==ミE=巴
伽一 一 『
一一 一,.̲̲
N = 8
睾9≡2
N
睾ヨ
≡ 2
一 m ︒, 一
E三
眉oぢ帽﹂ぢ﹂二
ca
ヨ
智葱﹂
Z = 配
くひ.bD匡
Neural Mechanisms Governing Distribution of Cardiac Output
2. Valve Dilator Neurons
Summary
1. The cardioarterial valves of five pairs of 1st to 5th lateral arteries (LA1 to LA5) arc innervated one—by—one by the 1st to 5th lateral cardiac nerves (LCN1
to LCN5) which are valve dilators.
2. The valve in LA5 is innervated by LCN5, which is composed of two axons arising one by one from the 2nd and 3rd abdominal ganglia. Neuronal cell
bodies of the two LCN5 axons were identified, by means of electrophysiology
and by dye injection, in a perineural sac—like structure of the ganglionic sheath at the site nearest the midline of the posterior half of the 2nd and 3rd abdominal ganglia.
3. The two cell bodies were often found to be located on ipsilateral or contralateral side to their axons running from the ganglion to the cardioarterial
valve.
4. The two LCN5 neurons were found to send both descending and ascending long interganglionic axons beyond the adjacent ganglia; to the 8th thoracic and 4th abdominal ganglia for one neuron in the 2nd abdominal ganglion, and the 1st abdominal and the 5th abdominal ganglia for the other neuron in the 3rd
abdominal ganglion.
5. It is likely that there is no direct physiological connection between the two LCN5 neurons, since activation of either one of the neurons did not alter the
impulse pattern of the other.
- 44
-Introduction
The anatomy of the innervation of the cardioarterial valves was reported by
Alexandrowicz, in studies of decapod (1932), stomatopod (1934) and isopod (1952) hearts. Recent studies have shown that the cardioarterial valves regulate distribution of haemolymph flow between components of the arterial system in crustaceans (Kihara et al. 1985; McMahon and Burnett 1990). This is quite different from the distribution control system found in molluscs and vertebrates, where regulation is carried out by arteriolar smooth muscle (for reviews, Brownell and Ligman 1992; Skelton et al. 1992), rather than by central valves. I have provided evidence that distribution of haemolymph flow in Bathynomus doederleini, during activity of specific organs, may also be controlled by the cardioarterial valves (see Part I, 1). Kihara et al. (1985) have shown that in Bathynomus the cardioarterial valves receive excitatory (constrictor) and/or inhibitory (dilator) innervation, from the central nervous system (CNS).
Fujiwara—Tsukamoto et aI. (1992) have reported the anatomy and physiology of the 1st to 5th lateral cardiac nerves (LCN1-5), which innervate the valves of the 1st to 5th lateral arteries (LA1-5) in Bathynomus. LCN2 and LCN3 consist of one axon each and LCN1, LCN4 and LCN5 consist of two axons. All axons in the LCNs are valve inhibitory axons which dilate the valves, increasing outflow of haemolymph to the arteries.
This section deals with LCN5 as an example of the nerves which innervate the cardioarterial valves. LCN5 innervates the valve of LA5, the artery which supplies the swimmerets. One of the two axons of LCN5 arises from the 2nd abdominal ganglion (AG2) and the other from the 3rd abdominal ganglion (AG3), via the 3rd root in each case (Fujiwara—Tsukamoto et al. 1992). The impulse rate
- 45
-of LCN5 increased during a period -of activation -of swimmeret movements (see Part I, 1) and also, tachycardia accompanied activation of swimmeret movements (Tanaka and Kuwasawa 1991a). This shows that valve nerves control haemolymph flow to target organs through individual arteries, according to their physiological demands. In order to analyze neuronal mechanisms of central control of haemolymph distribution from the heart to the arteries, it is necessary to identify neurons which innervate the cardiovascular system in the CNS. For the extrinsic cardiac nerves of Bathynonzus doederleini, three kinds of cardio—
regulatory neurons, two pairs of cardio—acceleratory neurons and one pair of cardio—inhibitory neurons, have recently been identified in the thoracic ganglia (Tanaka and Kuwasawa 1991b, 1991c), so valve motor neurons alone remain to be identified in the CNS. I will provide intraganglionic locations of inhibitory valve motor neurons to the cardioarterial valve of LAS of the isopod Bathynotnus doederleini.
- 46
-Materials and Methods
Before dissection, animals (Bathynomus doederleini, 9-15 cm in body length) were anesthetized with an isotonic (0.36 M) MgC12 solution injected into the body.
The animals were transected at an intersegmental region between the thorax and abdomen, and the ventral nerve cord together with a pair of LCN5s in the abdomen was exposed.
Back—filling by cobalt and nickel ions from the 3rd roots of AG2 and AG3 revealed about 20 candidates for LCN5 cell bodies in AG2 or AG3. I examined the effect on LCN5 activities of extracellular positive or negative DC stimulation, applied through a glass capillary suction electrode to perineural sac—like structures of the ganglionic sheath, which involved the candidates. Meanwhile, the proximal cut—stumps of right and left LCN5s at a point near the heart were drawn one—by—one into glass capillary suction electrodes to record their extracellular impulses. These procedures enabled to observe that the rate of LCN5 impulses was changed when stimuli were applied to a certain sac—like structure nearest the midline of the posterior halves of AG2 and AG3. Indeed, candidates for LCN5 cell bodies were found in the sac—like structures of the posterior halves of AG2 and AG3.
The sac—like structures of the ganglia were carefully desheathed. Intracellular recording and current injection were performed on LCN5 cell bodies using single 3 M KCl—filled glass microelectrodes (tip resistance, 10-20 MQ) connected with a bridge device. For electrophoretic injection of the dye, the candidate cell bodies were impaled with glass microelectrodes filled with 5 % Lucifer yellow dissolved in 1 M LiC1 solution, using negative current pulses, 5-15 nA, 1 sec in duration and at 0.5 Hz for 1-2 hr (cf. Stewart 1978). The preparations were incubated
- 47
-overnight at 4 °C, fixed with 4 % formaldehyde, dehydrated with graded series of ethanol and cleared with methyl salicylate. The dye—filled neurons were observed under a fluorescent microscope and photographed.
- 48
-Results and Discussion
Criteria employed for identification of LCN5 neurons were as follows.
1. Spontaneous intracellular impulses, recorded from the cell bodies, corresponded to extracellular impulses of one of two LCN5 axons in a one—to—one manner with a constant delay. 2. When the rate of impulses was altered by depolarizing or hyperpolarizing current, a one—to—one relationship was maintained between impulses of LCN5 and of the cell body. 3. Antidromically—conducted impulses from LCN5 to the cell body did not cease in a high Mg2+ saline (three times
[Mg2+] of normal sea water). 4. After the injection of Lucifer yellow, it was
shown that the axon from the cell body actually extended into the 3rd root of its own ganglion.
Intracellular recordings obtained from candidates for LCN5 cell bodies in AG2
(A) and AG3 (B) are shown in Figure 1. In the preparations from which the data were obtained, both the cell bodies were located at a site near the midline in the posterior halves of the ganglia on the contralateral side (cf. Fig. 2). Spontaneous intracellular impulses in the cell bodies corresponded to one of the two units of extracellular impulses of a right LCN5, in a one—to—one manner with a constant delay (compare the top traces with the middle traces). Al and B1 show the points at which depolarizing current was injected into a cell body. It may be seen that, as the rate of impulse firing increased in the cell body, so did the rate of one of the two units of extracellular impulses recorded from the right LCN5, in a one—to—one manner. Hyperpolarizing current (A2 and B2) stopped the intracellular impulses and one of the two units of extracellular LCN5 impulses. However, intracellular current injection into a left hemiganglion neuron did not affect the impulse patterns of the left LCN5, and that of the other unit (from AG3 in A or from AG2
- 49
-in B) of the right LCN5. When antidromic somatic impulses were generated by single stimuli applied to LCN5, during perfusion with high Mg2+ saline, the impulses corresponded to the stimuli in a one—to—one manner, with a constant latency (Fig. 1, A3 and B3).
Intracellular injection of Lucifer yellow into the cell bodies revealed their axonal processes (photographs in Fig. 2, A and B). Both the cell bodies (40 µm in diameter) were located in the posterior halves of right AG2 and AG3 hemiganglia contralateral to LCN5s. The morphological features of the two neurons appeared to be quite similar to each other. As shown in the schematic drawings of Figure 2, an axon arising from the cell body runs to the origin of the 3rd root, and divides into three major processes. Two of the three were
interganglionic processes ascending and descending beyond adjacent ganglia, and the other was one of the two LCN5 axons running to the periphery in the 3rd root of the ganglion.
Intracellular recordings and Lucifer Yellow staining show that the candidate neurons are the LCN5 neurons themselves. Among 12 cell bodies in 8 animals used in this experiment, 9 cell bodies in 7 animals were located on the side contralateral to the LCN5 carrying their axons, and 3 cell bodies in 3 animals were located on the side ipsilateral to their LCN5 periphery. However, the perineural sac—like structures of the ganglionic sheath, where LCN5 cell bodies were found, were always located at the site nearest the midline in the posterior halves of the 2nd and 3rd abdominal ganglia (AG2 and AG3). Fine processes ramifying from axons were observed (see schematic drawings of Fig.2). A descending process from AG2 and an ascending process from AG3 run very close to each other in the CNS (see Fig. 2A). However, direct current injected into either one of the neurons did not alter the impulse pattern of the other (Fig. 1, A and B, 1 and 2). It
- 50
-is likely that there is no direct physiological connection between the two LCN5
neurons.
Each of the 3rd roots of the 4th thoracic to 3rd abdominal ganglia contributes
an axon to the valves (Fujiwara—Tsukamoto et al. 1992). Since the two LCN5 neurons have similar intraganglionic locations in their individual neurons and have similar impulse characteristics, it may be possible to extrapolate the characteristics to the identification of cell bodies of LCNs other than LCN5, i.e. LCN1 to LCN4.
Identification of valve motor neurons allows analysis of the central cellular mechanisms of regulation of the distribution of cardiac output between components of the arterial system.
- 51
-References
Alcxandrowicz, J. S. (1932) The innervation of the heart of the Crustacea.
I. Decapoda. Q. J. Microsc. Sci. 75, 181-249.
Alexandrowicz, J. S. (1934) The innervation of the heart of Crustacea.
II. Stomatopoda. Q. J. Microsc. Sci. 76, 511-548.
Alcxandrowicz, J. S. (1952) Innervation of the heart of Ligia oceanica. J. Mar.
Biol. Assoc. U.K. 31, 85-96.
Brownell, P. H. and Ligman, S. H. (1992) Mechanisms of circulatory homeostasis
and response inAplysia. Experientia 48, 818-827.
Fujiwara—Tsukamoto, Y., Kuwasawa, K. and Okada, J. (1992) Anatomy and
physiology of neural regulation of haemolymph flow in the lateral arteries of the isopod crustacean, Bathynorus doederleini. In Phylogenetic Models in
Functional Coupling of the CNS and the Cardiovascular System, Comparative
Physiology, Vol.11, Edited by Hill, R. B. and Kuwasawa, K., Karger, Basel, pp70-85.
Kihara, A., Kuwasawa, K. and Yazawa, T. (1985) Neural control of the cardio—
arterial valves in an isopod crustacean, Bathynorus doederleini: Excitatory
and inhibitory junctional potentials. J. Comp. Physiol. A 157, 529-536.
McMahon, B. R. and Burnett, L. E. (1990) The crustacean open circulatory
system: A reexamination. Physiol. Zool. 63, 35-71.
Skelton, M., Alevizos, A. and Koester, J. (1992) Control of the cardiovascular system of Aplysia by identified neurons. Experientia 48, 809-817.
Stewart, W. W. (1978) Functional connections between cells as revealed by dye—
coupling with a highly fluorescent naphthalimide tracer. Cell 14, 741-759.
Tanaka, K. and Kuwasawa, K. (1991a) Central outputs for extrinsic neural control
- 52
of the heart in an isopod crustacean, Bathynomus doederleini: Neuroanatomy and electrophysiology. Comp. Biochem. Physiol. 98C, 79-86.
Tanaka, K. and Kuwasawa, K. (1991b) Identification of cardio—acceleratory neurons in the thoracic ganglion of the isopod crustacean Bathynomus
doederleini. Brain Res. 544, 311-314.
Tanaka, K. and Kuwasawa, K. (1991c) Identification of cardio—inhibitory neurons in the thoracic ganglion of the isopod crustacean Bathynomus doederleini . Brain Res. 558, 339-342.
- 53
-恵
じ
FigureLegends
Figure 1.
The records in A and B illustrate the methods used to identify LCN5 neurons in the 2nd (A) and 3rd (B) abdominal ganglia (AG2 and AG3). 1, 2. Intracellular records from the cell bodies (soma) of the left hemiganglion and extracellular records from right and left LCN5s (r and 1). Intracellular impulses corresponded to one of the two units of right LCN5 impulses one—to—one, with a constant delay.
The LCN5 cell bodies were depolarized (Al, B1) or hyperpolarized (A2, B2) during periods indicated by underlining. Large artifacts indicate the onset and end of current injections. A one—to—one relationship between intracellular and extracellular impulses continued during depolarization. 3. Superimposed records of antidromically—conducted impulses in cell bodies. Eight (A3) and ten (B3) single stimuli applied to an LCN5 generated antidromic impulses in the cell bodies of AG2 (A3) and AG3 (B3) in a one—to—one manner with a constant latency.
Downward deflections at the beginnings of records are stimuli. This record was obtained during perfusion of the preparation with high Mg2+ saline.
Figure 2.
Cell bodies and processes of LCN5 neurons stained with Lucifer yellow in the 2nd (A) and the 3rd (B) abdominal ganglia (AG2 and AG3) of the ventral nerve cord. Fluorescent micrographs show ventral views of the cell bodies (s, 40 µm in diameter) and the major processes arising from the cell bodies (arrows). A cell body of each LCN5 neuron was located in the posterior half of the contralateral hemiganglion in this preparation. An axon arising from the cell body divided into three major processes at the origin of the 3rd root. Two of the processes were interganglionic ascending and descending processes (arrows). The other one was the LCN5 axon, which runs out peripherally in the 3rd root. Arrowheads in AG2
- 55
-indicate the ascending process arising from the LCN5 neuron in AG3. Schematic drawings show the locations of LCN5 cell bodies and their processes. Broken lines show perineural sac—Iike structures of the ganglionic sheath on the ventral side.
- 56
-Figures
︒︒ の 6 言 ︒︒ ︒︒ E 9
的
ω 瞬 . O ︾亘 ︒, ∈ 9
的
> E の
N ﹂
窪 ︒ し︒ 言 ︒ 二
N ﹂
窪 O し︒
一 ﹂
窪Oω国
一
F
﹂
帽 E O ω
く
H.bD旧﹂