Copyright is owned by the Author of the thesis. Permission is given for a copy to be downloaded by an individual for the purpose of research and private study only. The thesis may not be reproduced elsewhere without the permission of the Author. RADIOIMMlJNOASSAY AND IMMllNOCYTOCHEMICAL STlJD(ES ON nm RECOVERY OF PfNl<:AL INNERVATION AND FUNCTION FOLLOWING UNILATERAL DENERVATION A thesis presented in partial fulfilment of the requirements for the Degree of Master of Science in Physiology at Massey University JOHN RICHARD PRANGLEY December, 1993 11 Thesis Ahstracl ·111L~ sym pathetic noradrcnergir nel1rons of the superior cen ,ica1 ganglia provide the major source o f innervation to the pinL!a l gland . Studies described in th.is thesis were rksigncd to fmt hcr investigate the initial decline ,md subsequent recovery of pineal rn cl.1t o ni11 :;ccrcto ry capacit y whiL·h has been report ed in sheep after unilateral superior cervical ganglioncctomy (Lapwood, 19~13). Fmihcr to tl1at. the compensatory mechanism proposed by Dornay, et al (I 085), of re-innervation of dencrvatcd tissue hy residual nerve fibres originating from the int.act SCG, was investigated /\ k la lonin secretory capacity is advocated J S a superior index o f pineal function with direct m easurement of pineal output. Radioinmnmoassay was used to m easure dark period plasm a level~ o f rndatonin prior to and at I, 3, 7, 14_ 21 and 28 clays afier unilat eral SCGX. Initial response to pa1tial dcnervation was a reduction in secretory capacity by 80% uf µre-operative kvds, follov,:c 5-l!T AT ATP AVP cAt1P CGRP ChAT CNS csr CST CV DJ\ B d.f DIC DSIP ECN GJ\f>-.n GFAI' GnRH HCI HIOMT HP hr ICC ICN IR llJ LI-I LHRH LP M mm NaCl NAT NE NGF NO nm NPY NSE NSE-LI OXT PKC PNMT Post-op Pre-op PVN List of Ahbrevlations 5-hydroxyl.rypyamine or serotonin Alpha h1buJin Adenosine triphosphate Arginine vasopressi11 cyclic Adenosine monophosphate Ca.lcitonin gene related protein choline acetyl transiera<;c Central nervous system Cerebrospinal fluid Cervical sympathetic trW1k Coefficients of variation Diarninobenz,,dine degrees of freedon 1 Differential contrast optics Delta sleep-inducing peptide External carotid nerve Growth associated protein (41 kD) Glial fibrilla.ry acidic protein Gona IR in pineal tissue collected from the control group (a) in addition to those from unilateral SCGX groups collected at 3 (b ), 14 ( c) and 28 days ( d) after surgery. Magni ti cation 313 X ----------------------------------------------- ----------------96 4 o GAl'-43 IR in pineal tissue collected 3 days aflc1 unilakral SCGX Magnifications (a) 313 X and (b) 400 X DIC ---------------------------------98 4. l O Negative (a) and positive (h) alpha t11hulin imrnunorcactivity in foetal ovine cerebellum . Magnification 125 X ---- ----- -----------------------------10 J 4.11 Pineal tissue collected 3 days after unilateral SCGX and used a5 negative control tissue. No alpha tubulin antiserum had been applied Dark spots are melanin. Magnification 125 X---- ---------------------------102 4.12 Representative examples of alpha tub11li11 JR in pineal tissue collected from the control group (a) in addition to those from unifatcral SCGX groups collected at 3 (b), 14 (c) and 28 days (d) aft er surgery Magnification 125 X-------------- --------- ---- ---------- --- ----------------------10:1 4.13 Representative examples of alpha tub11'in JR in pinc;d ti..,;;11 ,; colkcl cd fll,rn the control group (a) in addition to those u om unilateral SCGX groups collected at 3(.b), 14 (c) and 28 days (d) after smge1y. Magnification 125 X--------------------------------------------------------------105 4.14 Mean counl's (n=.,10) ofirnmunore::ictivc- (IR) or non- J\T IR pineal cells under 400 grid points superimposed on pineal tissue harvested from sheep 3, 14 & 28 days after unilateral SCGX, as well as from control animals --1 06 '1.15 Mean counts (1r0 l 0) of interccllular sp<1ce under 400 grid points; super­ imposed on pineal tissue harvested from sheep 3, 14 & 28 days aficr unilateral SCGX, as well as from control shecp-----------------------------107 5.1 Dual plot of: (1) Mean integrated melatonin secretory responses for 28 day group sheep calculated as percentages of pretreatment values. Measurements were recorded during 4 hrs of darkness, prior to and at 1, 3, 7, 14, 21 & 28 days after unilateral SCGX : (2) Mean integrated ICC intensity levels, as percentages of control animal values, of GAP-43 immunoreactivity in sheep pineal glands collected 3, 14 and 28 days after unilateral SCGX ----------------133 CHAPTER 1 Review of Literature l. Introduction 'llle pineal gland is an unpaired organ, situated in tl1e roof of the tllird ventricle of the brain, which controls a number of circadian and seasonal rhythms, through its secretion of melatonin at night (Reiter, et al, 1981). Its principle nerve supply is via post­ ganglionic sympatlletic fibres which originate in the superior cervical ganglia (Kappers, 1965) See Section 1.3.4. As discussed in section 1.3.5, one technique which has been used to study the control of pineal gland function, is denervation by bilateral superior cervical ganglionectomy. Occasionally unilateral SCGX has also been utilized. In one such sh1dy Lapwood (1993) found that while melatonin secretory capacity was abolished after bilateral SCGX and was reduced to 92% of pre-operative levels on day I after unilateral SCGX, it recovered to 77% by day 14 after surgery for tllat group. It was suggested that recovery of function after unilateral surgery, may have been due to re-innervation of denervated pineal endocrine cells (pinealocytes) by collateral sprouting of nerve tenninals originating from tlle remaining SCG. The experiment described in this thesis investigated whether full restoration of pineal melatonin secretory capacity occurred if the post-surgery period was extended to 28 days . In addition, a study was undertaken to investigate whether evidence of re-innervation of tlle pineal could be demonstrated. 11i.e aim of Chapter I is to provide an overview of the literature relating to both the pineal gland and the regeneration of nerves, as is pertinent to thi.s thesis. 2 1.1 Early history of pineal research Early anatomists held various views on the physiological function of the pineal in the human . This unique unpaired structure, that lies deeply recessed under the cerebral hemispheres of the brain, drew their attention and speculation. According to Kappers (1979) and Oksche (1984), Herophilas, an anatomist at the University of Alexandria in Egypt, was first to discover the pineal, around 300 BC. The philosopher Descartes considered it the "seat of the soul" . 1l1e possible physiological significance of the pineal was first recognised by Heubner in 1898, who noted precocious sexual maturity in a young boy whose pineal was destroyed by a tumor. Holmgren (1917/1918) noted that the cells of the pineal gland of an elasmobranch were sensory-like in nature: the pinealocytes resembled the sensory cells of the retina. Because some reptiles possess a prominent "third eye" the pineal of mammals was considered a vestige of this primitive visual organ. The obsenration that the human pineal may become calcified at an early age further consolidated thought that the pmeal was, indeed, a vestigial organ and therefore of little physiological consequence. However, in 1954, Kitay and Alt<;chule reviewed the literature on human pineal tumors and described clinical correlations of pineal dysfunction with evidence clearly revealing that the pineal may in some way be related to reproductive functions in humans. McCord and Allen (1917), interested in endocrme factors affecting morphogenesis, observed that bovine pineal extracts added to the water in which tadpoles swam caused the larvae to blanch. In 1958 dermatologist Aaron Lerner, in searching for a factor which might be responsible for vitiligo, was able to isolate and determine the structure of the bovine pineal extract as N-acetyl-5-methoxytryptamine, an indoleamine, which he named melatonin. Tilis molecule can now be readily synthesized and made available for a variety of physiological studies. 3 1.2 Seasonal adaptive changes mediated hy the pineal 1.2.J Seasonal Reproduction Many mammals in their natural habitat are seasonal breeders. Seasonal reproduction is one of the more conspicuous changes that natural populations of mammals rely on for their survival. It ensures the birth of the young during those seasons of the year in which their chances of survival are greatest (Bronson, 1988). Clearly, the most favourable seasons for supporting the survival of off.spring are those in which food is accessible and environmental conditions are mild, in the spring and summer seasons (Karsch., et al, 1984). Diverse species mate during various seasons of the year so that birth occurs during those favourable seasons. There are potent exogenous factors on which animals rely for the synchronization of their annual cycles. Most biometerological parameters change throughout the course of the year and animals could have selected any one of these to guide or determine their annual cycle of reproduction (Stonehouse, 1981). However, some factors change with greater regularity than others. One of the most dependably recurring phenomena in the environment is the photoperiod, consequently it has great predictive value in terms of anticipating the upcoming season. Hence, it is logical that many mammals have come to depend on the seasonal changes in photoperiod to synchronize their annual reproductive rhythms, as it is both essential and advantageous for these species to initiate reproduction at approximately the same time each year, before the optimal conditions for birth and rearing have arrived (Reiter, et al, 1981). 4 1.2.2 Photoperiod and the pineal gland Both circadian and circannual rhythms in the duration of daily photoperiods have been shown to be the major factors influencing the timing of reproductive activity in 2lmost all seasonally breeding mammalian species (Reiter, 1980). Central to seasonal reproductive adjustments in response to light is the pineal gland. Although the photic information is detected by the retinae of the lateral eyes (Moore, 1978), it is the pineal that tranduces (Wurtman, et al, 1968) the resultant neural information into a chemical signal that determines the level of reproductive activity. The pineal is a small organ located near the centre of the brain, that functions as an endocrine organ which secretes melatonin. As an end organ of the visual system in mammals, the pineal gland's production and secretion of melatonin arc affected by light which causes a drop in blood levels of the compound. As day length (and therefore night length) varies seasonally, the pineal gland, because of the secretion of melatonin., provides information concerning time of year to all other organs of the body. Thus in animals whose reproductive patterns fit into a specific seasonal scheme the pineal may play a pivotal role in the control of their gonadal function (Kauppila, et al, 1987). Hence the pineal gland is essential to the chronobiology that assists an animal in adapting to the external environment, both daily and seasonally (Reiter, 1991). 5 1.2.3 Reproductive seasonality in sheep Seasonally breeding animals which use photoperiod to time their reproductive activity can generally be divided into two groups - short and long day breeders. Short day breeders, such as domesticated sheep, use the decreasing daily photoperiod of autumn to time the initiation of breeding activity and generally have long gestation periods eventuating in spring parturition O~albandov, 1976). lhese modem sheep breeds have developed as a result of controlled breeding programmes intended to improve meat and wool production, and to increase fecw1dity (Carter & Cox, 1982). Marshall in (1937) was the first to define the reproductive cycle of sheep, with Hammond (1944) later establishing the importance of photoperiod in regulating the onset and termination of reproductive activities . Yeates (1947, 1949) in early studies investigating seasonal reproduction in sheep, concluded that, seasonal variation in the length of photoperiod was the predominating factor determining the time of onset and the duration of the breeding season. A change from increasing to decreasing photoperiod induced in both rams and ewes to commence behavioural characteristics associated with the onset of reproductive activity. Ram behaviour associated with increasing reproductive activity occurs in conjunction with elevated testosterone secretion from the testes. Characteristic behaviour includes increased libido, inter-male aggression and the occurence of fl.ehmen, the raising of the upper lip in order to facilitate the detection of olfactory stimuli originating from vaginal secretions (Lincoln & Short, 1980). Behavioural oestrus of ewes is characterised by sexual receptivity towards the ram, culminating in pro-active behaviour by some ewes. Conspicuous signs of behaviol.lf'ctl oestrus are however, mostly absent, with rams detecting oestrous ewes by pheremonal signals from their vaginal secretions (Smith, 1982). 6 Seasonal reproductive capacity may also be measured by hormonal, physical and physiological changes in hoth rams and ewes. The initiation and cessation of reproductive activity is a reflection of the changing secretory profiles of pituitary gonadotrophins and gonadal steroids (Lincoln, et al, 1977). Ram testis size is greatest during the breeding season and least during sexual quiescence (fulley & Burfening, 1983). Sperm output ancl quality (motility and percentage of live sperrnatazoa) (Dufour, et al, 1984; Boland, et al, 1985) and ejaculate volume (Sanford, et al, 1977; Barrell & Lapwood, l 978/l 979a: Boland, et al, 1985) are highest during the breeding season and lowest during sexual quiescence. For ewes, the onset of breeding activity is initiated by cyclic changes in ovarian hormones leading to follicle growth, ovulation and corpus luteurn development, swelling of the uterus and vagina, an increase in the secretory activity of glandular tissue within these structures, and an increase in the secretion of mucus from the cervix (reviewed by Smith, 1982). In addition to light and pheremonal factors influencing reproductive seasonality in sheep, both nutritional and temperature variations may be observed. Through effects of inhibition ofluteinizing hormone secretion, low levels of nutrition result in reduced levels of reproductive activity, delaying both the onset of puberty and of the breeding season. On the other hand high nutrition levels are associated with increased reproductive activity (Lindsay, et al, 1984; Bronson, 1988; Rhind, et al, 1989a). A study of the effects of temperature on the breeding cycle of Chm ewes has indicated that temperature, at least in this breed, may play a secondary, but important, role in timing the onset of breeding activity (Lees, 1971). 7 1.3 Pineal function 1.3.l Pineal development and morphology The vertebrate pineal, a part of the epithalamus, arises as a median evagination of the dienccphalic roof of the embryonic bra.in (Oksche, 1965). In some mammals the pineal gland moves away from the roof of the third ventricle and loses connection with the brain except for a thin 'pineal' stalk. The gland is richly perfused with blood vessel<; derived from the posterior cerebral arteries. The venous drainage of the gland is directly into large venous sinuses which surround the organ (Reiter, 1991). Pineal parenchyrnal cells, pinealocytes, are derived from the ependyrnal lining of the epithalamus; both light and dark parenchyrnal cells can be distinguished in the mammalian pineal gland (Oksche, 1%5). The dark cells contain pigment granules of an unkno>w11 nature, as well as glycogen deposits of undefined physiological significance. Dark pinealocytes are interconnected by tight junctions, suggesting that electrical signals may be communicated between the cells (Reiter, 1977). TI1e main body of the pinealocyte, the parikaryon, has either one or two processes emanating from it. These processes terminate in buds which he in close proximity to pericapillary spaces or inter­ cellular lacunae. The actual relationship of the terminals with the pericapillary space varies between species and is perhaps related to the mode of release of the secretory products (Reiter, 1977). The number of pinealocytes may decrease in advanced age, when calcium deposits, which can be visualised radiologically, also form in the gland (Reiter, 1991 ). Fibroblasts and glial cells make up the rest of the cellular components of the glandular mass which, in an adult sheep weighs about 60-80 mg and measures approximately 5-7rnm in length, and 3-5 mm in width (Barrell & Lapwood, 197811979b; Vollrath, 1981). 8 1.3.2 Pineal indoleamine blosynthesis 111c biochemistry of pineal indoleamine biosynthesis is well documented (Relkin, I 076; Sugden, 1989; Wurtman, et al, 1968). Indoleamine biosynthesis involves pinealocy1e uptake of the amino acid, L-tryptophan, from the blood (King & Steinlechner, 1985) Conversion by hydroxylation to 5-hydrnxytryptophan by the enzyme tryptophan hydroxylase follows. The aromatic enzyme 5-hydroxytryptophan decarboxylase acts on the hydroxylated derivative to form 5-hydroxytry1Jtamine. Serotonin concentrations are higher in the pineal than in any other organ or brain region (Quay, 1964). Serotonin is converted t,1 N-acetylserotonin by the action of N-acetyltransferase (Klein & Weller, 1970). The N-acetylserotonin produced is O-methylated by hydroxyindole-O- methyltransferase to fonn N-acetyl-5-methoxytryptaminc (melatonin) (A'-.yindoleacetic acid or reduced to 5-hydroxytryptophol. The latter compounds can then become O-methylated by HIOMT to give 5-methoxyindole acetic acid and 5-methoxytryptaphol (Wilson, 1978). The formation of melatonin may also occur from methoxytryptophan, although this is a minor synthetic pathway (Morton, 1987). 9 1.3.3 Effect of light on pineal lndoleamlnc blosynthesis Within the pineal conversion of serotonin to melatonin is a highly cyclic event which is closely related to the prevailing light : dark cycle to which animals are exposed. ln all animals thus far studied, melatonin production is greatest within the pineal gland during the dark phase of the light: dark cycle (Quay, 1964; Lynch, 1971; Panke, et al, 1978). Pineal serotonin levels also rewal marked diurnal changes with highest levels noted during daylight hours and depressed levels during darkness. Pineal enzyme activities are rapidly depressed by light (Reiter, et al, 1986). At night there is an increase in the activity of NAT in rat pineals which is 10- to nearly 100- fold greater than values in the light (Adrendt, 1985). The pineal concentration of N­ acetylserotonin is subsequently increased to values ten to thirty times greater than observed under day conditions. HIOMT activity also increases, which results in nocturnally elevated levels of pineal melatonin (Adrendt, 1985) ln experimental conditions reversal of external lighting periods reverses the rhythm of pineal enzyme activity and indolearnine biosynthesis. Thus a diurnal rhythm of pineal melatonin synthesis is observed but with maximum levels measured during the true day when lights are off. Shaw, et al (1988) observed a cessation of melatonin production in sheep exposed to continuous light, with normal night time levels recurring within 10 mins of lights off. Studies using monochromatic light have demonstrated that not all wavelengths are equally effective in suppressing pineal melatonin synthesis and secretion. Reiter (1985), in a review of the effect.s of light characteristics on the pineal, identified green wavelengths (510-550 nm) as being the most potent suppressors of pineal HIOMT activity. That review also reports between-species differences in effectiveness of various wavelengths in altering melatonin production. 10 In sheep, the intensity of light required to suppress nocturnal pineal melatonin levels in a dose-dependent manner has been shmvn to range between 1.02 to 88.60 Iux.., with 88.6 hD, producing a >80% reduction (Arendt & Ravau.lt, 1988). The duration of light exposure that can inhibit pineal melatonin synthesis during a period of darkness is very short, as little as 1 sec for the Syrian hamster. Return to night-time melatonin levels after a light pulse may take several hours in many rodent species, while in sheep there is a lag period of only 5-10 min (reviewed by Vollrath, 1981 ; Reiter, 199 J ) . 11 1.3.4 Neural control of pineal lndoleamine biosynthesis Melatonin is synthesised in response to norcpinephrine released from postganglionie sympathetic neurons originating from the SCG's. Thus the pineal is considered to be a neuroendocrine transducer, as neural input to this organ is converted into an endocrine output (Wurtman. et al, 1968). Postganglionic stimulation of the pinealocyte cells depends on the absence of light activation of the retina of the lateral eyes. Light information perceived by the eyes is transduced into a neural signal by the rcti.nal ganglion cells and then conveyed to the suprachiasmatjc nuclei of the brain by way of the retinohypothalamic tract. This pathway is always bilateral and decussates at the optic chiasma innervating the contraleral SCN (Mess & Ruzsas, 1986). Neuronal fibres from the SCN, which convey information to the pineal on the status of the environmental photoperiod, then course through the medial forebrain bundle down to the upper thoracic spinal cord. Axons from the preganglionic neurons, located in the intermediolateral cell columns of the spinal cord, synapse within the SCG. From these ganglia postganglionic fibres proceed to innervate the pinealocytes in the pineal gland (Mess & Ruzsas, 1986). Prior to their entrance into the gland many of the sympathetic fibres coalesce to form two bilaterally symmetrical nervi conarii which, in some mammals, fuse before entering the pineal. Within the pineal the fibres branch extensively and with the onset of darkness release noradrenaline from their terminals, followed by interaction of the catecholeamine with beta adrenergic receptors in the pinealocyte membrane (Panged, et al, 1990). Beta-adrenergic stimulation activates an adenylate cyclase enzyme via a stimulatory, guanine nucleotide-binding, regulatory protein (Spielgel, 1989). This results in a rapid and large (up to 60-fold in the rat pineal) increase in intracellular cyclic adenosine monophosphate. cAMP setves as a second messenger in the nocturnal elevation of melatonin biosynthesis by activating a cAMP 12 dependent protein kinase Transcription of mRNA follows, initiating an eventual rise in serotonin NAT activity (Sugden, 1989). ln addition to sympathetic innervation controlling pineal biosynthesis there is also evidence for a possible central innervation of sheep pineals. lmmunocytochemical studies have demonstrated irnmunoreactivity for vanous substances including somatostatin, GnRH, Substance P, CGRP, TRI-I, DSIP, NSE, A VP, OXT, GnRH NPY and VIP, as well a5 the enzymes ChAT and PNM!', within pineal nerve fibres, particularly in the stalk (Mockett, 199 I). Also electrophysiological (Schapiro & Salas, 1971; Dafuy, 1980: Reuss, et al, 1984: Reuss, 1987), retrograde neuron tracing (Guerillot, et al, 1982; Moller & Korf. 1983a, 1986) and lesion studies (Moller, et al, 1987b) indicate that various central structures have direct neural connections with the pineal in a range of species. For example, NPY-like immunoreactive nerve fibres projecting to the pineal from central nuclei have been demonstrated in the hypothalamus of cat, rat, monkey and golden hamster. Other peptidergic projections exhibited include rat arnygdala, monkey limbic regions and rat hippocarnpal region (reviewed by Ebadi, et al, 1989). Catecholaminergic neurons with a central origin have been demonstrated in the habenular area (Bjorkland, et al, 1972; Wiklund,. 197 4), brainstem (Moore & Bloom, 1979) and hypothalamus (Cuhnan, et al, 1987). Although no function has yet been ascribed to these central innervations, it appears possible that they may influence pineal function indirectly as demonstated in one lesion study conducted in rats in which disruption of central fibres from the PVN and hippocampus resulted in a significant reduction in nocturnal levels of NAT and HIOMT activity (Moller, et al, 1987b). Also, Morgan, et al (1988) demonstrated a VIP dose­ dependent effect on cAMP in sheep pineal homogenates, suggesting that modification of enzyme activity by centrally derived nerves is possible in this species. 13 Similar conclusions may be drawn from the findings of Mockett (1991) who clearly demonstrated the presence of NPY, VIP and PNMT immunoreactive nerve fibres v.:ithin the ovine pineal. While regulation of ovine pineal function is similar to that of most other mammalian species, in that it is primarily mediated by the sympathetic nervous system, it is unclear to what extent the two innervations interact to initiate or modify pineal secretory response. The various central structures having direct neural connections with the pineal are suggested to process or relay information about environmental or social conditions (e.g. visual processing by the dorsal nucleus of the lateral geniculate body), and hence may act as secondary routes fi.1r information of this nature to influence pineal function . 14 1.3.5 Techniques used to evaluate the control of pineal melatonin synthesis. That the pineal gland, through secretion of melatonin, is responsible for transducing photoperiodic cues into physiological adjustments to hormone profiles, is well established (Lincoln, 1980; Lincoln & Short, 1980; Karsch, et al, 1984). Four principal e>-.-perirnental methods have been used to investigate the relationships between control, function and biosynthesis associated with the gland. These arc pinealectomy, supenor cervical ganglionectomy, melatonin administration and photoperiod manipulation. Typical results from use of these methods, particularly with sheep, and the advantages and disadvantages of each, are given briefly in this section. Removal of the pineal gland has provided conclusive evidence that this gland is necessary for induction of the normal pattern of reproductive seasonality in sheep (Barrell & Lapwood, 1979b). Although pinealectomy is relatively difficult to perform in sheep (Roche & Dziuk, 1969), it does provide the only definitive means by which an animal's response to changes in environmental lighting, in the absence of pineal influences, can be measured. Also, since this gland is the major source of melatonin, pinealectomy abolishes the circadian pattern of melatonin secretion and allows for the study of reproductive function in the absence of variations in levels of this hormone (Bittman, et al, 1983). Denervation of the pineal gland by bilateral SCGX results in dysfunction of the pineal in sheep and as such has often been used as an alternative to pinealectomy (Lincoln, et al, 1989). Studies employing this technique have produced results similar to those of experiments using pinealectomy, that is, sheep no longer respond to changes in photoperiod and display cycles in gonadotrophin secretion and sexual behaviour which are less pronounced and occur at random relative to those recorded from control animals (Barrell & La.pwood, 1978/1979b~ Lincoln, 1979). The principal advantage of this smgical 15 method as compared to pinealectomy is the relative ease with which the SCG can be located and removed. In addition, the health of ganglionectomized sheep does not appear to be compromised As mentioned in the Jntroduction (page I), unilateral SCGX is a technique which occasionally has been used in studying the control of the pineal. More detail is given here on the results and conclusions from those studies, than for those using the other techniques mentioned in this section, because of the central place of unilateral SCGX in the studies described in this thesis. Reiter, et al (1978) established that unilateral SCGX caused pineal NAT and melatonin levels to be intermediate between those in sham-operated control rats and those in rats from which both ganglia had been removed. From that experiment it was concluded that unilateral SCGX impaired the ability of a portion of the pinealocytes to respond to darkness. Zigmond, et al (1981) measured a 75% decrease in NAT activity immediately after unilateral SCGX, however, subsequent measurements determined a recovery in pineal NAT activity to levels similar to those measured prior to surgery. That group proposed that residual nerve fibres originating from the intact SCG had a "reserve capacity" to stimulate denervated pinealocytes, whereby the loss of approximately 50% of sympathetic nerve fibres caused a reduction in nerve "re-uptake" of NE. Subsequent increases in NE levels were suggested to cause an increase in stimulation of denervated pinealocytes and hence the recovery in NAT levels. Later, Dornay, et al (1985) also measured a >50% decline, followed by a recovery, in both TH activity and NE uptake after unilateral SCGX. These results led to the proposal that the mechanism responsible for the recovery entailed "collateral sprouting" from residual nerve fibre terminals, culminating in the re-innervation of denervated pinealocytes. Anatomically, this process requires that nerve terminals of fibres from each 16 SCG are in close proximity to each other so that following unilateral SCGX and degeneration of the lesioned axon terminals, collateral sprouts developing from the intact nerve terminals were able to contact and re-innervate nearby denervated tissue. Evidence demonstrating that fibres from each SCG cross over to innervate the contralateral as well as the ipsilateral sides of the pineal, and that nerve terminals from each SCG are closely intermingled, has been demonstrated in rats (Dornay, et al, 1985; Zigmond, et al, 1981, 1985; Lingappa & Zigmond, 1985) and also in sheep (Mockett, 1991 ). Mockett and Lapwood (unpublished) recently found that atrophy of NSE-IR presumptive pinealocytes, which was observed alter bilateral SCGX., did not occur in animals killed 14 days after unilateral SCGX. In an attempt to explain why normal cell morphology was maintained after single SCGX, Lapwood (1993) took the physiological approach of measuring melatonin secretion profiles during exposure of sheep to darkness, before and periodically after unilateral SCGX. On the day after surgery there was a reduction of pineal melatonin secretory capacity to values 92% (P< 0.001) below those measured before surgery, however, succeeding measurements indicated a substantial recovery of that parameter of melatonin biosynthesis, to within 77% of pre-operative levels after 14 days. These results suggested that a full recovery in pineal secretory capacity was possible, despite a reduction of sympathetic nerve fibres in the pineal gland by approximately half. Reviews by Wurtman, et al (1963) and Reiter (1980) have addressed the topic of potent antigonadal effects on the mammalian reproductive system in response to exogenous melatonin injection. Several studies using administration of melatonin to sheep have focused on the effectiveness of that hormone in advancing the onset of ovarian cyclicity in seasonally anoestrus ewes. Collectively their results indicated that during the initial stages of 17 anestms, ewes were insensitive to the effects of melatonin and therefore attempts to advance the onset of the breeding season were unsuccessful. During the latter stages of anestrus, however, ewes again became sensitive to melatonin and implants or infusions at that time of year led to an advancement of the breeding season by 5-10 weeks (Kennaway, et al, 1982; Nowak & Rodway, 1985; English, et al, 1986). As a separate technique or in con_iunction with those described earlier, artificial manipulation of photoperiod has been an important feature of experiments designed to investigate the role of the pineal in mediating the effects of photoperiod on seasonal reproductive activity. Experiments described in this section., such as those by Hafez ( 1952), Barrell & Lapwood (1979a,b), Lincoln & Short, (1980) and Robinson & Karsh (1987), have involved combinations and/or variations of these techniques. Further studies investigating the role of light in sheep pineal physiology have experimented with a range of lighting regimes. Matthews, et al (1992) studying endogenous pacemaker effects on melatonin biosynthesis, housed sheep in conditions of acutely extended darkness with results suggesting a functional role for the suprachiasrnatic nuclei in sensitizing the photoreceptive system to seasonal changes in photoperiod. Shaw, et al (1988) have demonstrated, by placing sheep in constant darkness following exposure to prolonged continuous light for 28 days, that the response in melatonin production is rapid, often commencing within 10 min, regardless of what time of the day dark conditions were imposed. This not only indicates that the synthetic mechanisms which generate melatonin are maintained in an inducible state under constant illumination., but that under these conditions the pattern of melatonin secretion is not dominated by the historical photoperiod. Results derived from these experiments have all been instrumental in advancing our knowledge of pineal physiology, its control mechanisms and the gland's role in mediating the effects of photoperiod. 18 1.3.6 Post-ganglionic noradrenaline activity With the onset of darkness, sympathetic neurons increase their firing rate (Taylor & Wilson., 1970) causing a rise in NE levels (Levitt, et al, 1965) and a significant increase in NE turnover (Brownstein & Axelrod, 1974). NE ultimately derives from tyrosine, which, in the pineal, is produced primarily from dietary phenylalanine by the rate limiting enzyme tryptophan hydroxylase (Bensinger, et al, 1974). Tyrosine is converted to dihydroxyphenylalanine (DOPA) by tyrosine hydroxylase (Levitt, et al, 1974), the rate limiting step in NE synthesis. Tyrosine hydroxylase, which increases by more than 50% in the dark is highly concentrated in pineal tissue (McGeer & McGeer, 1966). Conversion of DOPA to dopamine is followed by the conversion of that compound to NE by the enzyme dopamine beta-hydroxylase (Laduron & Belpaire, 1968). On release into the axon terminal-pinealocyte cleft, NE can interact with the pineal's beta-adrenergic receptors. Deactivation occurs either by reuptake into the terminal neurons, diffusion away from the terminal-receptor environment, or through deactivating mechanisms related to enzymatic degradation to nonactive metabolites (reviewed by Backstrom & Wetterberg, l 972). A marked change in receptor numbers is apparent during each 24-hour period, with binding of labeled alprenolol (a beta-adrenergic drug) being twice as great at the end of the light period as it is at the beginning of the light period (Zatz, el al, 1976; Kebabian, el al., 1975). 19 1.3. 7 Melatonin circulation and excretion Plasma levels of melatonin in sheep and other animals (e.g. primates, pig, rat, cow, donkey, carnet chicken, salmon, lizard) are highest at middark and lowest at midlight periods (Vaughan, et al, 1978). ln addition to the circadian rhythm of melatonin levels in blood there is aLc;o a diurnal melatonin rhythm in the cerebrospinal fluid of primates (Reppert, et al, 1979) and sheep (Shaw, et al, 1989). In primates night peak values of melatonin in CSF are 2 to 15 times higher than day values. 'This increase occurs soon after lights are turned off, and the decrease towards day values occurs rapidly after lights are turned on again. Changes in CSF melatonin concentrations appear to reflect daily changes in plasma melatonin concentrations (Vaughan, et al, 1978). A 24 hour rhythm in the concentration of CSF melatonin suggests the possibility that the CSF may be an important route of communication between the pineal gland and other parts of the brain (Li, et al, 1989). HlOMT activity is also present in some blood cells, the harderian gland and the retina (Claustrat, et al, 1990), and melatonin may also be synthesized in other extrapineal sites such as the hypothalamus, retina and gastro-intestinal tract. These sites may assume some minor functional significance following pinealectomy. Melatonin is metabolized in the liver to 6-hydroxymelatonin by melatonin hydroxylase and then conjugated to sulfate or to glucuronide (Kopin, et aJ, 1961). In the brain, melatonin is metabolized to a hallucinogenic agent N-acetyl-5- methoxykenurenamine (Hirata, et al, 1974). Both the hepatic and neural metabolites are discharged into the blood and are eventually excreted in the urine along with small quantities of unchanged melatonin (Kopin, et al, 1961). 20 l .4. Circadian rhythms, photoperiodism and the pineal gland 1.4.1 General The majority of plants and animals have evolved, as part of their internal organisation, oscillating systems whose periods are closely matched to one or more of the major physical cycles in the environment. These oscillating systems, which include many behavioural and physiological processes, show significant daily (circadian) rhythms which arc thought to confer a selective advantage such that events occur at the correct time of the day. In mammals, light is the major entraining environmental cue for circadian and circannual rhythms, with temperature also being important to fishes, amphibians and reptiles (deVlamming & Olcese, 1981). 1.4.2 Formal properties of circadian rhythms Several formal properties characterise circadian rhythms. Firstly, they are entrained by the light/dark cycle and show a regular period of24 hours. Secondly , they do not vary greatly from the 24 hour period and resist entrainment to any other period. Finally, they are endogenously generated with a period close to 24 hours, even in the absence of entraining stimuli (Moore, 1978. Underwood, 1984). The light/dark cycle, therefore, imposes both period and phase on the endogenous rhythm. The demonstration that the daily pineal rhythms in serotonin, NAT and melatonin content continue under conditions of constant darkness, confirms that they are true circadian rhythms (Reiter, 1981). 21 1.4.3 Suprachiasmatic nuclei as circadian pacemakers Generation of circadian rhythms and their entrainment by the light/dark cycle is a unique feature of the CNS. Endogenously generated circadian rhythms require an oscillating system which is made up of one or more circadian pacemakers, of which the most important appears to be the suprachiasmatic nuclei (Moore, 1983), within the hypothalamus. Two observations provide evidence for the role of the SCN as a generator of photoperiod-entrained circadian rhythms. Firstly, a neural patl1way from the photoreceptors (eyes) to the SCN has been identified in the majority of fish, amphibians and mammals studied. This pathway, the retinohypothalamic tract, is always bilateral and usually innervates the contra.lateral SCN (Mess & Ruzsas, 1986). Additionally, the ability of light to alter the neural activity of the SCN (Nishino, et al, 1976) demonstrates that this pathway is functional and important. Secondly, following ablation of the SCN, there is a general loss of circadian rhythms, including drinking behaviour, locomotor activity and adrenal corticosterone secretion in rats and hamsters (Moore, 1983). Circadian rhythms in both neural activity (Inouye & Kawamura, 1979) and metabolism (Schwartz, et al, 1980) exist in the surgically isolated SCN, adding further support for the role of this structure as a circadian oscillator. The circadian rhythm in melatonin secretion is thought to be due to changes in synthesis generated by endogenous signals emanating from a neural centre located within the SCN (Rusak & Zucker, 1979). Light both suppresses melatonin synthesis and entrains the neural centre, so that the time of the increase and the decrease of melatonin secretion is related to the light/dark cycle (farnarkin, et el, 1985). 22 1.5 Nerve Regeneration Ute mechanisms underlying structural plasticity and regenerative growth in mature neurons, and the extraneural cues that regulate it remain largely undefined. As mentioned in Section 1.3.5., recovery of pineal gland function after unilateral SCGX, may occur as a result of pinealocyte innervation, by collateral sprouting of sympathetic nexve terminals, originating the intact SCG. Section l. 5 provides a review of current knowledge of nexve regeneration processes and environmental interactions as they pertain to an interpretation of results of the experimenlc; described in this thesis. 1.5.1 True regeneration Under normal circumstances nerves continue to maintain connections with their targets throughout the life of an animal, suggesting that some cellular behaviours associated with developmental growth must persist in mature neurons. They may continually remodel their connections (Purves & Voyvodic, 1987), produce and transport to the nerve terminals many of the molecules needed for nexve growth (Lasek, et al, 1984), and they elongate considerably during body growth. Elongation of axons and active remodelling of their terminal arbors underlies the differentiation of neural circuits during development (Collinridge & Bliss, 1987), contributes to some forms of neural plasticity in adult brains (Merzenich, 1987; Singer, 1987) and may determine the success or failure of nerve regeneration (Fawcett & Keynes, 1990). In the mature nervous system, with a few notable exceptions such as the olfactory system of most species (Graziadei, et al, 1980) and the vocal motor system of songbirds (Nottebohm, 1985), neurons do not develop either de novo nor migrate to new positions 23 (Mathew & Miller, 1990). However, new process outgrowth in the fom1 of axonal regeneration or sprouting does occur in response to neural trauma or pathology (Brown, 1984; Seil, 1988), albeit a comparatively slow process Developing sympathetic neurons explanted from fetal or neonatal animals can reinitiate neurite outgrowth in a culture dish wiL'un a few hours (Argiro & Johnson, 1982; Collins & Lee, 1982). In contrast, adult neurons explanted to identical culture conditions do not extend neurites for several days (Agranoff, et al, 1976; Collins & Lee, 1982). It is the cell soma that provides the essential metabolic machinery necessary for continued neuronal function at the onset of regeneration, through the expression of new genes and synthesis of proteins. Biochemical and structural changes, such as the swelling of nucleoli and proliferation of rough endoplasmic reticulurn, indicating increased RNA activity and protein synthesis, are amongst the first signs that regenerntion is starting (Stein, et al, 1974). Sequential gene expression is associated with commitment, migration., process outgrowth and synaptogenesis (Mathew & Miller, 1990). In general, the proteins produced during regeneration arc the same as those associated with axonal growth in development. Substances that are abundant in developing axons, such as growth associated proteins (GAPs) and tubulin, have their synthesis enhanced during the migratory phase of nerve regeneration., whereas neurofilament proteins associated with later developments, when axons have connected with their targets and have expanded radially, are decreased (Fawcett & Keynes, 1990). Primary terminology for nerve regeneration differentiates axonal growth which does not pass beyond the proximal surface of a lesion (Anderson., et al, 1971), and the actual reestablishment of point to point contacts (Guth, 1974). The former called collateral sprouting, the latter is true regeneration. For true regeneration proximal axons of a cut nerve may regenerate through to an intact distal endoneural tube at distances ofup to 1 cm, an attraction that is dependent on 24 the presence of live Schwarm cells in the distal endoneural tube (Kufller, 1986). Schwann cell multiplication is stimulated by machrophages and myelin debris, and is itself dependent on the successful degeneration of detached neural segments (Fawcett & Keynes, 1990). In response to axotomy penpheral nerve fibres distal to the lesion and detached from cell body metabolic machinery, degenerate, by the process of anterograde ("wallerian") degeneration (Hallpike, 1976; Allt, 1976). Anterograde degeneration leads to the removal and recycling of axonal and myelin-derived material, and prepares the environment through which regenerating proximal axons will grow. Both the axon and the myelin degenerate, leaving behind dividing Schwann cells inside a basal lamina tube that had surrounded the original nerve fibre (Cragg, 1970; Grafstein & McQuarrie, 1978); these columns of Schwann cell'>, surrounded by the basal lamina, are known as endoneural tubes. The proximal axon's initial response to axotomy is also degeneration, however, in contrast, the intact neuron then initiates a process of axonal regrowth along with its attendant metabolic changes ( Cragg, 1970; McQuanie, 1978), culminating m synaptogenesis with target tissue and, if successful, a return to operative functioning. Regenerative growth occurs with greater success following crush rather than cut injuries (Sutherland, 1978). Endoneural tubes and Schwarm cell basal lamina are left intact after crush injuries (Haftek & Thomas, 1968) and regenerating axons remaining in their parent tubes are guided directly back to their targets. If the tubes are disrupted, however, regenerating axons may enter inappropriate tubes in the distal stump, and so be guided to inappropriate targets (Muller, 1992) which may not be just ineffectual, but inhibitory. 25 1.5.2 Collateral regeneration Both myelinated and unmyelinated regenerating axons exhibit repetitive sprouting, such that each axon proximal to a site of axotomy can give rise to several processes distal to it (Jenq, et al, 1987). However, not all regenerative growth survives; over the months following a nexve repair, some axons will enlarge and return to an approximately normal diameter, whereas others will disappear (Cragg & 1nomas, 1964). Subsequent axon enlargement and maturation is dependent on connection with a target, and branches that disappear have presumably failed to do this (Aitken., 1949) In culture, foetal and neonatal axons initiate regenerative sprouting within a tew hours of axotomy (Argiro & Johnson., 1982; Collins & Lee, 1982). The first sprouts in myelinated axons are generally seen coming from the tenninal nodes of Ranvier, through the gap left by partial retraction of Schwann cells (Meller, 1987); unmyelinated axons sprout equally as rapidly (Bray, et al, 1972). While collateral sprouts are forming, the cut tip of the axon swells, inflated with smooth endoplasmic reticulum, mitochondria, and eventually microtubu.les (Fawcett & Keynes, 1990). A number of morphological variations of collateral regeneration are possible. Cotman & Nadler (1978) examining morphological variation in collateral sprouting have differentiated between 'paraterminal sprouting' and 'contact synaptogenesis'. In the former the terminal end-feet of an intact axon may simply enlarge and establish new synaptic contacts at sites left vacant as a result of damage to another neuron. Contrasted with 'contact synaptogenesis', in which an intact nerve axon in contact with a denervated cell would form new synapses at points of apposition where there were none previously. Alternatively, the formation of new synapses could follow a shift in position of an axon prior to creating new points of contact. These variations suggest that axon elongation 26 need not necessarily be a requirement for successful reinnervation if residual neurons were in close proximity to denervated cells. Electron microscopy studies investigating collateral sprouting following partial denervation of the septal ganglia, indicated that "intact fibres from one system could replace the axons of another system". Termed 'heterotypic sprouting', that mechanism of regeneration occurs when axons passing close to a deaITerented zone respond by developing and sending collaterals to occupy vacated synaptic sites, while retaining their own connections. Raisman (l 969) suggested that the response was triggered by the absence or removal of the normal afferents to the area, and a role for both glial cells and 'neural growth factors' were also implicated. 1.5.3 Axonal dynamics Although axons have some capacity for synthesis and processing of lipids and carbohydrates there is an absence of identifiable components for protein synthesis in neuronal processes (Koenig, 1967). Protein synthesis is generally thought to be restricted to cell bodies where protein synthesizing organelles are located. Sorting of proteins to specific membrane compartments occurs mostly at the level of the Golgi complex within the cell body (Stone & Hammerschlag, 1987). However, after assembly intracellular membranous organelles often must be moved considerable distances along an axonal or dendritic compartment before being delivered to terminal sites of action. During neuronal growth axonal dynamics encompass the synthesis, packaging, sorting, targeting, and translocation of the proteins needed for neuronal function and migration. Consequently, intracellular transport is highly developed in neurons with a wide variety of polypeptides being delivered to the axonal and dendritic tenninals. Brady 27 (1993) provides a review of the molecular mechanisms involved in nerve transport, including identification of potential mechanisms for regulating transport and characterization of the influence of different axonal environments on transport. ·nu-ee categories of cytoskeletal elements play distinct roles in both neuronal transport and growth: microtubules, neuroftlaments and microfilaments. Each consists of a diverse set of polypeptides that are confined in various configuration5 to fill specific functions within the nervous system. There are overlapping functions between classes, while isoforms exist with.in each class of cytoskeletal protein; these may subtly change local properties of the cytoskeleton (reviewed by Brady, 1988). 1.5.4 Microtubules Microtubules serve at least two main roles in the mature nervous system. First, they provide a structural framework for the axon (Freide & Samora_jski, 1970), playing an important role in determining the size and morphology of neuronal processes. In small unmyelinated fibres and in many dendrites, both of which have few neurofilaments, microtubules act as the primary structural element. A second critical role for microtubules is connected with intracellular transport, mediated by kinesins and dyne.ins, the molecular 'motors' of the neuronal transport system (Brady, 1991). During axon elongation and regeneration the synthesis and transport of proteins associated with growth is increased. The primary subunits of microtubules are alpha and beta tubulins, which are disassembled as part of the structural organisation of the cytoskeleton and then must be reassembled for neurite extension, such that the stability of assembled microtubules plays a critical role in effective regeneration (Brady & Black, 1986). 28 Tubulin is the major protein of vertebrate brain, representing 10-26% of total brain protein, whereas the protein content of other tissues is only 1-3% tubulin. Less than 5% of the protein derived from glia and other non-neural cells of the brain is tubulin, so most brain tuhulin is derived from neurons (Hiller & Weber, 1978) 'The importance of the tubulins in regeneration can he inferred from changes in axonal transport of tubulin associated with regeneration (McQuarrie & Lasek, 1989) and the up-regulation of specific tubulin genes during axonal growth or regeneration (Wong & Oblinger, 1990; Mathew & Miller, 1993). Relevant to this thesis, but not to this section on nerve regeneration, is the reported presence of microtubules and tubulin in pineal glands of some species. This matter is reviewed and discussed in Chapter 4. 1.5.5 Interactions between regenerating axons and t11elr environment A complex interplay between extrinsic influences and mechanisms intrinsic to the neuron initiate and determine the pattern ofregenerative processes. Based upon a number of observations, the control of regeneration is considered to be independent of cell body mechanisms (Fawcett & Keynes, 1990). There is evidence that sprouting of mouse motor terminals may occur even when axons are disconnected from their cell bodies (Brown & Lunn, 1988), and that an axon completely disconnected from its cell body can elaborate a new growth cone in response to axotomy (Mason & Muller, 1982). Also, the earliest regenerative sprouting at the proximal axon tip can occur ·within a few hours of axotomy (Argiro & Johnson, 1982), too rapidly for the cell body to have been informed by a retrograde signal and to have sent its response, even by fast axonal transport (Collins & Lee, 1982). Similar considerations apply during development, during which it would be impossible for the frequent changes in the rate of axonal growth 29 to be regulated sufficiently rapidly from the cell body (Cowan, 1979a). These observations suggest that there may he controlling mechanisms for axon elongation., acting rapidly and locally at the axon terminal.. From transplantation studies, both in vitro and in vivo (for review see Hatton, 1985), the accepted scenario for initiation of axon growth involves extensive interactions between an axon and multiple environmental cues. Olsen and Mal.mfors (l 970) showed that when transplanted into the anterior chamber of a host eye, a piece of iris, though deprived of its ovm nerve supply for 3 months, could evoke collateral sprouting from intact sympathetic nerves. Target tissue influence was implicated. Patterns of neural activity are also dependent on the availability of ghal cell surface molecules and extracellular matrix components (Muller, 1992). Glial cells function in the mainteance of ionic homeostasis (Kuffier, 1967; Walz & Hertz, 1983; Walz, 1989), the monitoring of presynaptic and post.synaptic neuronal activity (Kuffier, 1967; Orkand, et al, 1966; Wuttke, 1990), and in factors supporting and guiding growth and migration of regenerating nerves (Ard & Bunge, 1988; Hatten, et al, 1982; Levine & Card, 1987; Pixley, et al, 1987). The intrinsic ability of a neuron to respond to environmental cues, and the nature of its response to any particular cue, depend on the neuron's expression of relevent receptors, signal tranducing proteins, and on the presence of structural materials to execute a response. Purves (l 975) indicated that expression of some of the neural genes involved in axon growth and culminating in synaptogenesis, decreased sharply as neurons matured, so that the extracellular cues that stimulated axon elongation in developing neurons evoked different responses from mature neurons. Alternatively, repression of growth-related properties in many neurons during differentiation suggests that some of the genes involved in axon growth might be expressed transiently during development and be reinduced during successful neive regeneration. Relevant studies have 30 concentrated on the synthesis of proteins destined for transport into axons and their terminals, the population of neural proteins most directly involved in axon functions. 111..is has prompted a search for related genes using electrophoretic methods (Katz, et al, 1965) and phenotypic studies investigating expression of netve growth factors by means ofimmunocytochemistry (Fantini and Johansson., 1992). In almost all neurons screened in this way, it has been possible to identify one or more axonal proteins whose synthesis is specifically increased an order of magnitude during developmental or regenerative growtl1. 1.5.6 Neurnl growth factors Neuronal changes necessary for nerve regeneration generally lead to increases in the supply of substances necessary to rebuild fue growing neive fibre . Also there are extrinsic changes in the quantities of trophic (growth-promoting) factors that function in attracting axons to grow in tl1e proper direction (Finger & Stein, 1982). The first extrinsic factor to be isolated was nerve growth factor (Bueker, 1948; Levi­ Montalcini and Hamburger, 1951), which has been observed to influence adrenergic neurons in fue sympathetic nervous system and to promote an increase in the growth of axons from sympathetic ganglion towards a source of NGF (Campenot, 1977). Diamond, et al (1987) have demonstrated that intact sensory neurons sprout in response to denervation of adjacent sensory fields in fue skin, an effect that is inhibited by systematically administered antibodies to NGF. Furthermore, increasing local concentrations of NGF at the terminals of developing sympathetic neurons promotes increased neural growth, bofu in vivo (Edwards, et al, 1989) and in vitro (Campenot, 1982). Also, administration ofNGF to neonates causes both increased terminal sprouting 31 (Levi-Montalcini and Angeletti, 1968) and increased dendritic arborization of sympathetic neurons (Snider, 1988) While very little NGF or the receptors for NGF are found in the normal mature nervous system, both are apparent in areas of nerve regeneration. When a nerve is cut or crushed, the level of both the NGF molecule, NGF receptors and their respective mRNAs in the region distal to the injury, increases enormously (Heumann, et al, 1987). The expression of NGF by Schwann cells is probably contrnlled by axonal contact; NGF receptor levels increase in the absence of axonal contact, and decrease when contact is restored (faniushi, et al, l 988). Johnson and his colleagues suggest that the NGF receptor may act both in trophic support and as a cell adhesion molecule, the NGF having an association with NGF receptors on hoth axons and Schwann cells (Johnson, et al, 1988). 1.5. 7 Growth cones During target-directed outgrowth, when the nervous system is developing, or when a neuron is regenerating following injury, the growing axon elaborates a prominent, growth-specific structure at its leading edge. This 'growth cone' is able to translocate over a substrate and to make synaptic contacts with appropriate target neurons; it is highly specialised to perfonn a number of functions (for review, see Johnston & Wessels, 1980; Bunge, et al, 1983; Landis, 1983; Kater & Letourneau, 1985; Letourneau, 1985) including those of motility ( Letourneau, 1975; Tosney & Wessels, 1983; Argiro, et al, 1984; Bray & Chapman, 1985), pathfinding (Bastiani & Goodman, 1986; Bently & Toroian-Rayrnond, 1986; Caudy & Bently, 1986; Kuwada, 1986), adhesion to the substrate (Letourneau, 1975; Hammerback & Letourneau, 1986), and membrane addition (Bray, 1970, 1973; Pfenninger & Maylie-Pfenninger, 198la,b; Pfenninger & Johnson, 1983). 32 The molecular mechanisms that control growth cone formation and that regulate its function are unknown, however, a potential clue to the control of growth cone formation is that certain proteins, designated growth-associated proteins, or GAPs, are much more abundant in neurons with growing axons than in neurons where synaptic connections have already been established, suggesting that they may perform functions required at earlier stages during axonal growth. One of these, GAP-43, a phosphopeptide with a molecular weight of approximately 43 kDa (Basi, et al, 1987; Karns, et al, 1987), is synthesized and axonally transported at elevated levels during both developmental and regenerative axonal growth (Skene & Willard, I 98 la,b ). Subcellular fractionation, electrophoresis and immunocytochemical experiments have demonstrated that GAP-43 is a principal component of the growth cone, and is highly enriched there ( Mciri, et al, 1986; Skene, et al, 1986). Yet despite the extensive studies, both in vitro and in vivo, investigating GAP-43 associations with growth cones per se (Meiri, et al, 1986; Skene, et al, 1986; Meiri, el al, 1987; Goslin, et al, 1988; Goslin & Banker, 1990), to date only one investigation has focused on this protein's associations with axonal development during functional regeneration (Schreyer & Skene, 1993). Results from that experiment suggest that GAP-43 induction in the dorsal root ganglion of rats is caused by disconnection from target tissue and not by axon injury per se . 1.5.8 Growth associated protein- 43 (GAP-43) GAP-43 is an axonally transported phosphoprotein found on the inside of the membrane of regenerating and growing axons, particularly near the growth cone (Meiri, et al, 1988), and comprises of the order of 1% of the total protein in growth cone membranes (Skene, et al, 1986). Biochemical determination indicates that GAP-43 is 33 approximately 10-fold more concentrated in both developing and regenerating netvous tissue than in adult netvous tissue (Meiri, el al, 1986). The protein is induced very rapidly following axotomy, with levels increasing up to 100 times and then decreasing again on reinnervation (Skene, et al, 1986). Newly synthesized GAP-43 is rapidly transported down axons as part of the fast axonal transport system (Skene & Willard, 1981c), which uses membrane-bound vesicles (20-60 nm) to deliver material from its site of synthesis in the cell body (Pfenninger & Johnson, I 983) The time course of GAP-43 expression during axon regeneration is consistent with periods of axon elongation and the initial formation of an association between the terminal axon and target tissue. Meiri, et al (1988), working with a polyclonal antibody to GAP-43 in neonatal rat SCG tissue, reported a variation in the distribution of the protein as axons elongate. The cell body became progressively less irnmunoreactive, whereas the growth cone at the tip of the growing axon reacted more strongly. Finger & Stein (1982) suggested that the initial rise in GAP-43 synthesis coincided with, or slightly preceded, initiation of axon out.growth in fish and amphibian optic netves, but that the protein's synthesis did not peak or plateau until regeneration was well underway. In rat dorsal root ganglia, induction ofGAP-43 mRNA is shown to begin between l and 2 days after sciatic nerve injury, corresponding to the end of a lag period preceding ax.on outgrowth (McQuarrie, et al, 1977). However, induction of GAP-43 is not considered to be a secondary consequence of outgrowth, as application of colchicine to rat dorsal root ganglia at the time of netve injury, or at the end of a 2 day post-crush lag period, which should have prevented ax.on outgrowth, was found to have no effect on the time course or amplitude ofGAP-43 induction (Skene, 1989). Elevated GAP-43 expression continues throughout the period of axon elongation and also synaptogenesis in all developing and regenerating systems examined (Skene, 1989). GAP-43 induction in regenerating systems begins just early enough not to rule 34 out a role in the early phases of axon outgrowth, and persists just long enough not to rule out participation in later phases of synaptogenesis and maturation of the axon's terminal arbor. A slow decline in GAP-43 synthesis late in axon development or regeneration, could permit the protein to play some role in synaptogenesis, or in the active so11ing out of the terminal arbor. Localization of the protein to growth cones and the distal portions of outgrov.ring neurites, however, does argue against direct GAP-43 participation in maturation of the axon structure behind the immature axon sprouts, or in the slow growth in axon diameter, myelination and maturation of the axon's electrical properties (Goslin & Banker, 1990). 35 1.5.9 Structure and Biochemical Characterl~1ics ofGAP-43 Structural and biochemical studies of GAP-43 reveal a novel protein with an amino acid sequence that is highly conserved among mammals (Goslin & Banker, 1990) and that appears to interact extensively with several intracellular messenger systems (Goslin, et al, 1988). For a protein associated predominately with membranes, GAP-43 is surprisingly hydrophilic. GAP-43 is synthesized as a soluble protein, whose post-trdllSlational association with membranes is considered to be mediated by covalent attachment of fatty acid (Skene, 1989). A short hydrophobic region at the amino terminus of the protein is the most likely site of fatty acylation and membrane attachment (Basi, et al, 1987). The extreme hydrophilicity of GAJ>-43 and the nature of its membrane attachment are consistent with models that envision the protein extending away from the cytoplasmic surfaces of growth cones and synaptic membranes (Meiri, et al, 1988), in a position to interact with cytoplasmic or cytoskeletal proteins on one hand, and reversibly attached to the membrane on the other. GAP-43 has been identified as a uruque calmodulin-binding protein, binding calmodulin selectively in the absence of calcium, and releasing calmodulin at higher calcium concentrations. This 'reversed' calcium dependence for calmodulin binding contrasts to the usual mediation of calmodulin on the secondary messenger calcium in various physiological processes such as excitation-contraction coupling and excitation­ secretion coupling (for review see Cheung, 1980). On the basis of its abWldance, membrane binding properties, and reversed pattern of calmodulin binding; GAP-43 has been proposed to act Wlder low calcium conditions to sequester calmodulin in certain regions of the neural membrane, th.en releasing the calmodulin upon influx or mobiliz.ation of free calcium (Cimler, et al, 1987). However, because the calcium- 36 dependent antagonism of GAP-43 binding with calmodulin is strongly affected by ionic strength (Alexander, et al, 1987), it is not clear whether calcium regulates calmodulin binding to GAP-43 under any or all physiological conditions. An alternative regulator of calmodulin binding to GAP-43 is protein kinase C (PKC). Phosphorylation of GAP-43 by PKC strongly inhibits binding of that protein to calmodulin (Alexander, et al, 1987). Signalling across growth cone membranes is an essential feature of axon elongation and synaptogenesis, and numerous studies have implicated PKC in these events (Spinelli, el al, 1982; Murphy, et al, l 983; Hsu, et al, I 984; Mattson, el al, 1988; Hall, et al, 1988; Girard & Kuo, 1990). As a prominent substrate for PKC, a functional role for GAP-43 could be mediation of some of the effects of PKC on growth cone function (Meiri, et al, 1988; Nelson, et al, 1989). The identification ofGAP-43 as a major substrate of PKC and the induction of increased levels of GAP-43 in response to disconnection from target tissue (Schreyer & Skene, 1993), suggest that GAP-43 may be an element of, or may be regulated by, a transduction system that enables the growth cone to sample the environment and then to generate an internal response. 1.6 Aims of the present study The principal aims of the studies described in this thesis were to: (i) Confim1 the use of melatonin secretory responses, to darkness exposure of sheep, as a parameter of pineal fnnction. (ii) Examine the effects of unilateral SCGX on the profiles of melatonin secretory capacity over a period of28 days post-operatively, to explore whether or not fitl1 recovery of function occurred over that period. (iii) Investigate the occurence of new neural growth in pineal tissue as a response to partial denervation, using immunocytochemical localization ofGAP-43. (iv) Use localization of alpha tubulin to determine whether pinealocyte cell integrity was maintained after unilateral SCGX 37 38 CHAPTER 2 Materials and Methods AJI of the aims of thesis were investigated in a single experiment incorporating radioimmunoassay techniques to measure plasma melatonin levels, for which specific methods have been described in Chapter 3. In Chapter 4 the technique of immunocytochemistry was used to investigate the occurence of both GAP-43 and alpha tubulin in pineal tissue. With this approach it was aimed to establish whether or not functional recovery was related to neural growth, which has not previously been reported in the literature. While details of specific experimental techniques are described in Chapters 3 and 4, more general materials and methods (animal management, experimental design and surgical procedures) relevant to both those chapters, are given in Chapter 2. 2.1 Animal management and treatment groups 2.1.1 Animal management Romney ewes, approximately 9 months of age and weighing 30 - 35 kg , were used in the experiment detailed in this thesis. The natural photoperiod at the ti.me the animals were transferred to experimental conditions was 12 hr light : 12 hr dark (l2L:12D), with dawn occuring at approximately 0600 hrs (New Zealand Nautical Almanac, 1992-1993). 39 All animals were constrained in pens and housed indoors in a well ventilated, light- proof room maintained at a constant 15°C during autumn (May/June) 1993. Fluorescent lighting, which provided approximately 200 lux at the eye level of each animal, and was controlled by automatic time switches to provide an 8L: 16D lighting regime, with lights on from 0500 to 1300 hr. Feed consisted of approximately 800 grams ofluceme chaff per day, ·with water available ad libitum. All animals were acclimatized to the 8L:16D photoperiod for 21 days prior to surgery. 2.1.2 Treatment grou11s Twenty animals were randomly allocated to one of four groups (n = 5), with respect to a given time of sacrifice for harvesting of pineal glands. Respective animal groups were subjected to the following experimental procedures: Control Group: 15-20 min Saffan/Halothane anaesthesia, without sham SCGX. 3 Day Group: S/H anaesthesia and unilateral SCGX. 14 Day Group: S/H anaesthesia and unilateral SCGX. 28 Day Group: S/H anaesthesia and unilateral SCGX. All animals were bled in the dark for 4 hr on the day before surgery or anaesthesia (see Section 3.2.2) to test their melatonin secretory capacity. Subsequent bleedings for this purpose were conducted as follows (i) on days 1, 3, 7, 14, 21 and 28 after surgery or anaesthesia for the 28 day and control groups; and (ii) on days 3 and 14, respectively, for the 3 and 14 day groups. At the end of those last bleedings, all animals were euthanased and pineal glands collected for immunocytochemical evaluation (Chapter 4). 40 2.2 Sw·gical techniques 2.2.1 General surgical techniques a) Food was withheld for 24 hours preceding sw:gery. Water continued to be available ad libitum during this period. b) Anaesthesia was induced with intravenous 'Safran' (Glaxovet Ltd, Barefield, England) containing alphaxalone 0.9% w/v and alphadalonc acetate 0.3% w/v, average dose 3mg/kg body weight. Anaethesia was maintained with 2-3% (v/v) halothane ('Fluorothanc', ICl, Macclesfield, Cheshire, England) in oxygen, after intubation with a Magill endotracheal tube, and supplied at a rate of 2 I/min from a Fluotec 3 (Cyprane, Keighley, England) vapourizer. c) After induction of anaesthesia, unilateral SCGX was performed according to the method of Appleton and Waites (1957) (Section 2.2.2). All surgical procedures were conducted under sterile conditions. 2.2.2 Superior cervical gangllonedomy In this procedure a skin incision was started over the zygomatic process at a point mid-way between the canthus of the eye and the base of the ear, and was continued parallel with the mandible to the level of the thyroid cartilage. The thin subcutaneous platysma muscle was then incised along the same line, care being taken not to damage the ext.ernaljugular vein, which lies just deep to the muscle at the ventral end of the incision. 41 The depressor auriculae muscle was incised with the platysma muscle. After reflecting the cut edges of this muscle the parotid salivary gland, lying in fat in the dorsal area of the operative field, and the external jugular vein became visible. lbe few vessels which pass caudally from the extemaljugular and internal maxillary veins were cut between ligatures, an.d the caudal and ventral borders of the parotid gland were then freed from underlying tissue to enable this gland to be retracted rostro-dorsally. The common carotid artery could then be identified under fat, deep to the external jugular veirt, and its cranial course to where it is crossed by the hypoglossal nerve could be identified. Here the artery passes deep to the digastric muscle, dorsal to which is the fan-shaped jugulo-hyoid muscle. This latter muscle was partially cut through to enable retraction of the epihyoid bone. By retracting the bone and the cut ends of the muscle using self-retaining retractors, the superior cervical ganglion, the cervical sympathetic nerve and related structures could be readily freed from the fat which surrounds them. Following removal of the superior cervical ganglia the operation was completed by suturing the cut ends of the jugulo-hyoid muscle, suturing the caudal border of the parotid gland into place and closing the platysma muscle and skin in layers. Each operation was completed in approximately 20 mins. 2.2.3 Post surgical care Following recovery from anaesthesia, animals were placed back in their crates and given prophylactic antibiotic treatment consisting of an initial l O ml intramuscular injection of 'Streptopen' (Glaxo (NZ) Limited) followed by five consecutive daily treatments of 5 ml. This preparation contains procaine penicillin and dihydrostreptomycirt, each at a concentration of250 mg/ml. 42 2.3 Blood collection, pl'Ocessing and melatonin radioimmunoassay In Section 2 of Chapter 3 details are given for the periodic collection of blood, subsequent plasma extraction and radioimmunoassay protocol and data processing, used to determine pineal melatonin secretory capacity in response to unilateral SCGX. 2.4 lmmunocytochcmistry Section 2 of Chapter 4 details the collection of pineal tissue, its processing and the subsequent localization of GAP-43 and alpha tubulin using immunocytochemical techniques. 2.5 Statistical analyses. Analysis of variance (ANOVA) was used to assess data derived from both Chapters 3 and 4. In addition, paired !-tests were performed on data in Chapter 3. Details of statistical analyses are given in respective chapters. Levels of significance in all statistical analyses are denoted thus: * P < 0.05 ** P < 0.01 *** P < 0.001 43 Chapter 3. Effects of unllaternl superior cenrical gangllone<.1omy on ovine melatonin secretory responses to darkness exposure 3.1 Introduction As mentioned in Chapter 1, the pineal gland is innervated by noradrenergic sympathetic fibres originating from the right and left SCG's, the two nerves each providing approximately 50% of the pineal innervation, which is distributed equally over the two halves of the gland (Lingappa & Zigmond, 1987). Norepir1ephrine released at the nerve terminals regulates a number of aspects of pinealocyte biochemistry, with the most dramatic being the regulation of the synthesis of the hormone melatonin., during darkness (Reiter, 1976). Studies investigating pineal gland innervation commonly have used the technique of bilateral SCGX, which eliminates rodent pineal rhythms of NAT activity and melatonin content (Reiter, et al, 1979), as well as sheep pineal melatonin secretory capacity (e .g. Lincoln, et al, 1982; Lapwood, 1993). It also causes a reduction in pineal weight (Barrell & Lapwood, 1978-9) and pinealocyte atrophy (Mockett & Lapwood, unpublished). As discussed in section 1.3.5, the effects of unilateral SCGX also have been investigated, with most workers measuring pineal enzyme activity as their parameter of pineal function or secretory potential. Thus Reiter, et al (1979) reported that pineal NAT activity was reduced by half two days after surgery; Zigmond, et al (1981) on the other hand, found that while NAT levels were depressed 75% on the night after unilateral SCGX, effective recovery had occured the following night. Later Domay, et al (1985) measured pineal tyrosine hydroxy-lase activity 44 and tritiated noradrenaline uptake after unilateral SCGX; both were depressed by >>50% 2 days after surgery, hut increased to 80% of control levels by 10 days post-operatively. Until recently only one report appears to have been published on effects of unilateral SCGX on melatonin production: Reiter, et al (1979) found that pineal melatonin content was reduced about 50% two days post-surgery. Recently, Lapwood (1993) took the physiological approach of measuring melatonin secretion profiles during exposure of sheep to darkness, before and periodically after unilateral SCGX. On the day after surgery there was a reduction of pineal melatonin secretory capacity to values 92% (P< 0.001) below those measured before surgery, however, succeeding measurements indicated a substantial recovery of that parameter of melatonin biosynthesis, to within 77% of pre-operative levels after 14 days. This current study also used measurement of melatonin secretion profiles during exposure of sheep to darkness, but was designed to examine the degree of recovery in pineal melatonin secretory capacity which occured when an extended post-surgery survival time of28 days was utilized. 3.2 Materials and Methods 3.2.1 Group treatments Animal management and treatment group allocation is detailed in Sections 2.1 .1 and 2.1.2, respectively. Surgical techniques are described in detail in Section 2.2 45 3.2.2 Blood collection and processing Faint red illumination from a hand torch was used during the periods of darkness to aid in the blood sampling. After the acclimatization period of 3 weeks, on the day before surgery blood was collected twice prior to lights off then at 30 minute intervals for a four hour period in the dark, in order to measure pre-treatment melatonin secretory capacities. Blood samples were collected by jugular venipuncture into 7 ml vacutai.ner tubes containing 105 units of heparin, then immediately centrifuged at 3000 rpm for 15 ruins at 4°C. Care was taken to prevenl the haemolysis of blood samples as this has been rcp,lrted to lead to a spurious increase in apparent melatonin levels (Fraser, et al, 1983). After separation from cells, plasma was stored at -20°c until thawing 24 hours prior to assay. Blood sampling was again repeated at days I, 3, 7, 14, 21 and 28 days post-surgery or post-anaesthesia., using the above bleeding schedule, to study melatonin secretory capacities during darkness exposure, so as to measure responses after partial denervation or anaesthesia. Animals to be killed 3 and 14 days after surgery were bled in such a manner only on the day before surgery and on the day they were killed. Plasma samples were analysed for concentrations of melatonin as a direct measurement of the capacity of individual pineal glands to secrete that hormone. No account was taken of variable metabolic clearance rates for melatonin, consequentl