- Email: [email protected]
The injury response of the neurones of Periplaneta Americana
TISSUE El" CELL 1970 2 (3) 387-398 Published by Ofiver 8" Boyd Edinburgh, Printed in Great Britain
D. YOUNG,*$ DOREEN E. ASHHURST, t§ and M. J. COHEN*H
THE I N J U R Y R E S P O N S E OF THE N E U R O N E S OF PERIPLA N E TA A M E R I C A NA ABSTRACT. Bilateral pairs of specific neurones in the metathoracic ganglion of the cockroach were used to study the ultrastructural changes which f o l l o w axon injury. The appearance of these matched pairs of cells was compared three days after the axons on one side of the ganglion only had been severed. The control or uninjured cell body appeared normal, but changes were observed in the neurones whose axon had been cut. These changes included a general increase in the number of ribosomes t h r o u g h o u t the cytoplasm, dilatation of the cisternae of the endoplasmic reticulum and a reduction in size of the Golgi bodies. No ultrastructural features were seen which could correspond to the perinuclear ring of RNA observed histochemically with the light microscope. The possible nature of this RNA is discussed.
stein, 1957; Guthrie, 1962, 1967; Jacklet and Cohen, 1967). Variation in R N A staining properties is c o m m o n in both mammalian THE response of cockroach m o t o r neurones (Bodian, 1947) and insect neurones,.but in the to the severance of their axon has recently cockroach these differences are due to attracted attention, since in this insect a variation between specific cells and not bedense ring of R N A , 3 to 5/* wide, appears tween animals, since they are consistent in around the nucleus of the injured cell within the same cells from different ganglia. There is two days after injury and disappears again by an exact bilateral symmetry in the distribution the tenth day (Cohen and Jacklet, 1965); in of m o t o r neurones in cockroach ganglia and normal motor neurones the R N A is evenly the normal R N A staining properties of a dispersed throughout the cytoplasm (Ashbilateral pair of cells are very similar. Thus hurst, 1961; Wigglesworth, 1960). Later the injury response can be evaluated by regenerative changes include the outgrowth examining the bilateral pairs of cells in a of a new axon and the re-establishment of ganglion which has had the nerves severed on functional neuromuscular contacts (Bodenone sideonly; on the operated side, t h e r i n g o f R N A is present, on the other there is no ring. * Department of Zoology, Parks Road, Oxford, These differences in the staining properties England. are consistent and they have been employed -t"A.R.C. Unit of Insect Physiology, Department of Zoology, Parks Road, Oxford, England. together with methods for histological rePresent address: Research School of Biological Sciences, P.O. Box 475, Canberra City, A.C.T. construction, to construct three-dimensional 2601, Australia. cell maps of cockroach ganglia (Cohen and § Present address: Department of Anatomy, Jacklet, 1967; Young, 1969). In this way, it is Medical School Birmingham, 15, England. 11Present address: Department of Biology, Yale possible to deduce in which cell body the University, New Haven, Connecticut, U.S.A. axons in each nerve originate. Thus this response has proved to be a useful tool for Received 19 February 1970, 387 Introduction
388
YOUNG, ASHHURST & COHEN
many studies involving the central nervous system of cockroaches. The nature of the ultrastructural changes which accompany the formation of the perinuclear ring of R N A was not known, it was the aim of the experiments described here to investigate the changes in the injured neurone using the other member of the bilateral pair on the opposite side of the ganglion as the control cell. Since each neurone can be readily identified from the cell maps, a direct comparison of specific experimental and control cells from the same ganglion can be made and any differences in their structure may then be attributed to the injury of the axon of the experimental cell. The insect nervous system has, therefore, a distinct advantage for studies of this kind. Vertebrate ganglia cannot be mapped in the same way and so it is impossible to devise an experimental system in which two exactly corresponding cells are compared after the axon of one of them has been severed. Methods
All the experiments were performed on adult male cockroaches (Periplanetaamericana) between 1 and 4 days after their final moult. The cockroaches were first anaesthetized with CO2 and then a flap of cuticle on the ventral surface of the insect was lifted forward to expose the metathoracic ganglion. The peripheral nerves, that is, nerves 2 to 6, on the left side of the ganglion were sectioned approximately 0"25 mm from their point of emergence from the ganglion. The flap of cuticle was then replaced and sealed into position with low melting point dental wax. A solution of 0-025% benzachonium chloride (Roccal) was used as a local antiseptic on the external surface of the cuticle both before and after the operations. Sterile instruments were used throughout the procedure. The ganglia were fixed 3 days after the nerves were sectioned. The cockroaches were first anaesthetized with CO2. The flap of cuticle was then raised and the ganglion flooded with fixative. The ganglion was then cut longitudinally, that is into experimental
and control halves. The two parts of the ganglion were then removed from the cockroach and carried through the subsequent procedures separately. The fixative used was 1~ osmium tetroxide in veronal acetate buffer, pH 7.3 7-4, with 0.1 ml of a solution containing 1% calcium chloride added to each 10 ml of fixative. Fixation was carried out at either 4°C or room temperature; the difference in temperature appeared to have little effect on the fixation. A solution of 5% glutaraldehyde in cacodylate buffer, pH 7"4, followed by postfixation in 1% aqueous osmium tetroxide was also used as a fixative, but it was found to give very poor preservation with this material. The halved ganglia were fixed for approximately 2 hours in the buffered osmium fixative, and then transferred directly to 70% ethanol. They were subsequently dehydrated in graded ethanals, then passed through propylene oxide and embedded in araldite. The half ganglia were embedded in rectangular moulds according to the method described by Cheney and Ashhurst (1967). This enabled the half ganglia to be oriented accurately in the blocks. Since the neurones in this ganglion had been accurately mapped, and since certain neurones produced stronger rings of R N A than others (Cohen and Jacklet 1967), it was decided to locate particular cells in both the experimental and control halves of each ganglion. In order to locate the corresponding cells, the blocks were trimmed carefully and 1 ~ sections were taken at intervals. These were stained with toluidine blue and examined to ascertain the position reached in the half ganglion. When the level of the appropriate cell had been reached, the block was further trimmed and thin sections were cut for examination in the electron microscope. It should be emphasized that this procedure is extremely lengthy and there is considerable unavoidable spoilage of material involved in getting the paired cells from one ganglion. The sections were stained in either lead citrate or in 10% uranyl acetate in methanol, followed by lead citrate. They were examined in an A.E.I. EM6B electron microscope. TISSUE 8- CELL 1970 2 (3)
I N J U R Y RESPONSE OF N E U R O N E S Results The arrangement of the neurones within the metathoracic ganglion of P. americana is known in detail from the maps published by Cohen and Jacklet (1967). Matched pairs of cells, one from the uninjured, or control, side of the ganglion, the other from the injured, or experimental, side were examined in the electron microscope. The -previous studies had indicated that the ring of RNA around the nucleus is more pronounced in certain neurones; for this reason, most of the observations were based on cells 18 and 19 (see map, Cohen and Jacklet, 1967). In neurones 18 and 19, there is aparticularly clear differentiation between the perinuclear ring of R N A and the peripheral cytoplasm when the cells are stained by the pyronin-malachite green technique. An additional point of technical significance is that the identification of particular cells from sections is not easy, but in the case of cells 18 and 19, it is facilitated by their position immediately beneath the connective tissue sheath which surrounds the ganglion. The large motor neurones of Periplaneta are about 50 to 60t~ in diameter, with a nucleus of about 25 to 30~ in diameter. Previous electron microscope studies (Hess, 1958; Smith and Treherne, 1963) have indicated that the cytoplasm of the neurones contains numerous scattered inclusions; the control neurones fit their descriptions. The endoplasmic reticulum consists of randomly scattered, small tubular cisternae, each surrounded by a large number of ribosomes, some of which are in the form of organised groups, or polyribosomes (Figs. 5, 6). The distribution and variation in the size of these aggregations of endoplasmic reticulum can be seen in the distribution diagram (Fig. 2). The Golgi bodies, or dictyosomes, are large and crescent shaped and often have a large lysosome associated with them (Fig. 7). The mitochondria are very numerous and a few small microtubules are scattered in the cytoplasm. It was expected that a comparison of the fine structure of the control and experiTISSUE & CELL 1970 2 (3)
389 mental neurones might reveal differences in the endoplasmic reticulum of the perinuclear regions of the cells, which would reflect the presence of the ring of RNA observed by histochemical tests in the experimental cells (Cohen and Jacklet, 1965, 1967). The most obvious way in which this increased concentration of R N A might be shown would be in a marked increase in the amount of endoplasmic reticulum in a band up to 5~ wide around the nucleus, but no evidence for the presence of such a discrete ring of endoplasmic reticulum around the nucleus was seen (compare Figs. 3, 5). It appeared, however, that there might be a general increase in the amount of endoplasmic retieulum throughout the injured cell. In order to test this possibility, the distribution of the endoplasmic reticulum in the cytoplasm of four matched pairs of experimental and control cells was determined. As these cells are so large, it was necessary to make montages. For each cell, a montage of about ¼ of its total area was made; this included the cytoplasm from the nucleus to the cell membrane. The areas occupied by the endoplasmic reticulum were compared directly by making diagrams of these areas on tracing paper placed over the montages, This was necessary since even at the magnification of the prints making up the montages (17,000), it is difficult to distinguish the Golgi bodies from the endoplasmic reticulum, except on close observation. The maps, therefore, give a clear and direct comparison. No attempt was made to distinguish the ribosomes and cisternae in these maps; the total area of the endoplasmic reticulum was shaded. When the maps of the endoplasmic reticulum of the control and experimental cells are compared (Figs. 1, 2), it is at once apparent that there is an overall increase in the amount of endoplasmic reticulum in the experimental as compared to the control cell. This increase is spread uniformly throughout the entire cytoplasm; there is no suggestion of a discrete accumulation of endoplasmic reticulum within 5 ~ of the nucleus (1"2 cms. in Fig. 1). Hence, there is no "ring" of endoplasmic reticulum around
YOUNG, ASHHURST & COHEN
390 the nucleus which would correspond to the "ring" of R N A observed in the histochemical study. Small areas of the cytoplasm of the control and experimental cells can be seen at higher magnification in Figs. 3 and 5. The overall increase in the amount of endoplasmic
reticulum in the experimental cells is again clearly seen. The two components of the endoplasmic reticulum, the ribosomes and cisternae appear to increase in the same proportions, so that the ratio of the number of ribosomes to the number of cisternae is similar in both the control and experimental
Fig. 1. Diagram to show the distribution of endoplasmic reticulum and free ribosomes in a neurone from the experimental side of a metathoracic ganglion 3 days after the axon had been severed. The diagram was made from a montage of the neurone which contained 26 micrographs at a magnification of 17,000, The nucleus (N) and cell membrane ( C M ) are drawn to indicate the dimensions of the cell. The upper and right hand sides are cut off by a grid bar. The diagram illustrates the abundance of large masses of endoplasmie reticulum in the experimental neurones : An area 1.2 cms. wide around the nucleus corresponds to the position of the ring of RNA. x 2,500. TJSSUE ~ CELL 1970 2. (3)
I N J U R Y RESPONSE OF N E U R O N E S cells. One feature of the endoplasmic reticulum of the experimental cells is noteworthy; the cisternae are dilated into wide tubules, or, in some cases, vesicles (Fig. 4), whereas in the control ceils, the cisternae are quite narrow tubules (Fig. 6). This may indicate that the experimental ceils are actively engaged in a high rate of protein synthesis at this time. The majority of the ribosomes in both the control and experimental cells are at some distance from the cisternae, but they are
391 usually arranged in clusters, or polyribosomes. The Golgi complexes of the control cells are large dictyosomes with long flattened lamellae (Figs. 5, 7), but in the experimental cells, the complexes are nmch smaller and the lameIlae are slightly swollen or dilated (Figs. 3, 8). This difference may indicate an increase in the activity of the Golgi complexes of the experimental cells, correlated with the overall increase in protein secretion. There are no
Fig. 2. Diagram to show the distribution of endoplasmic reticulum and free ribosomes in a neurone from the control side of the same metathoracic ganglion as the cell in Fig. 1. The diagram was made from a montage of 19 micrographs. The nucleus (N) and cell membrane (CM) are indicated. The shaded areas, that is the endoplasmic reticulum, occupy less of the cytoplasm than in the experimental cell in Fig. 1. × 2,500. TISSUE 8 CELL 1970 2 (3)
~ , ~ i } ~ ¸¸,
ii I i~iiii:,~:{i¸~
~i~ii!i~ii~~,~I!":,'~i
,i ¸
i:i¸¸~II~ii,il¸~Iii}
YOUNG, A S H H U R S T & C O H E N
394 obvious differences in the lysosomes or mitochondria of the control and experimental cells, but no attempt was made to ascertain any changes in their size and number. Discussion
This analysis of the changes which follow axon injury in cockroach neurones is concerned mainly with the rough endoplasmic reticulum, since the initial aim of the experiments was to ascertain whether there is an ultrastructural change which can be correlated with the perinuclear ring of RNA seen after histochemical tests (Cohen, 1967; Cohen and Jacklet, 1965, 1967); it was thought that the 'ring' might be produced by an accumulation of rough endoplasmic reticulum around the nucleus, formed as a
direct response to axon injury. The findings reported here suggest that while there is an increase in the amount of rough endoplasmic reticulum in the neurone, this increase is spread evenly throughout the cell body and is not confined to the perinuclear ring. Thus the RNA ring cannot be attributed to a localised increase in the rough endoplasmic reticulum around the nucleus. A similar increase in endoplasmic reticulum in the neurones of cockroach giant fibres 5 days after nerve section has been reported by Farley and Milburn (1969). It appears that the formation of a perinuclear ring of R N A in response to the injury of an axon is not peculiar to the cockroaches. A similar perinuclear ring of RNA has been seen after axon injury in the cricket, A cheta domesticus (Edwards, personal
Fig, 3. Electron micrograph to show an area of the cytoplasm near the nucleus (N) of the experimental cell in Fig. 1. Irregular cisternae (C) and large areas of ribosomes (R) forming the endop[asmic reticulum occupy most of the cytoplasm. These often form very large aggregations. Mitochondria (M), and few small lysosomes (L) and Golgicomplexes (G) can also be seen in this area of the cytoplasm. The ring of RNA would be u p t o 12 cms. wide in this micrograph, x 23,000. Fig, 4. A small area of the endoplasmic reticulum of an experimental cell in which the dilatations of the cisternae (C) are clearly visible. In onearea (arrow), the section is glancing over the surface of a cisterna and ribosomes arranged in a spiral pattern can be seen lying on the membrane, x 51,000. Fig, 5. Electron micrograph to show an area of the cytoplasm near the nucleus (N) of the control cell in Fig. 2. Small irregular cisternae (C) and large areas of ribosomes (R) forming the endopiasmic reticulum are present, but these aggregations do not occupy as much of the cytoplasm as in the experimental cell in Fig. 3. Mitochondria (M), lysosomes (L) and a Golgi complex (G) are also present in this area of the cytoplasm. × 23,000. Fig, 6. A small area of endoplasmic reticulum from a control cell, The cisternae (C) are flattened and smaller than in the experimental cell shown in Fig. 4. × 51,000.
Fig. 7. A Golgi complex from a control cell. This has the typical dictyosome appearance commonly seen in the Golgi complexes of invertebrate neurones. The lamellae (LA) are long and flattened. Numerous small vesicles (V) are present on both sides of the complex. A large lysosome (L) is near the complex. × 49,000. Fig. 8, A Golgi complex from an experimental ceil. The lamellae (LA) are short and there are many vesicles (V) around them. The whole complex is smaller than in the control cell. × 62,000. TISSUE 8- CELL 1970 2 (3)
J~ 7~
zp
71
s~
r~
r~
r~
396 communication) and the mollusc, Anodonta cygnea (Sal~inki and Gubicza, 1969). In the crayfish and locust, Schistocerca gregaria, however, similar experimental procedures failed to induce the formation of a ring of R N A around the nucleus (Kennedy, personal communication; Young, unpublished observations). The explanation of this variation among the arthropods is not clear. All these species show degeneration and subsequent regeneration of severed axons similar to that in the cockroach, though the time course of these events is somewhat variable (Boulton, 1969; Edwards and Sahota, 1967; Hoy, 1969; Jacklet and Cohen, 1967; Usherwood, 1963); in the crayfish and locust regeneration occurs at both cut stumps leading to their fusion. A perinuclear ring of RNA has not been described in any vertebrate neurones. While the perinuclear ring of R N A in cockroach neurones cannot be correlated with a massing of ribosomes in the region, it is clearly a real phenomenon. Its explanation must lie in an accumulation, not of ribosomal RNA, but of a non-structural form. It may represent an outpouring of messenger RNA; such an increase of RNA passing from the nucleus might be expected in a cell about to engage in a period of very active protein synthesis. While changes occur in the distribution of the Nissl substance and hence the RNA in vertebrate neurones during chromatolysis, the total amount of R N A remains constant (BrattgS.rd, Edstr6m and Hyd6n, 1957; Watson, 1965). The cockroach may be peculiar in this respect as there appears to be an increase in RNA after the formation of the ring; further studies to elucidate this are necessary. The neurones of both insects and vertebrates have many similar ultrastructural features. The rough endoplasmic reticulum is scattered throughout the cell body, forming localised masses which consist of small flattened cisternae surrounded by large aggregations of ribosomes, many of which are free from the cisternal membrane; these Clumps of rough endoplasmic reticulum correspond to the Nissl substance of the light
YOUNG, ASHHURST & COHEN microscopists (Hyd6n, 1960; Palay, 1964; Smith and Treherne, 1963). Rosettes of free ribosomes, or polyribosomes, are typical of neurones (Palay, 1964; Ekholm and Hyd6n, 1965). Among the clumps of endoplasmic reticulum, many Golgi complexes, lysosomes and mitochondria are found. There is no precise pattern of organization of the organelles within the cytoplasm of neurones. The response to axon injury by vertebrate neurones is termed chromatolysis. Early histological studies showed a dispersion and movement of the Nissl substance to the periphery of the cell while the nucleus becomes eccentric (Bodian, 1947; BrattgS.rd, Edstr6m and Hyd6n, 1957). Ultrastructural studies of this process have confirmed the early studies. The tight clumps of rough endoplasmic reticulum are dispersed, and the cisternae and polyribosomes move peripherally (Smith, 1961; Pannese, 1963; Porter and Bowers 1963; Takano, 1964; Holtzman, Novikoff and Villaverde, 1967; Kirkpatrick, 1968), though Kirkpatrick also reports the presence of some perinuclear endoplasmic reticulum. Various changes in the numbers of ribosomes and cisternae have been suggested by several workers (Pannese, 1963; Porter and Bowers, 1963; Takano, 1964), but no quantitative studies have been made. The dispersion of rough endoplasmic reticulum is due to the scattering of the polyribosomes and the dilation of the cisternae to form, in extreme instances, rounded vacuoles (Pannese, 1963; Takano, 1964; Kirkpatrick, 1968). In some instances the nucleus becomes indented with an accumulation of endoplasmic reticulum in the indentation (Kirkpatrick, 1968). At the same time the Golgi lamellae appear swollen and there is an increase in the number of associated vacuoles (Pannese, 1963; Porter and Bowers, 1963; Takano, 1964), though there is some dispute as to whether this organelle increases in size; other authors report no change in the Golgi apparatus (Smith, 1961). A general increase in the number of both mitochondria and lysosomes has been observed (Hudson and Hartmann, 1961; Hudson, Lazarow and Hartman, 196I; Porter and Bowers, 1963; TISSUE & CELL 1970 2 (3)
INJURY
RESPONSE
397
OF NEURONES
Holtzman, N o v i k o f f and Villaverde, 1967) and there are reports o f both an increase and a decrease in the size of the mitochondria (Hudson, Lazarow and Hartman, 1961; Smith, 1961; Porter and Bowers, 1963; Takano, 1964). An increase in the number of neurofibrils has also been recorded (Kirkpatrick, 1968). These changes are at a peak 7 to 14 days after injury, after which time there is a gradual return to the normal condition. At first glance the cockroach neurone appears to react to injury very differently. There is no movement of the endoplasmic reticulum to the periphery of the cell and both ribosomes and cisternae appear to increase, rather than decrease, in overall numbers. The more detailed changes do, however, compare very closely with those observed in vertebrate neurones after injury. The cisternae of the endoplasmic reticulum are dilated in both. The Golgi lamellae swell and there is a general indication of increased activity. N o attempt was made to assess any changes in the number or size of either the mitochondria or lysosomes. Thus the responses of insect and vertebrate neurones to injury have much in c o m m o n ; it would be of interest to see the changes which occur in insect neurones in which no ring of R N A is formed. The changed appearance of the organdies after injury is compatible with the increased secretory activity which must occur as a
prerequisite for the regeneration of the axon. The dilated cisternae of the endoplasmic reticulum are comparable with those of actively secreting vertebrate and insect fibroblasts (Ross and Benditt, 1961; Ashhurst, 1964). The changed appearance of the Golgi complex has features in c o m m o n with that of other actively secreting cells; that the Golgi apparatus has been implicated in the formation of new cell membrane may also be pertinent in this context (Hicks, 1966). This suggests that the changes seen in the cockroach neurone three days after axon injury are true regenerative changes.
Acknowledgements We are indebted to Professor J. W. S. Pringle, F.R.S. for providing us with facilities in his Department. We should like to thank Dr. J. W. Jacklet for his assistance with some of the operations and dissections. Our thanks are also due to Mrs. B. M. Luke and Miss C. Williams for their help in m a k i n g the montages and diagrams, and to M r . J. S. Haywood and Mr. D. D. Turner for doing the photography. M. J. C. was a Guggenheim Fellow on leave from the University of Oregon when this study was initiated. The w o r k was supported in part by U.S. Public Health Service Grants N B 01624 and N B 08996 01 to M. J. C.
References ASHHURST, D. E. 1961. The cytology and histochemistry of the neurones of Periplaneta americana. Q. J1. rnicrosc. Sci., 102, 399-405. ASHHURST,D. E. 1964. Fibrillogenesis in the wax-moth, Galleria melone[la. Q. Jl. microsc. Sci., 105, 391-403. BODENSTEIN,D. 1957. Studies on nerve regeneration in Periplaneta americana. J. exp. Zool., 136, 89 115. BODIAN, D. 1947. Nucleic acid in nerve cell regeneration. Syrup. Soc. exp. Biol., 1, 163-178. BOULTON, P. S. 1969. Degeneration and regeneration in the insect nervous system. L Z.Zellfbrsch., 101, 98-118. BRATTGARD,S. O., EDSTROM, J. E. and HVD~N, H. 1957. The chemical changes in regenerating neurons. J. Neurochem., 1, 316-325. CnEN~V, R. A, and ASHHURST,D. E. 1967. A method for the orientation of tissues in epoxy-resin blocks and a design for a rotating stage for holding blocks during trimming. J. R. microsc. Soc. 86, 441-444. TISSUE 8- CELL 1970 2 (3)
398
YOUNG, ASHHURST
& COHEN
COHEN, M. J. 1967. Correlations between structure, function and RNA metabolism in central neurons of insects. In Invertebrate Nervous Systems, ed. C. A. G. Wiersma, pp. 65 78. Chicago: University of Chicago Press. COHEN, M. J. and JACKLET,J. W. 1965. Neurons of insects: RNA changes during injury and regeneration. Science, 148, 1237 1239. COHEN, M. J. and JACKLET,J. W. 1967. The functional organisation of motor neurons in an insect ganglion. Phil. Trans. R. Soc., (B), 252, 561-569. EDWARDS, J. S. and SAHOTA,T. S. 1967. Regeneration of a sensory system: the formation of central connections by normal and transplanted cerci of the house cricket, Acheta domesticus. J. exp. Zoo/., 166, 387-396. EKHOLM, R. and HYD~N, H. 1965. Polyribosomes in nerve cells. J. Uhrastruct. Res., 12, 239-240. FARLEY, R. D. and M~LBURN, N. S. 1969. Structure and function of the giant fibre system in the cockroach, Periplaneta americana. J. Insect Physiol., 15, 457-476. GUTHRIE, D. M. 1962. Regenerative growth in insect nerve axons. J. Insect Physiol., 8, 79-92. GUTHRIE, D. M. 1967. The regeneration of motor axons in an insect. J. Insect Physiol., 13, 1593-1611. HESS, A. 1958. The fine structure of nerve cells and fibres, neuroglia, and sheaths of the ganglion chain in the cockroach (Periplaneta americana). J. biophys, biochem. Cytol., 4, 731-742. HICKS, R. M. 1966. The function of the Golgi complex in transitional epithelium. Synthesis of the thick cell membrane. J. Cell Biol., 30, 623-643. HOLTZMAN,E,, NOVIKOFF,A. B. and VILLAVERDE,H. 1967. Lysosomes and GERL in normal and chromotolytic neurons of the rat ganglion nodosum. Y. Cell Biol., 33, 419 435. HoY, R. R. 1969. Degeneration and regeneration in abdominal flexor motor neurones in the crayfish. Y. exp. Zool., 172, 219 232. HUDSON, G. and HARTMANN, J. F. 1961. The relationship between dense bodies and mitochondria in motor neurones. Z, Zel[/brsch., 54, 147-157. HUDSON, G., LAZAROW, a . and HARTMANN, J. F. 1961. A qualitative electron microscope study of mitochondria in motor neurones following axon section. Exp. Cell Res., 24, 440-456. HYD6N, H. 1960. The neuron. In The Cell, Vol. IV, eds. J. Brachet and A. E. Mirsky, pp. 215 323. London & N.Y.: Academic Press. JACKLET,J. W. and COHEN, M. J. 1967. Nerve regeneration: correlation of electrical and behavioural events. Science, 156, 1640 1643. KIRKPATRlCK, J. B. 1968. Chromatolysis in the hypoglossal nucleus of the rat: an electron microscopic analysis, d. comp. Neurol., 132, 189-212. PALAY, S. L. 1964. The structural basis for neural action. In Brain function, Vol. II: RNA and brain Jhnction; memory and learning, ed. M. A. B. Brazier, pp. 69-108. Los Angeles: University of California Press. PANNESE, E. 1963. Investigations on the ultrastructural changes of the spinal ganglion neurons in the course of axon regeneration and cell hypertrophy. Z. Zell/brsch., 60, 711-740. PORTER, K. R. and BOWER, M. B. 1963. A study of chromatolysis in motor neurons of the frog Rana pipiens. Y. Cell Biol., 19, 56A-57A. Ross, R. and BEND~TT, E. P. 1961. Wound healing and collagen formation 1. Sequential changes in components of guinea pig skin wounds observed in the electron microscope. Y. hiophys, biochem. CytoI., 11, 677-700. SAL,~NKI, J. and GUB~CZA,A. 1969. Histochemical evidence of direct neuronal pathways in Anodonta cygnea L. Acta biol. Acad. Sci. hung., 20, 219~34. SMITH, D. S. and TREHERNE, J. E. 1963. Functional aspects of the organisation of the insect nervous system. In Advances in Insect Physiology, eds. J. E. Treherne, J. W. L. Beament and V. B. Wigglesworth, Vol, l, pp. 40l 484. SMITH, K. R. 1961. The fine structure of neurons of dorsal root ganglia after stimulating or cutting the sciatic nerve. J. cornp. Neurol., 116, 103 115. TAKANO, ]. 1964. Electron microscopic studies on retrograde chromatolysis in the hypoglossal nucleus and changes in the hypoglossal nerve, following its severance and ligation. Okajimas Fol. Anat. Jap., 40, 1-69. USUERWOOD,P. N. R. 1963. Response of insect muscle to denervation. I1. Changes in neuromuscular transmission. J. ins. Physiol., 9, 811 825. WATSON,W. E. 1965. An autoradiographic study of the incorporation of nucleic acid precursors by neurones and glia during nerve regeneration. J. Physiol., 180, 741 753. WIGGLESW0RTH, V. B. 1960. Axon structure and the dictyosomes (Golgi bodies) in the neurones of the cockroach, Periplaneta americana. Q. J1 microsc. Sci., 101, 381 388. YOUNG D. 1969. The motor neurons of the mesothoracic ganglion of Periplaneta americana. J. Insect Physiol., 15, 1175-1179. TISSUE 8" CELL 1970 2 (3)
D. YOUNG,*$ DOREEN E. ASHHURST, t§ and M. J. COHEN*H
THE I N J U R Y R E S P O N S E OF THE N E U R O N E S OF PERIPLA N E TA A M E R I C A NA ABSTRACT. Bilateral pairs of specific neurones in the metathoracic ganglion of the cockroach were used to study the ultrastructural changes which f o l l o w axon injury. The appearance of these matched pairs of cells was compared three days after the axons on one side of the ganglion only had been severed. The control or uninjured cell body appeared normal, but changes were observed in the neurones whose axon had been cut. These changes included a general increase in the number of ribosomes t h r o u g h o u t the cytoplasm, dilatation of the cisternae of the endoplasmic reticulum and a reduction in size of the Golgi bodies. No ultrastructural features were seen which could correspond to the perinuclear ring of RNA observed histochemically with the light microscope. The possible nature of this RNA is discussed.
stein, 1957; Guthrie, 1962, 1967; Jacklet and Cohen, 1967). Variation in R N A staining properties is c o m m o n in both mammalian THE response of cockroach m o t o r neurones (Bodian, 1947) and insect neurones,.but in the to the severance of their axon has recently cockroach these differences are due to attracted attention, since in this insect a variation between specific cells and not bedense ring of R N A , 3 to 5/* wide, appears tween animals, since they are consistent in around the nucleus of the injured cell within the same cells from different ganglia. There is two days after injury and disappears again by an exact bilateral symmetry in the distribution the tenth day (Cohen and Jacklet, 1965); in of m o t o r neurones in cockroach ganglia and normal motor neurones the R N A is evenly the normal R N A staining properties of a dispersed throughout the cytoplasm (Ashbilateral pair of cells are very similar. Thus hurst, 1961; Wigglesworth, 1960). Later the injury response can be evaluated by regenerative changes include the outgrowth examining the bilateral pairs of cells in a of a new axon and the re-establishment of ganglion which has had the nerves severed on functional neuromuscular contacts (Bodenone sideonly; on the operated side, t h e r i n g o f R N A is present, on the other there is no ring. * Department of Zoology, Parks Road, Oxford, These differences in the staining properties England. are consistent and they have been employed -t"A.R.C. Unit of Insect Physiology, Department of Zoology, Parks Road, Oxford, England. together with methods for histological rePresent address: Research School of Biological Sciences, P.O. Box 475, Canberra City, A.C.T. construction, to construct three-dimensional 2601, Australia. cell maps of cockroach ganglia (Cohen and § Present address: Department of Anatomy, Jacklet, 1967; Young, 1969). In this way, it is Medical School Birmingham, 15, England. 11Present address: Department of Biology, Yale possible to deduce in which cell body the University, New Haven, Connecticut, U.S.A. axons in each nerve originate. Thus this response has proved to be a useful tool for Received 19 February 1970, 387 Introduction
388
YOUNG, ASHHURST & COHEN
many studies involving the central nervous system of cockroaches. The nature of the ultrastructural changes which accompany the formation of the perinuclear ring of R N A was not known, it was the aim of the experiments described here to investigate the changes in the injured neurone using the other member of the bilateral pair on the opposite side of the ganglion as the control cell. Since each neurone can be readily identified from the cell maps, a direct comparison of specific experimental and control cells from the same ganglion can be made and any differences in their structure may then be attributed to the injury of the axon of the experimental cell. The insect nervous system has, therefore, a distinct advantage for studies of this kind. Vertebrate ganglia cannot be mapped in the same way and so it is impossible to devise an experimental system in which two exactly corresponding cells are compared after the axon of one of them has been severed. Methods
All the experiments were performed on adult male cockroaches (Periplanetaamericana) between 1 and 4 days after their final moult. The cockroaches were first anaesthetized with CO2 and then a flap of cuticle on the ventral surface of the insect was lifted forward to expose the metathoracic ganglion. The peripheral nerves, that is, nerves 2 to 6, on the left side of the ganglion were sectioned approximately 0"25 mm from their point of emergence from the ganglion. The flap of cuticle was then replaced and sealed into position with low melting point dental wax. A solution of 0-025% benzachonium chloride (Roccal) was used as a local antiseptic on the external surface of the cuticle both before and after the operations. Sterile instruments were used throughout the procedure. The ganglia were fixed 3 days after the nerves were sectioned. The cockroaches were first anaesthetized with CO2. The flap of cuticle was then raised and the ganglion flooded with fixative. The ganglion was then cut longitudinally, that is into experimental
and control halves. The two parts of the ganglion were then removed from the cockroach and carried through the subsequent procedures separately. The fixative used was 1~ osmium tetroxide in veronal acetate buffer, pH 7.3 7-4, with 0.1 ml of a solution containing 1% calcium chloride added to each 10 ml of fixative. Fixation was carried out at either 4°C or room temperature; the difference in temperature appeared to have little effect on the fixation. A solution of 5% glutaraldehyde in cacodylate buffer, pH 7"4, followed by postfixation in 1% aqueous osmium tetroxide was also used as a fixative, but it was found to give very poor preservation with this material. The halved ganglia were fixed for approximately 2 hours in the buffered osmium fixative, and then transferred directly to 70% ethanol. They were subsequently dehydrated in graded ethanals, then passed through propylene oxide and embedded in araldite. The half ganglia were embedded in rectangular moulds according to the method described by Cheney and Ashhurst (1967). This enabled the half ganglia to be oriented accurately in the blocks. Since the neurones in this ganglion had been accurately mapped, and since certain neurones produced stronger rings of R N A than others (Cohen and Jacklet 1967), it was decided to locate particular cells in both the experimental and control halves of each ganglion. In order to locate the corresponding cells, the blocks were trimmed carefully and 1 ~ sections were taken at intervals. These were stained with toluidine blue and examined to ascertain the position reached in the half ganglion. When the level of the appropriate cell had been reached, the block was further trimmed and thin sections were cut for examination in the electron microscope. It should be emphasized that this procedure is extremely lengthy and there is considerable unavoidable spoilage of material involved in getting the paired cells from one ganglion. The sections were stained in either lead citrate or in 10% uranyl acetate in methanol, followed by lead citrate. They were examined in an A.E.I. EM6B electron microscope. TISSUE 8- CELL 1970 2 (3)
I N J U R Y RESPONSE OF N E U R O N E S Results The arrangement of the neurones within the metathoracic ganglion of P. americana is known in detail from the maps published by Cohen and Jacklet (1967). Matched pairs of cells, one from the uninjured, or control, side of the ganglion, the other from the injured, or experimental, side were examined in the electron microscope. The -previous studies had indicated that the ring of RNA around the nucleus is more pronounced in certain neurones; for this reason, most of the observations were based on cells 18 and 19 (see map, Cohen and Jacklet, 1967). In neurones 18 and 19, there is aparticularly clear differentiation between the perinuclear ring of R N A and the peripheral cytoplasm when the cells are stained by the pyronin-malachite green technique. An additional point of technical significance is that the identification of particular cells from sections is not easy, but in the case of cells 18 and 19, it is facilitated by their position immediately beneath the connective tissue sheath which surrounds the ganglion. The large motor neurones of Periplaneta are about 50 to 60t~ in diameter, with a nucleus of about 25 to 30~ in diameter. Previous electron microscope studies (Hess, 1958; Smith and Treherne, 1963) have indicated that the cytoplasm of the neurones contains numerous scattered inclusions; the control neurones fit their descriptions. The endoplasmic reticulum consists of randomly scattered, small tubular cisternae, each surrounded by a large number of ribosomes, some of which are in the form of organised groups, or polyribosomes (Figs. 5, 6). The distribution and variation in the size of these aggregations of endoplasmic reticulum can be seen in the distribution diagram (Fig. 2). The Golgi bodies, or dictyosomes, are large and crescent shaped and often have a large lysosome associated with them (Fig. 7). The mitochondria are very numerous and a few small microtubules are scattered in the cytoplasm. It was expected that a comparison of the fine structure of the control and experiTISSUE & CELL 1970 2 (3)
389 mental neurones might reveal differences in the endoplasmic reticulum of the perinuclear regions of the cells, which would reflect the presence of the ring of RNA observed by histochemical tests in the experimental cells (Cohen and Jacklet, 1965, 1967). The most obvious way in which this increased concentration of R N A might be shown would be in a marked increase in the amount of endoplasmic reticulum in a band up to 5~ wide around the nucleus, but no evidence for the presence of such a discrete ring of endoplasmic reticulum around the nucleus was seen (compare Figs. 3, 5). It appeared, however, that there might be a general increase in the amount of endoplasmic retieulum throughout the injured cell. In order to test this possibility, the distribution of the endoplasmic reticulum in the cytoplasm of four matched pairs of experimental and control cells was determined. As these cells are so large, it was necessary to make montages. For each cell, a montage of about ¼ of its total area was made; this included the cytoplasm from the nucleus to the cell membrane. The areas occupied by the endoplasmic reticulum were compared directly by making diagrams of these areas on tracing paper placed over the montages, This was necessary since even at the magnification of the prints making up the montages (17,000), it is difficult to distinguish the Golgi bodies from the endoplasmic reticulum, except on close observation. The maps, therefore, give a clear and direct comparison. No attempt was made to distinguish the ribosomes and cisternae in these maps; the total area of the endoplasmic reticulum was shaded. When the maps of the endoplasmic reticulum of the control and experimental cells are compared (Figs. 1, 2), it is at once apparent that there is an overall increase in the amount of endoplasmic reticulum in the experimental as compared to the control cell. This increase is spread uniformly throughout the entire cytoplasm; there is no suggestion of a discrete accumulation of endoplasmic reticulum within 5 ~ of the nucleus (1"2 cms. in Fig. 1). Hence, there is no "ring" of endoplasmic reticulum around
YOUNG, ASHHURST & COHEN
390 the nucleus which would correspond to the "ring" of R N A observed in the histochemical study. Small areas of the cytoplasm of the control and experimental cells can be seen at higher magnification in Figs. 3 and 5. The overall increase in the amount of endoplasmic
reticulum in the experimental cells is again clearly seen. The two components of the endoplasmic reticulum, the ribosomes and cisternae appear to increase in the same proportions, so that the ratio of the number of ribosomes to the number of cisternae is similar in both the control and experimental
Fig. 1. Diagram to show the distribution of endoplasmic reticulum and free ribosomes in a neurone from the experimental side of a metathoracic ganglion 3 days after the axon had been severed. The diagram was made from a montage of the neurone which contained 26 micrographs at a magnification of 17,000, The nucleus (N) and cell membrane ( C M ) are drawn to indicate the dimensions of the cell. The upper and right hand sides are cut off by a grid bar. The diagram illustrates the abundance of large masses of endoplasmie reticulum in the experimental neurones : An area 1.2 cms. wide around the nucleus corresponds to the position of the ring of RNA. x 2,500. TJSSUE ~ CELL 1970 2. (3)
I N J U R Y RESPONSE OF N E U R O N E S cells. One feature of the endoplasmic reticulum of the experimental cells is noteworthy; the cisternae are dilated into wide tubules, or, in some cases, vesicles (Fig. 4), whereas in the control ceils, the cisternae are quite narrow tubules (Fig. 6). This may indicate that the experimental ceils are actively engaged in a high rate of protein synthesis at this time. The majority of the ribosomes in both the control and experimental cells are at some distance from the cisternae, but they are
391 usually arranged in clusters, or polyribosomes. The Golgi complexes of the control cells are large dictyosomes with long flattened lamellae (Figs. 5, 7), but in the experimental cells, the complexes are nmch smaller and the lameIlae are slightly swollen or dilated (Figs. 3, 8). This difference may indicate an increase in the activity of the Golgi complexes of the experimental cells, correlated with the overall increase in protein secretion. There are no
Fig. 2. Diagram to show the distribution of endoplasmic reticulum and free ribosomes in a neurone from the control side of the same metathoracic ganglion as the cell in Fig. 1. The diagram was made from a montage of 19 micrographs. The nucleus (N) and cell membrane (CM) are indicated. The shaded areas, that is the endoplasmic reticulum, occupy less of the cytoplasm than in the experimental cell in Fig. 1. × 2,500. TISSUE 8 CELL 1970 2 (3)
~ , ~ i } ~ ¸¸,
ii I i~iiii:,~:{i¸~
~i~ii!i~ii~~,~I!":,'~i
,i ¸
i:i¸¸~II~ii,il¸~Iii}
YOUNG, A S H H U R S T & C O H E N
394 obvious differences in the lysosomes or mitochondria of the control and experimental cells, but no attempt was made to ascertain any changes in their size and number. Discussion
This analysis of the changes which follow axon injury in cockroach neurones is concerned mainly with the rough endoplasmic reticulum, since the initial aim of the experiments was to ascertain whether there is an ultrastructural change which can be correlated with the perinuclear ring of RNA seen after histochemical tests (Cohen, 1967; Cohen and Jacklet, 1965, 1967); it was thought that the 'ring' might be produced by an accumulation of rough endoplasmic reticulum around the nucleus, formed as a
direct response to axon injury. The findings reported here suggest that while there is an increase in the amount of rough endoplasmic reticulum in the neurone, this increase is spread evenly throughout the cell body and is not confined to the perinuclear ring. Thus the RNA ring cannot be attributed to a localised increase in the rough endoplasmic reticulum around the nucleus. A similar increase in endoplasmic reticulum in the neurones of cockroach giant fibres 5 days after nerve section has been reported by Farley and Milburn (1969). It appears that the formation of a perinuclear ring of R N A in response to the injury of an axon is not peculiar to the cockroaches. A similar perinuclear ring of RNA has been seen after axon injury in the cricket, A cheta domesticus (Edwards, personal
Fig, 3. Electron micrograph to show an area of the cytoplasm near the nucleus (N) of the experimental cell in Fig. 1. Irregular cisternae (C) and large areas of ribosomes (R) forming the endop[asmic reticulum occupy most of the cytoplasm. These often form very large aggregations. Mitochondria (M), and few small lysosomes (L) and Golgicomplexes (G) can also be seen in this area of the cytoplasm. The ring of RNA would be u p t o 12 cms. wide in this micrograph, x 23,000. Fig, 4. A small area of the endoplasmic reticulum of an experimental cell in which the dilatations of the cisternae (C) are clearly visible. In onearea (arrow), the section is glancing over the surface of a cisterna and ribosomes arranged in a spiral pattern can be seen lying on the membrane, x 51,000. Fig, 5. Electron micrograph to show an area of the cytoplasm near the nucleus (N) of the control cell in Fig. 2. Small irregular cisternae (C) and large areas of ribosomes (R) forming the endopiasmic reticulum are present, but these aggregations do not occupy as much of the cytoplasm as in the experimental cell in Fig. 3. Mitochondria (M), lysosomes (L) and a Golgi complex (G) are also present in this area of the cytoplasm. × 23,000. Fig, 6. A small area of endoplasmic reticulum from a control cell, The cisternae (C) are flattened and smaller than in the experimental cell shown in Fig. 4. × 51,000.
Fig. 7. A Golgi complex from a control cell. This has the typical dictyosome appearance commonly seen in the Golgi complexes of invertebrate neurones. The lamellae (LA) are long and flattened. Numerous small vesicles (V) are present on both sides of the complex. A large lysosome (L) is near the complex. × 49,000. Fig. 8, A Golgi complex from an experimental ceil. The lamellae (LA) are short and there are many vesicles (V) around them. The whole complex is smaller than in the control cell. × 62,000. TISSUE 8- CELL 1970 2 (3)
J~ 7~
zp
71
s~
r~
r~
r~
396 communication) and the mollusc, Anodonta cygnea (Sal~inki and Gubicza, 1969). In the crayfish and locust, Schistocerca gregaria, however, similar experimental procedures failed to induce the formation of a ring of R N A around the nucleus (Kennedy, personal communication; Young, unpublished observations). The explanation of this variation among the arthropods is not clear. All these species show degeneration and subsequent regeneration of severed axons similar to that in the cockroach, though the time course of these events is somewhat variable (Boulton, 1969; Edwards and Sahota, 1967; Hoy, 1969; Jacklet and Cohen, 1967; Usherwood, 1963); in the crayfish and locust regeneration occurs at both cut stumps leading to their fusion. A perinuclear ring of RNA has not been described in any vertebrate neurones. While the perinuclear ring of R N A in cockroach neurones cannot be correlated with a massing of ribosomes in the region, it is clearly a real phenomenon. Its explanation must lie in an accumulation, not of ribosomal RNA, but of a non-structural form. It may represent an outpouring of messenger RNA; such an increase of RNA passing from the nucleus might be expected in a cell about to engage in a period of very active protein synthesis. While changes occur in the distribution of the Nissl substance and hence the RNA in vertebrate neurones during chromatolysis, the total amount of R N A remains constant (BrattgS.rd, Edstr6m and Hyd6n, 1957; Watson, 1965). The cockroach may be peculiar in this respect as there appears to be an increase in RNA after the formation of the ring; further studies to elucidate this are necessary. The neurones of both insects and vertebrates have many similar ultrastructural features. The rough endoplasmic reticulum is scattered throughout the cell body, forming localised masses which consist of small flattened cisternae surrounded by large aggregations of ribosomes, many of which are free from the cisternal membrane; these Clumps of rough endoplasmic reticulum correspond to the Nissl substance of the light
YOUNG, ASHHURST & COHEN microscopists (Hyd6n, 1960; Palay, 1964; Smith and Treherne, 1963). Rosettes of free ribosomes, or polyribosomes, are typical of neurones (Palay, 1964; Ekholm and Hyd6n, 1965). Among the clumps of endoplasmic reticulum, many Golgi complexes, lysosomes and mitochondria are found. There is no precise pattern of organization of the organelles within the cytoplasm of neurones. The response to axon injury by vertebrate neurones is termed chromatolysis. Early histological studies showed a dispersion and movement of the Nissl substance to the periphery of the cell while the nucleus becomes eccentric (Bodian, 1947; BrattgS.rd, Edstr6m and Hyd6n, 1957). Ultrastructural studies of this process have confirmed the early studies. The tight clumps of rough endoplasmic reticulum are dispersed, and the cisternae and polyribosomes move peripherally (Smith, 1961; Pannese, 1963; Porter and Bowers 1963; Takano, 1964; Holtzman, Novikoff and Villaverde, 1967; Kirkpatrick, 1968), though Kirkpatrick also reports the presence of some perinuclear endoplasmic reticulum. Various changes in the numbers of ribosomes and cisternae have been suggested by several workers (Pannese, 1963; Porter and Bowers, 1963; Takano, 1964), but no quantitative studies have been made. The dispersion of rough endoplasmic reticulum is due to the scattering of the polyribosomes and the dilation of the cisternae to form, in extreme instances, rounded vacuoles (Pannese, 1963; Takano, 1964; Kirkpatrick, 1968). In some instances the nucleus becomes indented with an accumulation of endoplasmic reticulum in the indentation (Kirkpatrick, 1968). At the same time the Golgi lamellae appear swollen and there is an increase in the number of associated vacuoles (Pannese, 1963; Porter and Bowers, 1963; Takano, 1964), though there is some dispute as to whether this organelle increases in size; other authors report no change in the Golgi apparatus (Smith, 1961). A general increase in the number of both mitochondria and lysosomes has been observed (Hudson and Hartmann, 1961; Hudson, Lazarow and Hartman, 196I; Porter and Bowers, 1963; TISSUE & CELL 1970 2 (3)
INJURY
RESPONSE
397
OF NEURONES
Holtzman, N o v i k o f f and Villaverde, 1967) and there are reports o f both an increase and a decrease in the size of the mitochondria (Hudson, Lazarow and Hartman, 1961; Smith, 1961; Porter and Bowers, 1963; Takano, 1964). An increase in the number of neurofibrils has also been recorded (Kirkpatrick, 1968). These changes are at a peak 7 to 14 days after injury, after which time there is a gradual return to the normal condition. At first glance the cockroach neurone appears to react to injury very differently. There is no movement of the endoplasmic reticulum to the periphery of the cell and both ribosomes and cisternae appear to increase, rather than decrease, in overall numbers. The more detailed changes do, however, compare very closely with those observed in vertebrate neurones after injury. The cisternae of the endoplasmic reticulum are dilated in both. The Golgi lamellae swell and there is a general indication of increased activity. N o attempt was made to assess any changes in the number or size of either the mitochondria or lysosomes. Thus the responses of insect and vertebrate neurones to injury have much in c o m m o n ; it would be of interest to see the changes which occur in insect neurones in which no ring of R N A is formed. The changed appearance of the organdies after injury is compatible with the increased secretory activity which must occur as a
prerequisite for the regeneration of the axon. The dilated cisternae of the endoplasmic reticulum are comparable with those of actively secreting vertebrate and insect fibroblasts (Ross and Benditt, 1961; Ashhurst, 1964). The changed appearance of the Golgi complex has features in c o m m o n with that of other actively secreting cells; that the Golgi apparatus has been implicated in the formation of new cell membrane may also be pertinent in this context (Hicks, 1966). This suggests that the changes seen in the cockroach neurone three days after axon injury are true regenerative changes.
Acknowledgements We are indebted to Professor J. W. S. Pringle, F.R.S. for providing us with facilities in his Department. We should like to thank Dr. J. W. Jacklet for his assistance with some of the operations and dissections. Our thanks are also due to Mrs. B. M. Luke and Miss C. Williams for their help in m a k i n g the montages and diagrams, and to M r . J. S. Haywood and Mr. D. D. Turner for doing the photography. M. J. C. was a Guggenheim Fellow on leave from the University of Oregon when this study was initiated. The w o r k was supported in part by U.S. Public Health Service Grants N B 01624 and N B 08996 01 to M. J. C.
References ASHHURST, D. E. 1961. The cytology and histochemistry of the neurones of Periplaneta americana. Q. J1. rnicrosc. Sci., 102, 399-405. ASHHURST,D. E. 1964. Fibrillogenesis in the wax-moth, Galleria melone[la. Q. Jl. microsc. Sci., 105, 391-403. BODENSTEIN,D. 1957. Studies on nerve regeneration in Periplaneta americana. J. exp. Zool., 136, 89 115. BODIAN, D. 1947. Nucleic acid in nerve cell regeneration. Syrup. Soc. exp. Biol., 1, 163-178. BOULTON, P. S. 1969. Degeneration and regeneration in the insect nervous system. L Z.Zellfbrsch., 101, 98-118. BRATTGARD,S. O., EDSTROM, J. E. and HVD~N, H. 1957. The chemical changes in regenerating neurons. J. Neurochem., 1, 316-325. CnEN~V, R. A, and ASHHURST,D. E. 1967. A method for the orientation of tissues in epoxy-resin blocks and a design for a rotating stage for holding blocks during trimming. J. R. microsc. Soc. 86, 441-444. TISSUE 8- CELL 1970 2 (3)
398
YOUNG, ASHHURST
& COHEN
COHEN, M. J. 1967. Correlations between structure, function and RNA metabolism in central neurons of insects. In Invertebrate Nervous Systems, ed. C. A. G. Wiersma, pp. 65 78. Chicago: University of Chicago Press. COHEN, M. J. and JACKLET,J. W. 1965. Neurons of insects: RNA changes during injury and regeneration. Science, 148, 1237 1239. COHEN, M. J. and JACKLET,J. W. 1967. The functional organisation of motor neurons in an insect ganglion. Phil. Trans. R. Soc., (B), 252, 561-569. EDWARDS, J. S. and SAHOTA,T. S. 1967. Regeneration of a sensory system: the formation of central connections by normal and transplanted cerci of the house cricket, Acheta domesticus. J. exp. Zoo/., 166, 387-396. EKHOLM, R. and HYD~N, H. 1965. Polyribosomes in nerve cells. J. Uhrastruct. Res., 12, 239-240. FARLEY, R. D. and M~LBURN, N. S. 1969. Structure and function of the giant fibre system in the cockroach, Periplaneta americana. J. Insect Physiol., 15, 457-476. GUTHRIE, D. M. 1962. Regenerative growth in insect nerve axons. J. Insect Physiol., 8, 79-92. GUTHRIE, D. M. 1967. The regeneration of motor axons in an insect. J. Insect Physiol., 13, 1593-1611. HESS, A. 1958. The fine structure of nerve cells and fibres, neuroglia, and sheaths of the ganglion chain in the cockroach (Periplaneta americana). J. biophys, biochem. Cytol., 4, 731-742. HICKS, R. M. 1966. The function of the Golgi complex in transitional epithelium. Synthesis of the thick cell membrane. J. Cell Biol., 30, 623-643. HOLTZMAN,E,, NOVIKOFF,A. B. and VILLAVERDE,H. 1967. Lysosomes and GERL in normal and chromotolytic neurons of the rat ganglion nodosum. Y. Cell Biol., 33, 419 435. HoY, R. R. 1969. Degeneration and regeneration in abdominal flexor motor neurones in the crayfish. Y. exp. Zool., 172, 219 232. HUDSON, G. and HARTMANN, J. F. 1961. The relationship between dense bodies and mitochondria in motor neurones. Z, Zel[/brsch., 54, 147-157. HUDSON, G., LAZAROW, a . and HARTMANN, J. F. 1961. A qualitative electron microscope study of mitochondria in motor neurones following axon section. Exp. Cell Res., 24, 440-456. HYD6N, H. 1960. The neuron. In The Cell, Vol. IV, eds. J. Brachet and A. E. Mirsky, pp. 215 323. London & N.Y.: Academic Press. JACKLET,J. W. and COHEN, M. J. 1967. Nerve regeneration: correlation of electrical and behavioural events. Science, 156, 1640 1643. KIRKPATRlCK, J. B. 1968. Chromatolysis in the hypoglossal nucleus of the rat: an electron microscopic analysis, d. comp. Neurol., 132, 189-212. PALAY, S. L. 1964. The structural basis for neural action. In Brain function, Vol. II: RNA and brain Jhnction; memory and learning, ed. M. A. B. Brazier, pp. 69-108. Los Angeles: University of California Press. PANNESE, E. 1963. Investigations on the ultrastructural changes of the spinal ganglion neurons in the course of axon regeneration and cell hypertrophy. Z. Zell/brsch., 60, 711-740. PORTER, K. R. and BOWER, M. B. 1963. A study of chromatolysis in motor neurons of the frog Rana pipiens. Y. Cell Biol., 19, 56A-57A. Ross, R. and BEND~TT, E. P. 1961. Wound healing and collagen formation 1. Sequential changes in components of guinea pig skin wounds observed in the electron microscope. Y. hiophys, biochem. CytoI., 11, 677-700. SAL,~NKI, J. and GUB~CZA,A. 1969. Histochemical evidence of direct neuronal pathways in Anodonta cygnea L. Acta biol. Acad. Sci. hung., 20, 219~34. SMITH, D. S. and TREHERNE, J. E. 1963. Functional aspects of the organisation of the insect nervous system. In Advances in Insect Physiology, eds. J. E. Treherne, J. W. L. Beament and V. B. Wigglesworth, Vol, l, pp. 40l 484. SMITH, K. R. 1961. The fine structure of neurons of dorsal root ganglia after stimulating or cutting the sciatic nerve. J. cornp. Neurol., 116, 103 115. TAKANO, ]. 1964. Electron microscopic studies on retrograde chromatolysis in the hypoglossal nucleus and changes in the hypoglossal nerve, following its severance and ligation. Okajimas Fol. Anat. Jap., 40, 1-69. USUERWOOD,P. N. R. 1963. Response of insect muscle to denervation. I1. Changes in neuromuscular transmission. J. ins. Physiol., 9, 811 825. WATSON,W. E. 1965. An autoradiographic study of the incorporation of nucleic acid precursors by neurones and glia during nerve regeneration. J. Physiol., 180, 741 753. WIGGLESW0RTH, V. B. 1960. Axon structure and the dictyosomes (Golgi bodies) in the neurones of the cockroach, Periplaneta americana. Q. J1 microsc. Sci., 101, 381 388. YOUNG D. 1969. The motor neurons of the mesothoracic ganglion of Periplaneta americana. J. Insect Physiol., 15, 1175-1179. TISSUE 8" CELL 1970 2 (3)