Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Growth-associated messenger ribonucleic acid expression in a model of successful central nervous system… Pataky, David Michael 1992

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1992_spring_pataky_david.pdf [ 7.38MB ]
JSON: 831-1.0086485.json
JSON-LD: 831-1.0086485-ld.json
RDF/XML (Pretty): 831-1.0086485-rdf.xml
RDF/JSON: 831-1.0086485-rdf.json
Turtle: 831-1.0086485-turtle.txt
N-Triples: 831-1.0086485-rdf-ntriples.txt
Original Record: 831-1.0086485-source.json
Full Text

Full Text

GROWTH-ASSOCIATED MESSENGER RIBONUCLEIC ACID EXPRESSION INA MODEL OFSUCCESSFUL CENTRAL NERVOUS SYSTEM REGENERATIONbyDAVID MICHAEL PATAKYB.Sc, The University of British Columbia, 1987A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCEinTHE FACULTY OF GRADUATE STUDIES(Department of Zoology)We accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAAPRIL 1992© David Michael PatakyIn presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)Department of  10 0 IA (., YThe University of British ColumbiaVancouver, CanadaDate 3o(41 9-1,DE-6 (2/88)ABSTRACTRecent experiments examining the development and plasticity of the chickembryonic spinal cord have described its inherent ability to recover from injurysuffered prior to embryonic day (E)13. Severing the spinal cord on or before E12resulted in complete anatomical and functional recovery, which defined thepermissive period for repair. Transections performed on E13 or later resulted inparaplegia characteristic of the restrictive period for repair. Previous studieshave described the expression of growth-associated proteins (GAPs) such asGAP 43 and a-tubulin which were expressed at high levels during axonextension, then down-regulated at the time of target contact. These proteinswere also expressed during abortive attempts at regeneration, and characterizeda genetic growth program which was recruited after injury.This study examined the neuronal response to injury as reflected bychanges in growth-associated gene expression in the hindbrain measured usingNorthern blotting. Levels of mRNA for GAP 43 and total a-tubulin duringnormal development were found to peak at E10-12, the period of maximaloutgrowth of brainstem-spinal projections. Complete spinal cord transectionsperformed on Ell (successful repair) or E14 (unsuccessful repair) did notdetectably alter total a-tubulin mRNA levels. GAP 43 mRNA levels were notdetectably altered after Eli transection. In contrast, transection on E14(unsuccessful repair) resulted in a maintained increase in GAP 43 mRNA levelsat least until 7 days post-transection, the longest survival period studied.Northern blotting was likely not sensitive enough to detect the fullcomplement of changes which occurred after injury. However, transectionduring the restrictive period for repair resulted in a maintained increase in GAP43 mRNA expression. These data suggested that some brainstem-spinaliprojection neurons injured on E14 retained the inherent ability to re-express atleast part of the axonal growth program, indicated by the appropriate re-expression of GAP 43 mRNA. This suggests that the presence of inhibitoryinfluences (or the absence of facilitatory influences) after E13 may haveprevented the re-growth of axons and the re-formation of appropriate synapses.iTABLE OF CONTENTSABSTRACT ..^.. ..iiTABLE OF CONTENTS .. ivLIST OF TABLES^.. ..vLIST OF FIGURES .. •• ..viACKNOWLEDGEMENTS .. •• ..viiINTRODUCTION ..^.. ..1MATERIALS AND METHODS ..14RESULTS^..^••^.. ..24DISCUSSION^..^.. ..40CONCLUSIONS^•• ..49REFERENCES^••iv..52LIST OF TABLES1. Sample Sizes used for the developmental studyand the transection study ..2. Analysis of Variance of log-transformed GAP 43expression levels after transection..^....15..33vLIST OF FIGURES1. Construction of the GAP 43 cDNA clone..^..202. Quantification of RNA amounts loadedin lanes^..^ ..253. Typical Northern blots of GAP 43 andtotal a-tubulin mRNA levels during development4. Densitometric analysis of GAP 43 mRNA levelsduring development^••5. iTigcl al developmental changes in GAP 43A levels from a selected blot ..6. Densitometric analysis of total a-tubulinmRNA levels during development7. Changes in GAP 43 expression afterEll transection^..^..^..^••8. Changes in GAP 43 expression afterE14 transection^..^..^..^..379.^Changes in GAP 43 expression after transectionsuperimposed upon typical developmentalexpression levels^..^•• ..39..28..29..30..31..36viACKNOWLEDGEMENTSI dedicate this thesis to my wife Soiban who was largely responsible formaintaining normalcy in my non-scientific life (heaps of understanding, moralsupport, help in the library, warm dinners, and clean laundry), and withoutwhom I would probably still be writing this. To a lesser degree, this also appliesto the rest of my family, whom I think still remember who I am.Special thanks to John Steeves for supporting my efforts in theseendeavours. Thanks are also deserved by Sohail Hasan for assistance with thetransections. I have a great number of colleagues throughout the department,besides everyone in the Steeves lab, whose helpful hints and equipment loanswere instrumental in the success of these experiments, and I thank them iINTRODUCTIONIt is well established in the literature that the central nervous system (CNS)of adult higher vertebrates is incapable of responding to injury with a regenerativeresponse culminating in the restoration of function (Ramon y Cajal, 1928; Barnardand Carpenter, 1950; Bjorklund et al, 1971). This is in marked contrast to theperipheral nervous system which is proficient at re-innervating targets after injuryin the same higher vertebrates (Ramon y Cajal, 1928; Hall, 1989). Thisdichotomous response to injury has been the focus of a great many studiesattempting to elucidate the factors responsible for allowing neurons to, orinhibiting them from re-connecting with their targets.This is not to say that the adult mammalian CNS is incapable of growth orplasticity in response to injury. Neurons within the CNS demonstrated non-regenerative terminal sprouting after injury (Crutcher and Marfurt, 1988) but thiswas usually insufficient to restore lost function. Given the correct environment,such as a peripheral nerve graft, CNS neurons of mammals have demonstrated arobust growth response (Tello, 1911; Richardson et al, 1980; Weinberg and Raine,1980; David and Aguayo, 1981; Benfey and Aguayo, 1982; Aguayo et al, 1983;Richardson et al, 1984; Vidal-Sanz et al, 1987; Carter et al, 1989). However, whenthis growth was re-directed into the adult CNS environment, it was abortedwithin close proximity to the graft/CNS border (David and Aguayo, 1981). Underthe right conditions then, some mammalian CNS neurons demonstrated the abilityto regenerate.It was clear that all adult mammalian CNS neurons did not enjoy thiscapacity for regeneration. The studies quoted above did not claim to rescue 100%of axotomized neurons with the peripheral nerve graft (Richardson et al, 1982;Villegas-Perez et al, 1988). Studies using the sensitive technique of in situ1hybridization to examine neuronal growth-associated gene expression in therubrospinal system (neurons previously thought incapable of regeneration) havedemonstrated a sub-population of neurons capable of mounting a regenerativeresponse qualitatively similar to that seen in the peripheral nervous system(Tetzlaff et al, 1991). Thus these neurons appeared to be capable of regenerationyet were unable to functionally re-connect with appropriate targets. The CNS is atissue of unimaginably complex connections, trophism, extracellular andintracellular factors which orchestrate its functions. Clearly there were manyelements to consider when trying to determine why CNS neurons failed tosuccessfully regenerate.There were yet other examples of regeneration observed in the developingCNS of higher vertebrates. The ultimate goal of regeneration, re-connecting aneuron with its targets, was similar to that of a developing nervous system whichbuilt these connections de novo. Initially, the developing CNS provided apermissive environment for neuronal growth, and therefore increased thelikelihood that it would be capable of a successful regenerative response. Thishypothesis has been tested in several experimental paradigms. Destruction of theceruleo-cerebellar projection showed a critical developmental period forsuccessful regeneration, beyond which repair ultimately failed (Schmidt et al,1980), as did the retinotectal pathway projecting through the superior brachium inneonate hamsters (So et al, 1981). A high level of plasticity in the corticospinaltract of the neonatal rat has also been confined to a critical period in development(Bernstein and Stelzner, 1983). Studies on neonatal cortical lesions in the rat byKolb and Whishaw (1989) also described a developmental window during whichsparing of function is greatest. These observations set a precedent fordevelopmental "windows" during which propensity for CNS repair was increased.2The experiments described here endeavoured to combine the sensitivity ofmolecular biological techniques with the study of regeneration in the embryonicCNS to examine gene expression in brainstem-spinal locomotor projectionneurons after spinal cord injury in the developing chick embryo. The embryonicchicken was an ideal model for studies of CNS regeneration for several reasons.The embryo was readily accessible and much more amenable to experimentalmanipulation than a mammalian embryo, as maternal complications wereavoided. There were striking similarities in the development and organization ofavian and mammalian locomotor systems, allowing direct comparisons to bemade. Most importantly, the embryonic spinal cord demonstrated a criticalperiod for successful repair of damage, allowing the comparison of regenerativeresponses in the permissive and restrictive periods for repair as defined below.A CRITICAL PERIOD FOR SPINAL CORD REGENERATIONIN THE EMBRYONIC CHICKRecent experiments have studied the development and plasticity of aviandescending brainstem-spinal projection neurons involved in the control oflocomotion. These brainstem-spinal projections originated from the red nucleus,vestibular nuclei, pontine and medullary reticular formation regions (Lawrenceand Kuypers, 1968a,b; Armstrong 1986; Garcia-Rill, 1986; Jordan 1986; McClellan,1986; Garcia-Rill and Skinner, 1987a,b; Steeves et al, 1987; Grinner and Dubuq,1988; Webster and Steeves, 1988). A large proportion of brainstem locomotoroutput was directed to the spinal cord by reticulospinal pathways originating inthe pons and medulla (Peterson, 1984; Steeves and Jordan, 1984; Armstrong, 1986;Garcia-Rill, 1986; Steeves et al, 1987). By studying the effect of complete highthoracic spinal cord transection on these pathways at different developmental3stages, a critical period for spinal cord repair has been defined in the chick embryo(Hasan et al, 1991; 1992)This critical period for repair has been characterized using a combination ofanatomical and physiological techniques on embryos which received completehigh thoracic spinal cord transections (Valenzuela et al, 1990; Hasan et al, 1991;1992). Anatomical recovery was assessed after transection by restricted injectionof a neuroanatomical retrograde tracing chemical into the lumbar cord. If thespinal cord transection was performed prior to embryonic day (E)13, thedistribution and number of retrogradely labelled brainstem-spinal neurons wasidentical to sham-operated or unoperated controls. Transections performed onembryos aged E13 or older resulted in reduced retrograde labelling of brainstemnuclei (Hasan et al, 1991).The functional recovery from mid-thoracic transection was assessed bybehavioural observation of the motor function of post-hatching chicks, and byfocal electrical stimulation of brainstem locomotor regions (Valenzuela et al, 1990;Hasan et al, 1991). Leg and wing electromyographic recordings were used tomonitor brainstem-evoked motor activity both in ovo (E18-E20) and after hatching.In comparison to sham- and un-operated control animals, complete functionalrecovery of motor function was evident for chicks with spinal cords transectedprior to E13. These findings were consistent for all methods of assessment(behavioural observation or brainstem stimulation), therefore embryonic spinalcord was capable of complete functional recovery from transections performedprior to E13. The quality of spinal cord repair diminished with transection at laterstages of development (E13 or older), resulting in paraplegia as in adult spinalcord-injured animals. These findings were reminiscent of the critical periods forrepair seen in the other developing systems mentioned previously (Schmidt et al,1980; So et al, 1981; Bernstein and Stelzner, 1983; Kolb and Whishaw, 1989).4One possible criticism of these observations arose from attributing therepair to the descent of late-developing pathways. However, studies of theontogeny of brainstem-spinal projection systems refuted this claim, as thesepathways were found to be constructed early in development. Reticular neuronsin the hindbrain were among the earliest to become post-mitotic, the earliestbecoming post-mitotic at stage 4 (Sechrist and Bronner-Fraser, 1991; stagingaccording to Hamburger and Hamilton, 1951). The earliest developing projectionshave reached the lumbar cord by E5 (Windle and Austin, 1936; Okado andOppenheim, 1985), and synaptogenesis occurred soon after the arrival ofsupraspinal projections (Shiga et al, 1991). All descending supraspinal pathwayswere largely complete by E11-12 (Okado and Oppenheim, 1985). These findings,combined with the physiological and anatomical evidence presented above,indicated that spinal repair was dependent on regeneration of axotimizedprojections as opposed to the subsequent descent of late developing pathways.Data from double retrograde labelling experiments supported theinvolvement of regeneration in the recovery process as opposed to subsequentdevelopment. In these experiments, brainstem spinal projections wereretrogradely labelled on E10, a transection performed on E12, then a second dyeinjected caudal to the transection site several days later, labelling those projectionswhich have grown through the transection site. The presence of both retrogradedyes within the same neuron supported the hypothesis that regeneration hadoccurred (Hasan et al, 1992). These data clearly defined a permissive period forsuccessful spinal cord regeneration (prior to E13), and a restrictive (unsuccessful)period (post-E13).These results, along with descriptions from other developing systems listedpreviously, suggested that for functional regeneration to occur in developing (ormature) CNS, there must be a permissive environment and the ability on the part5of the neurons to regenerate. During these "critical developmental periods" clearlythe first criterion was met. But what of the second? Were all developing neuronscapable of repairing injury? One approach to addressing this question was toexamine molecular events occurring during normal axonogenesis, then comparethem to events which occurred after injury. The underlying assumption was thatrebuilding an axon involves virtually the same mechanisms which supported theoriginal outgrowth. There was a great deal of evidence in the literature whichsuggested that cellular events occurring during regeneration were essentially arecapitulation of developmental processes (Price and Porter, 1972; Holder andClark, 1988). The following pages describe some molecular events occurringduring development, and what is known of their recapitulation duringregeneration.MOLECULAR CORRELATES OF AXONOGENESISStudies of axonogenesis have described many measurable events whichresulted in the formation of functionally appropriate synapses between a neuronand its target. The construction of neurites began with the elaboration ofstructures called growth cones at the leading edge of the neurite process. Thegrowth cone was thought to mediate interactions with the extracellularenvironmental cues which result in the directed extension of a neurite (Goldbergand Burmeister, 1989; Lipton and Kater, 1989; Gordon-Weeks, 1991; Strittmatterand Fishman, 1991).Axonogenesis required the accelerated synthesis of many structuralcomponents, such as a- and f3-tubulin for the synthesis of microtubules.Microtubules were directly involved in axonal transport and growth cone motility,and provided structure to the developing axon (Yamada et al, 1970; Bamburg et al,61986; Bray, 1987;). Other proteins whose presence correlated with axonogenesisbut whose functions were not so nearly well defined such as the growth-associated protein (GAP) 43, were also expressed at high levels duringaxonogenesis (Kalil and Skene, 1986; McGuire et al, 1988; Dani et al, 1991) andwere expressed in a developmentally regulated manner (Jacobson et al, 1986).GAP 43GAP 43 was originally identified by Skene and Willard (1981a,b) as one ofseveral proteins travelling via fast axonal transport in growing and regeneratingaxons. Subsequent isolation and purification revealed its true molecular weight tobe approximately 24kD, but its highly basic structure resulted in anomolousmigration on SDS-PAGE gels, hence the misnomer (Benowitz and Routtenberg,1987). This protein was simultaneously described by several other groupsexamining different aspects of neuronal function, and not until amino acidsequence data was available did they realize all were studying one and the sameprotein. Hence the literature contains references to pp46 (Meiri et al, 1986), Fl(Lovinger et al, 1985), B-50 (Dokas et al, 1990), and P-57 (Masure et al, 1986). It hasalso been proposed that it be referred to as neuromodulin (Liu and Storm, 1990).A review of its structure and cellular distribution provides some clues forthe reported function of GAP 43. It was highly enriched in growth cones ofdeveloping or regenerating axons (Meiri et al, 1986; 1988; Moya et al, 1989), butnot dendrites (Goslin et al, 1988, 1990). With the correct isolation procedure, itpurified with cytoskeletal elements (Allsop and Moss, 1989), although acytoplasmic pool has also been proposed (Estep et al, 1990). In the growth cone ithas been localized to the cytoplasmic side of the plasma membrane (Van LookerenCompagne et al, 1989; Gorgels et al, 1989) and in the adult brain it occupied a7similar site in presynaptic terminals of highly plastic cortical areas (Gispen et al,1985).GAP 43 has been implicated to play a role in neurotransmitter release andlong-term potentiation (LTP; Lovinger et al, 1985; Nelson and Routtenberg, 1985;Routtenberg and Lovinger, 1985; De Graan et al, 1990). As a substrate of proteinkinase C, it was one of the major proteins whose phosphorylation correlated withthe onset of LTP. It was not surprising that areas of the brain which continued toexpress GAP 43 in the mature CNS included the hippocampus and associativeareas of cortex (Neve et al, 1988; Benowitz et al, 1988; de la Monte et al, 1989;Benowitz et al, 1990).Studies of the structure of GAP 43 revealed a potential palmitylation sitenear the amino terminus which may allow interaction with the growth conemembrane via insertion of the fatty acyl tail into the lipid bilayer (Skene, 1989).The 10 N-terminal amino acids were thought to be sufficient to target the proteinto axonal growth cone membranes (Zuber et al, 1989). However, recent workindicated they may be necessary, but not sufficient to target protein products tothe axonal growth cone membrane (Liu et al, 1991). The molecule also had acalmodulin binding site at residues 43-51 with an adjacent phosphorylation site atserine residue 41. Phosphorylation at this site has been shown to block thecalmodulin binding activity of GAP 43, thus it may be a modulator of intracellularCa++ signalling by acting as a calmodulin "sponge" (Skene, 1990). Otherphosphorylation sites have also been described, but their functional significancewas unknown (Apel et al, 1991). GAP 43 also stimulated the binding of GTP-?-S toG., another way in which it may interact with second messenger systems(Strittmatter et al, 1990).Recent elegant work by Meiri et al (1991) using antibodies tophosphorylated and non-phosphorylated isoforms of GAP 43 indicated that8phosphorylation had the effect of stabilizing the growth cone cytoskeleton. Thusit likely acted as a part of the physiological "stop signal" when the appropiate cuewas received from the environment (such as contact with target cells). Theseresults, although exciting, were still preliminary and much work must be done todetermine the molecular interactions which produced this phenomenon.In spite of the work quoted above, it was still not clear how all of theobservations could be coalesced into a working model for the actual function ofGAP 43. Taken together, these observations have been proposed to indicate a rolein signal transduction modulation (perhaps extracellular cues) and/or integrationwith growth cone cortical cytoskeletal movement. It was clear, however, that thecorrelation with axon outgrowth could be exploited for studying developmentand regeneration in the nervous system (Benowitz et al, 1990).TUBULINDuring the neuronal process extension stage of neuronal development, theassembly of microtubules was required to provide the structure of the neurite(Daniels, 1972). Microtubules were assembled from the a- and P-tubulins (and themicrotubule-associated proteins), both of which have several isotypes derivedfrom different genes and different post-translational modifications (Cleveland andSullivan, 1985). Functional a-tubulin proteins arose from at least five genes in aneight member gene family in the chicken (Pratt et al, 1987; Pratt and Cleveland,1988; Valenzuela et al, 1981 ) producing five distinct polypeptide isotypes whichare then subjected to post-translational modification. The expression of at leastfour of these genes has been detected in the chick CNS (Pratt and Cleveland,1988).9Up-regulation of a-tubulin expression in the CNS during development hasbeen described in the chick and rat (Bamburg et al, 1973; Bhattacharya et al, 1987;Miller et al, 1987b). Differential developmental expression of a-tubulin isotypeshas been described in the rat CNS (Miller et al, 1987a). Tal tubulin (Tal, a neuronspecific isoform) was expressed at high levels during embryogenesis, correlatingwith the period of axon extension, and was then rapidly down-regulated whenaxonal growth was complete. This contrasted with T26 tubulin which wasconstitutively expressed (ie. not developmentally regulated, Miller et al, 1987b).This phenomenon has not been well studied in the chick, and it was unknown if asimilar situation exists.Tubulin mRNAs were subject to a rapid autoregulatory mechanismwhereby the presence of excess depolymerized tubulin subunit protein resulted inthe degradation of tubulin mRNA (Cleveland, 1988). The mechanism appeared toinvolve binding of unpolymerized (3-tubulin subunits to four amino-terminalamino acids of the nascent 13-tubulin polypeptide which resulted in cleaveage ofthe mRNA under translation (Cleveland, 1988; Cleveland, 1989) Hence cells onlyproduced as much tubulin protein as was needed at any time due to thetranslational control imposed by excess unpolymerized tubulin subunits. Thusthe resultant degradation of tubulin mRNA by this mechanism dictated that thelevels of intact tubulin mRNA detected by Northern blotting reflected the cellularneed for tubulin.The observations described above for GAP 43 and tubulin provided insightinto the developmental machinery required for axonogenesis. An understandingof these mechanisms underlying normal growth and development was necessaryfor unravelling the processes occurring during successful and unsuccessful CNSregeneration. In light of the CNS/PNS dichotomous response to injury, a10description of what was known about growth-associated gene expression duringregeneration follows.GENETIC GROWTH PROGRAMSAn examination of developmentally regulated mRNA from the nervoussystem provided one aspect of the description of a neuronal genetic "growthprogram" which can be re-activated in response to injury (Hoffman andCleveland, 1988; Miller and Geddes, 1990). During PNS development and repair,events such as axonal elongation were monitored by examining the expression ofspecific mRNAs for growth associated proteins such as GAP 43, and a-tubulin.This recapitulation of developmental processes (Holder and Clark, 1988) duringregeneration has been described in numerous studies of the molecular response toneuron injury.An example of this was the up-regulation of a-tubulin mRNA in responseto axotomy of facial and hypoglossal nerves (Pearson et al, 1988; Miller et al, 1989).Again, Tal was rapidly (within 4 hrs) upregulated in neurons of the facial nucleusin response to injury whereas T26 tubulin mRNA levels remained unchanged(Miller et al, 1989). As expected, there were parallel changes in the level of tubulinprotein synthesis in response to injury (Tetzlaff et al, 1988). The level of Tal (andtotal a-tubulin) mRNA was subsequently down-regulated at the time of targetcontact (Miller et al, 1989). Similarily, severing PNS axons initiated regenerativemechanisms including the re-expression of GAP 43 at high levels (Meiri et al, 1988;Van der Zee et al, 1989; Hoffman, 1989; Tetzlaff et al, 1991) which resulted inregrowth of the axon.Less was known about the expression of these molecules in the CNS inresponse to injury. Initially, it was thought that GAP 43 re-expression only11appeared at high levels during successful regeneration of the axon (Skene andWillard, 1981a,b; Skene, 1984; Kalil, 1988) and therefore would only be re-expressed in PNS after injury. A similar re-induction of GAP 43 and structuralelement mRNAs was seen in the rubrospinal and retino-tectal pathway afterinjury, and when allowed to grow into a peripheral nerve graft (Tetzlaff et al,1991; Doster et al, 1991). Recent reports, however, have indicated that GAP 43 andtotal a-tubulin were re-expressed after injury with or without regeneration (Bisby,1988; Doster et al, 1991; Tetzlaff et al, 1991), and were subsequently down-regulated after successful re-contact with appropriate targets.The pattern of expression of these mRNAs likely formed a part of thegenetic "growth program" which resulted in the successful linking of neuron andtarget during development. This "growth program" was also recruited after CNSinjury, even when regeneration was unsuccessful (Doster et al, 1991; Tetzlaff et al,1991). GAP 43 and total a-tubulin mRNA levels remained low in most areas of theadult CNS unless the appropriate signal for re-expression, such as injury to theaxon, was received. The fact that high levels of expression of these mRNAs wasrequired during attempted growth, taken together with their low levels ofexpression in the adult CNS, indicated that they may be excellent markers of thecapacity for CNS repair. However, no studies have been done on GAP expressionin a successful, functionally regenerating CNS model.The experiments described here provided for the first time an examinationof neuronal growth-associated mRNA expression in the successfully regeneratingembryorc CNS. This model also allowed comparison of the successful responseto injury suffered during the permissive period for repair with the responseduring the restrictive (unsuccessful) period for repair. This allowed thecorrelation of the regenerative responses seen at the molecular level with the12physiological success or failure of spinal cord repair in the chick embryo. Theexperiments were designed to address the following hypotheses:1) The normal developmental expression levels of GAP 43 and total a-tubulin mRNA in pons and medulla will follow the classic growth-associated pattern of expression and temporally correlate with previousanatomical data delineating the temporal development of brainstem-spinalaxonal projections.2) Previous anatomical and physiological data demonstrated successfulrepair of brainstem-spinal projections after E11 transection, therefore,complete thoracic spinal cord transection on Ell will result in a transientre-expression of total a-tubulin and GAP 43 mRNA, followed by a return tocontrol levels.3) Complete thoracic spinal cord transection on E14 does not result inanatomical or functional recovery, therefore E14 transection will result in amaintained re-expression of GAP 43 and total a-tubulin mRNA levels.Briefly, GAP 43 and total a-tubulin mRNA expression in the hindbrain ofthe chicken embryo peaked at E10-12, correlating with the time of maximal axonaloutgrowth of brainstem-spinal projections. There was no detectable change intotal a-tubulin mRNA levels after either permissive or restrictive period spinalcord transection. Transection during the permissive period for repair did notresult in a transient re-expression of GAP 43. However, transection during therestrictive period for repair did result in a maintained increase in GAP 43 mRNAlevels.13MATERIALS AND METHODSAnimalsFertilized White Leghorn chicken eggs (Gallus Domesticus) were obtainedfrom B&J Farm (Surrey, B.C.). The eggs were stored at 1000 for no longer thantwo weeks before incubation at 38°C with high humidity to allow germination tocontinue. All animals used in this study were staged at the time of sacrificeaccording to Hamburger and Hamilton (1951).Animals aged E4-E14 were sacrificed by rapid decapitation withoutanaesthetic, as the pain pathways resulting in the conscious perception of painwere not functional until later in development. Animals E15 or older wereanaesthetized by the application of 0.05m1 of Sodium Pentobarbital (MTCPharmaceuticals) to the air-sac membranes (embryo weight 30-50g) prior todecapitation. Animals younger than Ell required tissue sample pooling prior toRNA isolation as follows: E10 - 2 embryos; E9 - 3 embryos; E8 - 4 embryos; E6 -6 embryos; E4 - 10 embryos.The experimental animals which received transections were divided intotwo groups: the first received complete mid- to high-thoracic spinal cordtransections on Ell, the second received transections on E14. The young age ofthe transected animals abrogated the use of anaesthetic during the procedure.Transections were performed by first making a small window in the eggshell,peeling away the shell membrane. A small hole was then made in the chorion,and the embryo secured around the neck with a bent glass probe. A sharpenedpair of forceps were used to sever the spinal column, and a complete transectionwas ensured by passing a pin, calibrated to the depth of the cord for thatembryonic age, through the transection site. The window was sealed with14paraffin and a glass cover slip, and the egg returned to the incubator for 1, 3, 5 or7 days post-transection (DPT) survival periods. A 2DPT group was alsoincluded for the Ell transection study. GAP 43 and total a-tubulin expressionwas then measured using Northern blotting as described below, to examine thenormal developmental expression and compare transected animal groups withage-matched control group expression levels with sample sizes as indicated inTable I.DEVELOPMENTAL STUDY TRANSECTION STUDYAge at Sample Treatment Sample # ofSacrifice Size Group Size ControlsE4 3* E11T12 7 6E6 3* E11T13 5 4E8 4* E11T14 6 6E9 2* E11T16 6 4El0 7* EllT18 4 4Ell 2E12 7 El4T15 9 6E13 2 E14T17 8 7E14 7 E14T19 8 5E15 2 E14T21 5 5E16 5E17 3E18 3E19 3E21 3Table I: Sample sizes used for the developmental study and the transectionstudy. Transected animal groups are coded by the age when transected (ie.EllT) followed by the age at sacrifice. * indicates number of pooled tissuegroups as outlined above.15RNA IsolationAll chemicals and reagents were from the Sigma Chemical Co. (St. Louis,MO) unless otherwise indicated. The procedures described below were adaptedfrom Miller (1989). The pons and medulla (cerebellum excluded) were rapidlydissected out of the embryos in ice-cold phosphate buffered saline (PBS, 10mMNaPO4 pH 7.4, 7.5% NaC1) and pooled as described above if younger than Ell.The tissue was rinsed several times with PBS, then placed in 250m1 ice-cold lysisbuffer (20mM Tris[hydroxymethyl]aminomethane hydrochloride buffer pH 8.0,0.2M NaCI, 20mM MgC12) for homogenization with manually operated teflon-coated micro-homogenizers. The homogenate was spun in an Eppendorfmicrofuge at 40C (16000g) and the pellet discarded. An equal volume of "secondbuffer" (2% SDS, 0.1M NaCI, 40mM EDTA) was added to each sample followedby proteinase K (Boehringer Mannheim) to a final concentration of 200 µg/ml,the digestion allowed to proceed for twenty minutes at ambient temperature.The samples were extracted three times with an equal volume ofphenol/chloroform/isoamyl alcohol (25:24:1, BDH), then precipitated overnightat -200C by addition of one tenth sample volume of 3.0M sodium acetate and 3sample volumes of 95% ethanol (BDH).The following day, the precipitate was spun in a microfuge at 4 0C(16000g) for twenty minutes, the supernatant discarded and the pellet allowed todry for 10-15 minutes at ambient temperature. The pellet was resuspended in50111TE buffer (10mM Tris[hydroxymethyl]aminomethane hydrochloride bufferpH 8.0, 1mM EDTA) and re-precipitated by addition of 5111 3.0M sodium acetateand 15Oµ1 95% ethanol and stored at -20°C for at least one hour, then theprecipitation protocol was repeated once. After the final precipitation, the RNA16was resuspended in 20-40p,1 of TE, depending upon the yield as estimated by thefinal pellet size.The concentration of each sample was determined by spectrophotometricanalysis of a 4111 aliquot diluted to lml in water. The absorbance at 260nm and280nm was measured with a Beckman DU 64 spectrophotometer to determinethe concentration (A26o x 10) and purity (A26o / A280 ratio), then each samplediluted as required with TE to a final concentration of 3p.g/4.7111. Samples withan A26o / A280 ratio of less than 2 were subject to re-purification.The RNA was stored for no longer than two weeks at -20°C beforeproceeding to the electrophoresis procedure. If it was necessary to store samplesfor longer periods of time, the isolation procedure was halted during one of theprecipitation steps, and continued when the sample was needed.Denaturing Gel ElectrophoresisSamples were size separated in standard formaldehyde denaturingagarose gels. Samples were applied to wells in the gel to produce blots of adevelopmental series, or blots with transected animal samples and their age-matched controls. Each lane of the developmental series blots represented RNAisolated from a single animal, with the exception of samples from animalsyounger than Ell, which represented RNA isolated from pooled tissue asdescribed above. Each lane of the transection blots represented RNA isolatedfrom a single animal.The gels were prepared using MOPS buffer (20mM 3-IN-Morpholino]propane-sulfonic acid pH 7.4, 5mM sodium acetate, 1mM EDTA)with 1.1M formaldehyde, and agarose added to 1.5%. The running buffer wasalso MOPS buffer with 1.1M formaldehyde. A 4.7p1 (311g) aliquot of each sample17was diluted with 14p1 "mix" (50% formamide, 1.1M formaldehyde and MOPSbuffer, final concentrations). Samples were heated at 65°C for 10 minutes, thenplaced on ice. Two microlitres of loading dye were added to each sample priorto loading the gel. The loading dye was made by mixing 6X loading dye (15%Ficoll type 400, 0.25% Bromophenol Blue) and ethidium bromide (10mg/m1) in a10:1 ratio. Typically, gels were run at 60V (constant voltage conditions) for 4-5hrs with a Hoeffer submarine gel unit powered by a GPS 200 power supply(Pharmacia), then photographed with UV transillumination (FisherBiotech FBTI816) using type 57 Polaroid film.After equilibrating the gel in 10X SSC (1X SSC is 0.15M NaCl, .015Msodium citrate) for 30-60 min., a standard capillary transfer apparatus was usedto transfer the RNA from the gel to Nytran membrane (0.45p,m pore size,Schleicher and Schuell). Using 10X SSC as the transfer buffer, the gel was placedupside down supported on plexiglass above a 10X SSC buffer reservoir onWhatman #3 filter paper wicks arranged to draw fluid from the buffer reservoir.A piece of Nytran cut to the size of the gel and pre-soaked for 5 min. in distilledwater was layered on top of the gel, followed by 3 layers of #3 filter paper pre-soaked in distilled water and a stack of paper towels topped with a small glassweight. The entire set-up was draped in Saran Wrap (Safeway) to preventevaporation.The transfer was allowed to proceed overnight, then the Nytran and thegel were viewed under UV transillumination to ensure that the transfer wascomplete. The blots were stained using 0.02% methylene blue in 0.3M sodiumacetate for 60-75 seconds, and destained with three rapid rinses in ice-cold 2XSSPE (1X SSPE is 0.15M NaCl, 10mM NaPO4 pH 7.4, 2.5mM EDTA) thenimmediately photographed with type 57 polaroid film. Inspection of themethylene blue stained blots revealed lanes which appeared over- or under-18loaded based on the intensity of the stained ribosomal RNA bands. Samplesfrom these blots were re-run with slightly adjusted amounts of RNA loaded inlanes judged to be unequal on the initial blot. This was repeated until blots wereobtained which demonstrated equal amounts of RNA in each lane as judged bymethylen blue staining. The blots were baked at 80 00 under vacuum for onehour then stored in a sealed plastic bag at -20 0C until used for hybridizations.Northern HybridizationsPlasmids containing cDNAs for chicken GAP 43 (kind gift from Dr. LarryBaizer; Baizer et al, 1990) and rat a-tubulin (kind gift from Dr. Freda Miller;Miller et al, 1987a,b) were linearized by restriction endonuclease digestionasindicated in Fig. 1, then cleaned with two phenol/chloroform/isoamyl alcoholextractions and two sodium acetate/ethanol precipitations. cRNA antisenseprobes were generated using T7 (GAP 43) or SP6 (a-tubulin) bacteriophage RNApolymerases as described by the manufacturer of the in vitro transcription kit(Promega, Madison WI) using a 32P-labelled CTP (40mCi/ml, 800Ci/mmol;Amersham, Oakville Ont.). Unincorporated nucleotides were removed by spincolumn chromatography using Sephadex G-50 (Sambrook et al, 1989).Incorporation efficiency was crudely estimated by Geiger counter(Ludlum Measurements Inc.) measurement of the beta particle emissions fromthe purified probe compared with the emissions from the column containing theunincorporated nucleotides, and was usually 40-50%. Under the conditions ofthe reaction used, this routinely gave enough probe to hybridize withapproximately 2x106 cpm/ml of hybridization solution as determined by liquidscintillation counting of a 1i.d aliquot of the radiolabelled probe using a BeckmanLS 5000 TA liquid scintillation counter.19GAP 43 cDNA -pGEM3Z DFigure 1: Construction of the GAP 43 cDNA. This diagram indicates theorientation of the chicken GAP 43 cDNA cloned into the EcoRI site of theplasmid vector pGEM 3Z (Baizer et al, 1990). After linearizing with Pvull (cutsat base 403 of the 1000 by clone), in vitro transcription with T7RNA polymeraseand radiolabelled nucleotides produces an antisense cRNA probe. The total a-tubulin probe (Miller et al, 1987a,b) is in a similar vector, but oppositeorientation, requiring linearization with Hind DI and transcription with SP6RNA polymerase.20Before hybridization, blots were rinsed for 15 min. in 2X SSPE plus 2%SDS with agitation to remove any remaining methylene blue stain. The blotswere prehybridized in 5X PIPES buffer (25mM piperazine-N,Nt-bis[2-ethane-sulfonic acid] pH 7.4, 750mM NaCl, 25mM EDTA) containing 50% formamide,0.2% SDS, 2001.ts each of denatured salmon sperm DNA and baker's yeast tRNA(Boehringer Mannheim), 5X Denhardt's (1X Denhardt's is 0.02% Ficoll, 0.02%Polyvinylpyrrolidone, 0.02% Bovine Serum Albumin) for a minimum of 2 hoursat 65°C. Hybridization was done in buffer with the same composition, with theaddition of the labelled probe, overnight at 65°C. All blots were probed withboth GAP 43 and total a-tubulin cRNAs sequentially. The first probe wasstripped by incubating blots in 80% formamide at 80°C for 1 hour prior to re-probing with the second cRNA.Following the hybridization, blots were rinsed for 15 min. in 2X SSC +0.2% SDS at ambient temperature, then 0.5X SSC + 0.2% SDS at 55°C, then 0.1XSSC + 0.2% SDS at 65°C to remove any unbound and non-specifically boundprobe. Blots were wrapped in saran wrap and apposed to Kodak XAR x-ray filmat -700C. The x-ray film was pre-flashed using a Vivitar camera flash unit and 17Kimwipe tissues between the flash unit and the x-ray film at a distance of 170cmto generate a 0.15 absorbance unit change prior to exposing the film with theblots (Laskey and Mills, 1975; 1977).The x-ray films were developed using a Kodak X-omat automaticprocessor, and the density of the bands in each lane quantified using acomputerized scanning densitometer (Molecular Dynamics (MD) with MDImagequant 3.0 software). The densitometer provided a plot of the peakdensities of the bands in each lane, the area under each peak calculated byintegration of the density plot. Each number generated in this fashion thereforerepresented a single sampling from a single animal except for animals younger21than Ell used for the developmental study (tissue pooled prior to RNA isolationas described above).The initial blots run for this study included RNA size markers (BoehringerMannheim, data not shown) to determine the sizes of the mRNA bands detectedby the GAP 43 and total a-tubulin probes. The GAP 43 cDNA clone encodeschicken GAP 43, therefore the cRNA probe used here is 100% homologous to theGAP 43 mRNA. As the hybridizations were performed at high stringency (0.1XSSC + 0.2% SDS at 65°C final wash), the bands that were detected represent theauthentic GAP 43 mRNAs. The banding pattern described here is also identicalto that described by Baizer et al (1990) as being authentic GAP 43 mRNA.The total a-tubulin cDNA used here encodes rat a-tubulin, but the codingsequences of the a-tubulin gene families across species are very well conserved(Cleveland et al, 1980; Cleveland and Sullivan, 1985; Pratt and Cleveland, 1988).The high stringency conditions employed in this study ensured that theappropriately sized band (1.8kb) detected with the total a-tubulin cRNArepresented authentic chicken a-tubulin mRNA.Data from the developmental blots were presented as integrateddensitometric measurements from the GAP 43 and total a-tubulin cRNA probedblots.The integrated densitometric measurements obtained from the GAP 43probed blots of transected versus control animal RNA were subject to anadditional control for variations in the amount of RNA loaded in each lane. Thedata from the total a-tubulin re-hybridizations did not demonstrate changes as aresult of transection. The small variations in the total a-tubulin signal from laneto lane within a blot represented small variations in the amount of RNA loadedin each lane (see Figure 2). Therefore, data from the total a-tubulin re-hybridizations were used to transform the GAP 43 data to remove variations in22the detected GAP 43 mRNA levels resulting from unequal loading and transferof lanes in a gel. An adjustment factor for each lane of the GAP 43 blots wasgenerated by dividing the average total a-tubulin signal across all lanes ofanimals the same age by the total a-tubulin signal in that lane. The GAP 43densitometric data from the corresponding lane was then multiplied by thisratio. The adjusted GAP 43 data was then expressed as a ratio of experimental toaverage age-matched control values. The log of the experimental/control GAP43 expression ratio was tested by one-way analysis of variance to discriminatestatistically significant (p<0.05) variations from zero.23RESULTSRNA IsolationThe amount of RNA isolated from a given animal remained relativelyconstant in terms of gg/wet weight of hindbrain (approximately 0.5-11.1g/mg).Transection of the spinal cord did not consistently alter this parameter. Theamount of RNA isolated from animals older than Elf) was sufficient to run fiveor more Northerns, but animals EIO and younger required pooling of tissue fromtwo (E10) to ten (E4) animals to provide sufficient quantities of RNA.RNA Sample Quantification, Equality of Lane LoadingSpectrophotometric determination of RNA sample concentrations was notaccurate enough to ensure equal amounts of mRNA were loaded in each lane ofa Northern blot, as it did not account for pipetting errors while diluting samplesfor the spectrophotometer or loading the gel, or the inequities in RNA transfer toNytran. In an effort to control these sources of error, blots were stained after thetransfer step with methylene blue to evaluate the quantity of RNA loaded in eachlane of the blot. Methylene blue allowed the visualization of the major RNAspecies in the cytoplasmic extracts, the ribosomal RNA (rRNA) bands (Figure 2c).Gels containing lanes which appeared to be over- or underloaded based on theintensity of the methylene blue-stained rRNA bands were run again, with smallcorrections in the amount of RNA loaded in the lanes. This technique allowed aqualitative estimation of the gel loading and transfer efficiency for thedevelopmental expression autoradiograms.24E17 E17 E14T17E14T17 a28S18S6.35.0-2.0..1.5"28S-1.8Figure 2: Quantification of RNA amounts loaded in lanes. (a) Northern blot ofE17 control mRNA (first four lanes, E17) and E14 transected animals sacrificedon E17 (next five lanes, E14T17). The E14T17 animal group showed increasedGAP 43 mRNA levels compared to the age-matched controls. (b) The same blotas in (a) stripped of the GAP 43 cRNA and re-probed with the total a-tubulincRNA. Total a-tubulin levels were essentially equivalent in all lanes, providing aquantifiable assessment of the minor variations between lanes not detected bymethylene blue staining (c) or ethidium bromide staining (d). Non-specificbinding to 28S ribosomal RNA band is indicated in (b). Indicated sizes are inkilobases.25The results from the total a-tubulin probed transection blots (multiplelanes of RNA from animals the same age) equalized in this manner indicated thatthe variation of any one lane was not great. For example, the averagedensitometric value obtained from Figure 2b was 56.65 arbitrary units, and thegreatest variation from that value in any one lane was less than 30%. Thissuggested that the discrimination of a tenfold change in mRNA levels over theembryonic development period of the chick could easily be accomplished usingthe methylene blue staining technique. Ideally, it was necessary to re-probe thedevelopmental blots with a constitutively expressed mRNA probe, to provide aquantitative analysis of mRNA amounts in each lane. Such a probe wasunavailable for studying the development of the chick CNS, therefore methyleneblue staining was the best alternative.Both probes used in this study have been well characterized (Baizer et al,1990; Miller et al, 1987). The GAP 43 cRNA detected two mRNAs (6.3kb and1.5kb) as previously described (Baizer et al, 1990). Although the total a-tubulinprobe was derived from murine brain, the coding sequences of the a-tubulingene families across species were very well conserved (Cleveland et al, 1980;Cleveland and Sullivan, 1985). This cDNA was capable of detecting the Taltubulin isotype when cRNA probes were transcribed from the 3' untranslatedregion, however the cross-species variation in this region of the mRNA was toogreat between rat and chick to discriminate different a-tubulin isotypes in thisfashion (data not shown). However, under highly stringent washing conditions(0.1X SSC + 0.2% SDS at 650C), cRNA probes generated from the entire cDNAdid recognize a single mRNA band of the correct size (1.8kb) for all a-tubulinsexpressed in the chick hindbrain.26GAP 43 and a-Tubulin Expression During DevelopmentAnother problem arose from the absence of probing the developmentalblots for a constitutively expressed mRNA. Without such an internal standard, itwas not appropriate to combine data from different blots for several reasons.There were numerous variables associated with the production of anautoradiogram including variations in amount of mRNA loaded per lane, thespecific activity of the probe, final probe concentration, length of exposure timeof the x-ray film, and determination of the baseline for densitometric analysiswhich all affected the actual densitometric values obtained from blot to blot. Theinability to combine data from different developmental blots also precludes theuse of statistical analysis with the developmental study data, rendering itdescriptive at best. However, the raw densitometric values from all blots plottedon a single set of axes provided an indication of the trends which occurredduring development of the chick embryo hindbrain.Probing with the GAP 43 cRNA revealed a large (6.3kb) and a small(1.5kb) form of GAP 43 mRNA detected during the development of thehindbrain (Fig. 3a). The results from all developmental RNA blots shown inFigure 4 indicated the changes occurring in GAP 43 mRNA levels duringdevelopment. GAP 43 mRNA levels followed a growth-associated pattern ofexpression. Early in development levels were low, then rose to a peak at E10-12,then fell to near-adult levels by E21 (Figure 4). For simplicity, the data from arepresentative blot (Figure 5) was chosen to illustrate these changes, as it alsoindicated the relative amounts of the large and small GAP 43 mRNAs. Levels ofGAP 43 mRNA in the E4 embryo were almost below the level of detection ofthese techniques, but notably, the 1.5kb message was slightly more abundantthan the 6.3kb message at this stage of development (Figure 3a, 5). As276.3 kb.1.5 kb. •1.8 kbaGAP 43 TOTAL a-TUBULINI----- CD •:1-^C3)LLJ LLJ LLJ LLJ UJ LLI LLJC\1 M d- LO (1)0) 1r— T T T— T T T-LLJ LLI LLI LLJ LLI L1J LLI LLJFigure 3: Typical Northern blots of GAP 43 and Total a-Tubulin mRNA levelsduring development. (a) GAP 43 mRNA levels were almost undetectable at E4,rose to a peak at E10-12, then declined to near-adult levels by E21. (b) Total a-tubulin mRNA levels followed a similar time-course, but the peak expressionoccured over a broader range of developmental ages. Compare E10-E14expression levels for GAP 43 and total a-tubulin. Sizes of the transcripts were asindicated previously, developmental stage of embryos (days of incubation)indicated above each lane.28GAP 436.3kb mRNA1^14^6^81^112^141^1^1 16^18^201014013012011010090807060504030201001009080706050403020100 1^I^I^I^1^t^1^I 4^6^8^10^12^14^16^18^20GAP 431.5kb mRNAAGE OF EMBRYO (DAYS)Figure 4: Densitometric analysis of GAP 43 mRNA levels duringdevelopment. Each line in these plots represented an individual blot containinglanes of RNA isolated from animals of varying age, probed with the GAP 43cRNA. These data demonstrated the growth-associated pattern of GAP 43mRNA expression for the 6.3kb mRNA (a) and the 1.5kb mRNA (b) asdevelopment progressed.29• 6.3kb mRNA0 1.5kb mRNA120  110 -100 -90 -8070 -60 -50 -4030 -20 -10 -04^6^8^10^12^14^16^18^20AGE OF EMBRYO (DAYS)Figure 5: Typical developmental changes in GAP 43 mRNA levels from a selectedblot. Selected GAP 43 mRNA developmental expression data which illustrated thechanges in the relative amounts of 6.3kb and 1.5kb mRNA over the developmentalperiod examined (E4-E21).301009080706050403020100 4^6^8^10^12^14^16^18^20ACE OF EMBRYO (DAYS)Figure 6: Densitometric analysis of total a-tubulin mRNA levels duringdevelopment. Each line in this plot represented an individual blot containinglanes of RNA isolated from animals of varying age, probed with the total a-tubulin cRNA. This data demonstrated the growth-associated pattern of total cc-tubulin mRNA expression as development progressed.31development progressed, the 6.3kb GAP 43 message level increased 20-30 fold topeak on E10-12, while the 1.5kb message increased 5-10 fold over a similar timecourse. Levels of both message then declined to near adult levels by E21; the6.3kb species all but disappeared, and the 1.5kb species decreased toapproximately one-fifth of the peak expression levels (Figure 3a, 5).Probing with the total a-tubulin cRNA revealed a single mRNA band1.8kb in length. The developmental time course of a-tubulin expression wassimilar to GAP 43, but the peak expression occurred over a broader time period(Figure 3b, 6). Levels were almost undetectable at E4, then rose approximately10 fold by E8. There was a much smaller increase through to the peak expressionat E12, then a gradual decline (approximately 5 fold) to near adult levels by E21(Figure 6).Effects of Spinal Cord TransectionStatistical AnalysisThere was no detectable change in a-tubulin expression after any of thesurvival periods post-E11 or -E14 transection. For example, Figure 2a,bdemonstrated the GAP 43 and total a-tubulin mRNA levels in nine animals agedE17, five of which received spinal cord transections on E14. The methylene blue-and ethidium bromide-stained blots (Figure 2c,d) demonstrated the near-equivalence of RNA amounts in each lane. The GAP 43 signal from thetransected animals was greater than the control GAP 43 signal, yet the same blotre-probed with the total a-tubulin cRNA yielded only minor variation in the totala-tubulin mRNA levels across all the lanes (Figure 2a,b). The slight variations inthe total a-tubulin signal did not correlate with either control or experimental32Level Count AverageStnd. Error(internal)Stnd. Error(pooled s)95 Percent Confidenceintervals for mean1 7 -.087 .023 .033 -.154 -.021*2 7 -.046 .029 .033 -.112 .0193 5 -.061 .027 .039 -.139 .0164 5 -.094 .027 .039 -.172 -.015*5 6 .082 .033 .035 .010 .153*6 6 .059 .030 .035 -.012 .1307 6 .006 .019 .035 -.065 .0778 6 -.070 .018 .035 -.142 .00079 4 -.095 .023 .044 -.183 -.007*10 4 -.151 .009 .044 -.238 -.063*11 9 -.025 .025 .029 -.084 .03212 9 -.078 .022 .029 -.136 -.020*13 8 .437 .040 .031 .375 .499*14 8 .193 .030 .031 .131 .255*15 8 .402 .069 .031 .340 .463*16 8 .283 .036 .031 .221 .345*17 5 .477 .013 .039 .398 .555*18 5 .086 .016 .039 .007 .164*Total 116 .087 .008 .008 .071 .103Tabl e 2: Analysis of Variance of Log-transformed GAP 43 expression levelsafter transection. The ANOVA levels are coded as follows: 1, 2: E11T12 (Elltrap sect with E12 sacrifice) 6.3kb, 1.5kb mRNA respectively, as with all othergrolip;; 3,4: E11T13; 5,6: E11T14; 7,8: E11T16; 9,10: E11T18; 11,12: E14T15;13,14: E14T17; 15,16: E14T19; 17,18: E14T21. The count indicates the samplesize for each group. Asterisks indicate statistical significance (p<0.05).33groups, but were indicative of minor variations in the amount of RNA in eachlane which was undetectable with methylene blue or ethidium bromide staining.This allowed the use of the total a-tubulin data from transected animal blots asan internal control, and provided a quantitative assessment of the actualvariation in RNA amounts in each lane. This allowed an adjustment to be madeof the GAP 43 signal of a particular lane by multiplication with the ratio of theaverage total a-tubulin signal across the blot to the total a-tubulin signal for thatlane. The adjusted GAP 43 values in the transected animal lanes were thenexpressed as a ratio to the average adjusted age-matched control value, log-transformed, and subjected to one-way analysis of variance (ANOVA). Theresults from the ANOVA are presented in Table 2, with 95% confidence intervalscalculated for the (log-transformed) mean change over control from each group.If the 95% confidence interval included zero, the results were not statisticallysignificant.The adjusted GAP 43 expression levels after transection were alsoexpressed in terms of percent change from control values (Figures 7c, 8c), thenthese changes were superimposed upon the typical normal developmentalexpression graphs provided in Figure 5 to illustrate the relative change withrespect to the changing baseline which resulted from the normal developmentaldecline in GAP 43 mRNA levels over the one week survival period examined inthis study (Figure 9).34Effects of Spinal Cord Transection in the Permissive Period on GAP 43 mRNALevelsOne day following an Ell transection, average GAP 43 levels dropped17% for the higher size message, and 12% for the lower size message (Figure7a,c), but the variability in this transected animal group prevented the differencefrom being statistically significant for the smaller mRNA species. Two DPT(Figure 7c) the larger GAP 43 message was closer to the control values (12.5%drop), but the smaller message dropped to 18% below control values(significance level p<0.05). Message levels 3 DPT (Figure 7a,c) were 22% abovefor the larger size (p<0.05) and 16% above for the smaller size (not statisticallysignificant). By 5 DPT (Figure 7b,c), the larger size GAP 43 message hadreturned to control levels (2% difference) while the smaller size had dropped14% below control values (not statistically signigficant). Message levelscontinued to drop below control values 7 DPT (Figure 7c), with the larger GAP43 message 19% below control (p<0.05), and the smaller message 29% belowcontrol (p<0.05). These changes were very small and insignificant compared tothe developmental levels of GAP 43 mRNA being expressed during this period.These results suggest there was an increase in GAP 43 expression 3 DPT, whichreturned to control levels 5 DPT but this was inconclusive (Figure 7c).35Figure 7: Changes in GAP 43 Expression After Ell Transection. (a)Representative Northern blot which showed GAP 43 expression one (E11T12) andthree (E11T14) dayspost-E11 transection compared with age matched controls (E12and E14 respectively). Note the near-equivalence of 1 DPT and E12 control GAP 43mRNA levels, but 3 DPT animals showed an increase over E14 control mRNA levels.(b) GAP 43 mRNA expression 5 days post-E11 transection (E11T16) returned tocontrol levels. (c) Plot of percent change in experimental animal GAP 43 mRNAlevels (1,2,3,5 and 7 DPT) from age-matched control levels. (* indicates p<0.05, errorbars calculated from ANOVA 95% confidence intervals)36Figure 8: Change in GAP 43 Expression After E14 Transection. (a) RepresentativeNorthern blot which showed GAP 43 mRNA levels one (E14T15) and three DPT(E14T17) compared with age-matched controls (E15 and E17 respectively). GAP 43mRNA levels were equivalent to control levels 1 DPT but increased above controlamounts 3 DPT. (b) A similar Northern to (a) compared five DPT (E14T19) and sevenDPT (E14T21) to age matched controls (E19 and E21 respectively). GAP 43 mRNAlevels were still increased above control levels even 7 DPT. (c) Plot of percent changein experimental animal GAP 43 mRNA levels (1,3,5 and 7 DPT) from age-matchedcontrol levels. The data demonstrated a maintained re-expression of GAP 43 mRNA 3-7DPT. (* indicated p<0.05, error bars calculated from ANOVA 95% confidence intervals)37Effects of Spinal Cord Transection in the Restrictive Period on GAP 43 mRNALevelsOne DPT, GAP 43 message levels were slightly below control levels (-6%,-16% upper and lower respectively), but only the smaller mRNA signal wassignificantly different (p<0.05; Figure 8a,c). Animals allowed 3 DPT recoveryperiods (Figure 8a,c) showed a large increase in GAP 43 message levels, 174% forthe large size message (p<0.05) and 56% for the smaller message (p<0.05). Thisincrease was maintained 5 DPT, with 152% increase in the larger message(p<0.05) and a 92% increase in the smaller message (p<0.05) over control values(Figure 8b,c). An increase in GAP 43 message levels was still evident 7 DPT(Figure 8b,c), with 192% increase in the larger message (p<0.05) and a 30%increase in the smaller message (p<0.05). Thus transection in the restrictiveperiod for repair resulted in a maintained increase in GAP 43 mRNA levels(Figure 8).An assessment of the absolute magnitude of the change relative to thedecreasing baseline was obtained when the changes in GAP 43 expression aftertransection were related to the changing baseline of GAP 43 expression over theone week survival period. This figure was generated by applying the percentchanges from baseline presented in Figures 7 and 8 to the values of GAP 43expression indicated in the typical developmental plot illustrated in Figure 5.The magnitude of the observed increases 3 DPT was greater after E14 transectionthan after Ell transection, for both the 6.3kb (Figure 9a) and the 1.5kb (Figure 9b)mRNA. The magnitude of the peak change seen after E14 transection wassimilar to the peak changes seen over the course of normal development.3816014012010080604020012011010090807060504030201004^6^8^10^12^14^16^18^204^6^8^10^12^14^16^18^20AGE OF EMBRYO (DAYS)Figure 9: Changes in GAP 43 mRNA levels after transection superimposedupon typical developmental expression levels. This figure provides anindication of the magnitude of the changes in GAP 43 expression aftertransection relative to the decreasing baseline expression of the 6.3kb mRNA (a)and the 1.5kb mRNA (b). * indicates p<0.05, error bars calculated from theANOVA 95% confidence intervals.39DISCUSSIONThe results presented here illustrate a number of features of growth-associated protein expression in the developing hindbrain. Levels of expressionof growth-associated mRNA peaked during periods of maximal processoutgrowth from this brain region. After Ell transection, levels of GAP 43 andtotal a-tubulin mRNA expression did not increase above control levels despiteregrowth of axons. After E14 spinal cord injury, however, GAP 43 (but not totala-tubulin) mRNA levels showed a second peak in expression (3 days afterinjury) which was maintained until at least 7 DPT. A comprehensive discussionof these results follows.CRITIQUE OF THE METHODSOne of the most vital controls necessary with Northern blotting is toensure that comparisons made between lanes on a blot account for differences inthe relative amounts of mRNA in each lane. Some researchers have reported thatquantification of the amount of ribosomal RNA in each lane, either by ethidiumbromide staining of RNA in the gels (Bonini and Hofmann, 1991), or themembrane after transfer (Corea-Rotter et al, 1992) or hybridization with a 28Sribosomal RNA probe (de Leeuw et al, 1989) were accurate and reliable ways forassessing the equality of RNA loading across the lanes of a blot. Thesecalculations assume that the relative amounts, and susceptibility to degradation,of ribosomal and messenger RNA remain constant across and within animalgroups.These assumptions must be questioned when studying a developingsystem. For example, Figure 2 demonstrated some variability in the total a-40tubulin signal which was not as obviously represented in either the methyleneblue stained blot or the ethidium bromide stained gel. This variability did notcorrelate with the experimental treatment, but appeared to represent smalldifferences in mRNA loading and/or individual variation in mRNA/rRNA ratioamongst animals. The margin of error which remained after equalization bymethylene blue staining appeared to be on the order of ± thirty percent.Although this is almost an order of magnitude smaller than the changes seen inGAP 43 and total a-tubulin expression during normal development of thehindbrain, it is equal to or greater than the changes seen in expression followingEll spinal cord transection.Ideally, a constitutively expressed mRNA should also be probed andmeasured to allow a more quantitative analysis of the changes which occurredduring development. A constitutively expressed mRNA is defined asdemonstrating constant levels of expression in spite of experimentalmanipulation. Unfortunately, a neuron-specific, constitutively expressed mRNAfor which a chicken cDNA is available has not as yet been found, if one indeedexists. The next best alternative was to rely on the methylene blue staining for anapproximation of equality of loading across the lanes. Although this precluded aquantitative analysis, making the results descriptive at best, it did reveal thetrends which occurred during development.For the transection study, all data collected from experimental animalswere carefully compared to data collected from age-matched control animals.Minor variation, however, arising from intrinsic variation in the animals or as aresult of variation in the RNA isolation and Northern blotting procedures cannotbe completely discounted with this approach. Methylene blue staining was usedinitially to approximate the equality of loading from lane to lane, however amore quantitative analysis was possible due to the inability to detect any41significant changes with the total a-tubulin cRNA probe (see Figure 2). Therewas always minor variation in the a-tubulin signal, but this variation did notcorrelate with the experimental treatment, and is probably accounted for byindividual variation between animals and potential inequities in gel loading andtransfer. Although it was initially hypothesized that total a-tubulin mRNAlevels would change with transection, there are several possible explanations forwhy no changes were observed. It is most likely that the techniques employedare not sensitive enough to detect the changes in a-tubulin expression whichshould be occurring in the small subpopulation of cells in the hindbrain injuredby the transection. The high levels of a-tubulin being expressed in the majorityof cells in the hindbrain which were not injured probably diluted the signal fromthe injured neurons beyond the limits of detection. Alternatively, the levels of a-tubulin expression were already high enough that no further increases werenecessary or possible to accomodate the regenerative process. What ever thecase, in effect, the total a-tubulin mRNA levels acted as a constitutivelyexpressed message, allowing a quantitative evaluation of the differences in theamount of RNA loaded in each lane which the methylene blue staining wasunable to discriminate.Following all attempts to normalize data and remove variability due toextraneous factors, the GAP 43 mRNA levels, when compared between animalswithin any one group on a blot (either within transected or control groups of thesame age), still showed some minor variability. This undoubtedly reflects themany manipulations of the samples between the tissue isolation and the finaldensitometric measurement which introduces variation which is too difficult toeliminate. Hence, despite this effort, an accurate quantification of small changesin mRNA levels could not be attained with this technique.42Another major drawback of Northern blot analyses is that results can onlybe discriminated at the tissue level, with one datum obtained per animal (orpooled tissue sample). The data provides no clues to the cellular eventsunderlying the observed changes. This is a critical point to consider whenresults are interpreted from a heterogenous collection of cells such as thehindbrain. The in situ hybridization experiments now underway should addressthese concerns. Northern blotting, however, did provide useful informationconcerning temporal changes in mRNA expression.DEVELOPMENTAL EXPRESSION OF GROWTH-ASSOCIATED mRNAIt was interesting that the GAP 43 probe recognized two mRNAs duringdevelopment. The 1.5kb mRNA was described as the authentic GAP 43 messagefrom which the protein is transcribed, and the function of the larger speciesremains unknown (Baizer et al, 1990). The expression of multiple mRNAs for asingle protein is not isolated to GAP 43 expression in the chick, but has also beenreported from studies of nicotinic acetycholine receptor mRNA developmentalexpression in Drosophila CNS (Hermans-Borgmeyer et al, 1989). The largemRNA form found in that system was shown to contain intronic sequences usingintron-specific probes on Northern blots. It is unknown, however, if a similarmanifestation occurs with GAP 43 expression.The results shown in Figures 3, 4, 5 and 6 support the first hypothesis, anddemonstrated that the expression of GAP 43 and total a-tubulin duringdevelopment of the chick hindbrain follows the classical growth-associatedprotein pattern of expression. That is, they are developmentally regulated, withvery low levels of mRNA expression early in development, which rise to a peakcorrelating with the time period of axonal outgrowth and synaptogenesis and43then fall to near adult levels by hatching. These results are similar to thedescription of GAP 43 (Baiter et al, 1990) and total a-tubulin mRNA expression(Bhattacharya and Sarkar, 1991) in whole chick brain and the reporteddevelopmental expression of the 3D5 antigen, identified as chicken GAP 43protein (Allsopp and Moss, 1989). To date there have been no publishedaccounts of GAP 43 or total a-tubulin mRNA or protein expression in discretebrain regions of the chick.The measured expression of total a-tubulin mRNA in this study providedan average of all the a-tubulin production across the hindbrain, as all cell typesin this tissue produced a-tubulin mRNA(s) and the cRNA probe used to detecta-tubulin mRNA was not specific for any particular isotype. However, theobserved peak expression matched the time frame when maximum axonaloutgrowth, and hence tubulin requirement, occurred. The expression of tubulinmRNA has been found to be under strict auto-regulatory control (Cleveland,1988; Cleveland, 1989) meaning there was only as much tubulin message aroundas was needed to provide tubulin protein for cellular needs. Maximal axonoutgrowth from the hindbrain occurred between E8 and E13, when a-tubulinmRNA levels were very high. The peak expression was extended somewhat dueto the needs of other cell types such as oligodendrocytes, known to beginmyelination around E13 (Bensted et al, 1957; El-Eishi, 1967; State et al, 1977;Costa et al, 1981; Macklin and Weill, 1985). As well, there was a large increase inthe size of the embryos from E13 to E18. Although the brainstem-spinalprojections were not seeking new target sites during that period, total a-tubulinexpression would be required at a somewhat elevated level to supply themicrotubules necessary for axon extension to accommodate the growth of theanimal. These factors resulted in the broad peak expression observed for thetotal a-tubulin mRNA.44The GAP 43 peak expression was more defined than the a-tubulin peak,and occurred towards the end of the maximal axonal outgrowth phase ofhindbrain development. If GAP 43 was required for actual growth of axons as a-tubulin was, the peak expression should have occurred earlier and covered abroader range of developmental ages, as it did for a-tubulin expression (compareFigures 4 and 6). This suggested that the protein was not needed so much foraxonal outgrowth per se, but events occurring nearer the end of the axon growthphase such as target recognition or target evaluation and/or synaptogenesis.This is purely speculative but an attractive hypothesis considering the reportedgrowth cone/presynaptic functional correlations of GAP 43 in the developingand adult animal (Skene, 1989). A role for GAP 43 in signal transductionmechanisms has been postulated from the calmodulin binding properties, IP3regulatory functions, correlations with phosphorylation and LTP induction, andinteractions with G o as described in the introduction (Nelson and Routtenberg,1985; Lovinger et al, 1985; Skene, 1990; Strittmatter et al, 1990; De Graan et al,1990). During development (or regeneration) then, GAP 43 may play a role inthe process of evaluating potential targets by modulation of second messengersystem activity during the course of axon extension. The results of this studycorrelate the peak of GAP 43 expression with the time when these postulatedfunctions would be occurring. In this respect, mRNA levels did not peak untilafter axon outgrowth was well on its way, and levels dropped after E12 when alltargets had been contacted.45GROWTH-ASSOCIATED mRNA EXPRESSION AFTER TRANSECTIONTotal a-TubulinThe second and third hypotheses, that mRNA levels for total a-tubulinwould transiently increase after Ell transection and not change after E14transection, were not supported by the data obtained by Northern Blotting. In allblots of experimental animals probed with the total a-tubulin cRNA, there wasno consistent change in mRNA levels compared to age-matched controls. Therewas always minor variation in the a-tubulin signal which did not correlate withthe experimental treatment, but can be accounted for by individual variationbetween animals and potential inequities in gel loading and transfer. Althoughthere is a precedent in the literature for changes in a-tubulin levels after injury toCNS or PNS (Pearson et al, 1988; Miller et al, 1989; Miller and Geddes, 1990;Tetzlaff et al, 1991), there are a number of key differences between those reportsand this study.The work quoted above was done using the extremely sensitive techniqueof in situ hybridization with a neuron-specific a-tubulin isotype probe combinedwith retrograde tracing. This allowed a very precise cellular discrimination ofthe changes which occurred after injury only in those neurons that were injured.Northern blotting as performed in this study only discriminated changes at thetissue level, and essentially provided an assessment of the average expressionacross the tissue under study. Unfortunately the total a-tubulin cDNA, capableof discriminating the neuron-specific Tal tubulin isotype in the rat, failed torecognize a chicken homologue (data not shown). Therefore, the total a-tubulincRNA as employed here was detecting all isotypes of a-tubulin being producedby all cells of the hindbrain. Previous studies of a-tubulin expression in chicken46identified 4 isotypes which were expressed in brain (Pratt and Cleveland, 1988).Cal, ca3 and ca8 were expressed ubiquitously at low levels, although high levelsof cal and ca8 were found in brain. Ca5 was only found in brain,testis andthymus (Pratt and Cleveland, 1988). It remains unknown which of theseisotypes, if any, are expressed exclusively in neurons.Thus, it is possible that, as only a small subset of cells in the hindbrainwere neurons projecting to the spinal cord, the up-regulation of a-tubulin mRNAwhich was predicted to have occurred in those neurons suffering transectionswould not be enough to be discerned above the large background of a-tubulinproduction from the majority of cells in the hindbrain. A second possibility wasthat the normal developmental a-tubulin mRNA production in the hindbrainwas already high enough and no further increase was necessary or possible toaccomodate the regenerative processes. The latter scenario was only tenable atthe earlier stages of repair after transection when a-tubulin levels weredevelopmentally high, and can not explain why at E21, when the a-tubulinsignal was near adult levels in control animals, there were still no changes seenin, for example, the E14 transected, seven day survival group. Resolving thisquestion awaits the completion of the in situ hybridization experiments nowunder way.GAP 43The second hypothesis, that GAP 43 mRNA levels would transientlyincrease after Ell transection was not supported by the data from the Northernblotting experiments. Data from the Ell transection group did not provideconclusive evidence for changes in GAP 43 expression during successful47regeneration. Any changes seen were small and difficult to detect at the tissuelevel using Northern blotting relative to the amount of mRNA present from thenormal developmental expression.Initially, a small decrease in GAP 43 mRNA levels was observed after bothEll and E14 transection, more so after Ell transection (see Figure 9). This mayreflect the increased sensitivity to injury described for the embryonic CNS(Brodal, 1983). Previous experiments have not evaluated the extent of cell deathin chick embryos as a result of spinal cord transection at different developmentalstages. Perhaps the larger initial decrease in GAP 43 mRNA seen after Elltransection was a result of increased sensitivity to injury at that earlier age. Thesmaller initial decrease in GAP 43 seen for the E14 transection group suggestedthat more neurons survived the transection procedure, and resulted in a morerobust expression of GAP 43 mRNA similar to that seen during the normalcourse of development.Although there was a precedent in the literature for increased GAP 43mRNA levels after neuronal injury when regeneration was successful (Skene andWillard, 1981a,b; Skene, 1984; Kalil, 1988; Tetzlaff et al, 1991), it is possible thatthe techniques employed in the present study are just not sensitive enough todetect a change in the minority of neurons which received transections above thehigh background levels of expression in the majority of hindbrain neurons. Thenormal developmental expression levels from E12 to E14 were still high (butdecreasing) and increases in expression in the subpopulation of neuronsreceiving transections may simply have been masked. By E16 and E18, when thenormal GAP 43 developmental mRNA levels were lower, still no change wasdetectable after Ell transection. The results from previous anatomical andphysiological experiments indicate that regeneration is complete by this time(E16-E18, Hasan et al, 1991, 1992) thus GAP 43 mRNA levels in the transected48neurons had likely decreased as well. Another possibility was that GAP 43mRNA levels were already high enough at the time of transection (Ell) that nofurther increase was necessary or possible to accommodate the regeneration.This observation did not rule out the possibility of subsequent regenerationcontributing to the repair process, however, data from double retrograde tracingexperiments (Hasan et al, 1992) refute this possibility.This was in marked contrast to the mRNA expression levels after arestrictive period transection, which rose dramatically by three days post-transection and remained elevated compared to control animals throughout theperiod examined in this study. This supports the third hypothesis as it applies toGAP 43. Recent reports in the literature suggested that GAP 43 mRNA levelsincreased after neuronal injury whether or not regeneration was successful(Doster et al, 1991; Tetzlaff et al, 1991) but the time course of the responsediffered between the two paradigms. When regeneration was successful, the re-expression of GAP 43 correlated with the re-growth of the injured axons, andwas down-regulated at the time of target contact and formation of appropriatesynapses (Tetzlaff et al, 1991). When regeneration failed, the re-expression ofGAP 43 remained elevated long after the time it should have taken to grow toand re-connect with appropriate targets (Doster et al, 1991; Tetzlaff et al, 1991).The results from this study correlated well with these recent observations. Thusthe rise in GAP 43 mRNA levels observed here may indicate a futile return to anearlier developmental growth "mode" with a prolonged re-instatement of the6.3kb/1.5kb band density ratio reminiscent of the peak expression seen at E10.Perhaps it was the failure to re-connect with appropriate targets which did notallow the down-regulation of GAP 43 mRNA.Recent work by the Benowitz group has shed some light upon the cellularmechanisms responsible for regulation of GAP 43 mRNA levels. They used49Northwestern blotting to identify GAP 43 mRNA binding protein(s) found indeveloping neocortex and differentiating PC12 cells which appeared to confermessage stability (Irwin et al, 1991). These were small (17kDa) cytosolicprotein(s) which bound to defined sites in the 3' untranslated region of the GAP43 mRNA. Inhibition of protein synthesis in PC12 cells did not block theincreased stability of the GAP 43 message during induction of neurite outgrowth(Cansino et al, 1991), which indicated that the protein(s) were present butactivated only when increased GAP 43 message was required. The chicken GAP43 cDNA sequence incorporates two segments within the 3' untranslated regionwhich were highly conserved between avian and mammalian species (Balzer etal, 1990). One segment (nucleotides 934-959) was shown to produce a stem-loopstructure which may alter message stability (Reeves et al, 1986), the second(nucleotides 992-1028) contains an AUUUA motif which has been shown to affectmRNA stability (Cleveland and Yen, 1989). It was likely that a similarmechanism occurred in our system, that the developmental expression and up-regulation seen after transection was due to increased message stability ratherthan an increased transcription rate.50CONCLUSIONS AND FUTURE EXPERIMENTSThe data supported the first hypothesis, indicating that GAP 43 and a-tubulin mRNA expression in the hindbrain of the embryonic chick wasdevelopmentally regulated and correlated with periods of axon outgrowth andsynaptogenesis. Consistent changes in the hindbrain expression of a-tubulinafter injury to the spinal cord were not measurable using Northern blotting. Thiswas likely due to the insufficient sensitivity of the technique combined with thesmall proportion of neurons in the hindbrain which were injured by thetransection. The second and third hypotheses concerning a transient re-expression of total a-tubulin after Ell spinal cord injury and a maintained re-expression after E14 injury were not supported by the data. Total a-tubulinlevels after transection provided an internal control for measuring changes inGAP 43 expression. The data did not support the second hypothesis concerninga transient re-expression of GAP 43 after Ell spinal cord transection, as levels ofGAP 43 mRNA in the hindbrain did not detectably change after permissiveperiod spinal cord transection. In contrast, transection during the restrictiveperiod resulted in a maintained increase in GAP 43 mRNA levels, whichsupported the third hypothesis concerning GAP 43. These results suggested thatthe embryonic CNS retained the ability to appropriately re-express a part of thegenetic growth program after injury on E14 as indicated by the appropriateincrease in GAP 43 mRNA levels when there was no physiological or anatomicalevidence for successful regeneration.To further characterize the neuronal response to injury in this system,several experiments are required. A broader range of GAPs will be studied atthe cellular level using combined retrograde tracing and in situ hybridizationhistochemistry. This will allow a greater resolution of the response to injury of51identified brainstem-spinal projection neurons. Collaborations are underwaywith several groups supplying novel and identified GAP cDNAs for testing inthis system.It is unknown if the observed failure of regeneration seen after E13 is dueto the absence of some facilitatory influence such as neurotrophic factors or thepresence of some inhibitory factor such as myelin components. Growth factoradministration has been shown to enhance the regenerative ability of responsiveneurons within the CNS (Hefti, 1986; Williams et al, 1986 Kromer, 1987; Gage etal, 1988; Hagg et al, 1989), and our system can be used to rapidly screen the invivo consequences of putative CNS neuronal trophic factors such as theneurotrophins, fibroblast growth factors, etc. With the added sensitivity of usingin situ hybridization, very subtle effects of a variety of factors can be measured.Coincidentally, the process of myelination begins around E13 in the chickembryo (Bensted et al, 1957; Costa et al, 1981; El-Eishi, 1967; Macklin and Weill,1985; State et al, 1977). Evidence is accumulating for an inhibitory role played bymyelin contributing to the failure of CNS regeneration. CNS myelin (Savio andSchwab, 1989) and oligodendrocytes (Schwab and Caroni, 1988) are poorsubstrata for neurite outgrowth in vitro. Recently, several proteins have beenisolated from CNS myelin which contributed to its inhibitory properties (Caroniand Schwab, 1988). An elegant study by Schnell and Schwab (1990)demonstrated that immunochemical blockade of one of these proteins in vivoresulted in improved anatomical regrowth of severed adult rat corticospinalaxons.Recent efforts by a graduate student in our lab, Hans Keirstead, havedemonstrated that the onset of myelination in the chick embryo can be delayedwell into the restrictive period for repair. Injection of antibodies togalactocerebroside combined with serum complement proteins resulted in the52delay of the onset of myelination until E17. Preliminary experiments suggestthat the absence of myelin enhanced the ability of the spinal cord to repairdamage done on E15 (Keirstead et al, 1991). It may prove interesting to examinethe molecular response to injury in the dysmyelinated spinal cord injuredanimals.These experiments have opened the door to a careful molecularexamination of the neuronal response to injury in a system well-suited for thistype of study. Expanding the scope of GAPs in the repertoire and measuring theresponse at the cellular level with in situ hybridization should provide a sensitivesystem for examining the neuronal response to injury and the effectiveness ofexperimental intervention.53REFERENCESAguayo, A.J., Benfey, M, and David, S. (1983) A potential for axonal regenerationin neurons of the adult mammalian nervous system, In: Birth Defects: Original Article Series 19/4,  Haber, B., Perez-Pollo, J.R., Hashim, G.A. andGid-frida-Stella, A.M. (eds), Alan R. Liss, Ney York, pp. 327-340.Allsop, T.E. and Moss, J.D. (1989) A developmentally regulated chicken neuronalprotein associated with the cortical cytoskeleton. J. Neurosci. 9:13-24.Apel, E.D., Litchfield, D.W., Clark, R.H., Kreebs. E.G. and Storm, D.R. (1991)Phosphorylation of Neuromodulin (GAP-43) by casein Kinase II. J. Biol.Chem. 266:10544-10551.Armstrong, D.M. (1986) Supraspinal contributions to the initiation and control oflocomotion in the cat. Prog. Neurobiol. 26:273-361.Baetge, E.E. and Hammang, J.P. (1991) Neurite oitgrowth in PC12 cells deficientin GAP 43. Neuron 6:, 21-30.Baizer, L., Alkan, S., Stocker, K. and Ciment, G. (1990) Chicken growth-associated protein (GAP)-43: primary structure and regulated expressionof mRNA during embryogenesis. Mol. Brain Res. 7:61-68.Bamburg, J.R., Bray, D. and Chapman, K. (1986) Assembly of microtubules at thetip of growing axons. Nature 321:788-790.Bamburg, J.R., Shooter, E.M. and Wilson, L. (1973) Developmental changes inmicrotubule protein of chick brain. Biochemistry 12:1476-1482.Barnard, J.W. and Carpenter, W. (1950) Lack of regeneration in spinal cord of rat.J. Neurophysiol. 13:223-228.Benfey, M. and Aguayo, A.J. (1982) Extensive elongation of axons from rat braininto peripheral nerve grafts. Nature 296:150-152.Benowitz, L.I., Apostolides, P.J., Perrone-Bizzozero, N., Finklestein, S.P. andZwiers, H. (1988) Anatomical distribution of the growth-associatedprotein GAP-43/ B-50 in the adult rat brain. J. Neurosci. 8:339-352.Benowitz, L.I., Perrone-Bizzozero, N.I., Neve, R.L. and Rodriguez, W. (1990)GAP-43 as a marker for structural plasticity in the mature CNS. In:Progress in Brain Research vol. 86, Coleman, P., Higgins, G. and Phelps, C.(eds.) Elsevier Science Publishers, Amsterdam pp.309-320.Benowitz, L.I. and Routtenberg, A. (1987) A membrane phosphoproteinassociated with neural development, axonal regeneration, phospholipidmetabolism, and synaptic plasticity. TINS 10:527-532.Bensted, J.P.M., Dobbing, J., Morgan, R.S., Reid, R.T.W. and Wright, P.G. (1957)Neuroglial development and myelination in the spinal cord of the chickembryo. J. Embryol. Exp. Morph. 5:428-437.54Bernstein, D.R. and Stelzner, D.J. (1983) Plasticity of the corticospinal tractfollowing midthoracic spinal injury in the postnatal rat. J. Comp. Neurol.221:382-400Bhattacharya, B., Mandal, C., Basu, S. and Sarkar, P.K. (1987) Regulation of a-and 13-tubulin mRNAs in rat brain during synaptogenesis. Mol. Brain Res.2:159-162.Bhattacharya B. and Sarkar, P.K. (1991) Tubulin gene expression duringsynaptogenesis in rat, mouse and chick brain. Int. J. Dev. Neuro. 9:89-99.Bisby, M.A. (1988) Dependence of GAP 43 (B50, Fl) transport on axonalregeneration on rat dorsal root ganglion neurons. Brain Res. 458:157-161.Bjorklund, A., Katzman, R, Stenevi, U. and West, K.A. (1971) Development andgrowth of axonal sprouts from noradrenaline and 5-hydroxytryptamineneurones in the rat spinal cord. Brain Res. 31:21-33.Black, J.A., Waxman, S.G., Ransom, B.R. and Feliciano, M.D. (1986) Aquantitative study of developing axons and glia following alteredgliogenesis in rat optic nerve. Brain Res. 380:122-135.Bonini, J.A. and Hofmann, C. (1991) A rapid, accurate, nonradioactive methodfor quantitating RNA on agarose gels. Biotechniques 11:708-709.Bray, D. (1987) Growth cones: do they pull or are they pushed? TINS 10:431-434.Brodal, A (1983) Anterograde and retrograde degeneration of neurons in thecentral nervous system. In: Histology and Histopathology of the CentralNervous System, Haymacker, W. and Adams, R.D. (eds.) Charles C.Thomas, Springfield, Ill. pp. 276-362.Cansino, V.V., Kohn, D.T. and Perrone-Bizzozero, N.I. (1991) Post-transcriptionalmechjanisms during process outgrowth II. Stabilization of GAP-43 andother neuronal mRNAs in TPA induced PC12 cells. Soc. Neurosci. Abs.17:1310.Carmignoto, G., Maffei, L., Candeo, P., Canella, R. and Comelli, C. (1989) Effectof NGF on the survival of rat retinal ganglion cells following optic nervesection. J. Neurosci. 9:1263-1272.Caroni, P. and Schwab, M.E. (1988) Two membrane protein fractions from ratcentral myelin with inhibitory properties for neurite growth and fibroblastspreading. J. Cell Biol. 106:1281-1288.Carter, D.A., Bray, G.M. and Aguayo, A.J. (1989) Regenerated retinal ganglioncell axons can form well-differentiated synapses in the superior colliculusof adult hamsters. J. Neurosci. 9:4042-4050.Cleveland, D.W. (1988) Autoregulated instability of tubulin mRNAs: a noveleukaryotic regulatory mechanism. TIES 13:339-343.55Cleveland, D.W. (1989) Autoregulated control of tubulin synthesis in animalcells. Current Opinion in Cell Biol. 1:10-14.Cleveland, D.W., Lopata, M.A., MacDonald, R.J., Cowan, N.J., Rutter, W.J. andKirschner, M.W. (1980) Number and evolutionary conservation of a- and13-tubulin and cytoplasmic 13- and y-actin genes using specific clonedcDNA probes. CO 20:95-105.Cleveland, D.W. and Sullivan, K.F. (1985) Molecular biology and genetics oftubulin. Ann. Rev. Biochem. 54:331-365.Cleveland D.W. and Yen, T.J. (1989) Multiple determinants of eukaryotic mRNAstability. The New Biologist 1:121-126.Correa-Rotter, R, Mariash, C.N. and Rosenberg, M.E. (1992) Loading andtransfer control for Northern hybridization. Biotechniques 12:154-158.Costa, S.W., Engelhardt, E. and Engelhardt M. (1981) Identificacao e epoca demielinizacao de alguns feixes da medula cervical do Gallus domesticus.Rev. Brasil. Biol. 41:833-839.Crutcher, K.A. and Marfurt, C.F. (1988) Nonregenerative axonal growth withinthe mature mammalian brain: Ultrastructural identification ofsympathohippocampal sprouts. J. Neurosci. 8:2289-2302.Dani, J.W., Armstrong, D.M. and Benowitz, L.I. (1991) Mapping the developmentof the rat brain by GAP-43 immunocytochemistry. Neuroscience 40:277-Daniels, M.P. (1972) Colchicine inhibition of nerve fiber formation in vitro. J. CellBiol. 53:164-176.David, S. and Aguayo, A.J. (1981) Axonal elongation into peripheral nervoussystem "bridges" after central nervous system injury in adult rats. Science214:931-933.De Graan, P.N.E., Oestreicher. A.B., De Wit, M., Kroef, M., Schrama, L.H. andGispen, W.H. (1990) Evidence for the binding of calmodulin toendogenous B-50 (GAP-43) in native synaptosomal plasma membranes.Neurochem. la Monte, S.M., Federoff, H.J., Ng, S.-C., Grabczyk, E. and Fishman, M.C.(1989) GAP-43 gene expression during development: Persistence in adistinctive set of neurons in the mature central nervous system. Dev.Brain Res. Leeuw, W.J.F., Slagboom, P.E. and Vijg, J. (1989) Quantitative comparison ofmRNA levels in mammalian tissues: 28S ribosomal RNA lever as anaccurate internal control. Nucleic Acids Res. 17:10137-10138.Dokas. L.A., Pisano, M.R, Schrama, L.H., Zwiers, H. and Gispen, W.H. (1990)Dephosphorylation of B-50 in synaptic plasma membranes. Brain Res.Bull. 24:321-329.56Doster, S.K., Lozano, A.M., Aguayo, A.J. and Willard, M.B. (1991) Expression ofthe growth-associated protein GAP-43 in adult rat retinal ganglion cellsfollowing axon injury. Neuron 6:635-647.El-Eishi, H.I. (1967) Biochemical and histochemical studies on myelination in thechick embryo spinal cord. J. Neurochem. 14:405-412.Estep, R.P., Alexander. K.A. and Storm, D.R. (1990) Regulation of freecalmodulin levels in neurons by neuromodulin: Relationship to neuronalgrowth and regeneration. Current Topics in Cellular Regulation 31:161-180.Gage, F.H., Armstrong, D.M., Williams, L.R. and Varon, S. (1988) Morphologicalresponse of axotomized septal neurons to nerve growth factor. J. Comp.Neurol. 269:147-155.Garcia-Rill, E. (1986) The basal ganglia and the locomotor regions. Brain Res.Rev. 11:47-63.Garcia-Rill, E. and Skinner, RD. (1987a) The mesencephalic locomotor region. I.Activation of a medullary projection site. Brain Res. 411:1-12.Garcia-Rill, E. and Skinner, RD. (1987b) The mesencephalic locomotor region. II.Projections to reticulospinal neurons. Brain Res. 411:13-20.Gispen, W.H., Leunissen, J.L.M., Oestreicher, A.B., Verkleij, A.J. and Zwiers, H.(1985) Presynaptic localization of B-50 phosphoprotein: The (ACTH)-sensitive protein kinase substrate involved in rat brainpolyphosphoinositide metabolism. Brain Res. 328:381-385Goldberg, D.J. and Burmeister, D.W. (1989) Looking into growth cones. TINS12:503-506.Gordon-Weeks, P.R. (1991) Growth Cones: The mechanism of neurite advance.BioEssays 13:235-239.Gorgels, T.G.M.F., Van Lookeren Campagne, M., Oestreicher, A.B., Gribnau,A.A.M. and Gispen, W.H. (1989) B-50/GAP43 is localized at thecytoplasmic side of the plasma membrane in developing and adult ratpyramidal tract. J. Neurosci. 9:3861-3869.Goslin, K., Schreyer, D.J., Skene, J.H.P. and Banker, G. (1988) Development ofneuronal polarity GAP-43 distiguishes axonal from dendritic growthcones. Nature 336:672-674.Goslin, K., Schreyer, D.J., Skene, J.H.P. and Banker, G. (1990) Changes in thedistribution of GAP-43 during the development of neuronal. polarity. J.Neurosci. 10:588-602.Grillner, S. and Dubuc, R (1988) Control of locomotion in vertebrates: spinal andsupraspinal mechanisms. Adv. Neurol. 47:425-453.57Hagg, T., Fass-Holmes, B., Vahlsing, H.1., Manthorpe, M., Conner, J.M. andVaron, S. (1989) Nerve growth factor (NGF) reverses axotomy-induceddecreases in choline acetyltransferase, NGF receptor and size of medialseptum cholinergic neurons. Brain Res. 505:29-38.Hall, S.M. (1989) Regeneration in the peripheral nervous system. Neuropathol.Appl. Neurobiol. 15:513-529.Hamburger, V. and Hamilton, H.L. (1951) A series of normal stages in thedevelopment of the chick embryo. J. Morph. 88:49-92.Hasan, S.J., Keirstead, H.S., Muir, G.D. and Steeves, J.D. (1992) Axonalregeneration contributes to repair of injured brainstem-spinal neurons inembryonic chick. J. Neurosci. (submitted).Hasan, S.J., Valenzuela, J.I., Nelson, B.H., Shull, S.E., Ethell, D.W. and Steeves,J.D. (1991) Repair of transected spinal cord in embryonic chick. J. Restor.Neurol. 2:137-154Hefti, F. (1986) Nerve growth factor promotes survival of septal cholinergicneurons after fimbrial transections. J. Neurosci. 6:2155-2162.Hermans-Borgmeyer, I., Hoffmeister, S., Sawruk, E., Betz, H., Schmitt, B. andGunderfinger, E.D. (1989) Neuronal acetylcholine receptors in Drosophila:mature and immature transcripts of the ARD gene in the developingcentral nervous system. Neuron 2:1147-1156.Hoffman, P.N. and Cleveland, D.W. (1988) Neurofilament and tubulinexpression recapitulates the developmental program during axonalregeneration: induction of a specific 13-tubuhn isotype. Proc. Natl Acad.SQ. USA 85:4530-4533.Hoffman, P.N. (1989) Expression of GAP 43, a rapidly transported growth-associated protein, and class II beta tubulin, a slowly transportedcytoskeletal protein, are coordinated in regenerating neurons. J. Neurosci.9:893-897.Holder N. and Clarke, J.D.W. (1988) Is there a correlation between continuousneurogenesis and directed axon regeneration in the vertebrate nervoussystem? TINS 11:94-99.Irwin, N., Perrone-Bizzozero, N.I. and Benowitz, L.I. (1991) GAP 43 mRNAbinding proteins related to neural differentiation. Soc. Neurosci. Abs.17:1310.Jacobson, R.D. Virag, I. and Skene, J.H.P. (1986) A protein associated with axongrowth, GAP-43, is widely distributed and developmentally regulated inrat CNS. J. Neurosci. 6:1843-1855.Jordan, L.M. (1986) Initiation of locomotion from the mammalian brainstem. In:Neurobiology of Vertebrate Locomotion. S. Grillner (ed.), MacmillanPress, London, pp. 21-37.58Kalil, K. (1988) Regeneration of pyramidal tract axons. In: Advances in Neurology. vol. 47: Functional Recovery in Neurological Disease. Waxman, S.G. (ed.), Raven Press, New York, pp. 67-85.Kalil, K. and Skene, J.H.P. (1986) Elevated synthesis of an axonally transportedprotein correlates with axon outgrowth in normal and injured pyramidaltracts. J. Neurosci. 6:2563-2570.Keirstead, H.S., Hasan, S.J., Muir, G.D. and Steeves, J.D. (1991) Suppression ofspinal cord myelination extends the period for functional repair followingthoracic transection in embryonic chick. Fourth International Symposiumon Neural Regeneration, Abstract p.49Kolb, B. and Whishaw, I.Q. (1989) Plasticity in the neocortex: Mechanismsunderlying recovery from early brain damage. Prog. in Neurobiol. 32:235-276.Kromer, L.F. (1987) Nerve growth factor treatment after injury prevents neuronaldeath. Science 235:214-216.Laskey, R.A. and Mills, A.D. (1975) Quantitative film detection of 3H and 14C inpolyacrylamide gels by fluorography. Eur. J. Biochem. 56:335-341.Laskey, R.A. and Mills, A.D. (1977) Enhanced autoradiographic detection of 32pand I2Eising intensifying screens and hypersensitized film. FEBS Letters82:314-316.Lawrence, D.G. and Kuypers, H.G.J.M. (1968a) The functional organization of themotor system in the monkey. I. The effects of bilateral pyramidal lesions.Brain 91:1-14Lawrence, D.G. and Kuypers, H.G.J.M. (1968b) The functional organization ofthe motor system in the monkey. II. The effects of lesions of thedescending brainstem pathways. Brain 91:15-36.Lipton, S.A. and Kater, S.B. (1989) Neurotransmitter regulation of neuronaloutgrowth, plasticity and survival. TINS 12:265-270.Liu, Y., Chapman, E.R. and Storm, D.R. (1991) Targeting of Neuromodulin(GAP-43) fusion proteins to growth cones in cultured rat embryonicneurons. Neuron 6:411-420.Liu, Y. and Storm, D.R. (1990) Regulation of free calmodulin levels byneuromodulin: neuron growth and regeneration. TIPS 11:107-111.Lovinger, D.M., Akers, RF., Nelson, R.B., Barnes, C.A., McNaughton, B.L. andRouttenberg, A. (1985) A selective increase in phosphory1ation of proteinFl, a protein kinase C substrate, directly related to three day growth oflong term synaptic enhancement. Brain Res. 343:137-143.Macklin, W.B. and Weill, C.L. (1985) Appearance of myelin proteins duringdevelopment in the chick central nervous system. Dev. Neurosci. 7:170-178.59Masure, H.R., Alexander, K.A., Wakim, B.T. and Storm, D.R. (1986)Physicochemical and hydrodynamic characterization of P-57,neurospecific calmodulin binding protein. Biochemistry 25:7553-7560.McClellan, A.D. (1986) Command systems for initiating locomotion in fish andamphibians: parallels to initiation systems in mammals. In: Neurobiologyof Vertebrate Locomotion,  S. Grillner (ed.), Macmillan Press, London, pp.3-20.McGuire, C.B., Snipes, G.J. and Norden, J.J. (1988) Light-microscopicimmunolocalization of the growth- and plasticity-associated protein GAP-43 in the developing rat brain. Dev. Brain Res. 41:277-291.Meiri, K.F., Bickerstaff, L.E. and Schwob, J.E. (1991) Monoclonal antibodies showthat kinase C phosphorylation of GAP-43 during axonogenesis is bothspatially and temporally restructed in vivo. J. Cell Biol. 112:991-1005Meiri, K.F., Pfenninger, K.H. and Willard, M.B. (1986) Growth-associated protein,G AP -43, a polypeptide that is induced when neurons extend axons is acomponent of growth cones and corresponds to pp46, a major polypeptideof a subcellular fraction enriched in growth cones. Proc. Natl. Acad. Sci.USA 83:3537-3541.Meiri, K.F., Willard, M. and Johnson, M.I. (1988) Distribution and thephosphorylation of the growth-associated protein GAP-43 in regeneratingsympathetic neurons in culture. J. Neurosci. 8:2571-2581.Miller, F.D. (1989) Techniques in Cell and Molecular Biology Seminar Series,Session MI. "RNA Measurements". Dept. of Anatomy and Cell Biology,University of Edmonton.Miller, F.D. and Geddes, J.W. (1990) Increased expression of the major embryonica-tubulin mRNA, Tat During neuronal regeneration, sprouting, and inAlzheimer's disease. In: Progress in Brain Research vol. 86, Coleman, P.,Higgins, G. and Phelps, C. (eds.) Elsevier Science Publishers, Amsterdam,pp.321-330.Miller, F.D., Naus, C.C.G., Durand, M., Bloom, F.E. and Milner, R.J. (1987a)Isotypes of a-tubulin are differentially regulated during neuronalmaturation. J. Cell Biol. 105:3065-3073.Miller, F.D., Naus, C.C.G., Higgins, G.A., Bloom, F.E. and Milner, R.J. (1987b)Developmentally regulated rat brain mRNAs: molecular and anatomicalcharacterization. J. Neurosci. 7:2433 2444.Miller, F.D., Tetzlaff, W., Bisby, M.A., Fawcett, J.W. and Milner, R.J. (1989) Rapidinduction of the major embryonic a-tubulin mRNA, Tal, during nerveregeneration in adult rats. J. Neurosci. 9:1452-1463.Moya, K.L., Jhaveri, S., Schneider, G.E. and Benowitz. L.I. (1989)Immunohistochemical localization of GAP-43 in the developing hamsterretinofugal pathway. J.Comp. Neurol. 288:51-58.60Nelson, RB. and Routtenberg, A. (1985) Characterization of protein Fl (47 kDa,4.5 pI): A kinase C substrate directly related to neural plasticity. Exp.Neurol. 89:213-224.Neve, R.L., Finch, E.A., Bird, E.D. and Benowitz, L.I. (1988) Growth-associatedprotein GAP-43 is expressed selectively in associative regions of the adulthuman brain. Proc. Natl. Acad. Sci. USA 85:3638-3642.Okado, N. and Oppenheim, R.W. (1985) The onset and development ofdescending pathways to the spinal cord in the chick embryo. J. Comp.Neurol. 232:143-161.Pearson, R.C.A., Taylor, N. and Snyder, S.H. (1988) Tubulin messenger RNA: insitu hybridization reveals bilateral increases in hypoglossal and facialnuclei following nerve transection. Brain Res. 463:245-249.Peterson, B.W. (1984) The reticulospinal system and its role in the control ofmovement. In: Brainstem Control of Spinal Cord Function, C.D. Barnes(ed.), Academic Press, Orlando, Florida, pp. 28-86.Pratt, L.F. and Cleveland, D.W. (1988) A survey of the a-tubulin gene family inchicken: unexpected sequence heterogeneity in the polypeptides encodedby five expressed genes. EMBO J. 7:931-940.Pratt, L.F., Okamura, S. and Cleveland, D.W. (1987) A divergent testis-specific a-tubulin isotype that does not contain a coded c-terminal tyrosine. Mol.Cell. Biol. 7:552-555.Price, D.L. and Porter, K.R (1972) The response of ventral horn neurons toaxonal transection. J. Cell. Biol. 53:24-37.Ramon y Cajal, S. (1928) Degeneration & Regeneration of the Nervous System,Vol. 2, Oxford University Press, London, pp 484-516. CHECKReeves, R., Spies, A.G., Nissen, M.S., Buck, C.D. Weinberg, A.D., Barr, P.J.,Magnuson, N.S. and Magnuson, J.A. (1986) Molecular cloning of afunctional bovine interleukin 2 cDNA. Proc. Natl. Acad. Sci. U.S.A.83:3228-3232.Richardson, P.M., Issa, V.M.K. and Aguayo, A.J. (1984) Regeneration of longspinal axons in the rat, J.Neurocytol. 13:165-182Richardson, P.M., Issa, V.M.K. and Shemie, S. (1982) Regeneration andretrograde degeneration of axons in the rat optic nerve. J. Neurocytol.11:949-966.Richardson, P.M., McGuinness, U.M. and Aguayo, A.J. (1980) Axons from CNSneurons regenerate into PNS grafts. Nature (Lond) 284:264-265.Routtenberg, A. and Loving,er, D.M. (1985) Selective increase in phosphorylationof a 47-kDa protein (F1) directly related to long-term potentiation. Behay.Neural Biol. 43:3-11.61Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning. A Laboratory Manual. 2nd Ed. Cold Spring Harbor Laboratory Press, ColdSpring Harbor, New York, pp. E.37-E.38.Savio, T and Schwab, M.E. (1989) Rat CNS white matter, but not grey matter, isnonpermissive for neuronal cell adhesion and fiber outgrowth. J.Neurosci. 9:1126-1133.Schmidt, R.H., Kasik, S.A. and Bhatnagar, R.K. (1980) Regenerative criticalperiods for locus coeruleus in post-natal rat pups following intracisternal6-hydroxydopamine: A model of noradrenergic development. Brain Res.191:173-190.Schnell, L. and Schwab, M.E. (1990) Axonal regeneration in the rat spinal cordproduced by an antibody against myelin-associated neurite growthinhibitors. Nature 343:269-272.Schwab, M.E. and Caroni, P. (1988) Oligodendrocytes and CNS myelin arenonpermissive substrates for neurite growth and fibroblast spreading invitro. J. Neursci. 8:2381-2393.Sechrist, J. and Bronner-Fraser, M. (1991) Birth and differentiation of reticularneurons in the chick hindbrain: Ontogeny of the first neuronal population.Neuron 7:947-963.Shiga, T., Kunzi, R. and Oppenheim, R.W. (1991) Axonal projections andsynaptogenesis by supraspinal descending neurons in the spinal cord ofthe chick embryo. J. Comp. Neurol. 305:83-95.Skene, J.H.P. (1984) Growth-associated proteins and the curious dichotomies ofnerve regeneration. Cell 37:697-700.Skene, J.H.P. (1989) Axonal growth-associated proteins. Ann. Rev. Neurosci.12:127-156.Skene, J.H.P. (1990) GAP-43 as a 'calmodulin sponge' and some implications forcalcium signalling in axon terminals. Neuroscience Res. Supp.13:S112-S125.Skene, J.H.P. and Willard, M. (1981a) Axonally transported proteins associatedwith axon growth in rabbit central and peripheral nervous systems. J. CellBiol. 89:96-103.Skene, J.H.P. and Willard, M. (1981b) Changes in axonally transported proteinduring axon regeneration in toad retinal ganglion cells. T. Cell Biol. 89:86-95So, K.-F., Schneider, G.E. and Ayres, S. (1981) Lesions of the brachium of thesuperior colliculus in neonate hamsters: Correlation of anatomy withbehavior. Exp. Neurol. 72:379-400.State, F.A., El-Eishi, H.I. and Hamed, M.S. (1977) Myelination and cholinesteraseactivity of chick embryo spinal cord. Acta Anat. 99:469-476.62Steeves, J.D. and Jordan, L.M. (1984) Autoradiographic demonstration of theprojections from the mesencephalic locomotor region. Brain Res. 307:263-26.Steeves, J.D. Sholomenko, G.N. Webster, D.M.S. (1987) Stimulation of thepontomedullary reticular formation initiates locomotion in decerebratebirds. Brain Res. 401:205-212.Strittmatter, S.M., and Fishman, M.C. (1991) The neuronal growth cone as aspecialized transduction system. BioEssays 13:127-134.Strittmatter, S.M., Valenzuela, D., Kennedy, T.E., Neer, E.J. and Fishman, M.C.(1990) Go is a major growth cone peotein subject to regulation by GAP-43.Nature 344:836-841.Tello, F. (1911) La influencia del neurotropismo en la regeneracion de los centrosnerviosos. Trab. del Lab. de Invest. Biol. tomo 9:123-159.Tetzlaff, W., Alexander, S.W., Miller, F.D. and Bisby, M.A. (1991) Response offacial and rubrospinal neurons to axotomy: changes in mRNA expressionfor cytoskeletal proteins and GAP 43. J. Neurosci. 11:2528-2544Tetzlaff, W, Bisby, M.A. and Kreutzberg, G.W. (1988) Changes in cytoskeletalproteins in the rat facial nucleus following axotomy. J. Neurosci 8:3181-3189.Valenzuela, J.I., Hasan, S.J. and Steeves, J.D. (1990) Stimulation of the brainstemreticular formation evokes locomotor activity in embryonic chicken (inovo). Dev. Brain Res. 56:13-18.Valenzuela, P., Quiroga, M., Zaldivar, J., Rutter, W.J., Kirschner, M.W. andCleveland, D.W. (1981) Nucleotide and corresponding amino acidsequences encoded by a and [3 tubulin mRNAs. Nature 289:650-655.Van der Zee, C.E.E.M., Nielander, H.B. Vos, J.P., Lopes da Silva, S., Verhaagen, J.,Oestreicher, A.B., Schrama, L.H., Schotman, P. and Gispen, W.H. (1989)Expression of growth-associated protein B-50 (GAP43) in dorsal rootganglia and sciatic nerve during regenerative sprouting. J. Neurosci.9:3505-3512.Van Lookeren Campagne, M., Oestreicher, A.B., Van Bergen en Henegouwen,P.M.P. and Gispen, W.H. (1989) Ultrastructural immunocytochemicalloclalization of B-50/GAP43, a protein kinase C substrate, in isolatedpresynaptic nerve terminals and neuronal growth cones. J. Neurocytol.18:479-489.Vidal-Sanz, M., Bray, G.M., Villegas-Perez, M.P., Thanos, S. and Aguayo, A.J.(1987) Axonal regeneration and synapse formation in the superiorcolliculus by retinal ganglion cells in the adult rat. J. Neurosci. 7:2894-2909.Villegas-Perez, M.P., Vidal-Sanz, M., Bray, G.M. and Aguayo, A.J. (1988)Influences of peripheral nerve grafts on the survival and regrowth ofaxotomized retinal ganglion cells in adult rats. J. Neurosci. 8:265-280.63Webster, D.M.S. and Steeves, J.D. (1988) Origins of brainstem-spinal projectionsin the duck and goose. J. Comp. Neurol. 273:573-583.Weinberg, E.L. and Raine, CS. (1980) Reinnervation of peripheral nervesegments implanted into the rat central nervous system. Brain Res. 198:1-11.Williams, L.R, Varon, S., Peterson, G.M., Wictorin, K., Fischer, W., Bjorklund, A.and Gage, F.H. (1986) Continuous infusion of nerve growth factorprevents basal forebrain neuronal death after fimbria fornix transection.Proc. Natl. Acad. Sci. USA 83:9231-9235.Windle, W.F., and Austin, M.F. (1936) Neurofibrillar development in the centralnervous system of chick embryos up to 5 days' incubation. J. Comp.Neurol. 63:431-463Yamada, K.M., Spooner, B.S. and Wessels, N.K. (1970) Axon growth: Roles ofmicrofilaments and microtubules. Proc. Natl. Acad. Sci. USA 66:1206-1212.Zuber, M.X., Strittmatter, S.M. and Fishman, M.C. (1989) A membrane targettingsignal in the amino terminus of the neuronal protein GAP 43. Nature341:345-34864


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items