Vasoactive peptides and phenylephrine in isolated teleost hepatocytes
actions
THOMAS W. MOON AND THOMAS P MOMMSEN Department of Biology, University of Ottawa, Ottawa, Ontario KIN 6N5; and Department of Biochemistry and Microbiology, Universit) of Victoria, Victoria, British Columbia V8 W 2Y2, Canada
MOON, THOMAS W., AND THOMAS P. MOMMSEN. Vasoactive peptides and phenylephrine actions in isolated teleost hepatocytes. Am. J. Physiol. 259 (Endocrinol. Metab. 22): E644E649, 1990.-The metabolic actions of the vasoactive peptides vasotocin and isotocin and the a-agonist phenylephrine are examined in hepatocytes isolated from three teleost species: brown bullhead, rainbow trout, and American eel. These three compounds influenced hepatic gluconeogenesis and glycogenolysis with significant species differences. Vasotocin and isotocin affected only eel hepatocytes activating gluconeogenesis by 1.7-fold and glycogenolysis by 3-fold. Phenylephrine increased glycogenolysis by 7-fold in bullhead hepatocytes and gluconeogenesis by 1.4-fold in trout cells. Vasotocin and phenylephrine actions were correlated with increases in adenosine 3’,5’-cyclic monophosphate (CAMP). The vasotocin effects were unaffected by p- and a-antagonists supporting a &-type receptor on eel hepatocytes. Phenylephrine effects were abolished by propranolol and reduced by prazosin and yohimbine ( cyI- and cu2-antagonists, respectively). Phenylephrine, therefore, affected fish hepatocyte metabolism either by a mixed a/ ,&receptor mechanism emphasizing ,8-adrenoceptors or the classic CU/@agonist/antagonist scheme defined for mammals is not appropriate for these fish preparations.
gluconeogenesis; glycogenolysis; lactate oxidation; adenosine 3’,5’-cyclic monophosphate; cy- and P-adrenoceptors; V2 receptors; isotocin; vasotocin; trout; bullhead; American eel
of fish liver and isolated teleost hepatocytes to adrenergic substances is specific to different metabolic pathways. In general, ,&adrenoceptor agonists including epinephrine are known to initiate the depletion of endogenous liver glycogen by activating glycogen phosphorylase and activating metabolic flux through gluconeogenesis (4, 5, 11, 15, 24). In one species of Pacific salmon, norepinephrine increased lipolysis and therefore the availability of nonesterified fatty acids and glycerol (29, with the latter probably exerting indirect effects on the above pathways. Each of these effects is blocked by the general P-adrenoceptor antagonist propranolol but not a-adrenoceptor antagonists (e.g. phentolamine). Alterations in hepatocyte calcium by changes in extracellular concentrations, ionophores, or channel blockers have been found ineffectual in modifying epinephrinestimulated glycogenolysis in amphibians (14) and fish (4, 15). In fact, it has been stated that fish liver cells are devoid of receptors or specific cellular responses that are in mammals attributed to a-adrenoceptor responses (4, 15), although Brighenti et al. (5, 6) do not dismiss the
THE RESPONSE
E644
0193-1849/90
$1.50 Copyright
possibility of propranolol acting directly or indirectly on an a-mechanism in catfish liver cells. In addition to cyagonists, other hormones acting through the mobilization of calcium such as the vasoactive peptides (vasotocin, isotocin, and vasopressin) cause pronounced metabolic changes in mammalian liver cells (13) but are either ineffective [goldfish (15)] or increase [lungfish (11)] metabolic flux in fish liver. Our interest in the evolution of hormone receptor systems led us to analyze whether the variable metabolic effects of the vasoactive peptides and of the presumed a-adrenoceptor agonists were a common characteristic of teleost liver cells. In addition, we were puzzled by the unexplained finding of metabolic actions of the a-agonist phenylephrine allegedly acting through ,&adrenoceptors in liver cells of catfish (5) and carp (15). In mammals, cu-adrenergic phenomena generally are mediated through the mobilization of intracellular calcium and their metabolic actions are characterized by an activation of pyruvate dehydrogenase, a key enzyme in substrate funneling into the tricarboxylic acid cycle, which in turn results in increased rates of substrate oxidation (1, 7, 18). This study, therefore, presents a comparative analysis of the metabolic responses of hepatocytes from three species of teleost fishes: brown bullhead, rainbow trout, and American eel. We show that in the eel, but not in the other two species, both vasotocin and isotocin at nanomolar concentrations strongly activate gluconeogenesis and glycogenolysis, whereas the presumed CYagonist phenylephrine leads to the stimulation of both pathways in all three species. The use of specific agonists and antagonists provides evidence for the role of CY-and ,&adrenoceptors in these metabolic processes. MATERIALS
AND METHODS
Experimental animals were collected and maintained as described elsewhere (23). American eels (Anguilla rostrata) were collected in September and maintained without feeding at ambient water temperatures (November, 8°C; March, 3”C), brown bullheads (Ictalurus nebulosus) were fed and held at lZ°C, and rainbow trout (Oncorhynchus mykiss) were fed and held at ambient water temperatures on a 12:12 h light-dark photoperiod. Experiments were conducted either in November or the following March. Viable hepatocytes were obtained in good yield as outlined (23). Briefly, livers were perfused in situ with
0 1990 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 28, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
VASOACTIVE
PEPTIDES
AND
PHENYLEPHRINE
collagenase (0.015%) dissolved in the appropriately adapted Hanks’ salt solution with (eel) or without (bullhead, trout) 2 mM CaC12. After digestion of the liver for 20 (bullhead and trout) or 45 (eel) min at room temperature, cells were filtered through plankton netting (254 and 53 pm) and washed four times with Hanks’ solution containing 1.5 mM CaC12 and 2% defatted bovine serum albumin. Final suspensions contained 25-70 mg cells/ml. Rates of lactate gluconeogenesis and lactate oxidation were measured with a radiotracer technique (9). Glycogenolysis (in the absence of added lactate) in the three species and lactate gluconeogenesis in eel and bullhead cells were determined as glucose production by an enzymatic technique (23). The nonradiolabeled assay was found appropriate since lactate gluconeogenesis represented 50% (eel) and 69% (bullhead) of total glucose production (23). Cells were incubated with agonists, antagonists, or vehicle for a maximum of 60 min at 15OC. Vasoactive peptides and agonists/antagonists were dissolved in incubation medium before each experiment, and incubations were conducted in the dark. The following substances with their familiar mammalian classification (7) were used: agonist, L-phenylephrine (a& antagonists, prazosin ( cul; Pfizer), yohimbine (No), timo101(PI), atenolol (,&), and propranolol (p1 2). Unless noted otherwise, all substances were purchased from Sigma Chemical, St. Louis, MO. All chemical assays were conducted as previously reported (23). Experiments were conducted in a randomized block design that makes it possible to disregard the substantial fish-to-fish variation in actual metabolic rate and focus statistical analysis on hormone-induced alterations. No numbers were generated for empty cells, and statistical analysis always used paired comparisons. Specifically, analysis of variance (ANOVA) type III with fish-to-fish variation assigned as the second variable was generally employed. In some cases, orthogonal design was used. Therefore, rather than means, data are presented as percent increase (or decrease) over vehicle-treated control cells. RESULTS
Vasoactiue peptides. Of the three teleost preparations tested, only American eel hepatocytes consistently responded significantly to either vasotocin or isotocin; vasopressin was without effect on eel and bullhead cells (Table 1). Trout hepatocytes did demonstrate a small significant increase in glycogenolysis with isotocin and in gluconeogenesis with vasopressin (Table 1). Eel hepatocyte glycogenolysis was increased by 2.5and 3.6-fold by vasotocin and isotocin, respectively (Table 1). These effects were not altered significantly by the presence of either ,&- (propranolol, timolol, or ateno101) or cv- (prazosin, cyl, or yohimbine, cua)antagonists (data not shown). The glycogenolytic effects of the vasoactive peptides were seasonally dependent with effects in March reduced to -30% of the November values, a decrease that was approximately equivalent to that seen in the absolute rates of glycogenolysis in these cells. Isotocin did not significantly affect glycogenolysis in
IN
FISH
E645
LIVERS
1. Effects of vasoactive peptide hormones on glycogenolytic rates and lactate gluconeogenic and oxidation rates in isolated teleost hepatocytes
TABLE
Bull head
Trout
Glycogenolytic Control Vasotocin Isotocin Vasopressin
14.5-+1.57 100 107t2.9 109k3.2 10522.8
(12) (12) (4)* (5)
Vasotocin Isotocin Vasopressin
0.81t0.24 (10) 100 109t3.4 (6) ND 11355.2 (5)” Oxidation
Control Vasotocin Isotocin Vasopressin
5.42t0.80 (10) 100 lOOt1.7 (6) ND 102*1.8 (5)
Eel
9.02t1.61 100 245k24.7 364t39.0 104t5.8
(25) (23)” (16)” (4)
7.68kl.l
(26)
rate
3.11kO.37 100 115t14.5 103k4.2 98t6.2
Gluconeogenic Control
American
(29) (8) (4) (8)
rate
7.15t0.65 (24) 100 103t2.8 (8) ND llOzk2.3 (4)
100 lllk2.8 9721.0 108t3.3
(5)” (4) (4)
21.2t2.84 100 103t2.6 98k3.2 lOO_t3.1
(15)
rate 28.2t1.87 (24) 100 96k2.1 (8) ND lOO_t2.0 (4)
(4) (4) (3)
Values are expressed as percent of control (100%); number of experiments is in parentheses; control values have been reported previously (see Ref. 23). All experiments were conducted on fish during November. Concentrations of peptides were (in PM) vasotocin, 0.1; isotocin, 0.1; and vasopressin, 0.1. Glycogenolytic rates are expressed as pmol glucose produced from endogenous glycogen in 60 min/g cells k SE. Bullhead and eel hepatocytes were also tested in March and glycogenolytic rates were 122 t 4.3 (4) for isotocin (significantly different from vehicletreated controls, P < O.O5), but absolute control value was not significantly different from November value. Eel rates were 151 t 5.5 (16) for vasotocin and 126 -+ 8.1 (4) for isotocin (significantly different from vehicle-treated controls, P < 0.05), and both absolute control [2.63 t 0.39 (14)] and isotocin values were significantly different from the respective November values. For both rainbow trout (21) and eel (8), control glycogenolytic rates are a function of amount of glycogen present after isolation of hepatocytes. Lactate (2 mM, eel and trout; 4 mM, bullhead) gluconeogenic rates are expressed as pmol glucose produced in 60 min/g cells ~fi SE. Tentatively, rates of glucose production from lactate can be multiplied by 2 to give rate of lactate flux into gluconeogenesis; these rates added to rate of glycogenolysis will give total rate of glucose production. Lactate (2 mM, eel and trout; 4 mM bullhead) oxidation rates are expressed as pmol CO, formed in 60 min/ g cells k SE. Tentatively, rates of CO, production from lactate can be divided by 3 to yield actual flux of lactate into oxidative pathways. Statistics were done on nontransformed, not percent, values. ND, not done. * Significantly different from control (vehicle only) rates; P < 0.05.
bullhead cells isolated in November, but rates were significantly above control when assayed in March (Table 1). The sensitivity of glycogenolysis to these vasoactive peptides was in the 1 to 10 nM range (Fig. 1, A and B), and vasotocin increased hepatocyte adenosine 3’,5’cyclic monophosphate (CAMP) levels (Fig. lC, Table 2). 3-Isobutyl-l-methylxanthine (IBMX; 2.5 ,uM), a phosphodiesterase inhibitor, failed to augment CAMP accumulation at 1 PM vasotocin (Table 2). IBMX, added alone, significantly increased glycogenolysis in cells from both bullheads (173% of control) and eels (155% of control). Changes in CAMP were dose dependent (Table 2) and achieved a maximum within the first 10 min of hormone exposure (Fig. 1C). Again, the extent of CAMP increase with vasotocin was seasonally dependent with
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 28, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
E646
VASOACTIVE
PEPTIDES
AND
A Glycoged ysis , bso tocin
0
-6
1
-6
-4
(mol /L 1
loo
80 60 40
20
trino
a
Log concentration
(mol/L)
CAMP
o-O-0
Control I 0
I 5
Time
I IO
I
15
535+29(10)
min 30
10 559k53
(10)
532t31
1,012?167 1,245+150
(4) (8)
802+122(4)
1,175+93 1,595+238
(4) (4)
848k47 853k52 897t88
(10)
(8) (4) (4)
Values are expressed as pmol CAMP/g cells t SE; number of experiments is in parentheses. Cells were incubated at 15°C for indicated times. Cells were collected and processed for CAMP as indicated in MATERIALS AND METHODS. Experiments were done in a complete block design. IBMX, 3-isobutyl-l-methylxanthine (2.5 PM). All values are significantly different from time 0 control values (P < 0.05).
8 Glycogenolysis, lsotocin and Phenylephrine
1. American
LIVERS
2. Effects of vasotocin on concentration of CAMP in isolated American eel hepatocytes assayed in March
Control Vasotocin lo-’ M lo+ M lo-” M + IBMX lo-‘M
Log vosotocin
1
30
(minutes)
eel hepatocytes and effects of vasoactive peptide hormones and phenylephrine on glycogenolysis and increases in adenosine 3’,5’-cyclic monophosphate (CAMP) content with vasotocin addition. A: vasotocin and glycogenolysis. Composite plot, n = 4-14 experiments in unpaired design; values are means & SD; C, control (vehicle only). R: isotocin and phenylephrine sensitivity of glycogenolysis. Experiments were paired, n = 4; values are means t SD; C, control (vehicle only). C: time course of CAMP increases after vasotocin (Vt) addition. Mean of ‘2-4 experiments with SD not exceeding 10% of mean; Vt = lo-’ M. Open squares, controls for both November and March experiments. For actual CAMP concentrations, see Table 3. FIG.
FISH
Time,
c -K)
C
IN
TABLE
loo 80 60 40 20 0
0
PHENYLEPHRINE
values achieved in March -70% of those in November (Fig. 1C). Eel gluconeogenesis from lactate was affected by vasotocin but not by isotocin or vasopressin (Table 1). This effect was much reduced compared with that seen for glycogenolysis in the November experiments but was essentially equivalent in the March studies even though absolute gluconeogenic rates fell to 20% of the November values. The vasotocin-dependent activation of gluconeogenesis in the March experiments was not affected by either CY-or ,&antagonists or IBMX (data not shown). None of the vasoactive peptides tested significantly affected the rate of lactate oxidation in the hepatocyte preparations employed (Table 1). Phenylephrine. Phenylephrine stimulated glycogenolysis and lactate gluconeogenesis in hepatocytes of each species tested (Table 3). This effect was species and pathway dependent. On the one hand, glycogenolysis was activated to a larger extent than gluconeogenesis with bullhead hepatocytes showing the greatest increase in glycogenolysis followed by the eel with the trout a distant third. Phenylephrine-stimulated glycogenolysis was seasonally dependent; rates in the bullhead preparation were reduced to -30% in March compared with November values (Table 3). On the other hand, trout gluconeogenesis (Table 3) was much more sensitive to phenylephrine than either of the other two species. In each case, propranolol reduced this stimulation to levels found with propranolol alone (i.e., without phenylephrine addition); in fact, propranolol decreased gluconeogenic rates significantly below control values. Propranolol (a general ,8adrenoceptor blocker) more effectively blocked phenylephrine-activated glycogenolysis than either timolol or atenolol (,&-adrenoceptor blockers); this was especially obvious in the case of the bullhead (Table 3). In addition, the phenylephrine-stimulated effects on glycogenolysis were reduced by a-antagonists (Table 3) with a significant depression noted in bullhead cells. Phenylephrine sensitivity of these hepatocytes was in the range of 50-100 nM regardless of whether glycogenolysis (Figs. 1B and 2) or gluconeogenesis was examined (Fig. 3). CAMP concentrations rose in bullhead hepatocytes with phenylephrine addition, although increases were small (Fig. ZB). CAMP concentrations were not affected by the concurrent presence of ionomycin, a
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 28, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
VASOACTIVE
PEPTIDES
AND
PHENYLEPHRINE
IN
3. Effects of a-agonist phenylephrine and selective antagonists on glycogenolytic, lactate gluconeogenic, and oxidation rates in isolated teleost hepatocytes TABLE
Trout
Bullhead
A
American
Phenylephrine +Prazosin +Yohimbine +Propranolol +Timolol +Atenolol
(20)” (8)“t (8)*t (8)*t (4)*-F (4)“t
258t24.7 153-r-27.1 172t20.4 96t5.5 lOlt4.2 136t8.2
(17)” (5)” (5)* (10) t (7)t (7)“t
Phenylephrine +Prazosin +Yohimbine +Propranolol Propranolol
Gluconeogenic rate 135t7.8 (8)” 117t3.4 148t9.4 (4)* 118t3.7 144t9.3 (4)” 117t3.7 82t5.2 (3)* 98k3.3 86k3.3 (6)” lOlt2.8
(4)” (4)* (4)” (4)t (4)
109t1.5 108t2.5 llOt3.2 84t1.2 86kl.l
(7)” (5) (5) (4)“t (11)”
Phenylephrine +Prazosin +Yohimbine +Propranolol
106t3.2 110t5.0 106t3.6 9921.3
(4)”
104t2.0 103t0.6 107k1.7 98t5.2
(6) (4) (4)” (3)
(8) (4)” (4) (3)
E z -=
Eel
Glycogenolytic rate 125t2.1 (4)” 689t114 119t4.8 (4)” 393k65 119t4.7 (4)* 463t68 102k2.8 (4)” 127t9.4 ND 163t5.9 ND 253t26
Oxidation
FISH
g %
&0
E647
LIVERS
Total glucose
production
60
40 20 0 C -10
-8
-6
Log Phenylephrine 6
-4 (mol IL)
CAMP
rate 93t0.7 ND ND ND
Values are percent of control (vehicle only) (see Table 1); number of experiments is in parentheses. Except where noted (propranolol, gluconeogenic rate), no antagonist significantly affected control rates in absence of phenylephrine. Bullhead hepatocytes were also tested in March; glycogenolytic rates were 263 k 31 (8) [significantly different from controls (vehicle only); P < 0.051, but absolute control value was not significantly different from November value (see Ref. 23). Concentrations of agonists/antagonists were (in ,uM) phenylephrine, 10; prazosin, 10; yohimbine, 10; propranolol, 100; timolol, 10; and atenolol, 10. Antagonists were always added before agonists to cell incubates. Statistics were done on nontransformed, not percent, values. ND, not done. * Significantly different from control (vehicle only) rates, P < 0.05. t Significantly different from phenylephrine addition.
calcium ionophore, but were increased by IBMX. No major effects of phenylephrine were noted for lactate oxidation except for a minor depression with the bullhead cells (Table 3). DISCUSSION
After treatment with the vasoactive peptides, vasototin and isotocin, and with phenylephrine, two key metabolic pathways namely glycogenolysis and gluconeogenesis changed substantially in isolated teleost hepatocytes. A third pathway, lactate oxidation, remained essentially unaffected. In all three species analyzed, the rates of glycogenolysis and gluconeogenesis increased after incubation with phenylephrine, whereas vasotocin and isotocin exerted metabolic actions only in the American eel. Vasoactive peptides. Teleosts are reported to produce the two chemically related neurohypophysial peptides vasotocin and isotocin, which are analogous to mammalian vasopressin and oxytocin, respectively (2,3). Recent cDNA sequence data supported the common ancestral nature of this hormone superfamily (12). Vasotocin exerts a hyperglycemic action in coho salmon (20), but circulating concentrations of this hormone were undetectable in the American eel (3). Our results clearly show an increase in eel hepatocyte glucose production primarily from glycogen, and they also show that the effects of
I
I
1
0
IO
30
Time (minutes) 2. Bullhead hepatocytes and effects of phenylephrine on total glucose production and increases in CAMP content withphenylephrine addition. A: phenylephrine sensitivity of total glucose production. Represents total content of glucose in hepatocyte incubations after 1 h; values are means t SD, n = 4; C, control (vehicle only). B: time course of CAMP increases after phenylephrine addition. Open circles, phenylephrine at lo-” M; closed circles, phenylephrine + ionomycin (lo-” M); closed squares, phenylephrine + 3-isobutyl-l-methyl xanthine (IBMX) (2.5 X IO-’ M). Actual CAMP content of bullhead hepatocytes at 0 time was 391 X 10-l” mol/g (t 5, n = 4). FIG.
Lactate E
gluconeogenesis
100
g
80
-;
60
i .-x
40
1
Log Phenylephrine
(mols/L)
FIG. 3. Rainbow trout hepatocytes and effects of phenylephrine on lactate gluconeogenesis. Cells were incubated for 60 min with 2 mM Llactate (including 5 X 10” disintegrations/min (dpm) U-[‘“Cllactate per incubate) and analyzed for [14C]glucose (9); values are means k SD, n = 4; C, control (vehicle only).
isotocin are greater than that of vasotocin. Isotocin is neutral compared with the basic peptide vasotocin, a fact which could account for their different responsiveness (3.6- vs. 2.5fold increase in glycogenolysis, respectively; Table 1) and sensitivities (K, of -lo-” vs. IO-’ M, respectively; Fig. 1). Vasoactive peptide hormones in the laboratory rat
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 28, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
E648
VASOACTIVE
PEPTIDES
AND
PHENYLEPHRINE
stimulate hepatic glycogenolysis and gluconeogenesis through a CAMP-independent pathway, which involves increases in intracellular calcium (reviewed in Ref. 7). Our results imply that a CAMP-dependent mechanism is responsible for the activation of glycogenolysis and gluconeogenesis in the eel hepatocyte. CAMP concentrations increased in the presence of vasotocin (Fig. 1; Table Z), but neither of the classic ,& or o-adrenoceptor antagonists modified these biochemical responses. These results suggest the existence of a specific V-type receptor on the eel hepatocyte as recently reported for the trout hepatocyte (19) and gill membrane preparation (10). These authors, however, provided evidence for a V, receptor that is reported to inhibit CAMP accumulation and/or to increase calcium mobilization (classic mammalian vasopressin receptor) rather than the VZ receptor, which when activated leads to increased CAMP content. Our results showing an increased gluconeogenic flux and CAMP content with vasotocin support the existence of V2, not VI, receptors in the eel liver. The experiments reported by Lahlou et al. (19) showed that oxytocin, isotocin, and vasopressin but not vasotocin inhibited glucagon-stimulated CAMP increases. Whether their results were affected by the indirect experimental approach used or by some specific feature of the trout system not detected in our direct approach is unknown. Our studies, those of Janssens and Grigg (14) in toad liver, and indirectly those of McKeown et al. (20) on salmon support the existence of a V2 receptor much like that shown in the mammalian kidney responding to vasopressin (16). The rat may be unusual in possessing hepatic V1 receptors, as recent studies have shown the complete absence of V, vasopressin receptors in rabbit hepatocytes (26). Species, age, and sex differences have been frequently reported in both the existence and responsiveness to vasoactive peptides and adrenergic agonists in mammals (see Refs. 17 and 26). The absence of a hepatocyte response to vasoactive peptides in trout and catfish (present study) and carp (15), but its presence in eel (present study) and lungfish (11) hepatocytes, further demonstrates the distinct species specificity of these receptors within the animal kingdom. Phenylephrine. Phenylephrine, a classic al-adrenoceptor agonist of hepatic tissue in the laboratory rat (reviewed in Ref. 7), activated both glycogenolysis and gluconeogenesis in the three fish species tested here (Table 3). Bullhead hepatocyte glycogenolysis was the most responsive followed by eel glycogenolysis; gluconeogenie responses for both species were well below those of glycogenolysis. Trout, on the other hand, showed better gluconeogenic responses than glycogenolytic, but in both cases these were well below those of the other two species. The sensitivity of the hepatocytes to phenylephrine was low (50% of maximal response is achieved at micromolar phenylephrine concentrations) regardless of the biochemical response tested (Figs. 2 and 3). Previous studies reported tthat hepatocytes isolated from goldfish (4), carp (15), or catfish (5) responded as poorly to phenylephrine as the trout in our study, whereas bullhead and eel liver cells are even less sensitive to this compound. Unlike the laboratory rat where the phenylephrine response is mediated primarily by al-adrenoceptor mechanisms, this
IN
FISH
LIVERS
response in the three species used here is primarily by ,@adrenoceptor mechanisms, as indicated by propranolol blocking the response (Table 3). Timolol and atenolol, ,&antagonists (7), significantly reduced but did not abolish the increases; these results support a primary ,& effect for phenylephrine on the fish hepatocyte. An intriguing result is that both prazosin and yohimbine, cyl- and aZ-adrenoceptor antagonists, respectively (7), reduce the glycogenolytic response of phenylephrine in both bullhead and eel hepatocytes (Table 3). Prazosin is the more effective of the two antagonists. This could indicate a small cul-component (7) to phenylephrine action in these teleostean fishes, especially on glycogenolysis but possibly not on gluconeogenesis. Our data, therefore, suggest that phenylephrine has a mixed a/P response in the bullhead and eel systems, unlike that noted for carp (15). Brighenti et al. (5) could not completely exclude this possibility in the catfish hepatocyte. The absence of a phenylephrine-stimulated increased lactate oxidation also distinguishes the teleost hepatocyte response from that of rats. All gluconeogenic hormones including phenylephrine (1) stimulate mitochondrial respiration in mammals, but such effects in fish are equivocal (reviewed in Ref. 22). Kraus-Friedmann (18) proposed that a redistribution of cellular calcium could account for this increase. The absence of this effect in fish hepatocytes tested to date further supports the minor role calcium may play in these systems. The responsiveness of the three species used in our study to phenylephrine closely followed that of glucagon and GLP (23). Changes in CAMP were not found to correlate well with the size of the stimulation of pathway flux; glucagon stimulated bullhead glycogenolysis by -IO-fold, yet CAMP concentrations rose by only 1.4-fold, whereas in the eel CAMP increased by 20-fold with a 3fold increase in this pathway. Only small changes occurred in CAMP concentrations in bullhead hepatocytes with phenylephrine addition, even though glycogenolytic rates increased by approximately sevenfold (Fig. 2, Table 3). This apparent dissociation between CAMP and pathway responsiveness provides some support for a CAMPindependent transduction mechanism in these fish hepatocytes. In the rat hepatocyte, phenylephrine-stimulated metabolic changes are calcium dependent, but there is little evidence for such mechanisms in lower vertebrates (fish and amphibians): 1) calcium or calcium ionophores do not alter hepatocyte glucose release (4, 14, 15); 2) cr-adrenergic antagonists are unable to compete for iodocyanopindolol (a ,&adrenergic ligand) binding sites on carp liver cells (15); 3) phentolamine (an aantagonist) does not block agonist-stimulated glucose release by fish hepatocytes (4,5,15,24), CAMP increases (4, 6), or glycogen phosphorylase a increases (5, 15); and 4) there is no indication of phenylephrine increasing hepatocyte oxidation (Table 3) or pyruvate dehydrogenase activities (unpublished data), which are classic sites of action for phenylephrine in the rat liver (1, 18). It may be that the classic adrenergic antagonists used in rats are not appropriate in these fish systems, although even in rats the intepretation of antagonist data is not completely clear (see reviews in Refs. 7, 17). This study has shown that both of the vasoactive
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 28, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
VASOACTIVE
PEPTIDES
AND
I’HENYLEPHRINE
peptides vasotocin and isotocin and the adrenergic agonist phenylephrine stimulate glycogenolysis and gluconeogenesis in trout, bullhead, and eel hepatocytes. Although we show that increased CAMP levels accompany the metabolic effects in all cases, the precise nature of the coupling between these ligands and the changes in pathway flux remains unclear. Certainly the clear a-like effects reported in the rat hepatocyte system for both the vasoactive peptides and phenylephrine do not apply to these teleost systems. However, a small contribution of an a-mechanism to metabolic effects dominated by ,8effects (cf. propranolol effects) cannot be entirely excluded for the fish liver. We thank Drs. Jim Fenwick and Erika Plisetskaya for discussions of these data. Eels were provided by the joint cooperation of the Ontario Ministry of Natural Resources and Ontario Hydro. This study was supported by Natural Sciences and Engineering Research Council of Canada Grants OGPA6944 to T. Moon and OGPO041196 to T. Mommsen. Address for reprint requests: T. Moon, Dept. of Biology, Univ. of Ottawa, Ottawa, Ontario KlN 6N5, Canada. Received
16 January
1990; accepted
in final
form
28 June
1990.
REFERENCES 1. ASSIMACOPOULOS-JEANNET, F., J. G. MCCORMACK, AND B. JEANRENAUD. Effect of phenylephrine on pyruvate dehydrogenase activit,y in rat hepatocytes and its interaction with insulin and glucagon. FEBS Lett. 159: 83-88, 1983. 2. BATTEN, T. F. C. IJltrastructural characterization of neurosecretory fibers immunoreactive for vasotocin, isotocin, somatostatin, LHRH and CRF in the pituitary of teleost fish. Cell Tissue Res. 244: 661-671, 1986. 3. BENTLEY, P. J. Comparative Vertebrate Endocrinology (2nd ed.). Cambridge, UK: Cambridge Univ. Press, 1982, p. 485. 4. BIRNBAUM, M. J., J. SCHULTZ, AND J. N. FAIN. Hormone-stimulated glycogenolysis in isolated goldfish hepatocytes. Am. J. Physiol. 231: 191-197, 1976. L., A. C. PUVIANI, M. E. GAVIOLI, E. FABBRI, AND C. 5. BRIGHENTI, OTTOLENGHI. Catecholamine effect on cyclic adenosine 3’:5’-monophosphate level in isolated catfish hepatocytes. Gen. Comp. Endocrinol. 68: 216-223, 1987. L., A. C. PUVIANNI, M. E. GAVIOLI, AND C. OTTO6. BRIGHENTI, LEN(;HI. Mechanisms involved in catecholamine effect on glycogenolysis in catfish isolated hepatocytes. Gen. Comp. Endocrinol. 66: 306-313, 1987. 7. EXTON, J. H. Mechanisms involved in cu-adrenergic phenomena. Am. J. Physiol. 248 (Endocrinol. Metab. 11): E633-E647, 1985. G. D., AND T. W. MOON. The role of glycogen phospho8. FOSTER, rylase in the regulation of glycogenolysis by insulin and glucagon in isolated eel (Anguilla rostrata) hepatocytes. Fish Physiol.
IN
FISH
LIVERS
E649
Biochem.
‘* 10 ’ I’* 12.
13. 14.
15.
16. 17. 18. 1g
*
20.
21* 22.
23.
24.
25.
26.
In press. C. J., T. P. MOMMSEN, AND P. W. HOCHACHKA. Amino acid utilization in isolated hepatocytes from rainbow trout. Eur. J. Biochem. 113: 311-317, 1981. GUIBBOLINI, M., AND B. LAHLOU. Neurohypophysial peptide inhibition of adenylate activity in fish gills. The effect of environmental salinity. FEBS Lett. 220: 98-102, 1987. HANKE, W., AND P. A. JANSSENS. The role of hormones in regulation of carbohydraate metabolism in the Australian lungfish Neoceratodus forsteri. Gen. Comp. Endocrinol. 51: 364-369, 1983. HEIERHORST, J., S. D. MORELY, ?J. FI(XJEROA, C. KRENTI~ER, K. LEDERIS, AND D. RICHTER. Vasotocin and isotocin precursors from the white sucker, Catostomus commersoni: cloning and sequence analysis of the cDNAs. Proc. Natl. Acad. Sci. USA 86: 5242-5246, 1989. HEMS, D. A., AND P. D. WHITTON. Control of hepatic glycogenolysis. Physiol. Rev. 60: l-50, 1980. JANSSENS, P. A., AND J. A. GRIGG. Hormonal regulation of hepatic glycogenolysis in the toad, Xenopus laevis, is mediated by cyclic AMP and not Ca”. Gen. Comp. Endocrinol. 67: 227-233, 1987. JANSSENS, P. A., AND P. LOWREY. Hormonal regulation of hepatic glycogenolysis in the carp, Cyprinus carpio. Am. J. Physiol. 252 (Regulatory Integrative Comp. Physiol. 21): R653-R660, 1987. JARD, S., AND J. BOCKAERT. Stimulus-response coupling in neurohypophysial peptide target cells. Physiol. Rev. 55: 489-536, 1975. KRAUS-FRIEDMANN, N. Hormonal regulation of hepatic gluconeogenesis. Physiol. Rev. 64: 170-259, 1984. KRAUS-FRIEDMANN, N. What is the role of Ca” in the hormonal stimulation of gluconeogenesis ? Trends Biochem. Sci. 11: 276-279, 1986. LAHLOU, B., B. FOSSAT, J. PORTHE-NIBELLE, L. BIANCHINI, AND M. GUIBBOLINI. Inhibitory effect of neurohypophysial peptides on cyclic AMP accumulation by isolated hepatocytes of the trout. J. Endocrinol. 119: 439-445, 1988. MCKEOWN, B. A., T. M. JOHN, AND J. C. GEORGE. Effect of vasotocin on plasma GH, free fatty acids and glucose in Coho salmon (Oncorhynchus hisutsch). Endocrinol. Exp. 10: 45-51,1976. MOMMSEN, T. P. Comparative gluconeogenesis in hepatocytes from salmonid fishes. Can. J. 2001. 64: 1110-1115, 1986. MOMMSEN, T. P., AND T. W. MOON. Metabolic actions of glucagon-family hormones in liver. Fish Physiol. B&hem. 7: 279-288, 1989. MOMMSEN, T. P., AND T. W. MOON. Metabolic response of teleost hepatocytes to glucagon-like peptide and glucagon. J. Endocrinol. 126: 109-118, 1990. MOMMSEN, T. P., P. J. WALSH, S. F. PERRY, AND T. W. MOON. Interactive effects of catecholamines and hypercapnia on glucose production in isolated trout hepatocytes. Gen. Comp. Endocrinol. 70: 63-73, 1988. SHERIDAN, M. A. Effects of epinephrine and norepinephrine on lipid mobilization from Coho salmon liver incubated in vitro. Endocrinology 120: 2234-2239, 1987. VANDEKERCKHOVE, A., F. MIOT, S. KEPPENS, AND H. DE WUI,F. Lack of V, vasopressin receptors in rabbit hepatocytes. Biochem. J. 259: 609-611, 1989. FRENCH,
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 28, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
actions
THOMAS W. MOON AND THOMAS P MOMMSEN Department of Biology, University of Ottawa, Ottawa, Ontario KIN 6N5; and Department of Biochemistry and Microbiology, Universit) of Victoria, Victoria, British Columbia V8 W 2Y2, Canada
MOON, THOMAS W., AND THOMAS P. MOMMSEN. Vasoactive peptides and phenylephrine actions in isolated teleost hepatocytes. Am. J. Physiol. 259 (Endocrinol. Metab. 22): E644E649, 1990.-The metabolic actions of the vasoactive peptides vasotocin and isotocin and the a-agonist phenylephrine are examined in hepatocytes isolated from three teleost species: brown bullhead, rainbow trout, and American eel. These three compounds influenced hepatic gluconeogenesis and glycogenolysis with significant species differences. Vasotocin and isotocin affected only eel hepatocytes activating gluconeogenesis by 1.7-fold and glycogenolysis by 3-fold. Phenylephrine increased glycogenolysis by 7-fold in bullhead hepatocytes and gluconeogenesis by 1.4-fold in trout cells. Vasotocin and phenylephrine actions were correlated with increases in adenosine 3’,5’-cyclic monophosphate (CAMP). The vasotocin effects were unaffected by p- and a-antagonists supporting a &-type receptor on eel hepatocytes. Phenylephrine effects were abolished by propranolol and reduced by prazosin and yohimbine ( cyI- and cu2-antagonists, respectively). Phenylephrine, therefore, affected fish hepatocyte metabolism either by a mixed a/ ,&receptor mechanism emphasizing ,8-adrenoceptors or the classic CU/@agonist/antagonist scheme defined for mammals is not appropriate for these fish preparations.
gluconeogenesis; glycogenolysis; lactate oxidation; adenosine 3’,5’-cyclic monophosphate; cy- and P-adrenoceptors; V2 receptors; isotocin; vasotocin; trout; bullhead; American eel
of fish liver and isolated teleost hepatocytes to adrenergic substances is specific to different metabolic pathways. In general, ,&adrenoceptor agonists including epinephrine are known to initiate the depletion of endogenous liver glycogen by activating glycogen phosphorylase and activating metabolic flux through gluconeogenesis (4, 5, 11, 15, 24). In one species of Pacific salmon, norepinephrine increased lipolysis and therefore the availability of nonesterified fatty acids and glycerol (29, with the latter probably exerting indirect effects on the above pathways. Each of these effects is blocked by the general P-adrenoceptor antagonist propranolol but not a-adrenoceptor antagonists (e.g. phentolamine). Alterations in hepatocyte calcium by changes in extracellular concentrations, ionophores, or channel blockers have been found ineffectual in modifying epinephrinestimulated glycogenolysis in amphibians (14) and fish (4, 15). In fact, it has been stated that fish liver cells are devoid of receptors or specific cellular responses that are in mammals attributed to a-adrenoceptor responses (4, 15), although Brighenti et al. (5, 6) do not dismiss the
THE RESPONSE
E644
0193-1849/90
$1.50 Copyright
possibility of propranolol acting directly or indirectly on an a-mechanism in catfish liver cells. In addition to cyagonists, other hormones acting through the mobilization of calcium such as the vasoactive peptides (vasotocin, isotocin, and vasopressin) cause pronounced metabolic changes in mammalian liver cells (13) but are either ineffective [goldfish (15)] or increase [lungfish (11)] metabolic flux in fish liver. Our interest in the evolution of hormone receptor systems led us to analyze whether the variable metabolic effects of the vasoactive peptides and of the presumed a-adrenoceptor agonists were a common characteristic of teleost liver cells. In addition, we were puzzled by the unexplained finding of metabolic actions of the a-agonist phenylephrine allegedly acting through ,&adrenoceptors in liver cells of catfish (5) and carp (15). In mammals, cu-adrenergic phenomena generally are mediated through the mobilization of intracellular calcium and their metabolic actions are characterized by an activation of pyruvate dehydrogenase, a key enzyme in substrate funneling into the tricarboxylic acid cycle, which in turn results in increased rates of substrate oxidation (1, 7, 18). This study, therefore, presents a comparative analysis of the metabolic responses of hepatocytes from three species of teleost fishes: brown bullhead, rainbow trout, and American eel. We show that in the eel, but not in the other two species, both vasotocin and isotocin at nanomolar concentrations strongly activate gluconeogenesis and glycogenolysis, whereas the presumed CYagonist phenylephrine leads to the stimulation of both pathways in all three species. The use of specific agonists and antagonists provides evidence for the role of CY-and ,&adrenoceptors in these metabolic processes. MATERIALS
AND METHODS
Experimental animals were collected and maintained as described elsewhere (23). American eels (Anguilla rostrata) were collected in September and maintained without feeding at ambient water temperatures (November, 8°C; March, 3”C), brown bullheads (Ictalurus nebulosus) were fed and held at lZ°C, and rainbow trout (Oncorhynchus mykiss) were fed and held at ambient water temperatures on a 12:12 h light-dark photoperiod. Experiments were conducted either in November or the following March. Viable hepatocytes were obtained in good yield as outlined (23). Briefly, livers were perfused in situ with
0 1990 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 28, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
VASOACTIVE
PEPTIDES
AND
PHENYLEPHRINE
collagenase (0.015%) dissolved in the appropriately adapted Hanks’ salt solution with (eel) or without (bullhead, trout) 2 mM CaC12. After digestion of the liver for 20 (bullhead and trout) or 45 (eel) min at room temperature, cells were filtered through plankton netting (254 and 53 pm) and washed four times with Hanks’ solution containing 1.5 mM CaC12 and 2% defatted bovine serum albumin. Final suspensions contained 25-70 mg cells/ml. Rates of lactate gluconeogenesis and lactate oxidation were measured with a radiotracer technique (9). Glycogenolysis (in the absence of added lactate) in the three species and lactate gluconeogenesis in eel and bullhead cells were determined as glucose production by an enzymatic technique (23). The nonradiolabeled assay was found appropriate since lactate gluconeogenesis represented 50% (eel) and 69% (bullhead) of total glucose production (23). Cells were incubated with agonists, antagonists, or vehicle for a maximum of 60 min at 15OC. Vasoactive peptides and agonists/antagonists were dissolved in incubation medium before each experiment, and incubations were conducted in the dark. The following substances with their familiar mammalian classification (7) were used: agonist, L-phenylephrine (a& antagonists, prazosin ( cul; Pfizer), yohimbine (No), timo101(PI), atenolol (,&), and propranolol (p1 2). Unless noted otherwise, all substances were purchased from Sigma Chemical, St. Louis, MO. All chemical assays were conducted as previously reported (23). Experiments were conducted in a randomized block design that makes it possible to disregard the substantial fish-to-fish variation in actual metabolic rate and focus statistical analysis on hormone-induced alterations. No numbers were generated for empty cells, and statistical analysis always used paired comparisons. Specifically, analysis of variance (ANOVA) type III with fish-to-fish variation assigned as the second variable was generally employed. In some cases, orthogonal design was used. Therefore, rather than means, data are presented as percent increase (or decrease) over vehicle-treated control cells. RESULTS
Vasoactiue peptides. Of the three teleost preparations tested, only American eel hepatocytes consistently responded significantly to either vasotocin or isotocin; vasopressin was without effect on eel and bullhead cells (Table 1). Trout hepatocytes did demonstrate a small significant increase in glycogenolysis with isotocin and in gluconeogenesis with vasopressin (Table 1). Eel hepatocyte glycogenolysis was increased by 2.5and 3.6-fold by vasotocin and isotocin, respectively (Table 1). These effects were not altered significantly by the presence of either ,&- (propranolol, timolol, or ateno101) or cv- (prazosin, cyl, or yohimbine, cua)antagonists (data not shown). The glycogenolytic effects of the vasoactive peptides were seasonally dependent with effects in March reduced to -30% of the November values, a decrease that was approximately equivalent to that seen in the absolute rates of glycogenolysis in these cells. Isotocin did not significantly affect glycogenolysis in
IN
FISH
E645
LIVERS
1. Effects of vasoactive peptide hormones on glycogenolytic rates and lactate gluconeogenic and oxidation rates in isolated teleost hepatocytes
TABLE
Bull head
Trout
Glycogenolytic Control Vasotocin Isotocin Vasopressin
14.5-+1.57 100 107t2.9 109k3.2 10522.8
(12) (12) (4)* (5)
Vasotocin Isotocin Vasopressin
0.81t0.24 (10) 100 109t3.4 (6) ND 11355.2 (5)” Oxidation
Control Vasotocin Isotocin Vasopressin
5.42t0.80 (10) 100 lOOt1.7 (6) ND 102*1.8 (5)
Eel
9.02t1.61 100 245k24.7 364t39.0 104t5.8
(25) (23)” (16)” (4)
7.68kl.l
(26)
rate
3.11kO.37 100 115t14.5 103k4.2 98t6.2
Gluconeogenic Control
American
(29) (8) (4) (8)
rate
7.15t0.65 (24) 100 103t2.8 (8) ND llOzk2.3 (4)
100 lllk2.8 9721.0 108t3.3
(5)” (4) (4)
21.2t2.84 100 103t2.6 98k3.2 lOO_t3.1
(15)
rate 28.2t1.87 (24) 100 96k2.1 (8) ND lOO_t2.0 (4)
(4) (4) (3)
Values are expressed as percent of control (100%); number of experiments is in parentheses; control values have been reported previously (see Ref. 23). All experiments were conducted on fish during November. Concentrations of peptides were (in PM) vasotocin, 0.1; isotocin, 0.1; and vasopressin, 0.1. Glycogenolytic rates are expressed as pmol glucose produced from endogenous glycogen in 60 min/g cells k SE. Bullhead and eel hepatocytes were also tested in March and glycogenolytic rates were 122 t 4.3 (4) for isotocin (significantly different from vehicletreated controls, P < O.O5), but absolute control value was not significantly different from November value. Eel rates were 151 t 5.5 (16) for vasotocin and 126 -+ 8.1 (4) for isotocin (significantly different from vehicle-treated controls, P < 0.05), and both absolute control [2.63 t 0.39 (14)] and isotocin values were significantly different from the respective November values. For both rainbow trout (21) and eel (8), control glycogenolytic rates are a function of amount of glycogen present after isolation of hepatocytes. Lactate (2 mM, eel and trout; 4 mM, bullhead) gluconeogenic rates are expressed as pmol glucose produced in 60 min/g cells ~fi SE. Tentatively, rates of glucose production from lactate can be multiplied by 2 to give rate of lactate flux into gluconeogenesis; these rates added to rate of glycogenolysis will give total rate of glucose production. Lactate (2 mM, eel and trout; 4 mM bullhead) oxidation rates are expressed as pmol CO, formed in 60 min/ g cells k SE. Tentatively, rates of CO, production from lactate can be divided by 3 to yield actual flux of lactate into oxidative pathways. Statistics were done on nontransformed, not percent, values. ND, not done. * Significantly different from control (vehicle only) rates; P < 0.05.
bullhead cells isolated in November, but rates were significantly above control when assayed in March (Table 1). The sensitivity of glycogenolysis to these vasoactive peptides was in the 1 to 10 nM range (Fig. 1, A and B), and vasotocin increased hepatocyte adenosine 3’,5’cyclic monophosphate (CAMP) levels (Fig. lC, Table 2). 3-Isobutyl-l-methylxanthine (IBMX; 2.5 ,uM), a phosphodiesterase inhibitor, failed to augment CAMP accumulation at 1 PM vasotocin (Table 2). IBMX, added alone, significantly increased glycogenolysis in cells from both bullheads (173% of control) and eels (155% of control). Changes in CAMP were dose dependent (Table 2) and achieved a maximum within the first 10 min of hormone exposure (Fig. 1C). Again, the extent of CAMP increase with vasotocin was seasonally dependent with
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 28, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
E646
VASOACTIVE
PEPTIDES
AND
A Glycoged ysis , bso tocin
0
-6
1
-6
-4
(mol /L 1
loo
80 60 40
20
trino
a
Log concentration
(mol/L)
CAMP
o-O-0
Control I 0
I 5
Time
I IO
I
15
535+29(10)
min 30
10 559k53
(10)
532t31
1,012?167 1,245+150
(4) (8)
802+122(4)
1,175+93 1,595+238
(4) (4)
848k47 853k52 897t88
(10)
(8) (4) (4)
Values are expressed as pmol CAMP/g cells t SE; number of experiments is in parentheses. Cells were incubated at 15°C for indicated times. Cells were collected and processed for CAMP as indicated in MATERIALS AND METHODS. Experiments were done in a complete block design. IBMX, 3-isobutyl-l-methylxanthine (2.5 PM). All values are significantly different from time 0 control values (P < 0.05).
8 Glycogenolysis, lsotocin and Phenylephrine
1. American
LIVERS
2. Effects of vasotocin on concentration of CAMP in isolated American eel hepatocytes assayed in March
Control Vasotocin lo-’ M lo+ M lo-” M + IBMX lo-‘M
Log vosotocin
1
30
(minutes)
eel hepatocytes and effects of vasoactive peptide hormones and phenylephrine on glycogenolysis and increases in adenosine 3’,5’-cyclic monophosphate (CAMP) content with vasotocin addition. A: vasotocin and glycogenolysis. Composite plot, n = 4-14 experiments in unpaired design; values are means & SD; C, control (vehicle only). R: isotocin and phenylephrine sensitivity of glycogenolysis. Experiments were paired, n = 4; values are means t SD; C, control (vehicle only). C: time course of CAMP increases after vasotocin (Vt) addition. Mean of ‘2-4 experiments with SD not exceeding 10% of mean; Vt = lo-’ M. Open squares, controls for both November and March experiments. For actual CAMP concentrations, see Table 3. FIG.
FISH
Time,
c -K)
C
IN
TABLE
loo 80 60 40 20 0
0
PHENYLEPHRINE
values achieved in March -70% of those in November (Fig. 1C). Eel gluconeogenesis from lactate was affected by vasotocin but not by isotocin or vasopressin (Table 1). This effect was much reduced compared with that seen for glycogenolysis in the November experiments but was essentially equivalent in the March studies even though absolute gluconeogenic rates fell to 20% of the November values. The vasotocin-dependent activation of gluconeogenesis in the March experiments was not affected by either CY-or ,&antagonists or IBMX (data not shown). None of the vasoactive peptides tested significantly affected the rate of lactate oxidation in the hepatocyte preparations employed (Table 1). Phenylephrine. Phenylephrine stimulated glycogenolysis and lactate gluconeogenesis in hepatocytes of each species tested (Table 3). This effect was species and pathway dependent. On the one hand, glycogenolysis was activated to a larger extent than gluconeogenesis with bullhead hepatocytes showing the greatest increase in glycogenolysis followed by the eel with the trout a distant third. Phenylephrine-stimulated glycogenolysis was seasonally dependent; rates in the bullhead preparation were reduced to -30% in March compared with November values (Table 3). On the other hand, trout gluconeogenesis (Table 3) was much more sensitive to phenylephrine than either of the other two species. In each case, propranolol reduced this stimulation to levels found with propranolol alone (i.e., without phenylephrine addition); in fact, propranolol decreased gluconeogenic rates significantly below control values. Propranolol (a general ,8adrenoceptor blocker) more effectively blocked phenylephrine-activated glycogenolysis than either timolol or atenolol (,&-adrenoceptor blockers); this was especially obvious in the case of the bullhead (Table 3). In addition, the phenylephrine-stimulated effects on glycogenolysis were reduced by a-antagonists (Table 3) with a significant depression noted in bullhead cells. Phenylephrine sensitivity of these hepatocytes was in the range of 50-100 nM regardless of whether glycogenolysis (Figs. 1B and 2) or gluconeogenesis was examined (Fig. 3). CAMP concentrations rose in bullhead hepatocytes with phenylephrine addition, although increases were small (Fig. ZB). CAMP concentrations were not affected by the concurrent presence of ionomycin, a
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 28, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
VASOACTIVE
PEPTIDES
AND
PHENYLEPHRINE
IN
3. Effects of a-agonist phenylephrine and selective antagonists on glycogenolytic, lactate gluconeogenic, and oxidation rates in isolated teleost hepatocytes TABLE
Trout
Bullhead
A
American
Phenylephrine +Prazosin +Yohimbine +Propranolol +Timolol +Atenolol
(20)” (8)“t (8)*t (8)*t (4)*-F (4)“t
258t24.7 153-r-27.1 172t20.4 96t5.5 lOlt4.2 136t8.2
(17)” (5)” (5)* (10) t (7)t (7)“t
Phenylephrine +Prazosin +Yohimbine +Propranolol Propranolol
Gluconeogenic rate 135t7.8 (8)” 117t3.4 148t9.4 (4)* 118t3.7 144t9.3 (4)” 117t3.7 82t5.2 (3)* 98k3.3 86k3.3 (6)” lOlt2.8
(4)” (4)* (4)” (4)t (4)
109t1.5 108t2.5 llOt3.2 84t1.2 86kl.l
(7)” (5) (5) (4)“t (11)”
Phenylephrine +Prazosin +Yohimbine +Propranolol
106t3.2 110t5.0 106t3.6 9921.3
(4)”
104t2.0 103t0.6 107k1.7 98t5.2
(6) (4) (4)” (3)
(8) (4)” (4) (3)
E z -=
Eel
Glycogenolytic rate 125t2.1 (4)” 689t114 119t4.8 (4)” 393k65 119t4.7 (4)* 463t68 102k2.8 (4)” 127t9.4 ND 163t5.9 ND 253t26
Oxidation
FISH
g %
&0
E647
LIVERS
Total glucose
production
60
40 20 0 C -10
-8
-6
Log Phenylephrine 6
-4 (mol IL)
CAMP
rate 93t0.7 ND ND ND
Values are percent of control (vehicle only) (see Table 1); number of experiments is in parentheses. Except where noted (propranolol, gluconeogenic rate), no antagonist significantly affected control rates in absence of phenylephrine. Bullhead hepatocytes were also tested in March; glycogenolytic rates were 263 k 31 (8) [significantly different from controls (vehicle only); P < 0.051, but absolute control value was not significantly different from November value (see Ref. 23). Concentrations of agonists/antagonists were (in ,uM) phenylephrine, 10; prazosin, 10; yohimbine, 10; propranolol, 100; timolol, 10; and atenolol, 10. Antagonists were always added before agonists to cell incubates. Statistics were done on nontransformed, not percent, values. ND, not done. * Significantly different from control (vehicle only) rates, P < 0.05. t Significantly different from phenylephrine addition.
calcium ionophore, but were increased by IBMX. No major effects of phenylephrine were noted for lactate oxidation except for a minor depression with the bullhead cells (Table 3). DISCUSSION
After treatment with the vasoactive peptides, vasototin and isotocin, and with phenylephrine, two key metabolic pathways namely glycogenolysis and gluconeogenesis changed substantially in isolated teleost hepatocytes. A third pathway, lactate oxidation, remained essentially unaffected. In all three species analyzed, the rates of glycogenolysis and gluconeogenesis increased after incubation with phenylephrine, whereas vasotocin and isotocin exerted metabolic actions only in the American eel. Vasoactive peptides. Teleosts are reported to produce the two chemically related neurohypophysial peptides vasotocin and isotocin, which are analogous to mammalian vasopressin and oxytocin, respectively (2,3). Recent cDNA sequence data supported the common ancestral nature of this hormone superfamily (12). Vasotocin exerts a hyperglycemic action in coho salmon (20), but circulating concentrations of this hormone were undetectable in the American eel (3). Our results clearly show an increase in eel hepatocyte glucose production primarily from glycogen, and they also show that the effects of
I
I
1
0
IO
30
Time (minutes) 2. Bullhead hepatocytes and effects of phenylephrine on total glucose production and increases in CAMP content withphenylephrine addition. A: phenylephrine sensitivity of total glucose production. Represents total content of glucose in hepatocyte incubations after 1 h; values are means t SD, n = 4; C, control (vehicle only). B: time course of CAMP increases after phenylephrine addition. Open circles, phenylephrine at lo-” M; closed circles, phenylephrine + ionomycin (lo-” M); closed squares, phenylephrine + 3-isobutyl-l-methyl xanthine (IBMX) (2.5 X IO-’ M). Actual CAMP content of bullhead hepatocytes at 0 time was 391 X 10-l” mol/g (t 5, n = 4). FIG.
Lactate E
gluconeogenesis
100
g
80
-;
60
i .-x
40
1
Log Phenylephrine
(mols/L)
FIG. 3. Rainbow trout hepatocytes and effects of phenylephrine on lactate gluconeogenesis. Cells were incubated for 60 min with 2 mM Llactate (including 5 X 10” disintegrations/min (dpm) U-[‘“Cllactate per incubate) and analyzed for [14C]glucose (9); values are means k SD, n = 4; C, control (vehicle only).
isotocin are greater than that of vasotocin. Isotocin is neutral compared with the basic peptide vasotocin, a fact which could account for their different responsiveness (3.6- vs. 2.5fold increase in glycogenolysis, respectively; Table 1) and sensitivities (K, of -lo-” vs. IO-’ M, respectively; Fig. 1). Vasoactive peptide hormones in the laboratory rat
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 28, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
E648
VASOACTIVE
PEPTIDES
AND
PHENYLEPHRINE
stimulate hepatic glycogenolysis and gluconeogenesis through a CAMP-independent pathway, which involves increases in intracellular calcium (reviewed in Ref. 7). Our results imply that a CAMP-dependent mechanism is responsible for the activation of glycogenolysis and gluconeogenesis in the eel hepatocyte. CAMP concentrations increased in the presence of vasotocin (Fig. 1; Table Z), but neither of the classic ,& or o-adrenoceptor antagonists modified these biochemical responses. These results suggest the existence of a specific V-type receptor on the eel hepatocyte as recently reported for the trout hepatocyte (19) and gill membrane preparation (10). These authors, however, provided evidence for a V, receptor that is reported to inhibit CAMP accumulation and/or to increase calcium mobilization (classic mammalian vasopressin receptor) rather than the VZ receptor, which when activated leads to increased CAMP content. Our results showing an increased gluconeogenic flux and CAMP content with vasotocin support the existence of V2, not VI, receptors in the eel liver. The experiments reported by Lahlou et al. (19) showed that oxytocin, isotocin, and vasopressin but not vasotocin inhibited glucagon-stimulated CAMP increases. Whether their results were affected by the indirect experimental approach used or by some specific feature of the trout system not detected in our direct approach is unknown. Our studies, those of Janssens and Grigg (14) in toad liver, and indirectly those of McKeown et al. (20) on salmon support the existence of a V2 receptor much like that shown in the mammalian kidney responding to vasopressin (16). The rat may be unusual in possessing hepatic V1 receptors, as recent studies have shown the complete absence of V, vasopressin receptors in rabbit hepatocytes (26). Species, age, and sex differences have been frequently reported in both the existence and responsiveness to vasoactive peptides and adrenergic agonists in mammals (see Refs. 17 and 26). The absence of a hepatocyte response to vasoactive peptides in trout and catfish (present study) and carp (15), but its presence in eel (present study) and lungfish (11) hepatocytes, further demonstrates the distinct species specificity of these receptors within the animal kingdom. Phenylephrine. Phenylephrine, a classic al-adrenoceptor agonist of hepatic tissue in the laboratory rat (reviewed in Ref. 7), activated both glycogenolysis and gluconeogenesis in the three fish species tested here (Table 3). Bullhead hepatocyte glycogenolysis was the most responsive followed by eel glycogenolysis; gluconeogenie responses for both species were well below those of glycogenolysis. Trout, on the other hand, showed better gluconeogenic responses than glycogenolytic, but in both cases these were well below those of the other two species. The sensitivity of the hepatocytes to phenylephrine was low (50% of maximal response is achieved at micromolar phenylephrine concentrations) regardless of the biochemical response tested (Figs. 2 and 3). Previous studies reported tthat hepatocytes isolated from goldfish (4), carp (15), or catfish (5) responded as poorly to phenylephrine as the trout in our study, whereas bullhead and eel liver cells are even less sensitive to this compound. Unlike the laboratory rat where the phenylephrine response is mediated primarily by al-adrenoceptor mechanisms, this
IN
FISH
LIVERS
response in the three species used here is primarily by ,@adrenoceptor mechanisms, as indicated by propranolol blocking the response (Table 3). Timolol and atenolol, ,&antagonists (7), significantly reduced but did not abolish the increases; these results support a primary ,& effect for phenylephrine on the fish hepatocyte. An intriguing result is that both prazosin and yohimbine, cyl- and aZ-adrenoceptor antagonists, respectively (7), reduce the glycogenolytic response of phenylephrine in both bullhead and eel hepatocytes (Table 3). Prazosin is the more effective of the two antagonists. This could indicate a small cul-component (7) to phenylephrine action in these teleostean fishes, especially on glycogenolysis but possibly not on gluconeogenesis. Our data, therefore, suggest that phenylephrine has a mixed a/P response in the bullhead and eel systems, unlike that noted for carp (15). Brighenti et al. (5) could not completely exclude this possibility in the catfish hepatocyte. The absence of a phenylephrine-stimulated increased lactate oxidation also distinguishes the teleost hepatocyte response from that of rats. All gluconeogenic hormones including phenylephrine (1) stimulate mitochondrial respiration in mammals, but such effects in fish are equivocal (reviewed in Ref. 22). Kraus-Friedmann (18) proposed that a redistribution of cellular calcium could account for this increase. The absence of this effect in fish hepatocytes tested to date further supports the minor role calcium may play in these systems. The responsiveness of the three species used in our study to phenylephrine closely followed that of glucagon and GLP (23). Changes in CAMP were not found to correlate well with the size of the stimulation of pathway flux; glucagon stimulated bullhead glycogenolysis by -IO-fold, yet CAMP concentrations rose by only 1.4-fold, whereas in the eel CAMP increased by 20-fold with a 3fold increase in this pathway. Only small changes occurred in CAMP concentrations in bullhead hepatocytes with phenylephrine addition, even though glycogenolytic rates increased by approximately sevenfold (Fig. 2, Table 3). This apparent dissociation between CAMP and pathway responsiveness provides some support for a CAMPindependent transduction mechanism in these fish hepatocytes. In the rat hepatocyte, phenylephrine-stimulated metabolic changes are calcium dependent, but there is little evidence for such mechanisms in lower vertebrates (fish and amphibians): 1) calcium or calcium ionophores do not alter hepatocyte glucose release (4, 14, 15); 2) cr-adrenergic antagonists are unable to compete for iodocyanopindolol (a ,&adrenergic ligand) binding sites on carp liver cells (15); 3) phentolamine (an aantagonist) does not block agonist-stimulated glucose release by fish hepatocytes (4,5,15,24), CAMP increases (4, 6), or glycogen phosphorylase a increases (5, 15); and 4) there is no indication of phenylephrine increasing hepatocyte oxidation (Table 3) or pyruvate dehydrogenase activities (unpublished data), which are classic sites of action for phenylephrine in the rat liver (1, 18). It may be that the classic adrenergic antagonists used in rats are not appropriate in these fish systems, although even in rats the intepretation of antagonist data is not completely clear (see reviews in Refs. 7, 17). This study has shown that both of the vasoactive
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 28, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
VASOACTIVE
PEPTIDES
AND
I’HENYLEPHRINE
peptides vasotocin and isotocin and the adrenergic agonist phenylephrine stimulate glycogenolysis and gluconeogenesis in trout, bullhead, and eel hepatocytes. Although we show that increased CAMP levels accompany the metabolic effects in all cases, the precise nature of the coupling between these ligands and the changes in pathway flux remains unclear. Certainly the clear a-like effects reported in the rat hepatocyte system for both the vasoactive peptides and phenylephrine do not apply to these teleost systems. However, a small contribution of an a-mechanism to metabolic effects dominated by ,8effects (cf. propranolol effects) cannot be entirely excluded for the fish liver. We thank Drs. Jim Fenwick and Erika Plisetskaya for discussions of these data. Eels were provided by the joint cooperation of the Ontario Ministry of Natural Resources and Ontario Hydro. This study was supported by Natural Sciences and Engineering Research Council of Canada Grants OGPA6944 to T. Moon and OGPO041196 to T. Mommsen. Address for reprint requests: T. Moon, Dept. of Biology, Univ. of Ottawa, Ottawa, Ontario KlN 6N5, Canada. Received
16 January
1990; accepted
in final
form
28 June
1990.
REFERENCES 1. ASSIMACOPOULOS-JEANNET, F., J. G. MCCORMACK, AND B. JEANRENAUD. Effect of phenylephrine on pyruvate dehydrogenase activit,y in rat hepatocytes and its interaction with insulin and glucagon. FEBS Lett. 159: 83-88, 1983. 2. BATTEN, T. F. C. IJltrastructural characterization of neurosecretory fibers immunoreactive for vasotocin, isotocin, somatostatin, LHRH and CRF in the pituitary of teleost fish. Cell Tissue Res. 244: 661-671, 1986. 3. BENTLEY, P. J. Comparative Vertebrate Endocrinology (2nd ed.). Cambridge, UK: Cambridge Univ. Press, 1982, p. 485. 4. BIRNBAUM, M. J., J. SCHULTZ, AND J. N. FAIN. Hormone-stimulated glycogenolysis in isolated goldfish hepatocytes. Am. J. Physiol. 231: 191-197, 1976. L., A. C. PUVIANI, M. E. GAVIOLI, E. FABBRI, AND C. 5. BRIGHENTI, OTTOLENGHI. Catecholamine effect on cyclic adenosine 3’:5’-monophosphate level in isolated catfish hepatocytes. Gen. Comp. Endocrinol. 68: 216-223, 1987. L., A. C. PUVIANNI, M. E. GAVIOLI, AND C. OTTO6. BRIGHENTI, LEN(;HI. Mechanisms involved in catecholamine effect on glycogenolysis in catfish isolated hepatocytes. Gen. Comp. Endocrinol. 66: 306-313, 1987. 7. EXTON, J. H. Mechanisms involved in cu-adrenergic phenomena. Am. J. Physiol. 248 (Endocrinol. Metab. 11): E633-E647, 1985. G. D., AND T. W. MOON. The role of glycogen phospho8. FOSTER, rylase in the regulation of glycogenolysis by insulin and glucagon in isolated eel (Anguilla rostrata) hepatocytes. Fish Physiol.
IN
FISH
LIVERS
E649
Biochem.
‘* 10 ’ I’* 12.
13. 14.
15.
16. 17. 18. 1g
*
20.
21* 22.
23.
24.
25.
26.
In press. C. J., T. P. MOMMSEN, AND P. W. HOCHACHKA. Amino acid utilization in isolated hepatocytes from rainbow trout. Eur. J. Biochem. 113: 311-317, 1981. GUIBBOLINI, M., AND B. LAHLOU. Neurohypophysial peptide inhibition of adenylate activity in fish gills. The effect of environmental salinity. FEBS Lett. 220: 98-102, 1987. HANKE, W., AND P. A. JANSSENS. The role of hormones in regulation of carbohydraate metabolism in the Australian lungfish Neoceratodus forsteri. Gen. Comp. Endocrinol. 51: 364-369, 1983. HEIERHORST, J., S. D. MORELY, ?J. FI(XJEROA, C. KRENTI~ER, K. LEDERIS, AND D. RICHTER. Vasotocin and isotocin precursors from the white sucker, Catostomus commersoni: cloning and sequence analysis of the cDNAs. Proc. Natl. Acad. Sci. USA 86: 5242-5246, 1989. HEMS, D. A., AND P. D. WHITTON. Control of hepatic glycogenolysis. Physiol. Rev. 60: l-50, 1980. JANSSENS, P. A., AND J. A. GRIGG. Hormonal regulation of hepatic glycogenolysis in the toad, Xenopus laevis, is mediated by cyclic AMP and not Ca”. Gen. Comp. Endocrinol. 67: 227-233, 1987. JANSSENS, P. A., AND P. LOWREY. Hormonal regulation of hepatic glycogenolysis in the carp, Cyprinus carpio. Am. J. Physiol. 252 (Regulatory Integrative Comp. Physiol. 21): R653-R660, 1987. JARD, S., AND J. BOCKAERT. Stimulus-response coupling in neurohypophysial peptide target cells. Physiol. Rev. 55: 489-536, 1975. KRAUS-FRIEDMANN, N. Hormonal regulation of hepatic gluconeogenesis. Physiol. Rev. 64: 170-259, 1984. KRAUS-FRIEDMANN, N. What is the role of Ca” in the hormonal stimulation of gluconeogenesis ? Trends Biochem. Sci. 11: 276-279, 1986. LAHLOU, B., B. FOSSAT, J. PORTHE-NIBELLE, L. BIANCHINI, AND M. GUIBBOLINI. Inhibitory effect of neurohypophysial peptides on cyclic AMP accumulation by isolated hepatocytes of the trout. J. Endocrinol. 119: 439-445, 1988. MCKEOWN, B. A., T. M. JOHN, AND J. C. GEORGE. Effect of vasotocin on plasma GH, free fatty acids and glucose in Coho salmon (Oncorhynchus hisutsch). Endocrinol. Exp. 10: 45-51,1976. MOMMSEN, T. P. Comparative gluconeogenesis in hepatocytes from salmonid fishes. Can. J. 2001. 64: 1110-1115, 1986. MOMMSEN, T. P., AND T. W. MOON. Metabolic actions of glucagon-family hormones in liver. Fish Physiol. B&hem. 7: 279-288, 1989. MOMMSEN, T. P., AND T. W. MOON. Metabolic response of teleost hepatocytes to glucagon-like peptide and glucagon. J. Endocrinol. 126: 109-118, 1990. MOMMSEN, T. P., P. J. WALSH, S. F. PERRY, AND T. W. MOON. Interactive effects of catecholamines and hypercapnia on glucose production in isolated trout hepatocytes. Gen. Comp. Endocrinol. 70: 63-73, 1988. SHERIDAN, M. A. Effects of epinephrine and norepinephrine on lipid mobilization from Coho salmon liver incubated in vitro. Endocrinology 120: 2234-2239, 1987. VANDEKERCKHOVE, A., F. MIOT, S. KEPPENS, AND H. DE WUI,F. Lack of V, vasopressin receptors in rabbit hepatocytes. Biochem. J. 259: 609-611, 1989. FRENCH,
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 28, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
