The perikaryal surface of spinal ganglion neurons: differences between domains in contact with satellite cells and in contact with the extracellular matrix

&p.1:Abstract The perikaryal surface of spinal ganglionneurons undergoes dynamic changes throughout life. Inparticular, numerous slender projections develop andretract continuously from this surface. We showed previ-ously that the outgrowth of these projections, while anintrinsic property of spinal ganglion neurons, is also in-fluenced by the surrounding microenvironment. Sincethe latter consists of satellite cells and the extracellularmatrix, we sought to determine the relative contributionsof each of these components to the outgrowth of peri-karyal projections. To this end, we took advantage of alittle known characteristic of the satellite cell sheaths: inthe rabbit, these sheaths can exhibit gaps that leave thenerve cell body surface directly exposed to the extracel-lular matrix. We compared the surface domains coveredby satellite cells with those in direct contact with the ex-tracellular matrix. We found that the perikaryal projec-tions are abundant in the former domains but are absentin the latter. We also found that the perineuronal extra-cellular matrix of rabbit spinal ganglia contains lamininand fibronectin, two glycoproteins that have been report-ed to promote the growth of axonal processes from sen-sory ganglion neurons. Laminin and fibronectin werealso present at the level of the gaps in the satellite cellsheath. These results: (1) provide additional evidencethat environmental factors influence the outgrowth ofperikaryal projections from spinal ganglion neurons; (2)suggest that satellite cells permit the outgrowth of theseprojections; (3) suggest that in the spinal ganglia of adultrabbits the perineuronal extracellular matrix is not in it-self able to promote the outgrowth of these projections.This study provides a further example of the influencethat supporting neuroglial cells have on sensory ganglionneurons.

&kwd:Key words Peripheral neuroglia · Dorsal root ganglia ·Laminin · Fibronectin · Neuron-glia interactions&bdy:

Introduction

A number of observations suggest that the cell body sur-face of spinal ganglion neurons undergoes dynamicchanges throughout life (Obata and Inoue 1982; Panneseet al. 1983, 1996a; personal communication, Dr. G.Gallo). In particular, a large number of slender projec-tions (Pannese et al. 1983, 1985, 1990a, b) develop andretract continuously from this surface. Analysis of fac-tors that influence the development of these projectionsmay throw some light on the regenerative capacity of thebodies of spinal ganglion neurons in the adult.

While the outgrowth of these projections is an intrin-sic property of spinal ganglion neurons, it is also influ-enced by the perineuronal microenvironment (Pannese etal. 1994, 1995). This consists of a cellular component(satellite cells) and of the extracellular matrix (basal lam-ina and the connective tissue immediately adjacent to it).To investigate the relative contributions of each of thesecomponents to the outgrowth of perikaryal projections,in this study we took advantage of a little known charac-teristic of the satellite cell sheaths. In the rabbit, thesesheaths can exhibit gaps that leave the nerve cell bodysurface directly exposed to the extracellular matrix (Pan-nese 1981; Pannese et al. 1996b). We have, therefore,compared the surface domains covered by satellite cellswith those in direct contact with the extracellular matrix.

Our results suggest that satellite cells permit the out-growth of the perikaryal projections, whereas the peri-neuronal extracellular matrix does not, even though itcontains laminin and fibronectin – two glycoproteins thathave been reported to promote the growth of axonal pro-cesses from sensory ganglion neurons.

E. Pannese (✉)Viale S. Michele del Carso 15, I-20144 Milano, Italye-mail: [email protected], Tel.: +39-02-70600962, Fax: +39-02-70635928

P. Procacci · E. Berti · M. LeddaInstitute of Histology, Embryology and Neurocytology, University of Milan, Milan, Italy&/fn-block:

Anat Embryol (1999) 199:199–206 © Springer-Verlag 1999

O R I G I N A L A RT I C L E

&roles:Ennio Pannese · Patrizia Procacci · Emilio BertiMaria Ledda

The perikaryal surface of spinal ganglion neurons: differences between domains in contact with satellite cells and in contact with the extracellular matrix

&misc:Accepted: 30 July 1998

Materials and methods

Six rabbits (Oryctolagus cuniculus) of both sexes were used: threeyoung adults (12–15 months) and three old animals (60–79months).

Morphology

A young adult and an old rabbit were perfused transcardially witha solution containing 2% formaldehyde and 2% glutaraldehyde in0.1 M sodium cacodylate buffer (pH 7.3) under deep anaesthesiawith Nembutal (80 mg/kg i.p.). After fixation for about 3 h, tho-racic spinal ganglia were removed, washed in cacodylate buffer(0.2 M, pH 7.3) for 2 h and then postfixed at 0°C for 1.5 h in 2%OsO4, buffered with 0.1 M sodium cacodylate. The specimenswere washed in distilled water, stained with 2% aqueous uranylacetate, dehydrated in alcohol and embedded in Epon-Araldite res-in. Thin sections were examined under a Siemens Elmiskop 101electron microscope.

Light microscopic immunocytochemistry

Two rabbits, one young adult and one old, were perfused transcar-dially with phosphate-buffered-saline (PBS, always 0.1 M andpH 7.4) under deep anaesthesia with Nembutal (80 mg/kg i.p.).Thoracic spinal ganglia were quickly removed, fixed in 10% neu-tral buffered formalin for 4 h and embedded in paraffin accordingto conventional histological technique.

Paraffin sections, 5-µm-thick, were dewaxed in xylene and re-hydrated in a graded series of methanol; endogenous peroxidaseactivity was blocked with 5% H2O2 in methanol for 20 min, fol-lowed by rinsing in TRIS buffer saline (TBS) 0.5 M, pH 7.4. Thesections for incubation with anti-fibronectin antibody were treatedin a microwave oven (Panasonic, model NN 6371 WM) to retrieveantigen (Cattoretti et al. 1993; Gown et al. 1993): slides wereplaced in a plastic coplin jar filled with 0.01 M sodium citratebuffer (pH 6) and boiled (power 780 W) twice, 5 min each. Thesections were allowed to cool for 30 min in the same buffer atroom temperature and then placed in TBS.

All sections, both those for incubation with anti-laminin anti-body and those for incubation with anti-fibronectin antibody, wereexposed for 20 min to blocking serum [5% bovine serum albumin(BSA) in TBS] to prevent nonspecific binding. Sections were in-cubated overnight at 4°C with mouse anti-laminin monoclonal an-tibody (Boehringer Ingelheim) or with mouse anti-fibronectinmonoclonal antibody (Boehringer Mannheim) both diluted 1:50 inPBS containing 5% BSA and 0.05% sodium azide. Bound anti-body was then detected with peroxidase-conjugated rabbit anti-mouse IgG (Dako, diluted 1:50, in TBS) followed by peroxidase-conjugated swine antirabbit IgG (Dako, diluted 1:50, in TBS). In-cubations were performed for 1 h at room temperature.

Peroxidase activity was revealed by incubating the sections atroom temperature with a solution containing 3,3’-diaminobenzidi-ne (DAB) and 0.03% H2O2 in 0.05 M TRIS HCl buffer (pH 7.6).Sections were counterstained with hematoxylin and coverslippedwith Eukitt (Merck).

Control sections were incubated by replacing the primary anti-body with a nonspecific bovine antiserum.

Electron microscopic immunocytochemistry

A young adult and an old rabbit were perfused transcardially withPBS under deep anaesthesia with Nembutal (80 mg/kg i.p.). Tho-racic spinal ganglia were quickly removed and fixed in a freshlyprepared solution containing 4% paraformaldehyde and 0.075%glutaraldehyde in 0.1 M phosphate buffer (PB, pH 7.4) for 30 minat room temperature. After fixation, ganglia were washed in PBSfor 2 h and embedded in agar.

Ganglia were cut into 40–60µm sections with a vibratome.Sections thus obtained were incubated in PBS containing 3%H2O2 (30% solution), then washed in PBS for 10 min and pro-cessed as follows.

Sections were incubated overnight at 4°C in PBS containing8% BSA, 0.01% sodium azide and 0.05% Triton-X100 (to preventnonspecific binding and to facilitate the penetration of antibodies);then they were incubated overnight at 4°C with the monoclonalanti-laminin or monoclonal anti-fibronectin antibody used for lightmicroscopy (both diluted 1:50 in PBS containing 5% BSA and0.05% sodium azide). After rinsing (three times, 15 min each, inPBS containing 1% BSA), sections were incubated overnight at4°C in secondary antibody peroxidase-conjugated rabbit anti-mouse IgG (Dako) diluted 1:30 in PBS containing 5% BSA and0.05% sodium azide and then rinsed three times, 15 min each, inPBS containing 1% BSA. Subsequently, sections were incubatedovernight at 4°C in peroxidase-conjugated swine antirabbit IgG(Dako) diluted 1:30 in PBS containing 5% BSA and 0.05% sodi-um azide. Sections were then washed twice in PBS and once in0.1 M TRIS HCl buffer (pH 7.6) and developed at room tempera-ture for 15 min in a solution containing 3,3’-diaminobenzidine(DAB) and 0.03% H2O2 in 0.05 M TRIS HCl buffer (pH 7.6) toreveal peroxidase activity. Finally, sections were immersed for 1 hin 2.5% glutaraldehyde in PBS, washed for 30 min in PB (0.1 M,pH 7.4) and postfixed for 15 min in a 1% OsO4 solution in PB0.1 M, pH 7.4.

Sections thus treated were dehydrated in a graded series of eth-anol and embedded in Epon-Araldite resin. Thin sections werecounterstained with uranyl acetate and lead citrate and examinedunder a Siemens Elmiskop 101 electron microscope.

Control sections were first incubated with anti-laminin mono-clonal antibody or anti-fibronectin monoclonal antibody preab-sorbed with an excess of laminin (DBA) or fibronectin (DBA) andwere then processed as described.

Results

Morphology

We examined sections of 633 rabbit nerve cell bodies un-der the electron microscope, 302 from a young adult and331 from an old animal. Both in the young adult andaged animals the portions of nerve cell body surface cov-ered by satellite cells were highly irregular due to thepresence of numerous projections emerging from theneuronal perikaryon (Fig. 1a–c). By contrast, the surfaceportions directly exposed to the extracellular matrix weresmooth since they did not possess projections (Fig. 1b,c). Since perikaryal projections were numerous in thesurface domains covered by satellite cells and were ab-sent in the domains directly exposed to the extracellularmatrix, it was not necessary to carry out a quantitative

200

Fig. 1 a Satellite cell sheath entirely covering the surface of aportion of neuronal perikaryon. The neuronal surface is highly ir-regular due to the presence of numerous projections emergingfrom the perikaryon. Arrows indicate projections that in this sec-tion appear in continuity with the perikaryon; * marks projectionsappearing as isolated entities as they arise from the perikaryon atother levels. b, c Satellite cell sheaths showing gaps (between )that leave the neuronal surface directly exposed to the extracellularmatrix. The perikaryal projections (* ) are present on the neuronalsurface covered by satellite cells, while they are absent in the sur-face domains in direct contact with the extracellular matrix (Nneuronal perikaryon, scsatellite cells). Spinal ganglia of rabbits ata 12 months, b, c 79 months. ×14 250&/fig.c:

201

evaluation of the extent of projections in the two do-mains.

Light microscopic immunocytochemistry

Both in anti-laminin and in anti-fibronectin preparations,the immunostaining was evident around the entire con-tour of each satellite cell-nerve cell body unit. Lamininimmunoreactivity was linear and well-defined aroundeach unit (Fig. 2a), whereas fibronectin immunoreactivi-ty was more widely distributed in the interstitial connec-tive tissue (Fig. 2b).

Electron microscopic immunocytochemistry

Both in anti-laminin and in anti-fibronectin preparations,the intensity of the immunostaining decreased progres-sively from the periphery of the sample block to the inte-rior, as a result of limited penetration of the reagents.Despite some tissue damage from the processing proce-dure, both control and experimental samples were suffi-ciently well-preserved to permit identification of nervecell bodies and satellite cells and to discern their mutualrelations.

In anti-laminin preparations, the reaction product wasalways associated with the basal lamina (Fig. 3) along

202

Fig. 2 Light micrographs ofrabbit spinal ganglia showingimmunoreactivity for laminin(a) and fibronectin (b). Lami-nin immunoreactivity is linearand well-defined around eachsatellite cell-nerve cell bodyunit. Fibronectin immunoreac-tivity is more widely distribut-ed in the interstitial connectivetissue. Rabbit of 15 months. × 980&/fig.c:

the entire surface of the satellite cell-nerve cell body unitfacing the interstitial connective tissue of the ganglion.The portions of the basal lamina in direct contact withthe nerve cell body surface (at the level of the gaps in thesatellite cell sheath) were always laminin-positive(Fig. 3). The collagen fibrils close to the surface of thebasal lamina facing the interstitial connective tissue werealso labelled, although less intensely than the basal lami-na. This labelling of collagen fibrils is perhaps due todiffusion of reaction product away from the adjacentbasal lamina.

In anti-fibronectin preparations, the reaction productwas present over the entire surface of each satellite cell-nerve cell body unit, including the portions located at thelevel of the gaps in the satellite cell sheath. The reactionproduct for fibronectin was distributed more widely thanthat for laminin and was associated not only with thebasal lamina, but also with the amorphous matrix materi-al external to the basal lamina (Fig. 4).

Both in anti-laminin and anti-fibronectin preparations,immunostaining was never detected in nerve cell bodiesor in satellite cells. Both the intercellular spaces between

the adjacent satellite cells and the space between satellitecell sheath and enclosed nerve cell body were consistent-ly free of reaction product.

Differences in the distribution and abundance of thereaction product between young adult and aged rabbitswere not found, in either anti-laminin or anti-fibronectinpreparations. Reaction product was not detected in anycontrol preparations.

Discussion

Perikaryal projections were abundant on the neuronalsurface covered by satellite cells and were absent fromportions in direct contact with the extracellular matrix.This result: (1) is consistent with our previous conclu-sion (Pannese et al. 1994, 1995) that environmental fac-tors influence the outgrowth of perikaryal projectionsfrom the spinal ganglion neurons; (2) suggests that satel-lite cells permit the outgrowth of these projections bothin young adult and old rabbits; (3) suggests that in thespinal ganglia of adult rabbits the perineuronal extracel-

203

Fig. 3 Electron micrograph ofa rabbit spinal ganglion show-ing immunoreactivity for lami-nin. The reaction product is as-sociated with the basal laminaalong the surface of the satellitecell-nerve cell body unit facingthe interstitial connective tissueof the ganglion (arrows). Notethat the portion of the basallamina in direct contact withthe nerve cell body surface(gap in the satellite cell sheathbetween ) is laminin-positive(■■ marks satellite cell nuclei).Rabbit of 13 months. × 3800&/fig.c:

lular matrix is not in itself able to promote the outgrowthof these projections. We cannot definitively exclude thatthe extracellular matrix exerts some influence on the sat-ellite cells, which might in turn affect the nerve cell bodysurface. However, even if this hypothesis were correct, itwould not reduce the role of satellite cells in affectingthe outgrowth of projections.

In the spinal ganglia of the rat and guinea pig, theperineuronal extracellular matrix contains laminin and fi-bronectin (Bannerman et al. 1986; Schiff and Rosenbluth1986; Warburton and Santer 1997). Our study has shownthat these two glycoproteins are also present in the peri-neuronal extracellular matrix of rabbit spinal ganglia,both in young adult and old animals. It is noteworthy thatthese glycoproteins are also present at the level of thegaps in the satellite cell sheath, i.e. in close apposition tothe portions of nerve cell body surface that lack projec-tions. Numerous investigations have shown that lamininand fibronectin promote the growth of axonal processesfrom sensory ganglion neurons (Baron-Van Evercoorenet al. 1982; Carbonetto et al. 1983; Manthorpe et al.1983; Rogers et al. 1983; Hammarback et al. 1985; Mad-ison et al. 1985; Unsicker et al. 1985; Gundersen 1987;Orr and Smith 1988; Kuecherer-Ehret et al. 1990; Itoh etal. 1991; Delrée et al. 1993; Tomaselli et al. 1993; Tongeet al. 1997). By contrast, in normal conditions, these twoglycoproteins are not in themselves able to promote thegrowth of perikaryal projections from the sensory neu-

rons of spinal ganglia in the adult rabbit. It is likely thatthe outgrowth of axons and the outgrowth of perikaryalprojections are under the control of different mecha-nisms.

It is known that sensory ganglion neurons are able toinfluence their supporting neuroglial cells (satellite andSchwann cells). For example, these neurons regulate thenumber of surrounding satellite cells (Pannese 1960,1981, 1994; Humbertson et al. 1969) and the size of therelated neuroglial sheath (satellite cell sheath: Pannese etal. 1972, 1975; Schwann cell sheath: Pannese et al. 1987,1988); they also stimulate S100 production by the pre-cursors of their attendant neuroglial cells (Holton andWeston 1982). Other peripheral neurons have similar ac-tivities (e.g. see Pomeroy et al. 1996). Furthermore, sen-sory ganglion neurons regulate proliferation (Wood andBunge 1975), differentiation and maturation (Jessen etal. 1987; Dong et al. 1995), membrane organization(Despeyroux et al. 1994) and microtubule arrangement(Kidd et al. 1996) in their related Schwann cells. Finally,these neurons influence myelin formation (Weinberg andSpencer 1976), basal lamina deposition (Bunge et al.1982) and the expression of the transcription factor SCIP(Scherer et al. 1994) by their attendant Schwann cells.

These influences, however, are not unidirectional.Work over the past 15 years or so has shown that satelliteand Schwann cells can influence the neurons with whichthey are associated. For example, satellite and Schwann

204

Fig. 4 Electron micrograph ofa rabbit spinal ganglion show-ing immunoreactivity for fibro-nectin. The reaction product,which is present over the sur-face of the satellite cell-nervecell body unit, is associatedwith both the basal lamina andthe adjacent amorphous materi-al. Rabbit of 66 months. × 5660

cells can provide trophic support to sensory neurons (Va-ron et al. 1974; Heumann et al. 1987; Taniuchi et al.1988), can influence neuronal maturation (Mudge 1984),axon diameter (Windebank et al. 1985; Pannese et al.1988; de Waegh et al. 1992; Pannese 1994), the synthesisof specific enzymes (Fan and Katz 1993) as well as themolecular organization of the plasma membrane (Cooperand Lau 1986; Rosenbluth 1989; Joe and Angelides1992; Hinson et al. 1997; Rasband et al. 1998). Thesecells also prevent dendrite formation on the sensory neu-rons with which they are associated (De Koninck et al.1993). The present study provides a further example ofthe influence that supporting neuroglial cells have onsensory ganglion neurons. Taken together, these data in-dicate that there is reciprocal influence between the sen-sory ganglion neurons and their supporting neuroglialcells.

Further research is needed to establish whether satel-lite cells are only permissive or play a more active role ininfluencing the outgrowth of perikaryal projections.

&p.2:Acknowledgements The authors wish to thank Professor A. Cal-ligaro for his helpful suggestions, Professor G. Gabella for criticalreading of the manuscript and Mr. F. Redaelli for photographic as-sistance. This research was supported in part by a MURST grant,Italy.

References

Bannerman PGC, Mirsky R, Jessen KR, Timpl R, Duance VC(1986) Light microscopic immunolocalization of laminin, typeIV collagen, nidogen, heparan sulphate proteoglycan and fi-bronectin in the enteric nervous system of rat and guinea pig. JNeurocytol 15:733–743

Baron-Van Evercooren A, Kleinman HK, Ohno S, Marangos P,Schwartz JP, Dubois-Dalcq ME (1982) Nerve growth factor,laminin, and fibronectin promote neurite growth in human fe-tal sensory ganglia cultures. J Neurosci Res 8:179–193

Bunge MB, Williams AK, Wood PM (1982) Neuron-Schwann cellinteraction in basal lamina formation. Dev Biol 92:449–460

Carbonetto S, Gruver MM, Turner DC (1983) Nerve fiber growthin culture on fibronectin, collagen, and glycosaminoglycansubstrates. J Neurosci 3:2324–2335

Cattoretti G, Pileri S, Parravicini C, Becker MHG, Poggi S, Bi-fulco C, Key G, D’Amato L, Sabattini E, Feudale E, ReynoldsF, Gerdes J, Rilke F (1993) Antigen unmasking on formalin-fixed, paraffin-embedded tissue sections. J Pathol 171:83–98

Cooper E, Lau M (1986) Factors affecting the expression of ace-tylcholine receptors on rat sensory neurones in culture. J Phy-siol 377:409–420

De Koninck P, Carbonetto S, Cooper E (1993) NGF induces neo-natal rat sensory neurons to extend dendrites in culture afterremoval of satellite cells. J Neurosci 13:577–585

Delrée P, Ribbens C, Martin D, Rogister B, Lefebvre PP, Rigo JM,Leprince P, Schoenen J, Moonen G (1993) Plasticity of devel-oping and adult dorsal root ganglion neurons as revealed invitro. Brain Res Bull 30:231–237

Despeyroux S, Amédée T, Coles JA (1994) Axon contact is asso-ciated with modified expression of functional potassium chan-nels in mouse Schwann cells. Proc R Soc Lond B Biol Sci258:255–260

Dong Z, Brennan A, Liu N, Yarden Y, Lefkowitz G, Mirsky R,Jessen KR (1995) Neu differentiation factor is a neuron-gliasignal and regulates survival, proliferation, and maturation ofrat Schwann cell precursors. Neuron 15:585–596

Fan G, Katz DM (1993) Non-neuronal cells inhibit catecholamin-ergic differentiation of primary sensory neurons: role of leuke-mia inhibitory factor. Development 118:83–93

Gown AM, Wever N de, Battifora H (1993) Microwave-based an-tigenic unmasking. A revolutionary new technique for routineimmunohistochemistry. Appl Immunohistochem 1:256–266

Gundersen RW (1987) Response of sensory neurites and growthcones to patterned substrata of laminin and fibronectin in vit-ro. Dev Biol 121:423–431

Hammarback JA, Palm SL, Furcht LT, Letourneau PC (1985)Guidance of neurite outgrowth by pathways of substratum-ad-sorbed laminin. J Neurosci Res 13:213–220

Heumann R, Korsching S, Bandtlow C, Thoenen H (1987) Chang-es of nerve growth factor synthesis in nonneuronal cells in re-sponse to sciatic nerve transection. J Cell Biol 104:1623–1631

Hinson AW, Gu XQ, Dib-Hajj S, Black JA, Waxman SG (1997)Schwann cells modulate sodium channel expression in spinalsensory neurons in vitro. Glia 21:339–349

Holton B, Weston JA (1982) Analysis of glial cell differentiationin peripheral nervous tissue. I and II. Dev Biol 89:64–71 and72–81

Humbertson A Jr, Zimmermann E, Leedy M (1969) A chronologi-cal study of mitotic activity in satellite cell hyperplasia associ-ated with chromatolytic neurons. Z Zellforsch 100:507–515

Itoh T, Sobue G, Yasuda T, Mitsuma T, Takahashi A, Kimata K(1991) Geometry of adult rat sensory neurons in culture; itsmodulation by laminin. Neurosci Lett 123:212–216

Jessen KR, Mirsky R, Morgan L (1987) Axonal signals regulatethe differentiation of non-myelin-forming Schwann cells: animmunohistochemical study of galactocerebroside in transec-ted and regenerating nerves. J Neurosci 7:3362–3369

Joe E, Angelides K (1992) Clustering of voltage-dependent sodi-um channels on axons depends on Schwann cell contact. Na-ture 356:333–335

Kidd G, Andrews SB, Trapp BD (1996) Axons regulate the distri-bution of Schwann cell microtubules. J Neurosci 16:946–954

Kuecherer-Ehret A, Graeber MB, Edgar D, Thoenen H, Kreutz-berg GW (1990) Immunoelectron microscopic localization oflaminin in normal and regenerating mouse sciatic nerve. JNeurocytol 19:101–109

Madison R, Silva CF da, Dikkes P, Chiu T-H, Sidman RL (1985)Increased rate of peripheral nerve regeneration using biore-sorbable nerve guides and a laminin-containing gel. Exp Neu-rol 88:767–772

Manthorpe M, Engvall E, Ruoslahti E, Longo FM, Davis GE, Va-ron S (1983) Laminin promotes neuritic regeneration from cul-tured peripheral and central neurons. J Cell Biol 97:1882–1890

Mudge AW (1984) Schwann cells induce morphological transfor-mation of sensory neurones in vitro. Nature 309:367–369

Obata K, Inoue H (1982) Development of microspikes and neu-rites in cultured dorsal root ganglion cells. Neurosci Lett30:1–5

Orr DJ, Smith RA (1988) Neuronal maintenance and neurite ex-tension of adult mouse neurones in non-neuronal cell-reducedcultures is dependent on substratum coating. J Cell Sci 91:555–561

Pannese E (1960) Observations on the morphology, submicroscop-ic structure and biological properties of satellite cells (s.c.) insensory ganglia of mammals. Z Zellforsch 52:567–597

Pannese E (1981) The satellite cells of the sensory ganglia. AdvAnat Embryol Cell Biol, vol 65. Springer, Berlin HeidelbergNew York

Pannese E (1994) Neurocytology. Fine structure of neurons, nerveprocesses, and neuroglial cells. Thieme, Stuttgart New York

Pannese E, Bianchi R, Calligaris B, Ventura R, Weibel ER (1972)Quantitative relationships between nerve and satellite cells inspinal ganglia. An electron microscopical study. I. Mammals.Brain Res 46:215–234

Pannese E, Ventura R, Bianchi R (1975) Quantitative relationshipsbetween nerve and satellite cells in spinal ganglia: an electronmicroscopical study. II. Reptiles. J Comp Neurol 160:463–476

205

Pannese E, Gioia M, Carandente O, Ventura R (1983) A quantita-tive electron microscope study of the perikaryal projections ofsensory ganglion neurons. I. Cat and rabbit. J Comp Neurol214:239–250

Pannese E, Bianchi R, Gioia M, Ventura R (1985) A quantitativeelectron microscope study of the perikaryal projections of sen-sory ganglion neurons. II. Gecko and lizard. J Comp Neurol240:212–218

Pannese E, Rigamonti L, Procacci P, Ledda M, Arcidiacono G,Frattola D (1987) An electron microscope study of quantita-tive relationships between axon and Schwann cell sheath inmyelinated fibres of peripheral nerves. Anat Embryol 175:423–430

Pannese E, Arcidiacono G, Frattola D, Rigamonti L, Procacci P,Ledda M (1988) Quantitative relationships between axoplasmand Schwann cell sheath in unmyelinated nerve fibres. Anelectron microscope study. J Anat 159:49–56

Pannese E, Ledda M, Conte V, Procacci P, Matsuda S (1990a)Scanning electron-microscope observations of the perikaryalprojections of rabbit spinal ganglion neurons after enzymaticremoval of connective tissue and satellite cells. Cell TissueRes 260:167–173

Pannese E, Ledda M, Conte V, Procacci P (1990b) The perikaryalprojections of rabbit spinal ganglion neurons. A comparison ofthin section reconstructions and scanning microscopy views.Anat Embryol 181:427–432

Pannese E, Rigamonti L, Ledda M, Arcidiacono G (1994) Perik-aryal projections of spinal ganglion neurons: quantitative dif-ferences between membrane domains in contact with differentmicroenvironments. J Anat 185:497–502

Pannese E, Ledda M, Conte V, Rigamonti L, Procacci P (1995) OnThe influence of the perineuronal microenvironment on theoutgrowth of perikaryal projections of spinal ganglion neu-rons. J Submicrosc Cytol Pathol 27:303–308

Pannese E, Procacci P, Ledda M (1996a) Ultrastructural localiza-tion of actin in the cell body of rat spinal ganglion neurons.Anat Embryol 194:527–531

Pannese E, Procacci P, Ledda M, Conte V (1996b) Age-related re-duction of the satellite cell sheath around spinal ganglion neu-rons in the rabbit. J Neurocytol 25:137–146

Pomeroy SL, Zurakowski D, Khoxayo S, Endara M, Dikkes P(1996) Postnatal addition of satellite cells to parasympatheticneurons. J Comp Neurol 375:518–525

Rasband MN, Trimmer JS, Schwarz TL, Levinson SR, EllismanMH, Schachner M, Shrager P (1998) Potassium channel distri-bution, clustering, and function in remyelinating rat axons. JNeurosci 18:36–47

Rogers SL, Letourneau PC, Palm SL, McCarthy J, Furcht LT(1983) Neurite extension by peripheral and central nervoussystem neurons in response to substratum-bound fibronectinand laminin. Dev Biol 98:212–220

Rosenbluth J (1989) Role of Schwann cells in differentiation ofthe axolemma: consequences of myelin deficiency in spinalroots of the dystrophic mouse mutant. In: Scarpini E, FioriMG, Pleasure D, Scarlato G (eds) Peripheral nerve develop-ment and regeneration: recent advances and clinical applica-tions. Liviana Press, Padova, pp 39–53

Scherer SS, Wang D, Kuhn R, Lemke G, Wrabetz L, Kamholz J(1994) Axons regulate Schwann cell expression of the POUtranscription factor SCIP. J Neurosci 14:1930–1942

Schiff R, Rosenbluth J (1986) Ultrastructural localization of lami-nin in rat sensory ganglia. J Histochem Cytochem 34:1691–1699

Taniuchi M, Clark HB, Schweitzer JB, Johnson EM Jr (1988) Ex-pression of nerve growth factor receptors by Schwann cells ofaxotomized peripheral nerves: ultrastructural location, sup-pression by axonal contact, and binding properties. J Neurosci8:664–681

Tomaselli KJ, Doherty P, Emmett CJ, Damsky CH, Walsh FS, Rei-chardt LF (1993) Expression of β1 integrins in sensory neu-rons of the dorsal root ganglion and their functions in neuriteoutgrowth on two laminin isoforms. J Neurosci 13:4880–4888

Tonge DA, Golding JP, Edbladh M, Kroon M, Ekström PER, Eds-tröm A (1997) Effects of extracellular matrix components onaxonal outgrowth from peripheral nerves of adult animals invitro. Exp Neurol 146:81–90

Unsicker K, Skaper SD, Davis GE, Manthorpe M, Varon S (1985)Comparison of the effects of laminin and the polyornithine-binding neurite promoting factor from RN22 Schwannomacells on neurite regeneration from cultured newborn and adultrat dorsal root ganglion neurons. Brain Res Dev Brain Res17:304–308

Varon S, Raiborn C, Burnham P (1974) Comparative effects ofnerve growth factor and ganglionic nonneuronal cells on puri-fied mouse ganglionic neurons in culture. J Neurobiol 5:355–371

Waegh SM de, Lee VM-Y, Brady ST (1992) Local modulation ofneurofilament phosphorylation, axonal caliber, and slow ax-onal transport by myelinating Schwann cells. Cell 68:451–463

Warburton AL, Santer RM (1997) The hypogastric and thirteenththoracic ganglia of the rat: effects of age on the neurons andtheir extracellular environment. J Anat 190:115–124

Weinberg HJ, Spencer PS (1976) Studies on the control of myelin-ogenesis. II. Evidence for neuronal regulation of myelin pro-duction. Brain Res 113:363–378

Windebank AJ, Wood P, Bunge RP, Dyck PJ (1985) Myelinationdetermines the caliber of dorsal root ganglion neurons in cul-ture. J Neurosci 5:1563–1569

Wood PM, Bunge RP (1975) Evidence that sensory axons are mi-togenic for Schwann cells. Nature 256:662–664

206