A Histochemical Study of the Innervation of Cerebral Blood Vessels in the Turtle TADAHIKO IIJIMA Department ofAnatomy, School
of
Medicine, Fukuoka University, Fukuoka 814, Japan
ABSTRACT Specific histochemical techniques for the demonstration of noradrenaline and of acetylcholinesterase have been used to study t h e distribution of adrenergic and cholinergic nerves to t h e cerebral blood vessels of turtle, Geoclemys reeuesii. The major and medium-sized cerebral arteries were supplied with dense adrenergic nerve plexuses, the plexuses were particularly dense in t h e mediumsized pial arteries of very thick vascular wall, indicating the functional significance of these arteries in the cerebral circulation. The parenchymal arterioles and capillaries were also supplied with adrenergic nerves. On the other hand, t h e cholinergic innervation was less dense than the adrenergic one and the acetylcholinesterase activity of the nerve fibres was remarkably weak. By contrast, the parenchymal small arterioles and capillaries exhibited heavy acetylcholinesterase activity on the vascular wall and, in addition, the capillaries were supplied with t h e well-stained acetylcholinesterasepositive nerve fibres. These fibres and also t h e adrenergic fibres associating with the capillaries appear t o be of central origin. It is suggested t h a t the cholinergic mechanisms in t h e parenchymal small vessels also play a n important role in the cerebral circulation. The basophil leucocytes observed abundantly in t h e blood of turtle emitted a n intensive greenish yellow fluorescence after formaldehyde gas-treatment. The existence of adrenergic and cholinergic nerve terminals in t h e mammalian cerebral blood vessels has been thoroughly demonstrated by histochemical and electron microscopic studies (e.g., Nelson and Rennels, '70; Motavkin and Dovbish, '71; Rosenblum, '71; Owman e t al., '74) and t h e neurogenic influence on the cerebral circulation has also been documented by recent pharmacological studies (e.g., Mchedlishvili and Nikolaishvili, '70; Nielson and Owman, '71; Edvinsson e t al., '73b; Raichle e t al., '75). There are great many studies concerning t h e vertebrate autonomic nervous mechanisms in visceral and cardiovascular systems. I n his review Burnstock ('69)has described t h e mode of innervation in these systems with evolutionary terms. Comparative studies of innervation in t h e cerebral blood vessels, however, are very fragmentary. Recently, Iijima e t al. ('77) reported t h a t both adrenergic and cholinergic nerve plexuses in t h e cerebral J. COMP. NEUR.. 176: 307-314.
blood vessels of the snake show almost similar distribution as those in mammals. As a part of research project on t h e evolution of nervous system innervating t h e cerebral blood vessels of vertebrates, the turtle which is classified in a much lower category than t h e snake in the phylogeny of reptile has been chosen as a n experimental animal. The feature of adrenergic and cholinergic innervation has been investigated by fluorescence and cholinesterase histochemistry. I n t h e present paper, a disproportional development of both nerve plexuses; well-developed, brilliant adrenergic and ill-developed, faintly-stained cholinergic nerve meshwork, will be described. MATERIALS AND METHODS
Twelve turtles, Geoclemys reeuesii, were decapitated, and the brain was rapidly removed from t h e skull and placed in Ringer's solution. The vessels were carefully dissected from t h e brain, and were stretched on t h e glass slides
307
308
TADAHIKO I I J I M A
for fluorescence histochemistry (Falck, '62) or were fixed with 4% formaldehyde buffered to pH 7.4 for 30 minutes for cholinesterase histochemistry (Karnovsky and Roots, '64). For cholinesterase histochemistry with whole mount preparation, the cerebral vessels from eight animals were also fixed by perfusion of 44, formaldehyde through the aorta. To demonstrate the innervation in the parenchymal vessels, small blocks of brain from nine animals were frozen in dry-ice isopentane, and thin sections were prepared with cryostat or freeze-drying techniques. For fluorescence histochemistry, the slidemounted vessels were immediately transferred to a dessicator where they were dried in uucuo over phosphorus pentoxide for one hour. The slides were then treated in formaldehyde vapor from paraformaldehyde for one hour a t 8O"C, and were mounted for fluorescence microscopy. On the other hand, frozen blockes of the brain were dried in uucuo for six days a t - 4 5 ° C two days a t -20°C and one day at room temperature, then treated with formaldehyde gas for one hour at 80°C, and were infiltrated in uucuo with paraffin for ten minutes a t 60"C, and the sections (8-10 p m thick) were examined under the fluorescence microscope. For cholinesterase histochemistry, formalin-fixed vessels were washed thoroughly with distilled water, and maintained in substrate (acetylthiocholine iodide)-free Karnovsky's medium for 30 minutes a t 4"C, and then incubated in the complete medium, conM iso-OMPA (tetraisopropyltaining 2 x pyrophosphoramide) as a n inhibitor of nonspecific cholinesterases, for one hour a t 20°C. After washing with 50!% ethanol. they were stretched on microscope glass and dried in air, and were dehydrated with ethanol series followed with xylene and mounted in balsam. On the other hand, cryostat sections (15-30 p m thick) cut from frozen specimens of the brain were mounted on microscope glass and were fixed with 4% formaldehyde and then treated as described above. RESULTS
The internal carotid and the spinal (vertebral) arteries contributed to the blood supply of the turtle brain, and the branches of these arteries formed the circle of Willis on the ventral surface of t h e brain. The arteries leaving the circle showed a characteristic tortuosity in their course. A further characteris-
tic feature of the cerebral blood vessels in the turtle was the thickness of vascular wall. Although a n approximate diameter of major arteries was 110 p m , t h a t of pial arteries was 60 p m and t h a t of parenchymal arterioles and capillaries was 15 Fm, t h e thickness of vascular walls of both major and medium-sized pial arteries was 10-15 p m and t h a t of parenchymal arterioles and capillaries was 2-5 p m .
Adrenergic innervation A very dense network of fluorescent nerves occurred in t h e cerebral arteries of the turtle. The fluorescent fibres inclined to run longitudinally along the arterial axis, forming a longitudinally elongated meshwork. The nerve plexuses were particularly dense in the medium-sized pial arteries of 40-70 p m in diameter (asterisks in figs. 1,2, 5; fig. 4). The innervation in t h e major arteries, however, was somewhat less dense (figs. 1-3, 5 ) . I n cross sections, i t was clearly shown t h a t the tunica media of richly innervated, medium-sized pial arteries consisted of thick muscular layers (arrowheads in figs. 7 , 9 ) and t h a t those of less richly innervated, major cerebral arteries consisted of relatively thin musculature in contrast to a diameter (asterisk in fig. 7). The medium-sized pial arteries indicated with arrowheads in figures 7 and 9 were 40 and 60 p m in diameter, respectively. And the thickness of the vascular wall of both arteries was about 15 p m . On t h e other hand, the diameter of the major artery indicated by t h e asterisk in figure 7 was 120 p m and t h e wall thickness was about 15 pm. No distinct difference was observed between the density of innervation in the arteries of t h e internal carotid and that of the vertebro-basilar system. Small arteries and arterioles were sparsely innervated (fig. 6 ) , and some penetrating arteries and arterioles into the brain parenchyma were also supplied with adrenergic nerves (arrowheads in figs. 8, 10). Furthermore, the adrenergic nerves were observed on the wall of parenchymal capillaries (arrowheads in fig. 1l a ) . However, this study had failed to clarify the existence of t h e nerve fibre which passed from t h e capillary wall into the depth of the brain parenchyma. Cholinergc innervation T h e p l e x u s e s of a c e t y l c h o l i n e s t e r a s e (AChE)-positive nerve fibres were less dense t h a n t h e adrenergic ones. These plexuses were t h e most dense in the major cerebral arteries,
VASCULAR INNERVATION IN THE TURTLE BRAIN
309
Figs. 1-6 Adrenergic innervation in whole mount preparations. 1. Middle cerebral artery and its branch. 2. Anterior cerebral artery and its branch. 3. Posterior cerebral artery. 4. Branches of posterior cerebral artery. 5. Basilar artery and i ts branch. 6. Arterioles in distal portion of middle cerebral artery. Asterisks indicate richly innervated, medium-sized pial arteries. Slender arrowhead indicates basophil leucocyte. X 67.
and the density of nerves decreased rapidly a s the vessels diminished in diameter. Therefore, a striking unbalance between t h e adrenergic and cholinergic innervation was observed in the medium-sized arteries. In addition, t h e AChE activity of t h e nerves was remarkably weak, so t h a t i t was very difficult to demonstrate t h e AChE-positive nerves in whole mount preparations of major and mediumsized pial arteries and one was unable to visualize the nerves in small pial arteries and in cross sections of any cerebral artery. Prolongation of incubation time up to 90 minutes
or perfusion of fixative through the aorta to decrease the thickness of the vascular wall made i t possible to demonstrate the AChEpositive nerves. As the incubation time is prolonged, however, t h e preferential demonstration of cholinergic, adrenergic and sensory components might become more unreliable. For t h e feature of innervation in whole mount preparation, therefore, the descriptions were based exclusively upon the data in perfused animals. Although the activity of AChE was considerably weak, t h e finest fibres (about 1 F m
310
TADAHIKO IIJIMA
Figs. 7-11 Adrenergic innervation in paraffin sections of cerebral vessels. 7 . Richly innervated, mediumsized pial artery of thick vascular wall (arrowhead) and major cerebral artery (asterisk).8. Penetrating arteries into the brain parenchyma iarrowheads). 9. Richly innervated, medium-sized pial artery of thick vascular wall (arrowhead). 10. Parenchymal arterioles (arrowheads). 1l a . Parenchymal capillaries (arrowheads). l l b . dark^ field photomicrograph of l l a . Slender arrowheads indicate basophil leucocytes. 7-9 X 67. 10 X 135. 1 1 X 270.
thick) forming a meshwork were clearly demonstrated in the major arteries of both internal carotid (figs. 12, 13) and vertebro-basilar system (fig. 14). Arteries of both systems had plexuses of AChE-positive fibres with approximately the same density. A few thick nerve bundles (5-10 l m thick) which consisted of AChE-positive and -negative fibres were observed in t h e major arteries of both internal carotid (arrows in figs. 12, 13) and vertebrobasilar systems. On rare occasions, the wellstained fibres a s seen in mammalian cerebral arteries were observed, and these fibres were continuous with the faintly-stained ones (small arrows in fig. 13). Unlike those in pial vessels, the heavilystained AChE-positive fibres were observed in the parenchymal capillaries (fig. 15). These fibres passed from t h e vascular wall into t h e depth of brain parenchyma. As shown in figure 15, three sorts of AChE-positive fibres of which t h e diameter was 1 p m (arrow 11, 2 p m (arrow 2) and 4 p m (arrow 3) could be distinguished. The walls of parenchymal arterioles and capillaries also exhibited heavy AChE activity (fig. 15).
Basophil leucocytes The blood of t h e turtle contains abundant
basophil leucocytes. They emitted a n extensive greenish yellow fluorescence after the treatment with formaldehyde gas (slender arrowheads in figs. 2, 7-9, 11). But they did not exhibit such a fluorescence without formaldehyde treatment. DISCUSSION
I t has been demonstrated repeatedly t h a t mammalian cerebral arteries have dual adrenergic and cholinergic innervation as in t h e other blood vessels (Schenk and Badawi, '68), and in the cerebral arteries of mammals t h e cholinergic nerve plexuses showed approximately the same density as the adrenergic ones (Edvinsson, '75). Even in the cerebral arteries of t h e snake, the essential features of innervation were similar to those of mammals, though the adrenergic plexuses were more prominent than t h e cholinergic ones (Iijima et al., ' 7 7 ) . However, a remarkable unbalance of innervation was observed between adrenergic and cholinergic nerves in t h e cerebral arteries of t h e turtle; t h a t is, illdeveloped cholinergic plexuses with faintlystained fibres were coextensive with well-developed, brilliant fluorescent adrenergic plexuses. I t is difficult to distinguish exactly adren-
VASCULAR INNERVATION IN THE TUKI'1.E BRAIN
311
Figs. 12.14 AChE reaction in whole mount preparations 12, 13. Kamus anterior of cerebral artery. 14. bas^^ lar artery. Arrows indicate the thick nerve bundles. Small arrows indicate well- and I'aintly stained AChE- psi^ tive nerves which are continuous with each other. X 40. Fig. 15 AChE reaction in the parenchymal capillaries. Arrows 11-31 i n d ~ c a t et h e well-stained AChE~positive fibres. c. capillary x 160
ergic and sensory neurons from cholinergic ones solely by AChE activity. However, it is generally believed t h a t high concentrations of AChE are located in the cholinergic neurons and low or moderate levels of the enzyme activity are associated with the adrenergic and sensory ones (Koelle, ' 5 5 ; Ehinger, '66; Burnstock and Robinson, '67; Owman e t al., '74; Edvinsson, '75). It appears to the author t h a t an incubation of 60 minutes a t 20°C is a suitable period to demonstrate t h e cholinergic nerves, since the cholinergic fibres in cerebral arteries in mammals and snake are stained excellently at t h a t incubation period. In the turtle, however, nerves were stained very weakly at the same period. Besides these faintly-stained nerves, infrequent wellstained ones were also observed in t h e cerebral arteries of t h e turtle. If one considers only the well-stained nerves to be t h e cholinergic and the faintly-stained ones to be the adrenergic or sensory nerves, the present data mean t h a t the cerebral arteries of the turtle are still more scarcely innervated with the cholinergic than t h e adrenergic nerves. But
the author is inclined to think t h a t both the well- and faintly-stained nerves to be the cholinergic. This is based on the following observations: The well-stained fibres were continuous with the faintly-stained ones, and also in mammals, t h e faintly-stained fibres connecting with t h e well-stained cholinergic nerves in the major cerebral arteries were observed in the small branches of pial arteries (unpublished data). In t h e turtle thus it seems t h a t very scarce AChE-positive axones are present in the cholinergic fibres so t h a t the fibres are stained very weakly. In extracranial vascular beds, i t is well known t h a t smaller arterioles are more sensitive than larger ones to a variety of stimuli. The medium-sized pial arteries of thick musculature in the turtle had the most dense adrenergic innervation. This indicates t h e functional significance of these arteries in t h e cerebral circulation of the turtle. It has been reported t h a t t h e cerebrovascular nerves might enter the skull along t h e internal carotid and the vertebral arteries in mammals (cf. Edvinsson, '75) and in the
312
TADAHIKO IIJIMA
snake mainly along the cerebral carotid and the ophthalmic arteries (Iijima e t al., '77). In the turtle the thick nerve bundles which consisted of AChE-positive and -negative fibres were observed in the major arteries of both internal carotid and vertebro-basilar systems. Therefore, it seems that the cerebrovascular nerves in the turtle enter the skull along the internal carotid and the spinal (vertebral) arteries. The possible existence of influence on parenchymal micro-circulation by central adrenergic (non-sympathetic) neurons which originate directly from the parenchymal cell bodies (e.g., the locus coeruleus) was suggested in mammals (Falck et al., '65, '68; Hartman et al., '72; Edvinsson et al., '73a; Raichle e t al., '75; Rennels and Nelson, '75; Rosendorff et al., '76) and by both central adrenergic and cholinergic (non-parasympathetic) neurons in the snake (Iijima et al., '77). Many fluorescent varicosities were observed on the wall of the parenchymal capillaries of the turtle. In order for the peripheral adrenergic nerve fibres to reach the capillary wall, it would be necessary for the fibres to pass from the perivascular space into the brain by crossing the limiting membrane of the brain. Thus, although the adrenergic nerves which passed from the capillary wall into the depth of brain parenchyma could not be detected in the present research, it is likely that the adrenergic fibres associated with capillaries are of central origin. On the other hand, AChE-positive fibres which passed from the wall of parenchymal capillaries into the depth of brain parenchyma were frequently observed. As described above, pial arteries of the turtle had a much scarcer cholinergic innervation, and, in addition, the activity of AChE in the cholinergic fibres was very weak. Therefore, it is more reasonable to consider that the AChE-positive fibres associated with the vascular wall in the brain parenchyma are of central origin rather than to consider that they are of peripheral origin. It seems that the existence of these AChE-positive fibres and of the AChE exhibited on the wall of parenchymal small vessels may well cover a scarcity of the cholinergic innervation in the extraparenchymal vessels. I t is well known that mammalian mast cells contain heparin, histamine, serotonin and dopamine (Sagher and Even-Paz, '67; Olsson, '68) and histamine in particular is said to be
active in the regulation of the cerebral circulation (Rosenblum, '73; Bevan et al., '75). The blood of the turtle contains a considerable amount of histamine and distinct yellow fluorescence can be provoked by the o-phthalaldehyde method on the basophil leucocytes (blood mast cells) which are abundant in the blood (Takaya, '69). When studied with the method of Falck and Hillarp, however, histamine cannot emit a fluorescence sufficiently strong to interfere with t h e localization of tissue monoamines (Falck, '62), and a much longer exposure period to formaldehyde gas is required to produce a compound with intensive fluorescence (Van Orden, '70; Rost and Even, '71). The basophil leucocytes emitted an intensive greenish yellow fluorescence after exposure to formaldehyde gas for one hour a t 80°C. Excitation and emission spectra of formaldehyde gas-induced fluorescence in the basophil leucocytes of the turtle (the excitation and emission maxima a t 4701495 nm, unpublished data), however, do not coincide with those of histamine, serotonin and dopamine. ACKNOWLEDGMENTS
The author is very grateful to T. Wasano, Professor Emeritus of Kyushu University, Professor of Fukuoka University, for his helpful comments during the preparation of this manuscript. LITERATURE CITED Bevan, J. A., S. P. Duckles and T. J-F. Lee 1975 Histamine potentiation of nerve- and drug-induced responses of a rabbit cerebral artery. Circ. Res., 36: 647-653. Burnstock, G. 1969 Evolution of the autonomic innervation of visceral and cardiovascular systems in vertebrates. Pharmacol. Rev., 21: 247-324. Burnstock, G., and P. M. Robinson 1967 Localization of catecholamines and acetylcholinesterase in autonomic nerves. Circ. Res. (Suppl. 31, 21: 43-55. Edvinsson, L. 1975 Neurogenic mechanisms in the cerebrovascular bed. Autonomic nerves, amine receptors and their effects on cerebral blood flow. Acta Physiol. Scand. (Suppl.), 427: 1-35. Edvinsson, L., M. Lindvall, K. C. Nielsen and Ch. Owman 1973a Are brain vessels innervated also by central (nonsympathetic) adrenergic neurones? Brain Res., 63: 496-499. Edvinsson, L., K. C. Nielsen, Ch. Owman and K. A. West 1973b Evidence of vasoconstrictor sympathetic nerves in brain vessels of mice. Neurology, 23: 73-77. Ehinger, B. 1966 Cholinesterases in ocular and orbital tissues of some mammals. Acta Univ. Lund., 11, 2: 1-15, Falck, B. 1962 Observations on the possibilities of the cellular localization of monoamines by a fluorescence method. Acta Physiol. Scand. (Suppl. 1971, 56: 1-25. Falck, B., G. I. Mchedlishvili and Ch. Owman 1965 Histochemical demonstration of adrenergic nerves in cortexpia of rabbit. Acta Pharmacol. Toxicol., 23: 133-142.
VASCULAR INNERVATION IN THE TURTLE BRAIN Falck, B., K. C. Nielsen and Ch. Owman 1968 Adrenergic innervation of the pial circulation. Scand. J. Clin. Lab. Invest. (Suppl.), 102: 96-98. Hartman, B. K., D. Zide and S. Udenfriend 1972 The use of dopamine /3-hydroxylase a s a marker for the central noradrenergic nervous system in rat brain. Proc. Natl. Acad. Sci. (U.S.A.), 69: 2722-2726. Iijima, T., T. Wasano, T. Tagawa and K. Ando 1977 A histochemical study of the innervation of cerebral blood vessels in t h e snake. Cell Tissue Res., 179: 143-155. Karnovsky, M. J., and L. Roots 1964 A “direct-coloring’’ thiocholine method for cholinesterases. J. Histochern. Cytochem., 12: 219-221. Koelle, G. B. 1955 The histochemical identification of acetylcholinesterase in cholinergic, adrenergic and sensory neurons. J. Pharmacol. Exp. Ther., 114: 167-184. Mchedlishvili, G. I., and L. S. Nikolaishvili 1970 Evidence of a cholinergic nervous mechanism mediating t h e autoregulatory dilation of the cerebral blood vessels. Pfluegers Arch., 315: 27-37. Motavkin, P. A., and T. V. Dovbish 1971 Histochemical characteristics of acetylcholinesterase of the nerves innervating the brain vessels. Acta Morphol. Acad. Sci. Hung., 19: 159-173. Nelson, E., and M. Rennels 1970 Innervation of intracranial arteries. Brain, 93: 475-490. Nielsen, K. C., and Ch. Owman 1971 Contractile response and amine receptor mechanisms in isolated middle cerebral artery of the cat. Brain Res., 27: 33-42. Olsson, Y. 1968 Mast cells in t h e nervous system. Int. Rev. Cytol., 24: 27-70. Owman, Ch., L. Edvinsson and K. C. Nielsen 1974
313
Autonomic neuroreceptor mechanisms in brain vessels. Blood Vessels, 11: 2-31. Raichle, M. E., B. K. Hartrnan, J. 0. Eichling and L. G. Sharpe 1975 Central noradrenergic regulation of cerebral blood flow and vascular permeability. Proc. Natl. Acad. Sci. (U.S.A.),72: 3726-3730. Rennels, M. L., and E. Nelson 1975 Capillary innervation in the mammalian central nervous system: An electron microscopic demonstration (1). Am. J. Anat., 144: 233-241. Rosenblum, W. 1. 1971 Neurogenic control of cerebral circulation. Stroke, 2: 429-439. Rosendorff, C., G. Mitchell, D. R. L. Scriven and C. Shapiro 1976 Evidence for a dual innervation affecting local blood flow in the hypothalamus of the conscious rabbit. Circ. Res., 38: 140-145. Rost, F. W. O., and S. W. I. Even 1971 New methods for the histochemical demonstration of catecholamines, tryptamines, histamine and other arylethylamine by acidand aldehyde-induced fluorescence. Histochem. J., 3: 207-212. Sagher, F., and 2. Even-Paz 1967 Mastocytosis and the mast cell. S. Karger AG, Basal. Schenk, E. A,, and A. E. Badawi 1968 Dual innervation of arteries and arterioles. A histochemical study Z. Zellforsch., 91: 170-177. Takaya, K. 1969 The relationship between mast cells and histamine in phylogeny with special reference to reptiles and birds. Arch. Histol. Jpn., 30: 401-420. Van Orden 111, L. S. 1970 Quantitative histochemistry of biogenic amines. A simple microspectrofluorometer. Biochem. Pharmacol., 19: 1105-1117.
of
Medicine, Fukuoka University, Fukuoka 814, Japan
ABSTRACT Specific histochemical techniques for the demonstration of noradrenaline and of acetylcholinesterase have been used to study t h e distribution of adrenergic and cholinergic nerves to t h e cerebral blood vessels of turtle, Geoclemys reeuesii. The major and medium-sized cerebral arteries were supplied with dense adrenergic nerve plexuses, the plexuses were particularly dense in t h e mediumsized pial arteries of very thick vascular wall, indicating the functional significance of these arteries in the cerebral circulation. The parenchymal arterioles and capillaries were also supplied with adrenergic nerves. On the other hand, t h e cholinergic innervation was less dense than the adrenergic one and the acetylcholinesterase activity of the nerve fibres was remarkably weak. By contrast, the parenchymal small arterioles and capillaries exhibited heavy acetylcholinesterase activity on the vascular wall and, in addition, the capillaries were supplied with t h e well-stained acetylcholinesterasepositive nerve fibres. These fibres and also t h e adrenergic fibres associating with the capillaries appear t o be of central origin. It is suggested t h a t the cholinergic mechanisms in t h e parenchymal small vessels also play a n important role in the cerebral circulation. The basophil leucocytes observed abundantly in t h e blood of turtle emitted a n intensive greenish yellow fluorescence after formaldehyde gas-treatment. The existence of adrenergic and cholinergic nerve terminals in t h e mammalian cerebral blood vessels has been thoroughly demonstrated by histochemical and electron microscopic studies (e.g., Nelson and Rennels, '70; Motavkin and Dovbish, '71; Rosenblum, '71; Owman e t al., '74) and t h e neurogenic influence on the cerebral circulation has also been documented by recent pharmacological studies (e.g., Mchedlishvili and Nikolaishvili, '70; Nielson and Owman, '71; Edvinsson e t al., '73b; Raichle e t al., '75). There are great many studies concerning t h e vertebrate autonomic nervous mechanisms in visceral and cardiovascular systems. I n his review Burnstock ('69)has described t h e mode of innervation in these systems with evolutionary terms. Comparative studies of innervation in t h e cerebral blood vessels, however, are very fragmentary. Recently, Iijima e t al. ('77) reported t h a t both adrenergic and cholinergic nerve plexuses in t h e cerebral J. COMP. NEUR.. 176: 307-314.
blood vessels of the snake show almost similar distribution as those in mammals. As a part of research project on t h e evolution of nervous system innervating t h e cerebral blood vessels of vertebrates, the turtle which is classified in a much lower category than t h e snake in the phylogeny of reptile has been chosen as a n experimental animal. The feature of adrenergic and cholinergic innervation has been investigated by fluorescence and cholinesterase histochemistry. I n t h e present paper, a disproportional development of both nerve plexuses; well-developed, brilliant adrenergic and ill-developed, faintly-stained cholinergic nerve meshwork, will be described. MATERIALS AND METHODS
Twelve turtles, Geoclemys reeuesii, were decapitated, and the brain was rapidly removed from t h e skull and placed in Ringer's solution. The vessels were carefully dissected from t h e brain, and were stretched on t h e glass slides
307
308
TADAHIKO I I J I M A
for fluorescence histochemistry (Falck, '62) or were fixed with 4% formaldehyde buffered to pH 7.4 for 30 minutes for cholinesterase histochemistry (Karnovsky and Roots, '64). For cholinesterase histochemistry with whole mount preparation, the cerebral vessels from eight animals were also fixed by perfusion of 44, formaldehyde through the aorta. To demonstrate the innervation in the parenchymal vessels, small blocks of brain from nine animals were frozen in dry-ice isopentane, and thin sections were prepared with cryostat or freeze-drying techniques. For fluorescence histochemistry, the slidemounted vessels were immediately transferred to a dessicator where they were dried in uucuo over phosphorus pentoxide for one hour. The slides were then treated in formaldehyde vapor from paraformaldehyde for one hour a t 8O"C, and were mounted for fluorescence microscopy. On the other hand, frozen blockes of the brain were dried in uucuo for six days a t - 4 5 ° C two days a t -20°C and one day at room temperature, then treated with formaldehyde gas for one hour at 80°C, and were infiltrated in uucuo with paraffin for ten minutes a t 60"C, and the sections (8-10 p m thick) were examined under the fluorescence microscope. For cholinesterase histochemistry, formalin-fixed vessels were washed thoroughly with distilled water, and maintained in substrate (acetylthiocholine iodide)-free Karnovsky's medium for 30 minutes a t 4"C, and then incubated in the complete medium, conM iso-OMPA (tetraisopropyltaining 2 x pyrophosphoramide) as a n inhibitor of nonspecific cholinesterases, for one hour a t 20°C. After washing with 50!% ethanol. they were stretched on microscope glass and dried in air, and were dehydrated with ethanol series followed with xylene and mounted in balsam. On the other hand, cryostat sections (15-30 p m thick) cut from frozen specimens of the brain were mounted on microscope glass and were fixed with 4% formaldehyde and then treated as described above. RESULTS
The internal carotid and the spinal (vertebral) arteries contributed to the blood supply of the turtle brain, and the branches of these arteries formed the circle of Willis on the ventral surface of t h e brain. The arteries leaving the circle showed a characteristic tortuosity in their course. A further characteris-
tic feature of the cerebral blood vessels in the turtle was the thickness of vascular wall. Although a n approximate diameter of major arteries was 110 p m , t h a t of pial arteries was 60 p m and t h a t of parenchymal arterioles and capillaries was 15 Fm, t h e thickness of vascular walls of both major and medium-sized pial arteries was 10-15 p m and t h a t of parenchymal arterioles and capillaries was 2-5 p m .
Adrenergic innervation A very dense network of fluorescent nerves occurred in t h e cerebral arteries of the turtle. The fluorescent fibres inclined to run longitudinally along the arterial axis, forming a longitudinally elongated meshwork. The nerve plexuses were particularly dense in the medium-sized pial arteries of 40-70 p m in diameter (asterisks in figs. 1,2, 5; fig. 4). The innervation in t h e major arteries, however, was somewhat less dense (figs. 1-3, 5 ) . I n cross sections, i t was clearly shown t h a t the tunica media of richly innervated, medium-sized pial arteries consisted of thick muscular layers (arrowheads in figs. 7 , 9 ) and t h a t those of less richly innervated, major cerebral arteries consisted of relatively thin musculature in contrast to a diameter (asterisk in fig. 7). The medium-sized pial arteries indicated with arrowheads in figures 7 and 9 were 40 and 60 p m in diameter, respectively. And the thickness of the vascular wall of both arteries was about 15 p m . On t h e other hand, the diameter of the major artery indicated by t h e asterisk in figure 7 was 120 p m and t h e wall thickness was about 15 pm. No distinct difference was observed between the density of innervation in the arteries of t h e internal carotid and that of the vertebro-basilar system. Small arteries and arterioles were sparsely innervated (fig. 6 ) , and some penetrating arteries and arterioles into the brain parenchyma were also supplied with adrenergic nerves (arrowheads in figs. 8, 10). Furthermore, the adrenergic nerves were observed on the wall of parenchymal capillaries (arrowheads in fig. 1l a ) . However, this study had failed to clarify the existence of t h e nerve fibre which passed from t h e capillary wall into the depth of the brain parenchyma. Cholinergc innervation T h e p l e x u s e s of a c e t y l c h o l i n e s t e r a s e (AChE)-positive nerve fibres were less dense t h a n t h e adrenergic ones. These plexuses were t h e most dense in the major cerebral arteries,
VASCULAR INNERVATION IN THE TURTLE BRAIN
309
Figs. 1-6 Adrenergic innervation in whole mount preparations. 1. Middle cerebral artery and its branch. 2. Anterior cerebral artery and its branch. 3. Posterior cerebral artery. 4. Branches of posterior cerebral artery. 5. Basilar artery and i ts branch. 6. Arterioles in distal portion of middle cerebral artery. Asterisks indicate richly innervated, medium-sized pial arteries. Slender arrowhead indicates basophil leucocyte. X 67.
and the density of nerves decreased rapidly a s the vessels diminished in diameter. Therefore, a striking unbalance between t h e adrenergic and cholinergic innervation was observed in the medium-sized arteries. In addition, t h e AChE activity of t h e nerves was remarkably weak, so t h a t i t was very difficult to demonstrate t h e AChE-positive nerves in whole mount preparations of major and mediumsized pial arteries and one was unable to visualize the nerves in small pial arteries and in cross sections of any cerebral artery. Prolongation of incubation time up to 90 minutes
or perfusion of fixative through the aorta to decrease the thickness of the vascular wall made i t possible to demonstrate the AChEpositive nerves. As the incubation time is prolonged, however, t h e preferential demonstration of cholinergic, adrenergic and sensory components might become more unreliable. For t h e feature of innervation in whole mount preparation, therefore, the descriptions were based exclusively upon the data in perfused animals. Although the activity of AChE was considerably weak, t h e finest fibres (about 1 F m
310
TADAHIKO IIJIMA
Figs. 7-11 Adrenergic innervation in paraffin sections of cerebral vessels. 7 . Richly innervated, mediumsized pial artery of thick vascular wall (arrowhead) and major cerebral artery (asterisk).8. Penetrating arteries into the brain parenchyma iarrowheads). 9. Richly innervated, medium-sized pial artery of thick vascular wall (arrowhead). 10. Parenchymal arterioles (arrowheads). 1l a . Parenchymal capillaries (arrowheads). l l b . dark^ field photomicrograph of l l a . Slender arrowheads indicate basophil leucocytes. 7-9 X 67. 10 X 135. 1 1 X 270.
thick) forming a meshwork were clearly demonstrated in the major arteries of both internal carotid (figs. 12, 13) and vertebro-basilar system (fig. 14). Arteries of both systems had plexuses of AChE-positive fibres with approximately the same density. A few thick nerve bundles (5-10 l m thick) which consisted of AChE-positive and -negative fibres were observed in t h e major arteries of both internal carotid (arrows in figs. 12, 13) and vertebrobasilar systems. On rare occasions, the wellstained fibres a s seen in mammalian cerebral arteries were observed, and these fibres were continuous with the faintly-stained ones (small arrows in fig. 13). Unlike those in pial vessels, the heavilystained AChE-positive fibres were observed in the parenchymal capillaries (fig. 15). These fibres passed from t h e vascular wall into t h e depth of brain parenchyma. As shown in figure 15, three sorts of AChE-positive fibres of which t h e diameter was 1 p m (arrow 11, 2 p m (arrow 2) and 4 p m (arrow 3) could be distinguished. The walls of parenchymal arterioles and capillaries also exhibited heavy AChE activity (fig. 15).
Basophil leucocytes The blood of t h e turtle contains abundant
basophil leucocytes. They emitted a n extensive greenish yellow fluorescence after the treatment with formaldehyde gas (slender arrowheads in figs. 2, 7-9, 11). But they did not exhibit such a fluorescence without formaldehyde treatment. DISCUSSION
I t has been demonstrated repeatedly t h a t mammalian cerebral arteries have dual adrenergic and cholinergic innervation as in t h e other blood vessels (Schenk and Badawi, '68), and in the cerebral arteries of mammals t h e cholinergic nerve plexuses showed approximately the same density as the adrenergic ones (Edvinsson, '75). Even in the cerebral arteries of t h e snake, the essential features of innervation were similar to those of mammals, though the adrenergic plexuses were more prominent than t h e cholinergic ones (Iijima et al., ' 7 7 ) . However, a remarkable unbalance of innervation was observed between adrenergic and cholinergic nerves in t h e cerebral arteries of t h e turtle; t h a t is, illdeveloped cholinergic plexuses with faintlystained fibres were coextensive with well-developed, brilliant fluorescent adrenergic plexuses. I t is difficult to distinguish exactly adren-
VASCULAR INNERVATION IN THE TUKI'1.E BRAIN
311
Figs. 12.14 AChE reaction in whole mount preparations 12, 13. Kamus anterior of cerebral artery. 14. bas^^ lar artery. Arrows indicate the thick nerve bundles. Small arrows indicate well- and I'aintly stained AChE- psi^ tive nerves which are continuous with each other. X 40. Fig. 15 AChE reaction in the parenchymal capillaries. Arrows 11-31 i n d ~ c a t et h e well-stained AChE~positive fibres. c. capillary x 160
ergic and sensory neurons from cholinergic ones solely by AChE activity. However, it is generally believed t h a t high concentrations of AChE are located in the cholinergic neurons and low or moderate levels of the enzyme activity are associated with the adrenergic and sensory ones (Koelle, ' 5 5 ; Ehinger, '66; Burnstock and Robinson, '67; Owman e t al., '74; Edvinsson, '75). It appears to the author t h a t an incubation of 60 minutes a t 20°C is a suitable period to demonstrate t h e cholinergic nerves, since the cholinergic fibres in cerebral arteries in mammals and snake are stained excellently at t h a t incubation period. In the turtle, however, nerves were stained very weakly at the same period. Besides these faintly-stained nerves, infrequent wellstained ones were also observed in t h e cerebral arteries of t h e turtle. If one considers only the well-stained nerves to be t h e cholinergic and the faintly-stained ones to be the adrenergic or sensory nerves, the present data mean t h a t the cerebral arteries of the turtle are still more scarcely innervated with the cholinergic than t h e adrenergic nerves. But
the author is inclined to think t h a t both the well- and faintly-stained nerves to be the cholinergic. This is based on the following observations: The well-stained fibres were continuous with the faintly-stained ones, and also in mammals, t h e faintly-stained fibres connecting with t h e well-stained cholinergic nerves in the major cerebral arteries were observed in the small branches of pial arteries (unpublished data). In t h e turtle thus it seems t h a t very scarce AChE-positive axones are present in the cholinergic fibres so t h a t the fibres are stained very weakly. In extracranial vascular beds, i t is well known t h a t smaller arterioles are more sensitive than larger ones to a variety of stimuli. The medium-sized pial arteries of thick musculature in the turtle had the most dense adrenergic innervation. This indicates t h e functional significance of these arteries in t h e cerebral circulation of the turtle. It has been reported t h a t t h e cerebrovascular nerves might enter the skull along t h e internal carotid and the vertebral arteries in mammals (cf. Edvinsson, '75) and in the
312
TADAHIKO IIJIMA
snake mainly along the cerebral carotid and the ophthalmic arteries (Iijima e t al., '77). In the turtle the thick nerve bundles which consisted of AChE-positive and -negative fibres were observed in the major arteries of both internal carotid and vertebro-basilar systems. Therefore, it seems that the cerebrovascular nerves in the turtle enter the skull along the internal carotid and the spinal (vertebral) arteries. The possible existence of influence on parenchymal micro-circulation by central adrenergic (non-sympathetic) neurons which originate directly from the parenchymal cell bodies (e.g., the locus coeruleus) was suggested in mammals (Falck et al., '65, '68; Hartman et al., '72; Edvinsson et al., '73a; Raichle e t al., '75; Rennels and Nelson, '75; Rosendorff et al., '76) and by both central adrenergic and cholinergic (non-parasympathetic) neurons in the snake (Iijima et al., '77). Many fluorescent varicosities were observed on the wall of the parenchymal capillaries of the turtle. In order for the peripheral adrenergic nerve fibres to reach the capillary wall, it would be necessary for the fibres to pass from the perivascular space into the brain by crossing the limiting membrane of the brain. Thus, although the adrenergic nerves which passed from the capillary wall into the depth of brain parenchyma could not be detected in the present research, it is likely that the adrenergic fibres associated with capillaries are of central origin. On the other hand, AChE-positive fibres which passed from the wall of parenchymal capillaries into the depth of brain parenchyma were frequently observed. As described above, pial arteries of the turtle had a much scarcer cholinergic innervation, and, in addition, the activity of AChE in the cholinergic fibres was very weak. Therefore, it is more reasonable to consider that the AChE-positive fibres associated with the vascular wall in the brain parenchyma are of central origin rather than to consider that they are of peripheral origin. It seems that the existence of these AChE-positive fibres and of the AChE exhibited on the wall of parenchymal small vessels may well cover a scarcity of the cholinergic innervation in the extraparenchymal vessels. I t is well known that mammalian mast cells contain heparin, histamine, serotonin and dopamine (Sagher and Even-Paz, '67; Olsson, '68) and histamine in particular is said to be
active in the regulation of the cerebral circulation (Rosenblum, '73; Bevan et al., '75). The blood of the turtle contains a considerable amount of histamine and distinct yellow fluorescence can be provoked by the o-phthalaldehyde method on the basophil leucocytes (blood mast cells) which are abundant in the blood (Takaya, '69). When studied with the method of Falck and Hillarp, however, histamine cannot emit a fluorescence sufficiently strong to interfere with t h e localization of tissue monoamines (Falck, '62), and a much longer exposure period to formaldehyde gas is required to produce a compound with intensive fluorescence (Van Orden, '70; Rost and Even, '71). The basophil leucocytes emitted an intensive greenish yellow fluorescence after exposure to formaldehyde gas for one hour a t 80°C. Excitation and emission spectra of formaldehyde gas-induced fluorescence in the basophil leucocytes of the turtle (the excitation and emission maxima a t 4701495 nm, unpublished data), however, do not coincide with those of histamine, serotonin and dopamine. ACKNOWLEDGMENTS
The author is very grateful to T. Wasano, Professor Emeritus of Kyushu University, Professor of Fukuoka University, for his helpful comments during the preparation of this manuscript. LITERATURE CITED Bevan, J. A., S. P. Duckles and T. J-F. Lee 1975 Histamine potentiation of nerve- and drug-induced responses of a rabbit cerebral artery. Circ. Res., 36: 647-653. Burnstock, G. 1969 Evolution of the autonomic innervation of visceral and cardiovascular systems in vertebrates. Pharmacol. Rev., 21: 247-324. Burnstock, G., and P. M. Robinson 1967 Localization of catecholamines and acetylcholinesterase in autonomic nerves. Circ. Res. (Suppl. 31, 21: 43-55. Edvinsson, L. 1975 Neurogenic mechanisms in the cerebrovascular bed. Autonomic nerves, amine receptors and their effects on cerebral blood flow. Acta Physiol. Scand. (Suppl.), 427: 1-35. Edvinsson, L., M. Lindvall, K. C. Nielsen and Ch. Owman 1973a Are brain vessels innervated also by central (nonsympathetic) adrenergic neurones? Brain Res., 63: 496-499. Edvinsson, L., K. C. Nielsen, Ch. Owman and K. A. West 1973b Evidence of vasoconstrictor sympathetic nerves in brain vessels of mice. Neurology, 23: 73-77. Ehinger, B. 1966 Cholinesterases in ocular and orbital tissues of some mammals. Acta Univ. Lund., 11, 2: 1-15, Falck, B. 1962 Observations on the possibilities of the cellular localization of monoamines by a fluorescence method. Acta Physiol. Scand. (Suppl. 1971, 56: 1-25. Falck, B., G. I. Mchedlishvili and Ch. Owman 1965 Histochemical demonstration of adrenergic nerves in cortexpia of rabbit. Acta Pharmacol. Toxicol., 23: 133-142.
VASCULAR INNERVATION IN THE TURTLE BRAIN Falck, B., K. C. Nielsen and Ch. Owman 1968 Adrenergic innervation of the pial circulation. Scand. J. Clin. Lab. Invest. (Suppl.), 102: 96-98. Hartman, B. K., D. Zide and S. Udenfriend 1972 The use of dopamine /3-hydroxylase a s a marker for the central noradrenergic nervous system in rat brain. Proc. Natl. Acad. Sci. (U.S.A.), 69: 2722-2726. Iijima, T., T. Wasano, T. Tagawa and K. Ando 1977 A histochemical study of the innervation of cerebral blood vessels in t h e snake. Cell Tissue Res., 179: 143-155. Karnovsky, M. J., and L. Roots 1964 A “direct-coloring’’ thiocholine method for cholinesterases. J. Histochern. Cytochem., 12: 219-221. Koelle, G. B. 1955 The histochemical identification of acetylcholinesterase in cholinergic, adrenergic and sensory neurons. J. Pharmacol. Exp. Ther., 114: 167-184. Mchedlishvili, G. I., and L. S. Nikolaishvili 1970 Evidence of a cholinergic nervous mechanism mediating t h e autoregulatory dilation of the cerebral blood vessels. Pfluegers Arch., 315: 27-37. Motavkin, P. A., and T. V. Dovbish 1971 Histochemical characteristics of acetylcholinesterase of the nerves innervating the brain vessels. Acta Morphol. Acad. Sci. Hung., 19: 159-173. Nelson, E., and M. Rennels 1970 Innervation of intracranial arteries. Brain, 93: 475-490. Nielsen, K. C., and Ch. Owman 1971 Contractile response and amine receptor mechanisms in isolated middle cerebral artery of the cat. Brain Res., 27: 33-42. Olsson, Y. 1968 Mast cells in t h e nervous system. Int. Rev. Cytol., 24: 27-70. Owman, Ch., L. Edvinsson and K. C. Nielsen 1974
313
Autonomic neuroreceptor mechanisms in brain vessels. Blood Vessels, 11: 2-31. Raichle, M. E., B. K. Hartrnan, J. 0. Eichling and L. G. Sharpe 1975 Central noradrenergic regulation of cerebral blood flow and vascular permeability. Proc. Natl. Acad. Sci. (U.S.A.),72: 3726-3730. Rennels, M. L., and E. Nelson 1975 Capillary innervation in the mammalian central nervous system: An electron microscopic demonstration (1). Am. J. Anat., 144: 233-241. Rosenblum, W. 1. 1971 Neurogenic control of cerebral circulation. Stroke, 2: 429-439. Rosendorff, C., G. Mitchell, D. R. L. Scriven and C. Shapiro 1976 Evidence for a dual innervation affecting local blood flow in the hypothalamus of the conscious rabbit. Circ. Res., 38: 140-145. Rost, F. W. O., and S. W. I. Even 1971 New methods for the histochemical demonstration of catecholamines, tryptamines, histamine and other arylethylamine by acidand aldehyde-induced fluorescence. Histochem. J., 3: 207-212. Sagher, F., and 2. Even-Paz 1967 Mastocytosis and the mast cell. S. Karger AG, Basal. Schenk, E. A,, and A. E. Badawi 1968 Dual innervation of arteries and arterioles. A histochemical study Z. Zellforsch., 91: 170-177. Takaya, K. 1969 The relationship between mast cells and histamine in phylogeny with special reference to reptiles and birds. Arch. Histol. Jpn., 30: 401-420. Van Orden 111, L. S. 1970 Quantitative histochemistry of biogenic amines. A simple microspectrofluorometer. Biochem. Pharmacol., 19: 1105-1117.
