol/L formaldehyde.25 The RNA was transferred to a nylon membrane (GeneScreen Plus; DuPont) and probed with the 32P-labeled ANF cDNA. Transfer, hybridization, and washing were performed according to the manufacturer’s recommendations. Dot Blots. Formamide-denatured total RNA, 2 and 4 pg, was dot blotted onto two nylon membranes (GeneScreen Plus). One membrane was probed with the 32P-labeled cDNA for ANF; the other was probed with the 32P-labeled cDNA for 18s. Hybridization and washing were performed according to the manufacturer’s recommendations. Autoradiographs. Autoradiographs are obtained by exposing the membrane to Kodak X-Omat AR film (Eastman Kodak Co., Rochester, NY) for 18 hours or longer at -70°C with an intensifying screen. Autoradiographs were scanned with an LKB 222-020 UltroScan XL laser densitometer (LKB Instruments Inc., Gaithersburg, MD).

Statistics The unpaired Student’s t test was used to compare results; a P value of co.05 was considered significant. Receptor-binding studies were analyzed using LIGAND, a curve-fitting computer prononlinear, least-squares, gram.26*27

cGMP Generation One hundred seventy microliters of freshly isolated glomeruli were incubated at 37°C for 10 minutes. Then 10 pL of isobutylmethylxanthine (final concentration, 0.5 mmol/L) was added, followed 5 minutes later by 20 pL of 1.0 pmol/L ANF. The reaction was stopped exactly 60 seconds later by adding 50 pL of ice-cold 2% acetic acid and then immersing the solution in a boiling water bath for 3 minutes. The solution was evaporated to dryness, reconstituted in 200 pL of RIA buffer, and assayed using a commercially available RIA kit (Biomedical Technologies Inc., Stoughton, MA).

Results Animals We studied 10 BDL and 10 sham-operated rats (five pairs of each) an average of 38 days postoperatively (range, 31-44 days). Average daily food intake was essentially identical in both groups of rats (mean t- SD, 20.47 + 0.09 g/day for BDL vs. 20.50 + 0.05 g/day for sham-operated). Portal pressure was >14 cm H,O in all BDL rats (17.3 f 2.6; range,

ANF IN CIRRHOSIS 1359

April 1992

14.8-23.0) and 114 cm H,O in all sham-operated rats (13.6f 0.6;range, 12.5-14.0; P < 0.001). Histological examination of H&E- and Masson trichrome-stained liver sections showed cirrhosis or advanced fibrosis in each BDL rat and normal histology in each shamoperated rat. Ascites was not visibly present in the opened abdomen of any rat. Plasma ANF Levels The plasma level of ANF was not significantly different between BDL (54.9 + 20.9 pg/mL; range, 35-99 pg/mL) and sham-operated (48.3f 25.4pg/ mL; range, 15-78 pg/mL) rats (n = g/group). Receptor-Binding

Studies

Binding reached equilibrium after 60 minutes and remained constant for another 60 minutes. Results of the competitive binding inhibition experiments are shown in Figure 1. LIGAND analysis of the competitive binding inhibition data was most consistent with a single class of binding sites in each experiment. The equilibrium dissociation constant (I&) was unchanged following after bile duct ligation 0.51f 0.18nmol/L, vs. BDL, 0.63 f (sham-operated, 0.18 nmol/L) (Figure 2). However, ANF receptor density increased from 931 f 209 fmol/mg protein (range, 686-1258) in sham-operated rats to 1221 f 249 fmol/mg protein (range, 986-1593) in BDL rats (P < 0.05). Scatchard plots for one representative experiment comparing glomeruli from BDL and shamoperated rats is shown in Figure 2. The linear plots show that a single population of receptors is being determined in each case. Specific binding of ANF to isolated glomeruli supports the data obtained from

1 .OO

cl

0.00 0.00

,

0.05

0.10

ANF BOUND

0.15

0.20

0.25

(nmol/liter)

Figure 2. Representative Scatchard plots of competitive binding data for isolated glomeruli from sham-operated and BDL rats. “‘I-rat ANF,,,, concentration, approximately 30 pmol/L; glomerular protein concentration, 191 pg/mL (sham) and 166 pg/ mL (BDL); Kd, 0.55 nmoI/L (sham) and 0.60 nmoI/L (BDL); receptor density, 643 fmoI/mg protein (sham) and 1226 fmol/mg protein (BDL).

the Scatchard plots that receptor in BDL rats (Figure 3).

density is increased

cGMP Response cGMP generation in isolated glomeruli was quantitated by RIA after 1 minute of incubation with 100 nmol/L rat ANF,,_,,,. cGMP increased from a baseline value of 1273 + 475 fmol/mg protein without ANF to 10,930 f 2684 fmol/mg protein with 100 nmol/L ANF in BDL rats (9.4 f 4.3-fold increase; n = 5). In sham-operated rats, cGMP increased from a baseline value of 1166 + 399 fmol/mg protein to

-g

1500

Q)

0.80

.

r

BDL

T

5 t

1250

0.60 m

0.40

hnlabelled

ANF]

M

0.00

0.20

0.40

0.60

(ANFI lo-’ Figure 1. Competitive binding inhibition in isolated glomeruli from sham-operated and BDL rats. “‘I-rat ANF concentration, approximately 5 X 10’ dpm/250 uL (approximately 30 pmoI/L) (mean f SD; n = 5 per group).

0.80

1 .oo

M

Figure 3. Specific binding of increasing concentrations of rat ANF,.,,, to isolated glomeruli from sham-operated or BDL rats (mean + SEM; n = 5/group).

1360

MORGAN

ET AL.

GASTROENTEROLOGY

Figure 4. Northern analysis of RNA prepared from atria of a sham-operated rat and hybridized to a 32P-labeled rat ANF cDNA. Autoradiographs were exposed for 18 hours on Kodak XAR film with an intensifying screen; 10 pg of total cellular RNA was used. Arrows mark 28s RNA (4.8 kb) and 18s RNA (1.8 kb). The ANF cDNA probe hybridized with RNA of approximately 8.8 kb.

Vol. 102,No. 4

We found two alterations in the ANF hormonal system in cirrhosis without ascites: increased atria1 ANF mRNA level and increased density of ANF receptors in isolated glomeruli. Plasma ANF levels and cGMP generation in glomeruli in response to ANF stimulation were similar in cirrhotic and sham-operated rats. ANF mRNA levels are approximately double in atria from cirrhotic compared with sham-operated rats. This probably represents increased ANF synthesis in the atria of the cirrhotic rats. Although this is the first study to measure cardiac ANF mRNA in cirrhosis with or without ascites, it supports the limited information concerning cardiac ANF in cirrhosis with ascites. Gines et al. reported increased cardiac release of ANF in humans with cirrhosis and ascites, suggesting that increased ANF synthesis is present in this setting.’ Plasma ANF levels are similar in cirrhotic and control rats. Most but not all other investigators have reported unchanged plasma ANF level in compensated cirrhosis.‘3-‘5 In addition, investigators have reported unchanged plasma ANF levels in cirrhosis with ascites.” Our study suggests that plasma ANF is an insensitive measure of alterations in the ANF hormonal system in compensated cirrhosis. Increased atria1 ANF release with normal ANF plasma levels would suggest increased ANF clearance in cirrhosis. To date, little is known about changes in the site or rate of ANF clearance in cirrhosis with or without ascites. We observed only one binding affinity in isolated

13,047+ 2425fmol/mg protein after addition of ANF (12.4+ 6.5-fold increase; P > 0.15; n = 5).

BDL

Sham

Cardiac ANF mRNA A second set of pair-fed rats was used in the studies of ANF mRNA. Northern analysis of several atria1 samples from BDL and sham-operated rats showed a single band of approximately 0.9 kb (Figure 4). Dot-blot assays (Figure 5) showed significantly increased ANF mRNA levels in the atria of BDL rats [0.55+ 0.15density units (DU)/18s DU; n = 91 compared with sham-operated rats (0.26 + 0.03 DU/18s DU; P < 0.05; n = 10). No ANF mRNA was detected in the apex of the heart in either sham-operated or BDL rats (exposure for 1 week). Discussion Bile duct ligation is a frequently cirrhosis.28 Like humans with cirrhosis, velop portal hypertension and, in some In this study, however, rats had portal without ascites.

used model of BDL rats decases, ascites. hypertension

ANF

18s

Figure 5. Slot-blot analysis of rat atria1 RNA extracted from representative cirrhotic (BDL) and sham-operated rats. Four micrograms of total cellular RNA was loaded onto two separate screens. One screen was hybridized to a 32P-labeled rat ANF cDNA; the second was hybridized to a 3aP-labeled cDNA complementary to rat 18s ribosomal RNA, Autoradiographs were exposed for 3 days on Kodak XAR film with an intensifying screen. Atria from cirrhotic rats had approximately twice as much ANF RNA as atria from sham-operated rats (normalized for the amount of 18sribosomal RNA].

ANF IN CIRRHOSIS

April 1992

glomeruli from cirrhotic and sham-operated rats. Although we cannot exclude a second binding site with a K, of ~30 pmol/L in these experiments, we have looked closely for a second binding site in isolated glomeruli of many sham-operated rats using three concentrations of ANF/log and starting with concentrations as low as 0.1 pmol/L and have never found a second binding site with a different K,. Gerbes et al. recently reported finding two binding sites in glomeruli from sham-operated and BDL rats with ascites.” Differences in methodology may explain the discrepancy between the study of Gerbes et al. and the current study. Gerbes et al. performed their studies on a glomerular membrane preparation that was frozen before study. We used intact, whole glomeruli immediately after isolation. In addition, we acid washed (pH 5.0 for 2 minutes) the glomeruli to remove the remaining bound ANF from the receptors before study.‘g,30 Other investigators have found only one ANF-binding site in fresh, acid-washed, whole glomeruli.‘g The density of total ANF-binding sites is increased in glomeruli from cirrhotic rats. The absence of a significant difference between sham-operated and cirrhotic rats in cGMP generation in response to ANF stimulation would suggest that there is no change in the number of guanylate cyclase-containing receptors in glomeruli from cirrhotic rats. Our findings of increased density of total receptors with no change in guanylate cyclase-containing receptors suggest that the density of C receptors is increased in glomeruli from cirrhotic rats. This is consistent with the increased C-receptor density reported in glomeruli from BDL rats with ascites.” Increased C-receptor density could contribute to increased clearance of ANF. The mechanism for increasing glomerular ANFreceptor density after bile duct ligation is not known, primarily because of a lack of information about regulation of ANF receptors in health and disease. It is known that glomerular ANF-receptor density increases in normal rats subjected to sodium restriction or water deprivation31 However, sodium and water intake were essentially identical in our BDL and sham-operated rats. This study shows that the ANF hormonal system is altered in cirrhosis without ascites. In particular, it suggests that cardiac synthesis and release of ANF, a potent vasodilator, is increased in compensated cirrhosis, as has also been suggested in decompensated cirrhosis. These findings are consistent with the peripheral vascular theory of ascites formation, which predicts abnormalities in vasoactive and sodium-retaining hormones in compensated cirrhosis in the same direction as in decompensated cirrhosis.‘6 However, because the current study did not determine

1361

which change in the ANF hormonal system occurs first (i.e., increased atria1 synthesis or increased glomerular receptors), we cannot exclude the possibility that alterations in glomerular (or renal) receptors may lead to sodium retention, increased plasma volume, atria1 distention, and increased atria1 ANF release. In summary, BDL rats with cirrhosis and portal hypertension but without ascites have alterations in the ANF hormonal system consisting of increased atria1 ANF mRNA levels and increased glomerular ANF-receptor density. The increased atria1 ANF mRNA levels in cirrhosis suggest increased atria1 synthesis of ANF, a potent vasodilator, in compensated cirrhosis. The increased glomerular ANF-receptor density shows that changes in a regulator of sodium homeostasis occur in the kidney in compensated cirrhosis. Progressive alterations in renal ANF-receptor density or function in cirrhosis could contribute to urinary sodium retention. References 1. Inagami

T. Atria1 natriuretic factor. J Biol Chem 1989;264: 3043-3046. 2. Gines P, Jimenez W, Arroyo V, Navasa M, Lopez C, Tito L, Serra A, Bosch J, Sanz G, Rivera F, Rodes J. Atria1 natriuretic factor in cirrhosis with ascites: plasma levels, cardiac release and splanchnic extraction. Hepatology 1988;8:636-642. 3. Skorecki KL, Leung WM, Campbell P, Warner LC, Wong PY, Bull S, Logan AG, Blendis LM. Role of atria1 natriuretic factor in the natriuretic response to central volume expansion induced by head-out water immersion in sodium-retaining cirrhotic subjects. Am J Med 1988;85:375-382. 4. Morgan TR, Imada T, Hollister AS, Inagami T. Plasma human atria1 natriuretic factor in cirrhosis and ascites with and without functional renal failure. Gastroenterology 1988;95:16411647. A, Marco J, Cuadrado LM, Gutkowska J, 5. Fernandez-Cruz Rodriguez-Puyol D, Caramel0 C, Lopez-Novoa JM. Plasma levels of atria1 natriuretic factor in cirrhotic patients (letter). Lancet 1985;2:1439-1440, 6. Kawasaki H, Uemasu J, Maeda N, Hirayama C, Kobayashi T, Sakurai H. Plasma levels of atria1 natriuretic factor in patients with chronic liver disease. Am J Gastroenterol 1987;82:727731. 7. Laffi G, Pinzani E, Meacci, LaVilla G, Renzi D, Baldi E, Cominelli F, Marra F, Gentili P. Renal hemodynamic and natriuretic effects of human atria1 natriuretic factor infusion in cirrhosis with ascites. Gastroenterology 1989;96:167-177. 8. Salerno F, Badalamenti S, Incerti P, Capozza L, Mainardi L. Renal response to atria1 natriuretic factor in patients with advanced liver cirrhosis. Hepatology 1988;8:21-26. 9. Brabant G, Juppner H, Kirschner M, Boker K, Schmidt FW, Hesch RD. Human atria1 natriuretic factor (ANF) for the treatment of patients with liver cirrhosis and ascites. Klin Wochenschr 1986;64(Suppl 6):108-111. 10. Fyhrquist F, Totterman KJ, Tikkanen I. Infusion of atria1 natriuretic factor in liver cirrhosis with ascites (letter). Lancet 1985;2:1439. 11. Beutler J, Koomans H, Rabelink T, Gaillard CA, VanHattum JV, Boer P, Dorhout Mees EJ. Blunted natriuretic response and

1362 MORGAN ET AL.

low blood pressure after atria1 natriuretic factor in early cirrhosis. Hepatology 1989;10:148-153. 12. Gerbes AL, Kollenda MC, Vollmar AM, Reichen J, Vakil N, Scarborough RM. Altered density of glomerular binding sites for atria1 natriuretic factor in bile duct-ligated rats with ascites. Hepatology 1991;13:562-566. 13.Miyase S, Fujiyama S, Chikazawa H, Sato T. Atria1 natriuretic peptide in liver cirrhosis with mild ascites. Gastroenterol Jpn 1990;25:356-362. 14. Warner LC, Campbell PJ, Morali GA, Logan AG, Skorecki KL, Blendis LM. The response of atria1 natriuretic factor and sodium excretion to dietary sodium changes in patients with chronic liver disease. Hepatology 1990;12:460-466. 15. Salerno F, Badalamenti S, Moser P, Lorenzano E, Incerti P. Atria1 natriuretic factor in cirrhotic patients with tense ascites. Effect of large-volume paracentesis. Gastroenterology 1990;98:1063-1070. 16. Schrier RW, Arroyo V, Bernardi M, Epstein M, Henriksen JH, Rodes J. Peripheral arterial vasodilation hypothesis: A proposal for the initiation of renal sodium and water retention in cirrhosis. Hepatology 1988;8:1151-1157. 17. Horky K, Gutkowska J, Garcia R, Thibault GG, Genest J, Cantin M. Effect of different anesthetics on immunoreactive atria1 natriuretic factor concentrations in rat plasma. Biochem Biophys Res Comm 1985;129:651-657. 18. Misra RP. Isolation of glomeruli from mammalian kidneys by graded sieving. Am J Clin Path01 1972;58:135-139. 19. Ballermann BJ, Hoover RL, Karnovsky MJ, Brenner BM. Physiologic regulation of atria1 natriuretic factor receptors in rat renal glomeruli. J Clin Invest 1985;76:2049-2056. 20. Lee RT, Bloch KD, Pfeffer JM, Pfeffer MA, Neer EJ, Seidman CE. Atria1 natriuretic factor gene expression in ventricles of rats with spontaneous biventricular hypertrophy. J Clin Invest 1988;81:431-434. 21. Chan Y-L, Gutell R, Noller HF, Wool IG. The nucleotide sequence of a rat 18 S ribosomal ribonucleic acid gene and a proposal for the secondary structure of 18 S ribosomal ribonucleic acid. J Biol Chem 1984;259:224-230.

GASTROENTEROLOGY Vol. 102,No. 4

22. Feinberg AP, Vogelstein B. A technique for radiolabelling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 1983;132:6-13. 23. Feinberg AP, Vogelstein B. Addendum. Anal Biochem 1984;137:266-267, 24. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987;162:156-159, 25. Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a laboratory manual. 2nd ed. Cold Springs Harbor, NY: Cold Springs Harbor Laboratory, 1989. 26. Munson PJ, Rodbard D. LIGAND: a versatile computerized approach for characterization of ligand-binding systems. Anal Biochem 1980;107:220-239, 27. Munson PJ. LIGAND: a computerized analysis of ligand binding data. Methods Enzymol 1983;92:543-576. 28. Better OS. Bile duct ligation: an experimental model of renal dysfunction secondary to liver disease. In: Epstein M, ed. The kidney in liver disease. 2nd ed. New York: Elsevier, 1983:295311. 29. Gerbes AL, Arendt RM, Paumgartner G. Atria1 natriuretic factor. Possible implications in liver disease. J Hepatol 1987;5:123-132. 30. Martin ER, Lewicki JA, Scarborough RM, Ballermann BJ. Expression and regulation of ANP receptor subtypes in rat renal glomeruli and papillae. Am J Physiol 1989;257:F649-F657. 31. Gauquelin G, Garcia R, Carrier F, Cantin M, Gutkowska J, Thibault G, Schiffrin EL. Glomerular ANF receptor regulation during changes in sodium and water metabolism. Am J Physiol 1988;254:F51-F55.

Received June 24, 1991. Accepted September 4,199l. Address requests for reprints to: Timothy R. Morgan, M.D., Gastroenterology Section, Veterans Administration Medical CenterlllG, 5901 East Seventh Street, Long Beach, California 90822. Supported by an Irvine Faculty Research Fellowship and an RAGS grant from the Department of Veterans Affairs