- Email: [email protected]
Adrenergic nerves and receptors in the liver
Brain Research Bulletin, Vol. 9, pp. 709-714, 1982. Printed in the U.S.A.
Adrenergic Nerves and Receptors in the Liver E. MOGHIMZADEH,
Departments
of Histology,
A NOBIN AND E. ROSENGREN
Surgery and Pharmacology,
University of Lund, Lund, Sweden
E., A. NOBIN AND E. ROSENGREN. Adrenergic nerves and recqtors in the liver. BRAIN RES. BULL. 9(1-6) 70%714, 1982.-The finding of an adrenergic innervation of the liver parenchyma in many mammalian species, using the Falck-Hillarp fluorescence method, together with the electron microscopy documentation of synaptoid contacts between adrenergic nerve terminals and hepatocytes, constitute an important morphological basis for the possibility of a direct role of the sympathetic nervous system on the liver function. The results from the morphological studies are in accordance with physiological experiments, carried out in many laboratories, showing that direct electrical MOGHIMZADEH,
stimulation of the hepatic nerves in vivo leads to an increased release of glucose from the hepatocytes. These findings have augmented the demand for detailed studies of the receptor systems mediating the effects of catecholamines on liver metabolism. Our pharmacological studies on liver tissue in vitro have revealed a predominant role for adrenergic betareceptors mediating glucose output in man. Adrenergic innervation
Liver
Mammals
Histofluorescence
THE mammalian liver is richly supplied with autonomic nerves of both sympathetic and parasympathetic nature. Many early reports describe a hepatic innervation; however, a detailed analysis of the subject became feasible only with the introduction of specific morphological and pharmacological methods. The development of the Falck-Hillarp method in the early 60’s made it possible to selectively investigate the detailed distribution of adrenergic nerve terminals. The findings of a liver parenchymal sympathetic innervation greatly encourages further studies on a functional metabolic role of the hepatic nerves, postulated already by early physiologists. DISTRIBUTION
OF ADRENERGIC NERVES IN MAMMALIAN TISSUE
LIVER
The liver receives its adrenergic nerve supply via the splanchnic nerve, which emanates from the 7th to 10th thoracic spinal segments and reaches the liver via the celiac ganglia. The nerves enter the liver mainly in association with the hepatic artery and bile ducts [ 11. The intrahepatic distribution of adrenergic nerves has long been a question of debate. The prevailing concept has been that the adrenergic intrahepatic nerve supply occurred exclusively around blood vessels [26,27]. In our laboratory, by using the Falck-Hillarp method ([9,10], for technical details see [3]), we have found an intrinsic adrenergic nerve plexus in the walls of interlobular vessels of both portal and hepatic origin. The characteristic green-fluorescent varicose nerve fibers are especially abundant in the arteries. ‘Moreover, we have found adrenergic ‘Send reprint requests to E. Moghimzadeh,
Copyright
Department
0 1982 ANKHO
Electron microscopy
Receptors
nerve terminals in the liver parenchyma of man [20,21], as well as other mammalian species [17,21] (Fig. 1). The parenchymal adrenergic nerve fibers are seen to branch off from the intrinsic nerve plexus of the blood vessels and pass into the liver lobules. The fibers enter the lobules following the sinusoids; they are frequently seen to branch, and surround individual hepatocytes by interweaving between them. The nerve fibers are randomly distributed throughout the whole liver lobule, thus often extending all the way down to the central vein. We have found a marked species variation in the denstiy of the liver parenchymal innervation (in manuscript). Thus, human liver and liver from the rhesus monkey, baboon, cynomolgus monkey and guinea pig show a high density of parenchymal adrenergic nerves. The rabbit, cat, pig, cow and horse form an intermediate group, having a lower density of varicose adrenergic nerve fibers but an unequivocal distribution of these nerves to the liver parenchyma. The rat and the mouse are species with little or no parenchymal innervation. The adrenergic nature of the liver nerves has been established with microspectrofluorometric recordings [3]. The fluorescent fibers of liver parenchyma showed excitation/emission maxima at 410/480 nm, which is typical for catecholamines [20]. Chemical determinations of norepinephrine in liver tissue, using reversed phase high performance liquid chromatography with electrochemical detection [ 141, further support the view that norepinephrine is the neurotransmittor in these nerves. The results are shown in Fig. 2. The semiquantitative species differences in the density of ad-
of Histology, Biskopsgatan 5, S-223 62 Lund, Sweden.
International
Inc.-0361-9230/82/070709-06$03.00/O
MOGHIMZADEH,
710
NOBIN
AND KOSENtiKEh
FIG. I. Liver from three different mammalian species treated according to the Falck-Hillarp method. A. Cynomolguh monkey liver showing a dense intralobular network of fluorescent varicose adrenergic nerve fibers (X 140). B and C show norepincphrinecontaining varicose nerve fibers running in close contact with hepatocytes. B. Human liver (X 160). C. Guinea pig liver ! . 1601.
renergic parenchymal nerves are well correlated to the norepinephrine concentration of the liver. Thus, human liver, having the highest density of nerve fibers, has also the highest content of norepinephrine. Ultrastructural studies of human and monkey liver have been performed using 5-hydroxydopamine as a selective marker of adrenergic storage sites 141. Synaptoid contacts (neuro-effector distance of 20-100 nm) between adrenergic nerve terminals and hepatocytes were demonstrated, which provides a morphological basis for a functional relationship between the sympathetic nervous system and the hepatic parenchyma (Fig. 3) 1211.
200
400
600
800
1000
Human Rhesus
monkey
Cynomolgus
monkey
Baboon HOrSe COW Pig cat flabbIt Guinea
ROLE OF THE SYMPATHETIC NERVES FOR HEPATIC GLUCOSE RELEASE
The presence of hepatic autonomic nerves was suggested early by physiologists. Thus, electric stimulation of the hepatic nerves, besides its known effects on hepatic resistance-, capacitance-, and exchange vessels (for review see [ 13]), also influences hepatic glucose metabolism 116, 17, 19, 20, 24, 251. Electric stimulation of the splanchnic nerves may result in hyperglycemia via three different mechanisms: (a) adrenergic nerve stimulation of the pancreas resulting in an increased glucagon output concommitant with a decreased insulin output [5], (b) increased catecholamine output from the adrenal glands and (c) direct sympathetic stimulation of the hepatocytes [6]. The relative importance of each of these possible ways for controlling glucose homeostasis is unknown.
pig
Rat
FlG. 2. Liver tissue content of norepinephrine mammalian species. Mean value -+S.E.
(NE) in twelve
We have studied the functional significance of the hepatic parenchymal innervation in controlling blood glucose concentration in man, using peroperative stimulation [20]. Figure 4 shows that in response to selective electrical stimulation of the hepatic sympathetic nerves the arterial plasma glucose concentration increases promptly. These experiments indicate a role of the hepatic sympathetic nerves in release of glucose from the hepatocytes in man.
ADRENERGIC INNERVATION
OF THE LIVER
FIG. 3. Human liver treated with 5hydroxydopamine. An axon terminal embedded in a deep grove in the sinusoidal surface of a hepatocyte. Note small and large vesicles containing electron dense material. (~33,000). From A. Nobin et al. [28].
have been performed with cats [ 171. electrical stimulation of the hepatic sympathetic nerves in the adult, adrenalectomized cat evokes a small, insignificant increase (l-2 mM) of arterial plasma glucose concentration (Fig. 5). Figure 5 also illustrates the rise in arterial plasma glucose concentration when simultaneous electrical stimulation of both hepatic and pancreatic branches of the splanchnic nerve was performed. In this case the arterial plasma glucose concentration increased signilicantly by about 6 mM, together with a rise in the arterial plasma glucagon concentration as well as a marked decrease in the arterial plasma insulin concentration. This indicates that, at least in the cat, the pronounced hyperglycemic effect of activation of the sympathetic nervous system seems only to a minor extent to be mediated via a direct neurogenic effect on glucose release from the hepatocytes [ 171. Similar experiments
Arterial
plasma
glucose
(mM)
Selective
ADRENERGIC RECEPTORS IN LIVER TISSUE The sympathetic adrenergic nerves innervating the vascular bed are reported to mediate their vasomotor effects via alfa-adrenergic receptors (for reviews see [ 13,181).
The mechanisms by which catecholamines, from the adrenergic nerves and adrenal glands, exert their effects on the hepatocytes are much debated. The adrenergic receptor type involved in glycogenolysis and glucose release is, despite
Stim. 9.0-
8.0-
+io
io
Time (min)
3b
io
FIG. 4. Changes of arterial plasma glucose concentration in response to electrical stimulation (20 Hz, 50 V, 1 msec) of the hepatic sympathetic nerves in man. Mean value ?S.E. of 7 experiments are given. *Indicates a statistically significant increase above the control value (p
712
MOGHIMZADEH. INCREASE OF GLUCOSE (% OF CONTROL )
CHANGE OF ARTERIAL PLASMA GLUCOSE CONCENTRAT~N (m M)
NOBIN AND KOSENGKEN OUTPUT
l-_+IS0 HE 200
CHANGE OF ARTERIAL PLASMA INSULIN CONCENTRATION (pmOl/I
u
NE
0-e
PHE
_y)
1
150 CHANGE OF ARTERIAL PLASMA GLUCAGON CONCENTRATtON (pmOl/ 1 1
‘“”
100 FIG. 5. Changes of arterial plasma glucose, glucagon and insulin concentrations in response to electrical stimulation (10 Hz, 10 V. I msec) of the hepatic and pancreatic sympathetic nerves (3---O) or to stimulation of the hepatic sympathetic nerves selectively (O---O). The experiments were performed on three adrenalectomized cats and 60 min elapsed between the end of the first stimulation and the beginning of the second stimulation. Atropine (0.5 mgikg b. wt.) was given prior to both stimulation periods. Mean value tS.E. are given. Asterisks indicate a statistical difference between the two curves. “pcO.05: **p
50
lo-" many years of extensive studies, still not satisfactorily classified. Several species have been examined, and the predominant adrenergic receptor types have sometimes been classified as alpha-, sometimes as beta-, sometimes as both alpha- and beta-receptors or even as an unclassified adrenergic receptor. Species variations and even variations due to the nutritional status of the animals have been reported (for reviews see [7,15]). The relative importance of alpha- and beta-receptors is unclear, especially in the human liver. The problem has only been dealt with in a few in vivo experiments [2,22]. These studies indicate that the alfa-receptors predominate. Many of the problems in defining the adrenergic receptor in the liver arise from the use of the hyperglycemic response in the intact animal as an index of receptor sensitivity to catecholamines. This hyperglycemia is a complex response involving other factors than the liver glycogenolysis, e.g., glucose formation from lactate, insulin effects and tissue uptake of glucose [ 151. Zn vitro studies on human liver slices, carried out in our laboratory, using adrenergic agonists indicate that it is predominantly beta-receptors that mediate increased glucose output (in manuscript). Figure 6 shows the dose-dependent increase in rate of glucose release caused by different adrenergic agonists. The order of potency for the agonists is: isoprenaline > epinephrine > norepinephrine > phenylephrine. This series of potency is typical for a betareceptor of the bet&-subtype [12]. Furthermore, the presence of a specific beta-blocking agent causes a shift of the dose-response curve towards higher concentrations for the epinephrineinduced increase in glucose release (Fig. 7). The finding that a beta-adrenoceptor has a predominant role in the increased rate of glucose release from hepatocytes in the human liver indicates that this effect is mediated via
1o-7
lo+
105
AGONIST CONCENTRATION
10-e
10”
hl)
FIG. 6. The effects of various catecholamines on glucose release from human liver slices. The order of potency for the agonists is: isoprenaline (ISO) > epinephrine (E) > norepinephrine (NE) -a phenylephrine (PHE). Mean value +S.E.
adenosine 3’,5’-monophosphate (CAMP), in accordance with the second messenger hypothesis of hormone action of Sutherland and co-workers 1231. This hypothesis proposes that the primary action of catecholamines on the liver is the stimulation of adenyl cyclase and the formation of CAMP, which serves as the intracellular mediator. It should be born in mind that the existence of an alfareceptor-mediated, non-CAMP-dependent mechanism for catecholamine action on glucose release from liver tissue in some species is supported by many authors 181. COMMENTARY
The finding of an adrenergic innervation of the liver parenchyma in many mammalian species, using the FalckHillarp fluorescence method, togetherwith the electron microscopy documentation of synaptoid contacts between adrenergic nerve terminals and hepatocytes, constitute an important morphological basis for the possibility of a direct role of the sympathetic nervous system on liver function. The results of the morphological studies concur with the physiological experiments carried out in many laboratories, which show that direct electrical stimulation of the hepatic nerves irl viva leads to an increased release of glucose from the hepatocytes. These findings have increased the demand for detailed studies of the receptor systems mediating the effects
ADRENERGIC INNERVATION
713
OF THE LIVER
INCREASE OF GLUCOSE OUTPUT
04 0~ CONTROL)
“’ 1
H
E
c1-(7
PHA+E
m
PROP+E
E CONCENTRATION
(M)
FIG. 7. Interactions of phentohunine (PHA) 10eBM, an alpha-blocking agent, and propranolol (PROP) 10mBM, a beta-blocking agent, on the epinephrine (E)-induced release of glucose from human liver slices. Propranolol causes a shift of the dose-response curve towards higher concentrations. Phentolamine does not give any inhibition. Mean value +S.E.
of catecholamines on liver metabolism. Our pharmacological studies on liver tissue in vitro have revealed a predominant role for adrenergic beta-receptors mediating glucose output
ACKNOWLEDGEMENT
This study was supported by grants from the Swedish Medical Research Council (grant no. 056 and 712).
in man. REFERENCES 1. Alexander,
W. F. The innervation of the biliary system. J. 73: 357-370, 1940. Antonis, A., M. L. Clark, R. L. Hodge, M. Molony and T. R. E. Pilkington. Receptor mechanisms in the hyperglycaemic response to adrenaline in man. Lancer 1: 1135-1137, 1967. Bjorklund, A., B. Falck and C. Owman. Fluorescence microscopic and microspectrofluorometric techniques for the cellular localization and characterization of biogenic amines. In: Methods of Investigative and Diagnostic Endocrinology, edited by S. A. Berson, Vol. I: The Thyroid and Biogenic Amines, edited J. E. Rall and I. J. Kopin. Amsterdam: North-Holland Publishing Co., 1972, pp. 318-368. Bloom, F. E. Electron microscopy of catecholamine-containing structures. Catecholamines. In: Handbook of Experimental Pharmacology, vol. 33, edited by H. Blaschko, E. Muscholl. Berlin: Springer, 1972, pp. 46-78. Bloom, S. R., A. B. Edwards and N. J. A. Vaughn. The role of the sympathetic innervation in the control of plasma glucagon concentration in the calf. J. Physiol., Lond. 233: 457-466, 1973. Edwards, A. V. and M. Silver. Comparison of the hyperglycaemic and glycogenolytic responses to catecholamines with those to stimulation of the hepatic sympathetic innervation in the dog. J. Physiol., Lond. 223: 571-593, 1972. Ellis, S. Some explanations for species differences in hyperglycemic responses to adrenergic agents. Lifp Sri. 22: 12291236, 1978. camp. Neural.
2.
3.
4.
5. 6.
7.
8. Exton, J. H. Mechanisms involved in effects of catecholamines on liver carbohydrate metabolism. Biochem. Pharmac. 28: 2237-2240, 1979. 9. Falck, B. Observations
on the possibilities of cellular localization of monoamines by a fluorescence method. Acta physiol. stand. 56: Sup 1. 1, l-25, 1962. 10. Falck, B., N.- BP Hillarp, G. Thieme and A. Torp. Fluorescence of catecholamines and related compounds condensed with formaldehyde. J. Hisrochem. Cytochem. 10: 348-354, 1962. 11. Falck, B., S. Ingemansson, A. Nobin and E. Rosengren. The adrenergic innervation of the human liver. Acta endocr. Suppl. 199: 181, 1975. 12. Furchgott, R. F. The classification of adrenoceptors (adrenergic receptors). An evaluation from the standpoint of receptor theory. In: Handbook of Experimental Pharmacology. New series, vol. 33, edited by 0. Eichler, A. Farah, H. Herken and A. D. Welch. Berlin: Springer, 1972, pp. 283-335. 13. Greenway, C. V. and R. D. Starke. Hepatic vascular bed. Physiol. Rev. 51: 23-65, 1971. 14. Hansson, C., G. Agrup, H. Rorsman, A.-M. Rosengren and E. Rosengren. Analysis of cysteinyldopa, dopa, dopamine, noradrenalin and adrenalin in serum and urine using high-performance liquid chromatography and electrochemical detection. J. Chromat.
162: 7-22,
1979.
15. Hombrook, K. R. Adrenergic receptors for metabolic responses in the liver. Fedn Proc. 29: 1381-1385, 1970.
714
16. Jarhuh, J. Role of the sympathoadrenal system in haemorrhagic hyperglycemia. Acfrr phy,siol. .sc~cd. 93: 25-33, 1975. 17. Jarhuh, J., P.-O. Andersson, J. Hoist, E. Moghimzadeh and A. Nobin. On the sympathetic innervation to the cat’s liver and its role for hepatic glucose release. Acctr physiol. .sc~om/. 110: 5-l I. 1980. 18. Lautt, W. W. Hepatic nerves: a review of their functions and effects. C’trtt.J. P/r_rsio/. Phn,/rrtrc. 58: 105-123, 1980. 19. Lautt, W. W. and C. Wong. Hepatic glucose balance in response to direct stimulation of sympathetic nerves in the intact liver of cats. Ctr/r. J. Ph.~iol. Phonnrrc. 56: 1022-1028, 1978. 20. Nobin, A., B. Falck, S. Ingemansson, J. Jarhult and E. Rosengren. Organization and function of the sympathetic innervation of human liver. Actn phpiol. .scrrnd., Suppl. 452: 103-106, 1977. 21. Nobin, A., H. G. Baumgarten, B. Falck. S. Ingemansson, E. Moghimzadeh and E. Rosengren. Organization of the sympathetic innervation in liver tissue from monkey and man. Ccl/ Ti.v.\/tr R~.Y. 195: 371-380. 1978.
MOGHIMZADEH,
NOBIN
AND KOSENGKKN
22. Rizza, R. A., M. W. Haymond, J. M. Miles. C. .‘\. Ver-donh. P. E. Cryor and J. E. Gerich. Effect of tu-adrencrgic ctimulation and its blockade on glucose turnover in man. ,,\/u. .I /‘/,I~&/. 238: E467-E472, 1980. 23. Robinson, G. A., R. W. Butcher and E. W. Sutherland. c 1~lir AMP. New York: Academic Press, 1971. 24. Shimazu. T. Nervous control of peripheral metaholi\m. ;\( I<[ ph?‘siol. /x/l. 30: Suppl. 18, I-18, 1979. 25. Shimazu, T. and A. Amakavva. Regulation of glycogen metaholism in the liver by the autonomic nervous system: II. Neural control of glycogenolytic enzymes. Bioclrin/. hi~~ph~~\..A(/cr. 164: 349356, 1968. 26. Skaaring, P. and F. Bierring. On the intrinsic innervation of normal rat liver. Histochemical and scanning electron microscopic studies. Cc/l Ti.t.tru Rm. 171: 141-155, 1976. 27. Ungvary, G. and T. Donath. On the monoaminergic inner-vation of the liver. Ac,ttr c/no{. 72: 446459. 1969.
Adrenergic Nerves and Receptors in the Liver E. MOGHIMZADEH,
Departments
of Histology,
A NOBIN AND E. ROSENGREN
Surgery and Pharmacology,
University of Lund, Lund, Sweden
E., A. NOBIN AND E. ROSENGREN. Adrenergic nerves and recqtors in the liver. BRAIN RES. BULL. 9(1-6) 70%714, 1982.-The finding of an adrenergic innervation of the liver parenchyma in many mammalian species, using the Falck-Hillarp fluorescence method, together with the electron microscopy documentation of synaptoid contacts between adrenergic nerve terminals and hepatocytes, constitute an important morphological basis for the possibility of a direct role of the sympathetic nervous system on the liver function. The results from the morphological studies are in accordance with physiological experiments, carried out in many laboratories, showing that direct electrical MOGHIMZADEH,
stimulation of the hepatic nerves in vivo leads to an increased release of glucose from the hepatocytes. These findings have augmented the demand for detailed studies of the receptor systems mediating the effects of catecholamines on liver metabolism. Our pharmacological studies on liver tissue in vitro have revealed a predominant role for adrenergic betareceptors mediating glucose output in man. Adrenergic innervation
Liver
Mammals
Histofluorescence
THE mammalian liver is richly supplied with autonomic nerves of both sympathetic and parasympathetic nature. Many early reports describe a hepatic innervation; however, a detailed analysis of the subject became feasible only with the introduction of specific morphological and pharmacological methods. The development of the Falck-Hillarp method in the early 60’s made it possible to selectively investigate the detailed distribution of adrenergic nerve terminals. The findings of a liver parenchymal sympathetic innervation greatly encourages further studies on a functional metabolic role of the hepatic nerves, postulated already by early physiologists. DISTRIBUTION
OF ADRENERGIC NERVES IN MAMMALIAN TISSUE
LIVER
The liver receives its adrenergic nerve supply via the splanchnic nerve, which emanates from the 7th to 10th thoracic spinal segments and reaches the liver via the celiac ganglia. The nerves enter the liver mainly in association with the hepatic artery and bile ducts [ 11. The intrahepatic distribution of adrenergic nerves has long been a question of debate. The prevailing concept has been that the adrenergic intrahepatic nerve supply occurred exclusively around blood vessels [26,27]. In our laboratory, by using the Falck-Hillarp method ([9,10], for technical details see [3]), we have found an intrinsic adrenergic nerve plexus in the walls of interlobular vessels of both portal and hepatic origin. The characteristic green-fluorescent varicose nerve fibers are especially abundant in the arteries. ‘Moreover, we have found adrenergic ‘Send reprint requests to E. Moghimzadeh,
Copyright
Department
0 1982 ANKHO
Electron microscopy
Receptors
nerve terminals in the liver parenchyma of man [20,21], as well as other mammalian species [17,21] (Fig. 1). The parenchymal adrenergic nerve fibers are seen to branch off from the intrinsic nerve plexus of the blood vessels and pass into the liver lobules. The fibers enter the lobules following the sinusoids; they are frequently seen to branch, and surround individual hepatocytes by interweaving between them. The nerve fibers are randomly distributed throughout the whole liver lobule, thus often extending all the way down to the central vein. We have found a marked species variation in the denstiy of the liver parenchymal innervation (in manuscript). Thus, human liver and liver from the rhesus monkey, baboon, cynomolgus monkey and guinea pig show a high density of parenchymal adrenergic nerves. The rabbit, cat, pig, cow and horse form an intermediate group, having a lower density of varicose adrenergic nerve fibers but an unequivocal distribution of these nerves to the liver parenchyma. The rat and the mouse are species with little or no parenchymal innervation. The adrenergic nature of the liver nerves has been established with microspectrofluorometric recordings [3]. The fluorescent fibers of liver parenchyma showed excitation/emission maxima at 410/480 nm, which is typical for catecholamines [20]. Chemical determinations of norepinephrine in liver tissue, using reversed phase high performance liquid chromatography with electrochemical detection [ 141, further support the view that norepinephrine is the neurotransmittor in these nerves. The results are shown in Fig. 2. The semiquantitative species differences in the density of ad-
of Histology, Biskopsgatan 5, S-223 62 Lund, Sweden.
International
Inc.-0361-9230/82/070709-06$03.00/O
MOGHIMZADEH,
710
NOBIN
AND KOSENtiKEh
FIG. I. Liver from three different mammalian species treated according to the Falck-Hillarp method. A. Cynomolguh monkey liver showing a dense intralobular network of fluorescent varicose adrenergic nerve fibers (X 140). B and C show norepincphrinecontaining varicose nerve fibers running in close contact with hepatocytes. B. Human liver (X 160). C. Guinea pig liver ! . 1601.
renergic parenchymal nerves are well correlated to the norepinephrine concentration of the liver. Thus, human liver, having the highest density of nerve fibers, has also the highest content of norepinephrine. Ultrastructural studies of human and monkey liver have been performed using 5-hydroxydopamine as a selective marker of adrenergic storage sites 141. Synaptoid contacts (neuro-effector distance of 20-100 nm) between adrenergic nerve terminals and hepatocytes were demonstrated, which provides a morphological basis for a functional relationship between the sympathetic nervous system and the hepatic parenchyma (Fig. 3) 1211.
200
400
600
800
1000
Human Rhesus
monkey
Cynomolgus
monkey
Baboon HOrSe COW Pig cat flabbIt Guinea
ROLE OF THE SYMPATHETIC NERVES FOR HEPATIC GLUCOSE RELEASE
The presence of hepatic autonomic nerves was suggested early by physiologists. Thus, electric stimulation of the hepatic nerves, besides its known effects on hepatic resistance-, capacitance-, and exchange vessels (for review see [ 13]), also influences hepatic glucose metabolism 116, 17, 19, 20, 24, 251. Electric stimulation of the splanchnic nerves may result in hyperglycemia via three different mechanisms: (a) adrenergic nerve stimulation of the pancreas resulting in an increased glucagon output concommitant with a decreased insulin output [5], (b) increased catecholamine output from the adrenal glands and (c) direct sympathetic stimulation of the hepatocytes [6]. The relative importance of each of these possible ways for controlling glucose homeostasis is unknown.
pig
Rat
FlG. 2. Liver tissue content of norepinephrine mammalian species. Mean value -+S.E.
(NE) in twelve
We have studied the functional significance of the hepatic parenchymal innervation in controlling blood glucose concentration in man, using peroperative stimulation [20]. Figure 4 shows that in response to selective electrical stimulation of the hepatic sympathetic nerves the arterial plasma glucose concentration increases promptly. These experiments indicate a role of the hepatic sympathetic nerves in release of glucose from the hepatocytes in man.
ADRENERGIC INNERVATION
OF THE LIVER
FIG. 3. Human liver treated with 5hydroxydopamine. An axon terminal embedded in a deep grove in the sinusoidal surface of a hepatocyte. Note small and large vesicles containing electron dense material. (~33,000). From A. Nobin et al. [28].
have been performed with cats [ 171. electrical stimulation of the hepatic sympathetic nerves in the adult, adrenalectomized cat evokes a small, insignificant increase (l-2 mM) of arterial plasma glucose concentration (Fig. 5). Figure 5 also illustrates the rise in arterial plasma glucose concentration when simultaneous electrical stimulation of both hepatic and pancreatic branches of the splanchnic nerve was performed. In this case the arterial plasma glucose concentration increased signilicantly by about 6 mM, together with a rise in the arterial plasma glucagon concentration as well as a marked decrease in the arterial plasma insulin concentration. This indicates that, at least in the cat, the pronounced hyperglycemic effect of activation of the sympathetic nervous system seems only to a minor extent to be mediated via a direct neurogenic effect on glucose release from the hepatocytes [ 171. Similar experiments
Arterial
plasma
glucose
(mM)
Selective
ADRENERGIC RECEPTORS IN LIVER TISSUE The sympathetic adrenergic nerves innervating the vascular bed are reported to mediate their vasomotor effects via alfa-adrenergic receptors (for reviews see [ 13,181).
The mechanisms by which catecholamines, from the adrenergic nerves and adrenal glands, exert their effects on the hepatocytes are much debated. The adrenergic receptor type involved in glycogenolysis and glucose release is, despite
Stim. 9.0-
8.0-
+io
io
Time (min)
3b
io
FIG. 4. Changes of arterial plasma glucose concentration in response to electrical stimulation (20 Hz, 50 V, 1 msec) of the hepatic sympathetic nerves in man. Mean value ?S.E. of 7 experiments are given. *Indicates a statistically significant increase above the control value (p
712
MOGHIMZADEH. INCREASE OF GLUCOSE (% OF CONTROL )
CHANGE OF ARTERIAL PLASMA GLUCOSE CONCENTRAT~N (m M)
NOBIN AND KOSENGKEN OUTPUT
l-_+IS0 HE 200
CHANGE OF ARTERIAL PLASMA INSULIN CONCENTRATION (pmOl/I
u
NE
0-e
PHE
_y)
1
150 CHANGE OF ARTERIAL PLASMA GLUCAGON CONCENTRATtON (pmOl/ 1 1
‘“”
100 FIG. 5. Changes of arterial plasma glucose, glucagon and insulin concentrations in response to electrical stimulation (10 Hz, 10 V. I msec) of the hepatic and pancreatic sympathetic nerves (3---O) or to stimulation of the hepatic sympathetic nerves selectively (O---O). The experiments were performed on three adrenalectomized cats and 60 min elapsed between the end of the first stimulation and the beginning of the second stimulation. Atropine (0.5 mgikg b. wt.) was given prior to both stimulation periods. Mean value tS.E. are given. Asterisks indicate a statistical difference between the two curves. “pcO.05: **p
50
lo-" many years of extensive studies, still not satisfactorily classified. Several species have been examined, and the predominant adrenergic receptor types have sometimes been classified as alpha-, sometimes as beta-, sometimes as both alpha- and beta-receptors or even as an unclassified adrenergic receptor. Species variations and even variations due to the nutritional status of the animals have been reported (for reviews see [7,15]). The relative importance of alpha- and beta-receptors is unclear, especially in the human liver. The problem has only been dealt with in a few in vivo experiments [2,22]. These studies indicate that the alfa-receptors predominate. Many of the problems in defining the adrenergic receptor in the liver arise from the use of the hyperglycemic response in the intact animal as an index of receptor sensitivity to catecholamines. This hyperglycemia is a complex response involving other factors than the liver glycogenolysis, e.g., glucose formation from lactate, insulin effects and tissue uptake of glucose [ 151. Zn vitro studies on human liver slices, carried out in our laboratory, using adrenergic agonists indicate that it is predominantly beta-receptors that mediate increased glucose output (in manuscript). Figure 6 shows the dose-dependent increase in rate of glucose release caused by different adrenergic agonists. The order of potency for the agonists is: isoprenaline > epinephrine > norepinephrine > phenylephrine. This series of potency is typical for a betareceptor of the bet&-subtype [12]. Furthermore, the presence of a specific beta-blocking agent causes a shift of the dose-response curve towards higher concentrations for the epinephrineinduced increase in glucose release (Fig. 7). The finding that a beta-adrenoceptor has a predominant role in the increased rate of glucose release from hepatocytes in the human liver indicates that this effect is mediated via
1o-7
lo+
105
AGONIST CONCENTRATION
10-e
10”
hl)
FIG. 6. The effects of various catecholamines on glucose release from human liver slices. The order of potency for the agonists is: isoprenaline (ISO) > epinephrine (E) > norepinephrine (NE) -a phenylephrine (PHE). Mean value +S.E.
adenosine 3’,5’-monophosphate (CAMP), in accordance with the second messenger hypothesis of hormone action of Sutherland and co-workers 1231. This hypothesis proposes that the primary action of catecholamines on the liver is the stimulation of adenyl cyclase and the formation of CAMP, which serves as the intracellular mediator. It should be born in mind that the existence of an alfareceptor-mediated, non-CAMP-dependent mechanism for catecholamine action on glucose release from liver tissue in some species is supported by many authors 181. COMMENTARY
The finding of an adrenergic innervation of the liver parenchyma in many mammalian species, using the FalckHillarp fluorescence method, togetherwith the electron microscopy documentation of synaptoid contacts between adrenergic nerve terminals and hepatocytes, constitute an important morphological basis for the possibility of a direct role of the sympathetic nervous system on liver function. The results of the morphological studies concur with the physiological experiments carried out in many laboratories, which show that direct electrical stimulation of the hepatic nerves irl viva leads to an increased release of glucose from the hepatocytes. These findings have increased the demand for detailed studies of the receptor systems mediating the effects
ADRENERGIC INNERVATION
713
OF THE LIVER
INCREASE OF GLUCOSE OUTPUT
04 0~ CONTROL)
“’ 1
H
E
c1-(7
PHA+E
m
PROP+E
E CONCENTRATION
(M)
FIG. 7. Interactions of phentohunine (PHA) 10eBM, an alpha-blocking agent, and propranolol (PROP) 10mBM, a beta-blocking agent, on the epinephrine (E)-induced release of glucose from human liver slices. Propranolol causes a shift of the dose-response curve towards higher concentrations. Phentolamine does not give any inhibition. Mean value +S.E.
of catecholamines on liver metabolism. Our pharmacological studies on liver tissue in vitro have revealed a predominant role for adrenergic beta-receptors mediating glucose output
ACKNOWLEDGEMENT
This study was supported by grants from the Swedish Medical Research Council (grant no. 056 and 712).
in man. REFERENCES 1. Alexander,
W. F. The innervation of the biliary system. J. 73: 357-370, 1940. Antonis, A., M. L. Clark, R. L. Hodge, M. Molony and T. R. E. Pilkington. Receptor mechanisms in the hyperglycaemic response to adrenaline in man. Lancer 1: 1135-1137, 1967. Bjorklund, A., B. Falck and C. Owman. Fluorescence microscopic and microspectrofluorometric techniques for the cellular localization and characterization of biogenic amines. In: Methods of Investigative and Diagnostic Endocrinology, edited by S. A. Berson, Vol. I: The Thyroid and Biogenic Amines, edited J. E. Rall and I. J. Kopin. Amsterdam: North-Holland Publishing Co., 1972, pp. 318-368. Bloom, F. E. Electron microscopy of catecholamine-containing structures. Catecholamines. In: Handbook of Experimental Pharmacology, vol. 33, edited by H. Blaschko, E. Muscholl. Berlin: Springer, 1972, pp. 46-78. Bloom, S. R., A. B. Edwards and N. J. A. Vaughn. The role of the sympathetic innervation in the control of plasma glucagon concentration in the calf. J. Physiol., Lond. 233: 457-466, 1973. Edwards, A. V. and M. Silver. Comparison of the hyperglycaemic and glycogenolytic responses to catecholamines with those to stimulation of the hepatic sympathetic innervation in the dog. J. Physiol., Lond. 223: 571-593, 1972. Ellis, S. Some explanations for species differences in hyperglycemic responses to adrenergic agents. Lifp Sri. 22: 12291236, 1978. camp. Neural.
2.
3.
4.
5. 6.
7.
8. Exton, J. H. Mechanisms involved in effects of catecholamines on liver carbohydrate metabolism. Biochem. Pharmac. 28: 2237-2240, 1979. 9. Falck, B. Observations
on the possibilities of cellular localization of monoamines by a fluorescence method. Acta physiol. stand. 56: Sup 1. 1, l-25, 1962. 10. Falck, B., N.- BP Hillarp, G. Thieme and A. Torp. Fluorescence of catecholamines and related compounds condensed with formaldehyde. J. Hisrochem. Cytochem. 10: 348-354, 1962. 11. Falck, B., S. Ingemansson, A. Nobin and E. Rosengren. The adrenergic innervation of the human liver. Acta endocr. Suppl. 199: 181, 1975. 12. Furchgott, R. F. The classification of adrenoceptors (adrenergic receptors). An evaluation from the standpoint of receptor theory. In: Handbook of Experimental Pharmacology. New series, vol. 33, edited by 0. Eichler, A. Farah, H. Herken and A. D. Welch. Berlin: Springer, 1972, pp. 283-335. 13. Greenway, C. V. and R. D. Starke. Hepatic vascular bed. Physiol. Rev. 51: 23-65, 1971. 14. Hansson, C., G. Agrup, H. Rorsman, A.-M. Rosengren and E. Rosengren. Analysis of cysteinyldopa, dopa, dopamine, noradrenalin and adrenalin in serum and urine using high-performance liquid chromatography and electrochemical detection. J. Chromat.
162: 7-22,
1979.
15. Hombrook, K. R. Adrenergic receptors for metabolic responses in the liver. Fedn Proc. 29: 1381-1385, 1970.
714
16. Jarhuh, J. Role of the sympathoadrenal system in haemorrhagic hyperglycemia. Acfrr phy,siol. .sc~cd. 93: 25-33, 1975. 17. Jarhuh, J., P.-O. Andersson, J. Hoist, E. Moghimzadeh and A. Nobin. On the sympathetic innervation to the cat’s liver and its role for hepatic glucose release. Acctr physiol. .sc~om/. 110: 5-l I. 1980. 18. Lautt, W. W. Hepatic nerves: a review of their functions and effects. C’trtt.J. P/r_rsio/. Phn,/rrtrc. 58: 105-123, 1980. 19. Lautt, W. W. and C. Wong. Hepatic glucose balance in response to direct stimulation of sympathetic nerves in the intact liver of cats. Ctr/r. J. Ph.~iol. Phonnrrc. 56: 1022-1028, 1978. 20. Nobin, A., B. Falck, S. Ingemansson, J. Jarhult and E. Rosengren. Organization and function of the sympathetic innervation of human liver. Actn phpiol. .scrrnd., Suppl. 452: 103-106, 1977. 21. Nobin, A., H. G. Baumgarten, B. Falck. S. Ingemansson, E. Moghimzadeh and E. Rosengren. Organization of the sympathetic innervation in liver tissue from monkey and man. Ccl/ Ti.v.\/tr R~.Y. 195: 371-380. 1978.
MOGHIMZADEH,
NOBIN
AND KOSENGKKN
22. Rizza, R. A., M. W. Haymond, J. M. Miles. C. .‘\. Ver-donh. P. E. Cryor and J. E. Gerich. Effect of tu-adrencrgic ctimulation and its blockade on glucose turnover in man. ,,\/u. .I /‘/,I~&/. 238: E467-E472, 1980. 23. Robinson, G. A., R. W. Butcher and E. W. Sutherland. c 1~lir AMP. New York: Academic Press, 1971. 24. Shimazu. T. Nervous control of peripheral metaholi\m. ;\( I<[ ph?‘siol. /x/l. 30: Suppl. 18, I-18, 1979. 25. Shimazu, T. and A. Amakavva. Regulation of glycogen metaholism in the liver by the autonomic nervous system: II. Neural control of glycogenolytic enzymes. Bioclrin/. hi~~ph~~\..A(/cr. 164: 349356, 1968. 26. Skaaring, P. and F. Bierring. On the intrinsic innervation of normal rat liver. Histochemical and scanning electron microscopic studies. Cc/l Ti.t.tru Rm. 171: 141-155, 1976. 27. Ungvary, G. and T. Donath. On the monoaminergic inner-vation of the liver. Ac,ttr c/no{. 72: 446459. 1969.