- Email: [email protected]
Morphological changes in gland cells and axons resulting from stimulation of the salivary nerves of the Cockroach, Nauphoeta cinerea
TISSUE &CELL 1 1981 Longman
0040-8166/81/00120141$02.00
1981 13 (1) 141-151 Group Ltd
DAVID
J. MAXWELL
MORPHOLOGICAL CHANGES IN GLAND CELLS AND AXONS RESULTING FROM STIMULATION OF THE SALIVARY NERVES OF THE COCKROACH, NAUPHOETA ClNEREA ABSTRACT. The salivary glands of the cockroach, Nauphoeta rineren (Olivier, 1789), are innervated and there is considerable evidence to suggest that dopamine is the neurotransmitter at the neuroglandular junction. As the gland is a bilaterally symmetrical structure it was possible to electrically stimulate the salivary nerve supplying the ipsilateral side of the gland whilst the contralateral side of the gland served as a convenient control. Saliva elicited from the glands by electrical stimulation of these nerves was collected and used to monitor the physiological state of the tissue. Glands were fixed for light and electron microscopy during secretion and it was observed that the ductules in peripheral acinar cells were distended in stimulated sides of the glands but not in contralateral unstimulated sides. This evidence implies that peripheral cells are responsible for the initiation of salivary fluid secretion. Changes were also observed in the catecholamine containing axons that innervate the glands. In stimulated axons a statistically significant reduction in numbers of small agranular vesicles was observed when compared with contralateral unstimulated controls and freshly fixed tissue. This was not the case with the larger granular vesicles of the same axons which showed no reduction in number as a result of stimulation. In addition it was also noted that the small agranular vesicles tended to aggregate and change their shapes in response to nerve stimulation. These results imply that the small agranular vesicles play a role in transmitter release.
Introduction
ducts and branch profusely over the surface of the gland (Bowser-Riley, 1978). Axons associated with the acini are morphologically of two distinct types (Maxwell, 1978); type A axons which possess large granular vesicles and small agranular vesicles which probably contain dopamine (Maxwell, 1980) and type B axons which only contain large granular vesicles. After stimulation of the salivary nerves an electrical potential can be recorded from the acini (House, 1973). This response is characterized by a latency of 1 set, a hyperpolarization of 30 mV and a time course of about 10 sec. It is accompanied by a rise in membrane potassium conductance (Ginsborg et al., 1974). Electrical stimulation of the salivary nerves also elicits secretion from the gland (Smith and House, 1977). Stimulation with 50 V pulses of 0.5 msec duration at 10 Hz
THE acini of the cockroach salivary gland consist of two disctinct types of cell (Bland and House, 1971); peripheral cells which possess an invaginated ductule surrounded by microvilli, infoldings of the basal plasma membrane and numerous mitochondria, and central cells which possess secretory granules which are thought to contain amylase (Bland and House, 1971). These acini are richly innervated by the salivary nerves which have been shown (with methylene blue stain) to arise from the suboesophageal ganglion, run down the reservoir
University of Edinburgh, Department of Veterinary Physiology, Royal (Dick) School of Veterinary Studies, Summerhall, Edinburgh EH9 IQH, Scotland. Received 7 May 1970. Revised 15 September 1980. 141
I42
MAXWELL
produces a maximal rate of secretion of 80 nl/min until fatigue occurs. This fatigue has been attributed to a failure of transmitter release as perfused dopamine is able to elicit secretion for considerably longer periods of time than nerve evoked secretion. Secretory experiments along with electrophysiological experiments have now provided a considerable body of evidence suggesting that dopamine is the neurotransmitter at the cockroach salivary gland (for reviews see House, 1977; House and Ginsborg, 1979). The present study attempts to investigate morphological changes that occur in acinar cells and in the axons that innervate these cells as a consequence of electrical stimulation of the salivary nerves. The salivary gland of the cockroach is eminently suitable for the investigation of such phenomena as it is a bilaterally symmetrical structure which receives separate innervation to each side. Only a very limited amount of contralateral innervation occurs in the midline region (Ginsborg and House, 1976). In view of the bilateral arrangement one gland may be stimulated while the other serves as a control. Materials and Methods
in it to allow the saliva to escape. Saliva elicited by nerve stimulation was collected with the aid of a micropipette and the diameters of spherical droplets of saliva were measured in the paraffin pool under the eyepiece graticule of a Zeiss dissecting microscope. (A discussion of the accuracy of this method is given in House and Smith, 1978.) For each animal the physiological state of the tissue was monitored by plotting rate of secretion against time as in Fig. 1. In Fig. 1 it can be seen that in response to constant electrical stimulation of the salivary nerves the gland initially produces a copious flow of secretion which dissipates after about 40 min. In some experiments, fixative (Maxwell, 1978) was introduced into the bath during this phase of copious secretion and in others it was not introduced until secretion had ceased. In this way it was possible to fix the stimulated and unstimulated sides of the same gland under the same conditions. Electron microscopy Tissue was processed for electron microscopy by the method described in Maxwell (1978). Sections 1 t&m thick were also prepared for light microscopy from much of the
Electrical stimulation Isolated preparations of cockroach salivary glands were made from adult cockroaches, Nauphoeta cinerea (Olivier, 1789), and placed in disposable perfusion chambers after the method of Smith and House (1977). Saline (Smith and House, 1977) was delivered to the chamber at a rate of 2.5 ml/min by a Watson-Marlowe flow inducer. Salivary nerves were stimulated by drawing them, along with the reservoir duct, into a glass suction electrode. Only one of the salivary nerves was stimulated; the other served as a control. The nerve to be stimulated was chosen at random. Electrical pulses of 40 V and 5 msec duration were delivered at a frequency of 5 Hz by a Grass stimulator (SD 5). A stimulus of this magnitude usually produces a half maximal response (House and Smith 1978). Secretion was collected by drawing the ipsilateral secretory duct through a small hole in the perfusion chamber into a pool of liquid paraffin. The secretory duct was secured with a piece of silver wire and a small incision was made
Stimulation
21
cc
0
40V 5H.z
,Isb, 30
60
90
120
Time (mins) Fig, 1. The characteristic response of a cockroach salivary gland to continuous stimulation of one of the salivary nerves. In this case saliva was collected at 10 min intervals from the ipsilateral salivary duct. Typically, there is an initial copious flow of saliva which dissipates after roughly 40 min of stimulation. Glands were left to rest in the chamber for ca. 30 min before the application of electrical stimulation in each experiment to eliminate the possibility of spontaneous secretion. The methods and responses of secretory experiments in this preparation have been discussed fully by Smith and House (1977) and House and Smith (1978).
STIMULATION
tissue; these blue dye.
OF COCKROACH
were
stained
with
SALJVARY
toluidine
i-lnalysis sf micrographs Every axonal profile that contained vesicular material was photographed under the electron microscope and subsequently analysed. Care was taken to ensure that no profile was measured more than once. Linear data, such as vesicular diameters, were gathered from micrographs with the aid of a measuring device which was able to store them directly into the files of a PDP-12 computer. The measuring device consists of a linear transducer arranged like a set of vernier calipers. The operator is able to send analogue signals to the computer which are interpreted and returned to the user as scaled linear measurements and simultaneously stored in a file (Clark Ensor and Maxwell, unpublished). Information that was gathered in this manner was then processed by additional programs which could access these files. Cross-sectional areas of axons, excluding areas of mitochondria and other large organelle structures, were calculated directly from micrographs using a Magiscan image analysis system (Joyce Loebl Ltd). With this system it is possible to view micrographs on the television screen of the Magiscan and to demarcate the area for calculation with a light pen. The program could also calculate the vesicular densities of these axons (expressed as number of vesicles per square micron) as it was possible to count the number of vesicles enclosed within the axonal profiles using the light pen. A white point appeared on the screen indicating that the vesicular profile had been counted. Results Structure &the salivary nerves The paired salivary nerves of the cockroach salivary gland are attached to the reservoir ducts by means of some connective tissue. Each nerve consists of two large axons of an average diameter of 7 pm (Fig. 2) which are surrounded by glia. The entire nerve is contained within a basement membrane. In addition to these two large axons a number of smaller axons, often accompanied with glial cells, are found within the basement membrane.
NERVES
14.7
The @ects of nerve stimulation upon acinar cells If glands were fixed during the period of copious flow of secretion (i.e. about 10 min after the commencement of stimulation, see Fig. I) it was noted that peripheral cell ductules from the stimulated sides of glands were distended (Fig. 3). The unstimulated side of glands was normal in appearance (Fig. 4) and only on one occasion was a group of acini observed that had distended ductules. This is in keeping with the idea that there is only a limited contralateral innervation of the midline region of the gland (Ginsborg and House, 1976). The mean size of the ductule in the stimulated side of glands was 12.81 ? 3.4 pm (N= 48) and that from the unstimulated side was 6.73 + 2.09 pm (N= 25). A t-test for unpaired samples indicates that this is a highly significant change (P
144
MAXWELL
Ahhreviationr
Ax BM C D
for /ig;gsres
Axon Basement membrane Cenrral cell Duct
GI GV P SD
Glial cell Granular vesicle Peripheral cell Septate desmosomes
Fig. 2. The salivary nerve of the cockroach. Note the two large axons that arc turrounded by glia. The arrows indicate several smaller axons within the basement membrane region of the nerve. x 7000.
Fig. 3. A 1pmthick section from a stimulated gland that was fixed during fluid secretion. An arrow indicates one of the many distended peripheral cell ductules. x 350. Fig. 4. A lprn thick section from the unstimulared side of the gland illustrated Fig. 3. In this case no distended peripheral cell ductules are visible. x 350.
in
Fig. 5. An electron mxrograph of a peripheral cell with a. distended duct& that was fixed during stimulation of the ipsilateral salivary nerve. Note the infoldings of the basal plasma membrane (arrow), the numerous mitochondria. the mlcrovilli surrounding the duct and (although not resolved at this magnification) the highly convoluted septate desmosomes that link the cells. x 7200.
146
MAXWELL
agranular vesicles occurs after stimulation of axons when compared with axons from the unstimulated sides of glands or with those from freshly fixed tissue. No such reduction in number could be found for the granular vesicles that are also present in these axons. A change in shape of the small agranular vesicles was also commonly detected after stimulation (e.g. Figs. 8, 10, 12, 14, 16). A previous study has shown that these vesicles normally come from a population of prolate rotary ellipsoids (Maxwell, 1978). Shapes of vesicles were compared by plotting the logarithm of the maximum diameter divided by the minimum diameter (measured at right angles to each other), as a histogram (Fig. 18). This method illustrates that the distributions for freshly fixed tissue (Fig. 18a) and for unstimulated controls (Fig.
18~) are similar, thus suggesting that the vesicles are a similar shape. However, vesicles from stimulated axons have a narrower distribution tending towards zero (Fig. 18b). This suggests that these vesicles are more circular in their profiles than those from unstimulated tissue. No changes could be detected in type B axons after electrical stimulation. Discussion Electron microscopy has revealed the existence of a nerve that is associated with the reservoir duct of the cockroach salivary apparatus. Electrophysiological (Ginsborg and House, 1976), secretory (Smith and House, 1977) and methylene blue (BowserRiley, 1978) studies all indicate that this is one of the paired salivary nerves which
Fig. 6. An axon from the unstimulated control taneously with the stimulated side. x 45,000. Fig. 7. An axon that was stimulated small vesicles (arrow). x 52,500. Fig. 8. As a result of stimulation around a dense membrane structure present (asterisk). x 45,000.
side of a gland that was fixed simul-
for several minutes.
Note the aggregation
of
the small vesicles of this axon have aggregated (arrow). An invagination of the axolemma is also
Fig. 9. A stimulated axon illustrating the aggregation of the small vesicles (large arrow) and infoldings of the axolemma (small arrows). x 50,000. Figs. IO-17 illustrate stimulated axons (even numbers) from four different experiments and control axons from the unstimulated contralateral glands in each experiment (odd numbers). Fig. IO. A stimulated axon illustrating the aggregation around a dense membrane structure (arrow). x 60,000. Fig.
I I. Control.
vesicles
x 60,000.
Fig. 12. In this stimulated axon there is an aggregation and an infolding of the axolemma (asterisk). x 60,000. Fig. 13. Control.
of small agranular
of agranular
vesicles (arrow)
x 60,000.
Fig. 14. Two stimulated axons, In one axon aggregations of agranular vesicles may be seen (arrows). One of these aggregations is around a dense membrane structure (small arrow). Both axons contain small agranular vesicles that are unusually circular in their profiles. x 40,000. Fig. 15. Control.
x 40,000.
Fig. 16. This stimulated axon still contains a high density of small agranular but many of them are more circular in their profiles than usual. x 45,000. Fig. 17. Control.
x 45,000.
vesicles
STIMULATION
OF
COCKROACH
SALIVARY
Table 1. Means and standard deviations (S. D.) of groups of control, stimulated and freshly fixed axons which contain granular and agranular vesicles. The densities of vesicles in each group are expressed in terms of number of vesicles per square micron
Numbers
Means (vesicles/ ELM)
S.D.
Controls Stimulated Freshly fixed
19 27 24
78.1 45.2* 96 1
+35 +41 +54
Granular vesicles Controls Stimulated Freshly fixed
18 25 23
5.4 5.4 5.3
k4.3 k 6.5 + 5.4
Treatment
149
NERVES
apical membranes of the acini as the presence of ouabain in the bathing medium does not inhibit secretion. The model suggests that chloride ions passively accompany sodium and that an accumulation of sodium chloride occurs inside the acinar ductules. Such a mechanism could cause the
a
Agranular vesicles
. . *. *.
. . * . . .
rop;..i;.;..;...;
..... ..... ..._.
* Significantly different from both the control and the freshly fixed mean values at the 0.01 level using Student’s t-test. No other differences were found to be significant. b
arise from the suboesophageal ganglion and innervate the acini. It is not known if both the large and small axons are responsible for the innervation of the gland although if the large axons are allowed to degenerate some axonal degeneration profiles are also observed in the glands (Maxwell, unpublished observation). Evidence has been presented that demonstrates distension of peripheral cell ductules in response to electrical stimulation of the salivary nerves. Such a distension has also been observed with Nomarski optics in live peripheral cells during stimulation of this preparation (R. K. Smith, unpublished). This suggests that peripheral cells are responsible for the initiation of fluid secretion in response to nervous stimulation, as distension was not seen in unstimulated sides of the same glands. Smith and House ( 1980) have demonstrated that in the absence of sodium ions in the bathing medium, the cockroach salivary gland is unable to secrete in response to applied dopamine. They have developed a model of salivary secretion in this gland which depends upon the passive entry of sodium ions into the basal membranes of the cells and active pumping of sodium ions at the apical membranes in the ductules. Any ATP dependent pump must be situated on the
ii
2% ::, ~............_.____._,..,,._.__ .
r. . l
.
I f
.f
f
.
**
.
f,
.a *a
f *
**t. .ff.
0.
** . . (I*
1
0
Log Ratio
max/rnin
Fig. 18. Histograms of the logarithm of the maximum diameter divided by the minimum diameter of the small agranular vesicles observed in cockroach salivary gland axons. Fig. 18a is a plot of vesicle dimensions taken from several freshly fixed axons. Fig. 18b represents similar data collected from stimulated axons. Fig. 1%~ is data collected from unstimulated axons in the contralateral glands to Fig. 18b. Figs. 18a and 18c have similar distributions whereas Fig. 18b is narrower suggesting that vesicles have changed in shape as a result of stimulation.
150
movement of water into peripheral cell ductules and hence distension of them. Supporting evidence for this idea may be inferred from the structure of peripheral cells. First, spaces between the densely packed microvilli could provide small extracellular channels in which sodium chloride could be concentrated and thus a standing osmotic gradient set up. (Diamond and Tormey, 1966; Diamond, 1980). Secondly, peripheral cells are crammed with mitochondria suggesting that their energy requirements are high. Finally, invaginations of the peripheral cell basal plasma membrane could provide low-resistance routes for water transport. Such a function has been ascribed (Green, 1979) to similar structures in the Malpighian tubules of Arachnocampa luminosa. It seems improbable that an extracellular route of water transport is of much importance in this gland as the septate desmosomes which occupy these extracellular spaces have been described as ‘barriers to diffusion’ (NoirotTimothee and Noirot, 1980; Noirot-Timothee et al., 1978). No differences were detected between the central cells of stimulated and unstimulated glands. In the rat parotid gland it is possible to reduce the numbers of secretory granules as a consequence of stimulation of the sympathetic nerve supply (Garrett et al., 1979). This result suggests that the amylase containing secretory granules of the cockroach salivary gland are released by some mechanism that is independent of the salivary nerves. Ever since the observation of quanta1 release of neurotransmitter (e.g. Del Castillo and Katz, 1954) synaptic vesicles have been designated the morphological ‘packets’ of such quanta. Direct evidence for such a relationship has proved to be elusive, as few exocytotic profiles have been observed by conventional electron microscopal techniques (e.g. Osborne, 1977), although methods rapid freezing of stimulated involving tissue combined with freeze-fracture techniques have recently produced promising results (Heuser et al., 1979). Several authors have suggested that a reduction in numbers of synaptic vesicles after stimulation may imply release of transmitter; e.g. in the cat sympathetic ganglion (Pysh and Wiley, 1974); in axon varicosities of the mouse vas
MAXWELL
deferens (Basbaum and Heuser, 1979); in the hatchet fish medulla (Model et al., 1975); at the crayfish neuromuscular junction (Atwood and Lang, 1972) and at the locust neuromuscular junction (McKinlay and Usherwood, 1973; Botham et al., 1978). The present study also indicates that the small agranular vesicles of cockroach salivary gland axons are reduced in number as a consequence of electrical stimulation. In keeping with the observations of Basbaum and Heuser (1979) on aminergic axon varicosities in the mouse vas deferens, the large granular vesicles are not reduced in number. It may be suggested that the reduction in number of the small agranular vesicles represents a release of dopamine from the axons as other data indicate that these small vesicles contain dopamine (see Maxwell, 1980). The role of the larger granules is not known but the evidence would suggest that they are not influenced by neural stimulation. On several occasions vesicles in stimulated axons were seen to aggregate around dense membrane structures. Such structures have been infrequently observed in freshly fixed tissue (Maxwell, 1978). Previous authors have referred to these structures as ‘active zones’ (e.g. Rheuben and Kammer, 1980) and the increase in frequency of such zones in this study supports the idea that they are involved in transmitter release. Aggregation of vesicles after electrical stimulation of locust motor neurones has also been observed by McKinlay and Usherwood (1973). The tendency for vesicles to adopt circular profiles after stimulation is difficult to explain. Osborne (1977) also noted that there was an interchange between flattened and vesicles during stimulation of circular photoreceptor cells. Acknowledgements Financial support was provided by the Science Research Council. 1 wish to thank Drs K. P. Bland, C. R. House and K. L. Prince for their constructive comments upon this manuscript. Thanks are also due to Dr R. J. Houchin for her help with the Magiscan system, to Mr C. M. Warwick for preparation of the photographic plates and to Mrs L. Manson for typing the manuscript.
S-l IMULATION
OF
COCKROACH
SALIVARY
IS1
NERVES
References Arwooo. H. L.. and LANG, F. 1972. Synpatic vesicles: selective depletion in crayfish excitatory and inhlbitory axons. Science, 176. 1353-l 355. BASBAUM,C. B. and HEUSER,J. E. 1979. Morphological studies of stimulated adrenergic axon varicosities in the mouse vas deferens. J. Cell Biul.. 80, 310-325. BLAND, K. P. and HOUSE, C. R. 1971. Function of the salivary glands of the cockroach Nauphortn cinererr. .I. Insect Physiol., 17, 2084-2096. BOTHAM, R. P., BEADLE,D. J., HART, R. J., POTTER, C. and WILSON, R. G. 1978. Stimulation induced depletion of the synaptic vesicles in excitatory motor nerve terminals of the locust, Locmtrr migratooricz L. Exprrentia, 34, 207-208. BOWSER-RILEY, F. 1978. The salivary glands of the cockroach Nauphorta cinerea (Olivier): a study of 11s innervation by light and scanning electron microscopy. Cell Tim. Res., 187, 525-534. DEL CASTILLO, J. and KATZ, B. 1954. Quanta1 components of the end plate potential. J. Phwiol.. 124, 560.. 537. DIAMONU, J. M. and TORMEY, J. McD. 1966. Role of long extracellular channels in fluid transport across epithelia. Nature, Lond. 210, 817-820. DIAMOND,J. M. 1980. Osmotic water flow in leaky epithelia. J. Membrane Biol.. 51, 195-216. GARRETT.J. R., HARROP, T. J., KIDS, A. and THULIN, A. 1979. Nerve-induced secretory changes in salivary glands. In Nrrces and the gut (eds. F. B. Brooks and P. W. Evers), pp. 1440. C.B.S. Inc., New Jersey. GINSBORG, B. L., HOUSE, C. R. and SILKINSKY, E. M. 1974. Conductance changes associated with the secretory potential in the cockroach salivary gland. J. Physiol., 236, 723-731.
GINSBORG, B. L. and HOUSE, C. R. 1976. The response to nerve stimulation cinrrra
Olirier.
J. Physiol.,
ofthe salivary gland of N~oph~~<~fo
262, 447-487.
GREEN, L. F. B. 1979. Regional specialization in the Malpighian tubules of the New Zealand glow-worm Arachnocampa /uminosa (Diptera: Mycetophilidae). The structure and function of type I and II cells. Timur
& Cell, 11, 673-702.
HEUSER. J. E., REESE, T. S., DENNIS, M. J., JAN, Y., JAN, L. and EVANS. L. 1979. Synaptic vesicle exocytosis captured by quick freezing and correlated with transmitter release. J. Cell Biol., 81, 275-300. HOUSE, C. R. 1973. An electrophysiological study of neuroglandular transmission in the isolated salivary gland of the cockroach. J. Exp. Biol., 58, 29-43. House. C. R. 1977. Cockroach salivary gland: secretory epithelium with a dopaminergic innervation. In Trmsport of Jons and Water in Animal.\ (eds. B. L. Gupta, R. B. Moreton. J. L. Oschman and B. J. Wall). Academic Press, London. HOUSE, C. R. and SMITH, R. K. 1978. On the receptors involved in the nervous control of salivary secretion by Nauphorta cinrrea (Olivier). J. Physiol., 279, 457-471. HOUSE, C. R. and GINSBORG, B. L. 1979. Pharmacology of cockroach salivary secretion. Camp. Biochent. Ph.)..~ioI., 63C, l-6. MCKINLAY, R. G. and USHERWOOD,P. N. R. 1973. The role of synaptic vesicles in transmission at the insect nerve-muscle junction. Lifp Sri., 13, 1051-1056. MAXWELL, D. J. 1978. Axons associated with the salivary apparatus of the cockrocah Nauphwtcr cinrrtw Tissur & Cell, 10, 699-706. MAXWELL, D. J. 1980. Histochemical properties of axons associated with the salivary apparatus of the cockroach Nuuphorta cinrwa (Olivier). Tissue & Cell, 12, 703-71 I. MODEL, P. G., HIGHSTEIN, S. M. and BENNETT, M. V. L. 1975. Depletion of vesicles and fatigue of transmission at a vertebrate central synapse. Bruin Res., 45, 209-228. NOIRIOT-TIMOTHBE,C., SMITH, D. S.. CAYER. M. L. and NOIROT, C. 1978. Septate junctions in insects: comparison between intracellular and intramembranous structures. Tksue & Cell, 10, 125-136. NOIROT-TIMOTHBE.C. and NOIROP. C. 1980. Septate and scalariform junctions in arthropods. Jnr. Rrt. Cytol., 63, 97- 139. OSBURNE, M. P. f977. Role of vesicles with some observations on vertebrate sensory cells. In Swrapwv feds. G. A. Cottrell and P. N. R. Usherwood). Blackie, London and Glasgow. PYSH, J. J. and WILEY, R. G. 1974. Synaptic vesicle depletion and recovery in cat sympathetic ganglia electrically stimulated in vim. J. Cell Biol., 60, 365-374. RHEUBEN. M. B. and KAMMER,A. E. 1980. Comparison of slow larval and fast adult muscle innervated by the same motor neurone. J. exp. Biol., 84, 103-I 18. SMITH, R. K. and HOUSE, C. R. 1977. Fluid secretion by isolated cockroach salivary glands. Exprrienticr, 33, 1182-l 183.
SMITH, R. K. and HOUSE. C. R. 1980. Ion and water transport Mrmhrme Biol., 51, 325-346
by isolated
cockroach
salivary
glands. J.
0040-8166/81/00120141$02.00
1981 13 (1) 141-151 Group Ltd
DAVID
J. MAXWELL
MORPHOLOGICAL CHANGES IN GLAND CELLS AND AXONS RESULTING FROM STIMULATION OF THE SALIVARY NERVES OF THE COCKROACH, NAUPHOETA ClNEREA ABSTRACT. The salivary glands of the cockroach, Nauphoeta rineren (Olivier, 1789), are innervated and there is considerable evidence to suggest that dopamine is the neurotransmitter at the neuroglandular junction. As the gland is a bilaterally symmetrical structure it was possible to electrically stimulate the salivary nerve supplying the ipsilateral side of the gland whilst the contralateral side of the gland served as a convenient control. Saliva elicited from the glands by electrical stimulation of these nerves was collected and used to monitor the physiological state of the tissue. Glands were fixed for light and electron microscopy during secretion and it was observed that the ductules in peripheral acinar cells were distended in stimulated sides of the glands but not in contralateral unstimulated sides. This evidence implies that peripheral cells are responsible for the initiation of salivary fluid secretion. Changes were also observed in the catecholamine containing axons that innervate the glands. In stimulated axons a statistically significant reduction in numbers of small agranular vesicles was observed when compared with contralateral unstimulated controls and freshly fixed tissue. This was not the case with the larger granular vesicles of the same axons which showed no reduction in number as a result of stimulation. In addition it was also noted that the small agranular vesicles tended to aggregate and change their shapes in response to nerve stimulation. These results imply that the small agranular vesicles play a role in transmitter release.
Introduction
ducts and branch profusely over the surface of the gland (Bowser-Riley, 1978). Axons associated with the acini are morphologically of two distinct types (Maxwell, 1978); type A axons which possess large granular vesicles and small agranular vesicles which probably contain dopamine (Maxwell, 1980) and type B axons which only contain large granular vesicles. After stimulation of the salivary nerves an electrical potential can be recorded from the acini (House, 1973). This response is characterized by a latency of 1 set, a hyperpolarization of 30 mV and a time course of about 10 sec. It is accompanied by a rise in membrane potassium conductance (Ginsborg et al., 1974). Electrical stimulation of the salivary nerves also elicits secretion from the gland (Smith and House, 1977). Stimulation with 50 V pulses of 0.5 msec duration at 10 Hz
THE acini of the cockroach salivary gland consist of two disctinct types of cell (Bland and House, 1971); peripheral cells which possess an invaginated ductule surrounded by microvilli, infoldings of the basal plasma membrane and numerous mitochondria, and central cells which possess secretory granules which are thought to contain amylase (Bland and House, 1971). These acini are richly innervated by the salivary nerves which have been shown (with methylene blue stain) to arise from the suboesophageal ganglion, run down the reservoir
University of Edinburgh, Department of Veterinary Physiology, Royal (Dick) School of Veterinary Studies, Summerhall, Edinburgh EH9 IQH, Scotland. Received 7 May 1970. Revised 15 September 1980. 141
I42
MAXWELL
produces a maximal rate of secretion of 80 nl/min until fatigue occurs. This fatigue has been attributed to a failure of transmitter release as perfused dopamine is able to elicit secretion for considerably longer periods of time than nerve evoked secretion. Secretory experiments along with electrophysiological experiments have now provided a considerable body of evidence suggesting that dopamine is the neurotransmitter at the cockroach salivary gland (for reviews see House, 1977; House and Ginsborg, 1979). The present study attempts to investigate morphological changes that occur in acinar cells and in the axons that innervate these cells as a consequence of electrical stimulation of the salivary nerves. The salivary gland of the cockroach is eminently suitable for the investigation of such phenomena as it is a bilaterally symmetrical structure which receives separate innervation to each side. Only a very limited amount of contralateral innervation occurs in the midline region (Ginsborg and House, 1976). In view of the bilateral arrangement one gland may be stimulated while the other serves as a control. Materials and Methods
in it to allow the saliva to escape. Saliva elicited by nerve stimulation was collected with the aid of a micropipette and the diameters of spherical droplets of saliva were measured in the paraffin pool under the eyepiece graticule of a Zeiss dissecting microscope. (A discussion of the accuracy of this method is given in House and Smith, 1978.) For each animal the physiological state of the tissue was monitored by plotting rate of secretion against time as in Fig. 1. In Fig. 1 it can be seen that in response to constant electrical stimulation of the salivary nerves the gland initially produces a copious flow of secretion which dissipates after about 40 min. In some experiments, fixative (Maxwell, 1978) was introduced into the bath during this phase of copious secretion and in others it was not introduced until secretion had ceased. In this way it was possible to fix the stimulated and unstimulated sides of the same gland under the same conditions. Electron microscopy Tissue was processed for electron microscopy by the method described in Maxwell (1978). Sections 1 t&m thick were also prepared for light microscopy from much of the
Electrical stimulation Isolated preparations of cockroach salivary glands were made from adult cockroaches, Nauphoeta cinerea (Olivier, 1789), and placed in disposable perfusion chambers after the method of Smith and House (1977). Saline (Smith and House, 1977) was delivered to the chamber at a rate of 2.5 ml/min by a Watson-Marlowe flow inducer. Salivary nerves were stimulated by drawing them, along with the reservoir duct, into a glass suction electrode. Only one of the salivary nerves was stimulated; the other served as a control. The nerve to be stimulated was chosen at random. Electrical pulses of 40 V and 5 msec duration were delivered at a frequency of 5 Hz by a Grass stimulator (SD 5). A stimulus of this magnitude usually produces a half maximal response (House and Smith 1978). Secretion was collected by drawing the ipsilateral secretory duct through a small hole in the perfusion chamber into a pool of liquid paraffin. The secretory duct was secured with a piece of silver wire and a small incision was made
Stimulation
21
cc
0
40V 5H.z
,Isb, 30
60
90
120
Time (mins) Fig, 1. The characteristic response of a cockroach salivary gland to continuous stimulation of one of the salivary nerves. In this case saliva was collected at 10 min intervals from the ipsilateral salivary duct. Typically, there is an initial copious flow of saliva which dissipates after roughly 40 min of stimulation. Glands were left to rest in the chamber for ca. 30 min before the application of electrical stimulation in each experiment to eliminate the possibility of spontaneous secretion. The methods and responses of secretory experiments in this preparation have been discussed fully by Smith and House (1977) and House and Smith (1978).
STIMULATION
tissue; these blue dye.
OF COCKROACH
were
stained
with
SALJVARY
toluidine
i-lnalysis sf micrographs Every axonal profile that contained vesicular material was photographed under the electron microscope and subsequently analysed. Care was taken to ensure that no profile was measured more than once. Linear data, such as vesicular diameters, were gathered from micrographs with the aid of a measuring device which was able to store them directly into the files of a PDP-12 computer. The measuring device consists of a linear transducer arranged like a set of vernier calipers. The operator is able to send analogue signals to the computer which are interpreted and returned to the user as scaled linear measurements and simultaneously stored in a file (Clark Ensor and Maxwell, unpublished). Information that was gathered in this manner was then processed by additional programs which could access these files. Cross-sectional areas of axons, excluding areas of mitochondria and other large organelle structures, were calculated directly from micrographs using a Magiscan image analysis system (Joyce Loebl Ltd). With this system it is possible to view micrographs on the television screen of the Magiscan and to demarcate the area for calculation with a light pen. The program could also calculate the vesicular densities of these axons (expressed as number of vesicles per square micron) as it was possible to count the number of vesicles enclosed within the axonal profiles using the light pen. A white point appeared on the screen indicating that the vesicular profile had been counted. Results Structure &the salivary nerves The paired salivary nerves of the cockroach salivary gland are attached to the reservoir ducts by means of some connective tissue. Each nerve consists of two large axons of an average diameter of 7 pm (Fig. 2) which are surrounded by glia. The entire nerve is contained within a basement membrane. In addition to these two large axons a number of smaller axons, often accompanied with glial cells, are found within the basement membrane.
NERVES
14.7
The @ects of nerve stimulation upon acinar cells If glands were fixed during the period of copious flow of secretion (i.e. about 10 min after the commencement of stimulation, see Fig. I) it was noted that peripheral cell ductules from the stimulated sides of glands were distended (Fig. 3). The unstimulated side of glands was normal in appearance (Fig. 4) and only on one occasion was a group of acini observed that had distended ductules. This is in keeping with the idea that there is only a limited contralateral innervation of the midline region of the gland (Ginsborg and House, 1976). The mean size of the ductule in the stimulated side of glands was 12.81 ? 3.4 pm (N= 48) and that from the unstimulated side was 6.73 + 2.09 pm (N= 25). A t-test for unpaired samples indicates that this is a highly significant change (P
144
MAXWELL
Ahhreviationr
Ax BM C D
for /ig;gsres
Axon Basement membrane Cenrral cell Duct
GI GV P SD
Glial cell Granular vesicle Peripheral cell Septate desmosomes
Fig. 2. The salivary nerve of the cockroach. Note the two large axons that arc turrounded by glia. The arrows indicate several smaller axons within the basement membrane region of the nerve. x 7000.
Fig. 3. A 1pmthick section from a stimulated gland that was fixed during fluid secretion. An arrow indicates one of the many distended peripheral cell ductules. x 350. Fig. 4. A lprn thick section from the unstimulared side of the gland illustrated Fig. 3. In this case no distended peripheral cell ductules are visible. x 350.
in
Fig. 5. An electron mxrograph of a peripheral cell with a. distended duct& that was fixed during stimulation of the ipsilateral salivary nerve. Note the infoldings of the basal plasma membrane (arrow), the numerous mitochondria. the mlcrovilli surrounding the duct and (although not resolved at this magnification) the highly convoluted septate desmosomes that link the cells. x 7200.
146
MAXWELL
agranular vesicles occurs after stimulation of axons when compared with axons from the unstimulated sides of glands or with those from freshly fixed tissue. No such reduction in number could be found for the granular vesicles that are also present in these axons. A change in shape of the small agranular vesicles was also commonly detected after stimulation (e.g. Figs. 8, 10, 12, 14, 16). A previous study has shown that these vesicles normally come from a population of prolate rotary ellipsoids (Maxwell, 1978). Shapes of vesicles were compared by plotting the logarithm of the maximum diameter divided by the minimum diameter (measured at right angles to each other), as a histogram (Fig. 18). This method illustrates that the distributions for freshly fixed tissue (Fig. 18a) and for unstimulated controls (Fig.
18~) are similar, thus suggesting that the vesicles are a similar shape. However, vesicles from stimulated axons have a narrower distribution tending towards zero (Fig. 18b). This suggests that these vesicles are more circular in their profiles than those from unstimulated tissue. No changes could be detected in type B axons after electrical stimulation. Discussion Electron microscopy has revealed the existence of a nerve that is associated with the reservoir duct of the cockroach salivary apparatus. Electrophysiological (Ginsborg and House, 1976), secretory (Smith and House, 1977) and methylene blue (BowserRiley, 1978) studies all indicate that this is one of the paired salivary nerves which
Fig. 6. An axon from the unstimulated control taneously with the stimulated side. x 45,000. Fig. 7. An axon that was stimulated small vesicles (arrow). x 52,500. Fig. 8. As a result of stimulation around a dense membrane structure present (asterisk). x 45,000.
side of a gland that was fixed simul-
for several minutes.
Note the aggregation
of
the small vesicles of this axon have aggregated (arrow). An invagination of the axolemma is also
Fig. 9. A stimulated axon illustrating the aggregation of the small vesicles (large arrow) and infoldings of the axolemma (small arrows). x 50,000. Figs. IO-17 illustrate stimulated axons (even numbers) from four different experiments and control axons from the unstimulated contralateral glands in each experiment (odd numbers). Fig. IO. A stimulated axon illustrating the aggregation around a dense membrane structure (arrow). x 60,000. Fig.
I I. Control.
vesicles
x 60,000.
Fig. 12. In this stimulated axon there is an aggregation and an infolding of the axolemma (asterisk). x 60,000. Fig. 13. Control.
of small agranular
of agranular
vesicles (arrow)
x 60,000.
Fig. 14. Two stimulated axons, In one axon aggregations of agranular vesicles may be seen (arrows). One of these aggregations is around a dense membrane structure (small arrow). Both axons contain small agranular vesicles that are unusually circular in their profiles. x 40,000. Fig. 15. Control.
x 40,000.
Fig. 16. This stimulated axon still contains a high density of small agranular but many of them are more circular in their profiles than usual. x 45,000. Fig. 17. Control.
x 45,000.
vesicles
STIMULATION
OF
COCKROACH
SALIVARY
Table 1. Means and standard deviations (S. D.) of groups of control, stimulated and freshly fixed axons which contain granular and agranular vesicles. The densities of vesicles in each group are expressed in terms of number of vesicles per square micron
Numbers
Means (vesicles/ ELM)
S.D.
Controls Stimulated Freshly fixed
19 27 24
78.1 45.2* 96 1
+35 +41 +54
Granular vesicles Controls Stimulated Freshly fixed
18 25 23
5.4 5.4 5.3
k4.3 k 6.5 + 5.4
Treatment
149
NERVES
apical membranes of the acini as the presence of ouabain in the bathing medium does not inhibit secretion. The model suggests that chloride ions passively accompany sodium and that an accumulation of sodium chloride occurs inside the acinar ductules. Such a mechanism could cause the
a
Agranular vesicles
. . *. *.
. . * . . .
rop;..i;.;..;...;
..... ..... ..._.
* Significantly different from both the control and the freshly fixed mean values at the 0.01 level using Student’s t-test. No other differences were found to be significant. b
arise from the suboesophageal ganglion and innervate the acini. It is not known if both the large and small axons are responsible for the innervation of the gland although if the large axons are allowed to degenerate some axonal degeneration profiles are also observed in the glands (Maxwell, unpublished observation). Evidence has been presented that demonstrates distension of peripheral cell ductules in response to electrical stimulation of the salivary nerves. Such a distension has also been observed with Nomarski optics in live peripheral cells during stimulation of this preparation (R. K. Smith, unpublished). This suggests that peripheral cells are responsible for the initiation of fluid secretion in response to nervous stimulation, as distension was not seen in unstimulated sides of the same glands. Smith and House ( 1980) have demonstrated that in the absence of sodium ions in the bathing medium, the cockroach salivary gland is unable to secrete in response to applied dopamine. They have developed a model of salivary secretion in this gland which depends upon the passive entry of sodium ions into the basal membranes of the cells and active pumping of sodium ions at the apical membranes in the ductules. Any ATP dependent pump must be situated on the
ii
2% ::, ~............_.____._,..,,._.__ .
r. . l
.
I f
.f
f
.
**
.
f,
.a *a
f *
**t. .ff.
0.
** . . (I*
1
0
Log Ratio
max/rnin
Fig. 18. Histograms of the logarithm of the maximum diameter divided by the minimum diameter of the small agranular vesicles observed in cockroach salivary gland axons. Fig. 18a is a plot of vesicle dimensions taken from several freshly fixed axons. Fig. 18b represents similar data collected from stimulated axons. Fig. 1%~ is data collected from unstimulated axons in the contralateral glands to Fig. 18b. Figs. 18a and 18c have similar distributions whereas Fig. 18b is narrower suggesting that vesicles have changed in shape as a result of stimulation.
150
movement of water into peripheral cell ductules and hence distension of them. Supporting evidence for this idea may be inferred from the structure of peripheral cells. First, spaces between the densely packed microvilli could provide small extracellular channels in which sodium chloride could be concentrated and thus a standing osmotic gradient set up. (Diamond and Tormey, 1966; Diamond, 1980). Secondly, peripheral cells are crammed with mitochondria suggesting that their energy requirements are high. Finally, invaginations of the peripheral cell basal plasma membrane could provide low-resistance routes for water transport. Such a function has been ascribed (Green, 1979) to similar structures in the Malpighian tubules of Arachnocampa luminosa. It seems improbable that an extracellular route of water transport is of much importance in this gland as the septate desmosomes which occupy these extracellular spaces have been described as ‘barriers to diffusion’ (NoirotTimothee and Noirot, 1980; Noirot-Timothee et al., 1978). No differences were detected between the central cells of stimulated and unstimulated glands. In the rat parotid gland it is possible to reduce the numbers of secretory granules as a consequence of stimulation of the sympathetic nerve supply (Garrett et al., 1979). This result suggests that the amylase containing secretory granules of the cockroach salivary gland are released by some mechanism that is independent of the salivary nerves. Ever since the observation of quanta1 release of neurotransmitter (e.g. Del Castillo and Katz, 1954) synaptic vesicles have been designated the morphological ‘packets’ of such quanta. Direct evidence for such a relationship has proved to be elusive, as few exocytotic profiles have been observed by conventional electron microscopal techniques (e.g. Osborne, 1977), although methods rapid freezing of stimulated involving tissue combined with freeze-fracture techniques have recently produced promising results (Heuser et al., 1979). Several authors have suggested that a reduction in numbers of synaptic vesicles after stimulation may imply release of transmitter; e.g. in the cat sympathetic ganglion (Pysh and Wiley, 1974); in axon varicosities of the mouse vas
MAXWELL
deferens (Basbaum and Heuser, 1979); in the hatchet fish medulla (Model et al., 1975); at the crayfish neuromuscular junction (Atwood and Lang, 1972) and at the locust neuromuscular junction (McKinlay and Usherwood, 1973; Botham et al., 1978). The present study also indicates that the small agranular vesicles of cockroach salivary gland axons are reduced in number as a consequence of electrical stimulation. In keeping with the observations of Basbaum and Heuser (1979) on aminergic axon varicosities in the mouse vas deferens, the large granular vesicles are not reduced in number. It may be suggested that the reduction in number of the small agranular vesicles represents a release of dopamine from the axons as other data indicate that these small vesicles contain dopamine (see Maxwell, 1980). The role of the larger granules is not known but the evidence would suggest that they are not influenced by neural stimulation. On several occasions vesicles in stimulated axons were seen to aggregate around dense membrane structures. Such structures have been infrequently observed in freshly fixed tissue (Maxwell, 1978). Previous authors have referred to these structures as ‘active zones’ (e.g. Rheuben and Kammer, 1980) and the increase in frequency of such zones in this study supports the idea that they are involved in transmitter release. Aggregation of vesicles after electrical stimulation of locust motor neurones has also been observed by McKinlay and Usherwood (1973). The tendency for vesicles to adopt circular profiles after stimulation is difficult to explain. Osborne (1977) also noted that there was an interchange between flattened and vesicles during stimulation of circular photoreceptor cells. Acknowledgements Financial support was provided by the Science Research Council. 1 wish to thank Drs K. P. Bland, C. R. House and K. L. Prince for their constructive comments upon this manuscript. Thanks are also due to Dr R. J. Houchin for her help with the Magiscan system, to Mr C. M. Warwick for preparation of the photographic plates and to Mrs L. Manson for typing the manuscript.
S-l IMULATION
OF
COCKROACH
SALIVARY
IS1
NERVES
References Arwooo. H. L.. and LANG, F. 1972. Synpatic vesicles: selective depletion in crayfish excitatory and inhlbitory axons. Science, 176. 1353-l 355. BASBAUM,C. B. and HEUSER,J. E. 1979. Morphological studies of stimulated adrenergic axon varicosities in the mouse vas deferens. J. Cell Biul.. 80, 310-325. BLAND, K. P. and HOUSE, C. R. 1971. Function of the salivary glands of the cockroach Nauphortn cinererr. .I. Insect Physiol., 17, 2084-2096. BOTHAM, R. P., BEADLE,D. J., HART, R. J., POTTER, C. and WILSON, R. G. 1978. Stimulation induced depletion of the synaptic vesicles in excitatory motor nerve terminals of the locust, Locmtrr migratooricz L. Exprrentia, 34, 207-208. BOWSER-RILEY, F. 1978. The salivary glands of the cockroach Nauphorta cinerea (Olivier): a study of 11s innervation by light and scanning electron microscopy. Cell Tim. Res., 187, 525-534. DEL CASTILLO, J. and KATZ, B. 1954. Quanta1 components of the end plate potential. J. Phwiol.. 124, 560.. 537. DIAMONU, J. M. and TORMEY, J. McD. 1966. Role of long extracellular channels in fluid transport across epithelia. Nature, Lond. 210, 817-820. DIAMOND,J. M. 1980. Osmotic water flow in leaky epithelia. J. Membrane Biol.. 51, 195-216. GARRETT.J. R., HARROP, T. J., KIDS, A. and THULIN, A. 1979. Nerve-induced secretory changes in salivary glands. In Nrrces and the gut (eds. F. B. Brooks and P. W. Evers), pp. 1440. C.B.S. Inc., New Jersey. GINSBORG, B. L., HOUSE, C. R. and SILKINSKY, E. M. 1974. Conductance changes associated with the secretory potential in the cockroach salivary gland. J. Physiol., 236, 723-731.
GINSBORG, B. L. and HOUSE, C. R. 1976. The response to nerve stimulation cinrrra
Olirier.
J. Physiol.,
ofthe salivary gland of N~oph~~<~fo
262, 447-487.
GREEN, L. F. B. 1979. Regional specialization in the Malpighian tubules of the New Zealand glow-worm Arachnocampa /uminosa (Diptera: Mycetophilidae). The structure and function of type I and II cells. Timur
& Cell, 11, 673-702.
HEUSER. J. E., REESE, T. S., DENNIS, M. J., JAN, Y., JAN, L. and EVANS. L. 1979. Synaptic vesicle exocytosis captured by quick freezing and correlated with transmitter release. J. Cell Biol., 81, 275-300. HOUSE, C. R. 1973. An electrophysiological study of neuroglandular transmission in the isolated salivary gland of the cockroach. J. Exp. Biol., 58, 29-43. House. C. R. 1977. Cockroach salivary gland: secretory epithelium with a dopaminergic innervation. In Trmsport of Jons and Water in Animal.\ (eds. B. L. Gupta, R. B. Moreton. J. L. Oschman and B. J. Wall). Academic Press, London. HOUSE, C. R. and SMITH, R. K. 1978. On the receptors involved in the nervous control of salivary secretion by Nauphorta cinrrea (Olivier). J. Physiol., 279, 457-471. HOUSE, C. R. and GINSBORG, B. L. 1979. Pharmacology of cockroach salivary secretion. Camp. Biochent. Ph.)..~ioI., 63C, l-6. MCKINLAY, R. G. and USHERWOOD,P. N. R. 1973. The role of synaptic vesicles in transmission at the insect nerve-muscle junction. Lifp Sri., 13, 1051-1056. MAXWELL, D. J. 1978. Axons associated with the salivary apparatus of the cockrocah Nauphwtcr cinrrtw Tissur & Cell, 10, 699-706. MAXWELL, D. J. 1980. Histochemical properties of axons associated with the salivary apparatus of the cockroach Nuuphorta cinrwa (Olivier). Tissue & Cell, 12, 703-71 I. MODEL, P. G., HIGHSTEIN, S. M. and BENNETT, M. V. L. 1975. Depletion of vesicles and fatigue of transmission at a vertebrate central synapse. Bruin Res., 45, 209-228. NOIRIOT-TIMOTHBE,C., SMITH, D. S.. CAYER. M. L. and NOIROT, C. 1978. Septate junctions in insects: comparison between intracellular and intramembranous structures. Tksue & Cell, 10, 125-136. NOIROT-TIMOTHBE.C. and NOIROP. C. 1980. Septate and scalariform junctions in arthropods. Jnr. Rrt. Cytol., 63, 97- 139. OSBURNE, M. P. f977. Role of vesicles with some observations on vertebrate sensory cells. In Swrapwv feds. G. A. Cottrell and P. N. R. Usherwood). Blackie, London and Glasgow. PYSH, J. J. and WILEY, R. G. 1974. Synaptic vesicle depletion and recovery in cat sympathetic ganglia electrically stimulated in vim. J. Cell Biol., 60, 365-374. RHEUBEN. M. B. and KAMMER,A. E. 1980. Comparison of slow larval and fast adult muscle innervated by the same motor neurone. J. exp. Biol., 84, 103-I 18. SMITH, R. K. and HOUSE, C. R. 1977. Fluid secretion by isolated cockroach salivary glands. Exprrienticr, 33, 1182-l 183.
SMITH, R. K. and HOUSE. C. R. 1980. Ion and water transport Mrmhrme Biol., 51, 325-346
by isolated
cockroach
salivary
glands. J.