Cell Tissue Res (1989) 258:425428

a n d T'tssue Resealv_h 9 Springer-Verlag 1989

Innervation of the arteriovenous anastomoses in the dog tongue T. Iijima, T. Kondo, K. Nishijima, and T. Tanaka Department of Oral Anatomy, Faculty of Dentistry, Kyushu University, Fukuoka, Japan

Summary. Profiles of nerve plexuses in the arteriovenous anastomoses of the dog tongue were investigated by both transmission and scanning electron microscopy. Three-dimensional morphology of the vascular nerves was examined after removal of the connective tissue components by the HCl-hydrolysis method. Tight bending and a rich nerve supply were the most characteristic features of the anastomosing channels. The tunica media consisted of an outer circular layer of typical smooth-muscle cells and an inner region containing longitudinal plicae of ramified smoothmuscle cells. The tunica adventitia was exclusively occupied by nerve bundles; fibroblasts were poorly developed. Numerous nerve bundles of variable size were coiled around the anastomosing channels, and occasional bundles ran crosswise over the U-shaped bent vessels. Key words: Arteriovenous anastomoses - Vascular innervation - Tongue - Transmission electron microscopy - Scanning electron microscopy Dog

There are numerous arteilovenous anastomoses (AVAs) in the tunica propria mucosae of the dog tongue (Brown 1937; Pilchard and Daniel 1953; Kishi et al. 1988). The existence of a neural control of AVAs, which regulates the lingual blood flow, has been suggested by the data from physiological studies (Pleschka et al. 1979; Thomson and Pleschka 1980), and Brown (1937) and Iijima et al. (1987) have revealed the presence of a rich nerve supply to AVAs, by light microscopy and transmission electron microscopy (TEM), respectively. Since the introduction of the HCl-collagenese method to selectively remove connective tissue components (Evan et al. 1976), the three-dimensional organization of the vascular wall has been studied by many investigators using this technique in combination with scanning electron microscopy (SEM). Literature concerning the vascular nerves, however, is sparse (Uehara et al. 1981; Fujiwara and Uehara 1984; Ushiki and Ide 1988). Nerves innervating the blood vessels run in the tunica adventitia and form terminal varicosities in the media-adventitia border. These terminals, however, do not constitute any special junctional complexes with muscle cells (Burnstock 1975). Thus, it is very difficult Send offprint requests to: Dr. T. Iijima, Department of Oral Anatomy, Faculty of Dentistry, Kyushu University 61, Fukuoka 812, Japan

for the complete network of vascular nerves to remain intact after removal of the connective tissue components by the HCl-hydrolysis method. In the present study, we have used a modification of Evan's method to investigate the characteristic features of the nerve plexuses in dog-tongue AVAs after HCl-hydrolysis.

Materials and methods Three male mongrel dogs, each weighing 9-12 kg, were killed under anesthesia with Ketalar 50 (ketamine hydrochloride, 57.6 mg/ml) at a dosage of 1.0 ml/kg. Small pieces of the tongue, including the dorsal mucous membrane and the superficial layer of the intrinsic muscles, were removed. For TEM, the tissue pieces were fixed with 3% glutaraldehyde in 0.1 M sodium cacodylate buffer (pH 7.2) for 2 h at 4 ~ C, and then post-fixed with 2% OsO4 in the same buffer for 1 h at 4 ~ C. After dehydration with ethanol, they were embedded in Epon 812. Thin sections were stained with uranyl acetate and lead nitrate. For SEM, pieces of tongue tissue were first digested in 0.25% trypsin solution (Difco trypsin 1:250) in 0.1 M phosphate buffer for 1.5 h at 36 ~ C, then fixed with 3% glutaraldehyde in 0.1 M sodium cacodylate buffer for 2 h at 4 ~ C. They were next treated with 8 N HC1 for 50-90 min at 60 ~ C and agitated to remove the muscle layer and the connective tissue elements surrounding the blood vessels, and were then dehydrated in a graded series of ethanol, immersed in isoamyl acetate, critical-point dried with CO2, and sputter-coated with platinum.

Results Transmission electron microscopy

The tunica intima was composed of a single, continuous endothelium of highly irregular outline. No elastic membranes could be observed in the wall of the anastomosing channels. The tunica media was composed of an inner, longitudinal plicae of ramified smooth-muscle cells and an outer circular layer of typical smooth-muscle cells (Fig. 1), 1-3 cells thick. The structure of the tunica media was essentially the same as that reported for rabbit-ear AVAs (Iijima et al. 1988). However, in dog-tongue AVAs, the smooth-muscle cells in the outer layer usually exhibited an oblique rather than a circular arrangement, because of the tight bending

426

Fig. 1, Transmission electron micrograph showing a typical profile of a transverse section of the dog-tongue arteriovenous anastomoses completely closed. The solid line indicates the basal surface of the endothelium; and the dotted line, the media-adventitia border. Note that the tunica adventitia (A) is composed exclusively of nerve bundles (NB), and that fibroblasts (F) were poorly developed. Clusters of dense cytoplasmic processes are of the ramified smooth-muscle cells forming longitudinal plicae (1-4). Arrows indicate vascular lumen completely closed; asterisk, nerve bundle covered with a perineurium; double asterisks, smooth-muscle cell showing the characteristics of "epitheloid cell" (clear cytoplasm and clear, round nucleus). OC outer circular layer of the tunica media. Bar: 10 gm. x 2800

of the AVAs. In the anastomosing channels, where longitudinal plicae contracted to close the vascular lumen completely, cytoplasmic processes of the ramified smooth-muscle cells revealed a higher electron density than those of relaxed-muscle cells.

The tunica adventitia was very thick and compact, and was exclusively occupied by the nerve bundles of variable size (Fig. 1). Several small axons were gathered into thin bundles, but there were also thicker bundles consisting of 20-30 axons. Axons were often crowded with numerous synaptic vesicles (Iijima et al. 1987). Thick-nerve bundles which were covered by perineurium and contained several unmyelinated axons, were occasionally found in the periphery of the tunica adventitia (Fig. 1). No myelinated fibers, however, were observed r u n n i n g along the vessels. Fibroblasts, which are a predominant feature in the tunica adventitia of the rabbit-ear AVAs (Iijima et al. 1988), were poorly developed in dog-tongue AVAs.

Scanning electron microscopy After trypsin-digestion and HCl-hydrolysis, the connective tissue elements such as fibroblasts, collagen fibrils and amorphous intercellular substances were sufficiently removed to allow visualization of the vascular nerves. The

427

Figs. 2-5. Scanning electron micrographs showing the nerve plexuses in blood vessels in the dog tongue. Note thick-nerve bundles running over fibroblast coverings (F) along the axis of the artery (asterisks in Fig. 2), thin bundles penetrate into the fibroblast coverings or run just over the tunica media of the artery (arrows in Figs. 2 and 3), and numerous bundles of variable sizes coil over U-shape arteriovenous anastomoses (A VAs in Figs. 3-5). Arrows in Fig. 5 indicate nerve bundles running crosswise over U-shape anastomosing channel. Asterisk in Fig. 4 indicates a smooth-surfaced nerve bundle. A artery; V vein.

Bar. 10 ~tm. Figs. 2, 3: • 1000; Figs. 4, 5: • 1300

a m o u n t o f connective tissue elements removed was dependent on such factors as the size o f the tissue specimens, intensity o f shaking, and d u r a t i o n o f HC1 treatment. Fig. 2 shows an artery still covered with considerable a m o u n t s o f connective tissue. Thick-nerve bundles were seen to run along the vessel axis out of the fibroblast coverings (asterisks). Thin-nerve bundles ran obliquely or circularly to the vessel axis (arrows), and also were apt to penetrate deeply into the fibroblast coverings. W h e n the fibrobrast coverings o f arteries were completely removed, thinnerve bundles were observed forming coarse meshworks on the tunica media o f the artery (Fig. 3). The vessels were characterized by their tight bending and an extremely dense nerve supply in the region o f the anastomosing channels (Figs. 3-5). These nerve bundles were o f variable size and formed dense, circularly-arranged plexuses a r o u n d the anastomosing channels. Incomplete fibroblast coverings were often observed surrounding or intermingling with the nerve plexuses (Figs. 3, 4). Nerve bundles running crosswise over the U - s h a p e d A V A s were another characteristic feature o f the innervation (Fig. 5). The smooth muscle cells o f the tunica media were usually hidden under an abundance o f nerve bundles. The thick-nerve bundles o f the smooth surface (Fig. 4), were sometimes observed at the periphery o f the tunica adventitia. These bundles seemed to correspond to the nerve bundles ensheathed in perineurium, noted under T E M (Fig. 1). A l t h o u g h most veins did not survive the HCl-treatment, it seemed that they were only sparsely supplied with

nerves, an observation that is in accordance with d a t a rep o r t e d in a previous study (Iijima et al. 1987). Discussion

The present T E M d a t a demonstrate that the tunica adventitia in dog-tongue A V A s is p r e d o m i n a n t l y c o m p o s e d o f nerve bundles, and fibroblasts are p o o r l y developed. Threedimensionally, these dense nerve plexuses consist o f nerve bundles o f variable size coiled a r o u n d the AVAs. It seems that this unique construction o f the tunica adventitia facilitated the survival o f the structural features o f the nerve plexuses, after removal o f the connective tissue elements. A n o t h e r characteristic feature o f these nerve plexuses is the occurrence o f occasional nerve bundles running crosswise over the U-shape AVAs. U e h a r a et al. (1981) and F u j i w a r a and U e h a r a (1984) have demonstrated the presence o f coarse meshworks o f a u t o n o m i c nerves in the small vessels o f rat m a m m a r y glands and skeletal muscles, and have noted that the axons show some varicose swellings suggesting an existence o f effective neural control o f the small vessels. In the present, S E M failed to reveal any varicosities in the dense perivascular nerve plexuses. Physiological studies have d e m o n s t r a t e d that lingual b l o o d flow increases a b o u t 4 times with the onset o f p a n t i n g (Pleschka et al. 1979), and that the t e m p e r a t u r e - d e p e n d e n t increase in b l o o d flow through the tongue is exclusively a result o f changes in the b l o o d flow t h r o u g h A V A s (Kr6-

428 nert et al. 1980). M o s t A V A s in the d o g t o n g u e m a y thus be closed in n o r m a l conditions. T h e u n i q u e structural bending o f A V A s m a y play a role in a t t a i n i n g their efficient closure, a n d the nerves r u n n i n g crosswise o v e r these U shaped A V A s , which has been n o t e d in the present study, m a y p l a y a role in the b e n d i n g o f these vessels.

References Brown ME (1937) The occurrence of arterio-venous anastomoses in the tongue of the dog. Anat Rec 69 : 287-297 Burnstock G (1975) lnnervation of vascular smooth muscle: histochemistry and electron microscopy. Clin Exp Pharmacol Physiol [Suppl] 2: 7-20 Evan AP, Dail WG, Dammrose D, Palmer C (1976) Scanning electron microscopy of cell surfaces following removal of extracellular material. Anat Rec 185:433 446 Fujiwara T, Uehara Y (1984) The cytoarchitecture of the wall and the innervation pattern of the microvessels in the rat mammary gland: A scanning electron microscopic observation. Am J Anat 170 : 3%54 lijima T, Kondo T, Hasegawa K (1987) Autonomic innervation of the arteriovenous anastomoses in the dog tongue. A histochemical and ultrastructural study. Cell Tissue Res 247:167-177

Iijima T, Hasegawa K, Hirose H (1988) Wall structure of arteriovenous anastomoses in the rabbit ear. Combined light-, scanningand transmission electron-microscopic studies. Cell Tissue Res 252: 1-8 Kishi Y, So S, Harada Y, Takahashi K (1988) Three-dimensional SEM study of arteriovenous anastomoses in the dog's tongue using corrosive resin casts. Acta Anat 132:17-27 Kr6nert H, Wurster RD, Pierau Fr-K, Pleschka K (1980) Vasodilatory response of arteriovenous anastomoses to local cold stimuli in the dog's tongue. Pfliigers Arch 388:17 19 Pleschka K, Kiihn P, Nagai M (1979) Differential vasomotor adjustments in the evaporative tissues of the tongue and nose in the dog under heat load. Pflfigers Arch 382:255-262 Prichard MML, Daniel PM (1953) Arterio-venous anastomoses in the dog. J Anat 87:6(~74 Thomson EM, Pleschka K (1980) Vasodilatory mechanisms in the tongue and nose of the dog under heat load. Pflfigers Arch 387:161-166 Uehara Y, Desaki J, Fujiwara T (1981) Vascular autonomic plexuses and skeletal neuromuscular junctions: A scanning electron microscopic study. Biomed Res 2 [Suppl] 139-143 Ushiki T, Ide C (1988) Autonomic nerve networks in the rat exocrine pancreas as revealed by scanning and transmission electron microscopy. Arch Histol Cytol 51:71-81 Accepted June 15, 1989