Reelin immunoreactivity in the larval sea lamprey brain

Journal of Chemical Neuroanatomy 23 (2002) 211–221

Reelin immunoreactivity in the larval sea lamprey brain

Emma Perez-Costas a, Miguel Melendez-Ferro a, Ysabel Santos b, Ramon Anadon a,M. Celina Rodicio a, Hector J. Caruncho a,*

a Department of Fundamental Biology, Faculty of Biology, Uni�ersity of Santiago de Compostela, 15782 Santiago de Compostela, Spainb Department of Microbiology and Parasitology, Faculty of Biology, Uni�ersity of Santiago de Compostela,

15782 Santiago de Compostela, Spain

Received 1 June 2001; accepted 6 December 2001

Abstract

In order to analyze the presence of a reelin-like protein in the brain of a primitive vertebrate with a laminar-type brain, suchas the sea lamprey, Western blot and immunohistochemical approaches were employed by using the G10 and 142 reelin-specificmonoclonal antibodies. Western blots of lamprey brain extracts showed bands of about 400 kDa, 180 kDa and others below 100kDa; similar bands were observed in samples from rat cerebellum. In different larval stages there was a prominent reelinimmunolabeling associated with the olfactory bulb, pallial regions, habenula, hypothalamus and optic tectum. In addition, theolfactory and optic tracts, as well as the afferent and efferent (fasciculus retroflexus) tracts of the habenular ganglion, also showedimmunopositivity in these stages. Interestingly, the highest level of labeling was observed in premetamorphic larvae, just prior toentering the metamorphic stage. These data indicate that reelin expression is also prominent in brains of primitive vertebrateswithout layered cortical regions, suggesting that some physiological roles of reelin not related to the regulation of neuronalmigration in layered cortical regions (i.e. involvement in axon pathfinding, synaptogenesis, dendritic arborization and neuronalplasticity) might have appeared earlier in evolution. © 2002 Elsevier Science B.V. All rights reserved.

Keywords: Agnathans; Extracellular matrix proteins; Laminar brain; Neuronal migration; Axon pathfinding; Dendritic arborization; Synaptogen-esis

www.elsevier.com/locate/jchemneu

1. Introduction

Reelin, a large extracellular glycoprotein of about400 kDa, has been shown to play an important role inregulating neuronal migration towards layered corticalregions in developing amniotic vertebrates (see as re-views Hirotsune et al., 1995; Rakic and Caviness, 1995;Frotscher, 1997, 1999; Curran and D’Arcangelo, 1998;Lambert de Rouvroit and Goffinet, 1998a,b; Rice andCurran, 1999a; Gilmore and Herrup, 2000). In additionto the regulation of migration towards layered corticalareas, reelin also appears to be involved during devel-opment in axon pathfinding and synaptogenesis in thehippocampus (Del Rıo et al., 1997; Borrell et al., 1999)and in regulating the pattern of dendritic arborizationin Purkinje cells (Miyata et al., 1996, 1997).

Reelin shows a widespread expression during braindevelopment of amniotic vertebrates, being present inthe olfactory bulb, cortical marginal zone, cerebellargranule cells and retina (Bar et al., 1995; D’Arcangeloet al., 1995; Ogawa et al., 1995; Ikeda and Terashima,1997; Schiffmann et al., 1997; Alcantara et al., 1998;Meyer and Goffinet, 1998; Pesold et al., 1998; Bernieret al., 1999, 2000; Goffinet et al., 1999; Meyer andWahle, 1999; Rice and Curran, 1999b), and continuesto be highly expressed in the adult brain, where it mightbe involved in regulating neuronal plasticity (Im-pagnatiello et al., 1998; Pesold et al., 1998, 1999;Guidotti et al., 2000; Rodrıguez et al., 2000).

The sea lamprey, Petromyzon marinus, belongs to themost primitive group of living vertebrates, the jawlessfishes or Agnatha. The anadromous sea lamprey passesthrough a life cycle which comprises several years ofburrowing as a filter feeder, and after a transformingphase the adult will migrate to the sea, where it feeds

* Corresponding author. Tel.: +34-981-563100x13296; fax: +34-981-596904.

E-mail address: [email protected] (H.J. Caruncho).

0891-0618/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved.PII: S 0 8 9 1 -0618 (01 )00156 -9

E. Perez-Costas et al. / Journal of Chemical Neuroanatomy 23 (2002) 211–221212

parasitically and grows prior to its upstream spawningmigration. Its life cycle, together with their phylogeneticposition and relative simplicity, makes lampreys anattractive model for the study of the development,structure and function of the vertebrate central nervoussystem. The histogenesis of the lamprey central nervoussystem follows the general vertebrate pattern (Nieuwen-huys and Nicholson, 1998; Pombal and Puelles, 1999),therefore it could be expected that fundamentalmolecules implied in brain development are alreadyexpressed in this group.

In order to study the expression and possible roles ofreelin in a primitive line of vertebrates with a laminar-type brain [term indicating that the vast majority ofneuronal somata remain in the periventricular region,extending their dendritic tree to lateral areas occupiedmostly by neuropil (see as a review Nieuwenhuys andNicholson, 1998)], we carried out Western blot andimmunohistochemical experiments on the expression ofthis protein in the sea lamprey brain. In this work, wedemonstrated the expression of a reelin-like protein indifferent brain regions, being the highest level of im-munostaining found in premetamorphic larval brains.

2. Materials and methods

Larval lampreys from 25 to 120 mm in length (fivespecimens of each length studied), and adult sea lam-preys (twenty-four specimens for Western blot, and twospecimens for immunohistochemistry) (Petromyzonmarinus, L.), caught at the Ximonde Biological Station(Ulla River, Galicia, Spain), with permission of theOffice for Environmental Affairs of the Xunta de Gali-cia, were used in this study. Larval lampreys werecaptured by dredging of the bottom of the river, whileadult individuals were captured in the traps of theBiological Station. Water temperature of the riverranged from 16 to 20 °C. Larval lampreys were kept inaerated aquaria with a bed of sediment from the river.Adult lampreys were immediately sacrificed and storedafter capture. All experiments conformed to the Eu-ropean Community guidelines on animal care andexperimentation.

2.1. Antibodies

Two monoclonal antibodies (a generous gift from DrA. Goffinet, University of Namur, Belgium) were em-ployed. Both antibodies (142 and G10) recognized epi-topes close to the amino terminal region of reelin. The142 antibody recognizes the protein SP region (epitopelocated between aminoacids 164–189), while the G10antibody recognizes the protein H region (epitope lo-cated between aminoacids 164–245) (De Bergeyck etal., 1998). In addition, the G10 antibody has been used

in Western blots and immunohistochemical experimentsin murine brain (De Bergeyck et al., 1998; Lambert deRouvroit and Goffinet, 1998a,b; Pesold et al., 1998,1999), while the 142 antibody has been used to studyreelin distribution in the human and non-human pri-mate brains (Impagnatiello et al., 1998; Meyer andGoffinet, 1998; Rodrıguez et al., 2000), in reptiles(Bernier et al., 1999; Goffinet et al., 1999), and also inamphibians and fishes (Perez-Garcıa et al., 2001).

2.2. Western blots

The very small size of larval brain implies the use ofa high number of specimens (�40 premetamorphicbrains) for each protein extraction cycle. Due to theunavailability of such a high number of larval brains,we have used adult brains for this assay.

Adult sea lampreys (a minimum of eight specimensper each extraction cycle) were sacrificed with an over-dose of benzocaine, and their brains were removed andimmediately frozen at −80 °C. Additionally,Sprague–Dawley rats were sacrificed with an overdoseof chloroform, and their cerebella were also removedand stored at −80 °C. Samples were freeze-dried andhomogenized at 4 °C in 6 volume extraction buffer (50mM Tris, 5 mM EDTA, 150 mM NaCl, 2 mM Phenyl-methylsulfonylfluoride, 10 mM N-ethylmaleimide, pH7.6) and centrifuged at 20 000×g at 4 °C. The proteinspresent in the supernatants were precipitated with 6volume of 100% methanol followed by 100×g centrifu-gation at 4 °C. The extract samples (a minimum con-centration of 50 �g of total protein for lampreysamples, and 15 �g for rat samples) were resolved on6% acrylamide gel, and then electroblotted onto 0.2 �mpolyvinylidene difluoride membranes (Biorad, Hercules,CA). After transfer, non-specific binding sites on themembrane were blocked by incubating in 5% non-fatpowdered milk dissolved in Tris-buffered saline (TBS)for 1 h. After blocking, the membranes were rinsed inTBS containing 0.05% Tween 20 and incubatedovernight with monoclonal anti-reelin antibodies (142or G10) diluted 1:1000–5000 and 1:1000 in TBS, re-spectively. The membranes were processed according tothe peroxydase-antiperoxydase (PAP) procedure anddeveloped with 3-3�diaminobenzidine.

2.3. Immunohistochemistry

Larvae of 25, 50, 75 mm, premetamorphic (�100mm) were deeply anesthetized with benzocaine (0.05 gper liter of fresh water) and decapitated. Whole headswere fixed overnight by immersion in freshly preparedcold 4% paraformaldehyde in 0.1 M phosphate buffer(PB), pH 7.4. Adult lampreys were deeply anaesthetizedas above, and perfused through the aorta with lampreyringer in 2 mM Hepes buffer, pH 7.4, following by the

E. Perez-Costas et al. / Journal of Chemical Neuroanatomy 23 (2002) 211–221 213

same fixative for larval specimens. In addition, some17 days-old rat embryos were sacrificed with an over-dose of chloroform, decapitated and their brains wereimmediately removed from the skull and immersed inthe same cold fixative as above.

Whole heads (larval lampreys) and brains (adultlampreys and rat embryos) were rinsed in 0.1 M PB,pH 7.4, and cryoprotected in 30% sucrose in PB, em-bedded in OCT compound (Sakura, Torrance, USA)and frozen in liquid nitrogen-cooled isopentane. Coro-nal or horizontal sections (12 �m in thickness) werecut on a cryostat, mounted on chromalum gelatinizedslides, and processed for immunolabeling. Larval lam-prey sections were blocked in 10% normal goat serumand incubated with a monoclonal antibody againstreelin in a humid chamber (142, diluted 1:500; or G10diluted 1:1000), overnight at room temperature. Afterwashing in phosphate-buffered saline, the sections wereincubated for 1 h with a goat anti-mouse IgG diluted1:50 (Dako, Glostrup, Denmark). Sections were pro-cessed with the PAP-mouse complex diluted 1:500(Sigma, St. Louis, USA), and developed with 3-3�-di-aminobenzidine. In controls where the primary anti-body was omitted no immunostaining was observed.Adult brain sections were also processed by the PAPprocedure using both antibodies against reelin (142antibody dilution ranged from 1:100 to 1:1000, andG10 antibody dilution ranged from 1:50 to 1:1000).

Coronal sections from embryonic rat brains wereprocessed using the G10 antibody in the same condi-tions previously described for immunohistochemistryin larval lamprey brains.

3. Results

3.1. Western blots

In adult sea lamprey brain extracts both the 142 andG10 antibodies evidenced a band of about 400 kDa, asecond one of about 180 kDa, and others of molecularweights lower than 100 kDa (Fig. 1). Similar bands ofabout 400, 180 kDa, and others of molecular weightlower than 100 kDa, were observed in rat cerebellarextracts processed at the same time. In samples thathave been frozen and thawed several times the inten-sity of the 400 kDa and that of 180 kDa bands stain-ing was decreased while those of less than 100 kDawas increased, suggesting that they could be prote-olytic fragments (data not shown).

3.2. Immunohistochemistry

3.2.1. Immunolabeling featuresBoth the G10 and 142 antibodies showed a good

immunolabeling in larval lamprey brains. The same

brain regions that showed immunostaining when usingthe 142 antibody (Fig. 2A) were also revealed as posi-tive with the G10 antibody (Fig. 2B) and vice-versa,while control experiments by omitting the primary an-tibody showed no immunolabeling (Fig. 2C).

We have tried to perform an immunolabeling experi-ment in adult brain sections. However, with the condi-tions used for larval immunohistochemistry (Section 2)we were unable to detect any consistent reelin im-munoreactivity, although Western blot experiments(see above) clearly show positive bands.

3.2.1.1. Specificity of immunolabeling in fiber tracts. Inthe developing lamprey brain some fiber tracts wereseen as immunopositive (Fig. 2D), while others wereclearly negative. The appearance of the immunoposi-tive tracts in the lamprey brain (Fig. 2D) is similar tothat observed in some tracts of the rodent brain duringdevelopment (Fig. 2E), validating the observations inlamprey brains. In addition, the omission of the pri-mary antibody shows a complete lack of labeling infiber tracts (Fig. 2C).

3.2.1.2. Diffuse immunolabeling. In some regions ofthe lamprey brain we found a diffuse immunostainingthat showed a punctate appearance similar to thatobserved in some areas of the developing rat brain(Fig. 2F–G).

Fig. 1. Western blotting of reelin in sea lamprey and rat brainsamples. Four positive bands are clearly seen in lamprey (with the 142and G10 antibodies) as well as in rat (with the G10 antibody)samples, at about 400, 180, 65 and 55 kDa. MWM, Molecular weightmarkers; L, Labeling of laminin as a marker of 400 kDa; S, Standardmolecular weight markers.

E. Perez-Costas et al. / Journal of Chemical Neuroanatomy 23 (2002) 211–221214

Fig. 2. Comparative images of reelin immunoreactivity. (A–C) Comparison of results obtained using two anti-reelin antibodies; 142 (A), G10 (B),and a control section omitting the primary antibody (C) in adjacent coronal sections through the habenular ganglion of a 75 mm larva. Note thepresence of the reelin immunopositive afferent tract of the habenular ganglion (arrows) in (A) and (B), and the absence of immulabeling in thenegative control (asterisk in C). (D and E) Specificity of immunolabeling in fiber tracts as shown by comparison of the stainings obtained withthe G10 antibody in a premetamorphic larval lamprey (D), and an embryonic rat brain (E): (D) Coronal section of a premetamorfic larval lampreyshowing the immunopositive optic tract (arrows). (E) Coronal section of a 17 days-old embryo rat brain showing the immunopositive striamedullaris tract (arrows). (F and G) High magnification micrographs showing the similarity of the diffuse immunolabeling as observed in lampreyand embryonic rat brains: (F) Coronal section through the lateral pallium of a premetamorphic larval lamprey processed with the 142 antibody,showing some immunopositive neuronal somata (open arrows), surrounded by a punctate diffuse reelin immunostaining (white arrowheads). (G)Coronal section of a 17 days-old embryo rat cortex processed with the G10 antibody, showing a reelin immunopositive Cajal-Retzius cell (openarrow), surrounded by a punctate diffuse reelin immunostaining (white arrowheads). Note that some immunolabeled somata also display animmunopositive process (black arrowheads) both in lamprey (F) and rat (G) brains. Scale bars: 100 �m in (A–C); 50 �m in (D); 25 �m in (E);10 �m in (F) and (G).

3.2.2. Distribution of reelin immunoreacti�ity in the lam-prey brain

During larval lamprey development, there are differ-ences in the distribution of reelin immunoreactivity inseveral brain regions. In general, the highest level ofreelin immunoreactivity was observed in premetamor-phic larval brains.

3.3. Telencephalon

3.3.1. Olfactory bulbThe earliest larval stages studied (25 and 50 mm),

showed a low level of reelin immunostaining in thelateral regions associated with the incoming of theolfactory nerves, being this immunolabeling slightly

E. Perez-Costas et al. / Journal of Chemical Neuroanatomy 23 (2002) 211–221 215

more evident in larvae of 50 mm. In larvae of 75 mm,as well as in premetamorphic stages, there was positivityassociated with the olfactory nerves and also a high levelof staining in the glomerular layer and a weaker labelingin other layers (Fig. 3A). High magnification micro-graphs allowed to detect immunopositive somata in theperiglomerular region as well as a punctate diffuselabeling in the glomeruli (Fig. 3B).

3.3.2. PalliumLarvae of 25–50 mm showed a very weak immunola-

beling in the lateral pallium, however no immunopositivesomata were evident at these stages. Larvae of 75 mm andpremetamorphic stages showed some immunopositiveneuronal somata in deep layers of the lateral pallium(Fig. 2F, Fig. 3D), as well as a diffuse immunolabelingsurrounding them, and also laterally, in cell-poor areas(Fig. 3C–D). The lamina of neuronal somata adjacentto the ventricle (subventricular zone) evidenced a lack ofimmunostaining (see star in Fig. 3D).

There was no evidence of reelin immunolabeling in themedial pallium till the premetamorphic stages, wherethere was a weak diffuse immunoreactivity surroundingimmunonegative neuronal somata.

3.4. Diencephalon

3.4.1. Hypothalamic regions

3.4.1.1. Preoptic and infundibular hypothalamic regions.In the hypothalamus there was no evidence of labelingin larvae of less than 75 mm. However, in 75 mm andpremetamorphic larval stages there was a diffuse im-munolabeling close to the ventricle in the preoptic area(Fig. 4A), as well as positive periventricular fiber bundles(Fig. 4F).

3.4.1.2. Optic chiasm region. In 25 and 50 mm larvae therewas a weak labeling associated with the developing opticnerve and chiasm. In larvae of 75 mm there was amoderate staining of the optic nerve and tract, whereasin the premetamorphic stage there was a prominentlabeling of both the nerve and ascending lateral optictract (Fig. 2D, Fig. 4A–C).

3.4.2. Epithalamus

3.4.2.1. Stria medullaris and rostral habenula. In 25 mmlarvae there was a very weak labeling of the afferent

Fig. 3. Reelin immunoreactivity in coronal sections through the olfactory bulb and rostral forebrain. (A) Coronal section through the olfactorybulb of a 75 mm larva processed with the 142 antibody, showing an intense staining of the glomerular layer (g) and also a diffuse labelingthroughout the olfactory bulb (asterisk). (B) High magnification of the inset shown in (A). Note the presence of some immunoreactive neuronalsomata (open arrows) as well as a punctate diffuse immunostaining in the glomerular layer (g). (C) Coronal section of a premetamorphic larvaprocessed with the 142 antibody, showing an intense diffuse immunostaining in the lateral pallium (asterisks). (D) High magnification of the insetin (C) pointing out the presence of immunoreactive neuronal somata in the lateral pallium (open arrows) and the punctate diffuse immunolabelingsurrounding them. Observe the lack of immunostaining in the somata of the subventricular zone (star). Note the presence in some neurons of areelin immunoreactive process (arrowhead). Scale bars: 100 �m in (A) and (C); 10 �m in (B) and (D).

E. Perez-Costas et al. / Journal of Chemical Neuroanatomy 23 (2002) 211–221216

Fig. 4. Reelin immunoreactivity in the caudal forebrain. (A) Horizontal section of a premetamorphic larval lamprey at the level of the preopticrecess (asterisk) processed with the G10 antibody, showing diffuse immunoreactivity surrounding it (arrowheads).Note that the fibers of the medialand lateral optic tracts are also immunopositive (arrows). (B) Horizontal section processed with the G10 antibody, showing reelin labeling in theoptic chiasm (arrows) and optic nerve (open arrows) of a premetamorphic larva. (C) Coronal section through the diencephalon of apremetamorphic larva processed with the 142 antibody, showing the reelin immunoreactive afferent tract of the habenular ganglion (blacktriangles), and the presence of reelin labeling in the optic tract (arrows), as well as in the optic nerve (open arrows). (D) High magnification ofthe inset in (C) demonstrating reelin immunoreactive fibers in the afferent tract of the habenular ganglion (black triangles). Note the nearbyimmunonegative neuronal somata (asterisks). (E) Caudal diencephalic region of a premetamorphic larva processed with the 142 antibody, showingdiffuse reelin immunoreactivity in the habenular ganglion (arrowhead), as well as in the fasciculus retroflexus (arrows). Inset: Detail of the caudalhabenula showing an immunopositive neuron (open arrow) surrounded by a diffuse labeling. (F) Image of the caudal diencephalon of apremetamorphic larva processed with the 142 antibody, showing a very intense reelin immunoreactivity in the posterior commissure (open arrows),as well as positive fiber bundles (arrowheads) that course between periventricular cellular layers in the thalamic (th) and hypothalamic (hyp)regions. Note also the presence of the immunopositive fasciculus retroflexus (black arrows). Scale bars: 200 �m in (C), (F), and (E); 100 �m in(A) and (B); 50 �m in (D); 10 �m in (E) inset.

E. Perez-Costas et al. / Journal of Chemical Neuroanatomy 23 (2002) 211–221 217

tract of the habenula, the so called ‘stria medullaristract’, which courses from telencephalic pallialregions following a dorso-caudal pathway towards therostral habenula. In 50 mm larvae the immuno-reactivity of this pathway was slightly increased. Larvaeof 75 mm as well as those in premetamorphicstages showed an intense labeling of the stria medullaris(Fig. 2A–B, Fig. 4C–D), but no positive neuronalsomata were found in this area (see asterisks inFig. 4D).

3.4.2.2. Caudal habenula and fasciculus retroflexus. Bothlarvae of 25 or 50 mm showed a weak labeling in thefasciculus retroflexus. In 75 mm and premetamorphiclarval brains a conspicuous reelin diffuse immunolabel-ing was found in the caudal habenula (Fig. 4E) sur-rounding immunopositive neuronal somata (Fig. 4Einset). In addition, there was an intense labeling of thefibers in the fasciculus retroflexus (Fig. 4E) that areoriginated from caudal habenula positive neurons (Fig.4E inset).

3.4.3. Thalamus

3.4.3.1. Posterior commissure and dorsal thalamus. Inlarvae of 25 or 50 mm there was a weak immunoreac-tivity of fibers crossing at the posterior commissurelevel, while the dorsal thalamic region appeared asimmunonegative. In 75 mm and premetamorphiclarval stages there was a very intense staining of poste-rior commissure fibers (see open arrows in Fig. 4F).Besides this, there were some dorsal thalamic im-munopositive fiber bundles that coursed between im-munonegative neuronal somata of periventricular layers(Fig. 4F).

3.5. Mesencephalon

3.5.1. Pretectal and tectal regionsLarval lampreys of 25 or 50 mm showed no labeling

in the pretectal area. Samples of 75 mm larval brainspresented a weak diffuse immunolabeling lateral to themedial cellular region. In pretectal areas of premeta-morphic larvae there were immunoreactive fiber bun-dles between cells of the periventricular stratum,as well as a diffuse immunolabeling with a punctateappearance (Fig. 5A–B). The optic tectum of premeta-morphic larvae showed an intense immunoreactivityassociated laterally with incoming optic tract fibers,that can be followed towards the medial region wherethey are seen between cellular layers close to the ventri-cle (Fig. 5C–D). There also was an intense diffuseimmunoreactivity surrounding neurons located awayfrom the ventricle (Fig. 5C–D), some of these neuronsshowed an immunopositive somata (see open arrows inFig. 5D).

3.6. Rombencephalon

In general, this area showed a very low level ofstaining in the lamprey brain. Only in premetamorphiclarvae some weakly immunopositive reticular and oc-tavolateral neurons, as well as a faint diffuse immuno-labeling were observed.

4. Discussion

The present report shows that a protein recognizedby the G10 and 142 monoclonal antibodies and withsimilar molecular weight than mammalian reelin is ex-pressed in the lamprey brain. These antibodies recog-nize epitopes between amino acids 164–245 (G10) and164–189 (142) (De Bergeyck et al., 1998), and havebeen used in immunohistochemical studies of reelinexpression in the murine brain (G10), in the human andnon-human primate, reptilian, amphibian and fishbrains (142) (Impagnatiello et al., 1998; Meyer andGoffinet, 1998; Pesold et al., 1998, 1999; Bernier et al.,1999; Goffinet et al., 1999; Meyer and Wahle, 1999;Rodrıguez et al., 2000; Perez-Garcıa et al., 2001).

4.1. Reelin protein expression

Western blots from adult lamprey brain samplesusing the 142 or G10 antibodies clearly showed bandsof about 400 and 180 kDa, as well as other bands withmolecular weights lower than 100 kDa. Similar bands(400, 180 kDa, and others lower than 100 kDa), werealso observed in samples from rat cerebellar extractsprocessed together. In addition, samples that werefrozen and thawed several times showed a smalleramount of the 400 and 180 kDa bands and a higheramount of those of lower molecular weight, possiblyindicating that these bands represent reelin proteolyticfragments. The existence of different reelin proteolyticfragments was recently described for purified reelinprotein obtained from cell cultures (Quattrocchi et al.,2001). The finding of all of these bands in lampreybrain samples with the two different monoclonal anti-bodies strongly indicates that, in fact, they are recogniz-ing a reelin-like protein in the lamprey brain.

The existence of higher and lower molecular weightbands of about 400 and 180 kDa (full-length reelin anda reelin cleaved product) has been shown in mice, ratand human brains, as well as in cerebellar granule celland mouse reelin transfected human kidney (HEK293T) cell cultures (Impagnatiello et al., 1998; Lambertde Rouvroit et al., 1999a,b; Lacor et al., 2000; Quat-trocchi et al., 2001).The presence of the 180 kDaprotein might be due to the processing of the full-lengthreelin by a metalloproteinase after secretion or well at apost-endoplasmic reticulum compartment (Lambert de

E. Perez-Costas et al. / Journal of Chemical Neuroanatomy 23 (2002) 211–221218

Fig. 5. Reelin immunolabeling in pretectal and tectal areas. (A) Coronal section at the level of the rostral mesencephalon of a premetamorphiclarva processed with the 142 antibody, showing diffuse reelin immunolabeling, which is stronger in the dorsal and dorsolateral parts (whiteasterisk), as well as immunopositive fiber bundles (black arrows) coursing between pretectal neurons. (B) High magnification of (A) showing apunctate appearance of the diffuse immunostaining (white arrowheads). (C) Coronal section at the level of the optic tectum of a premetamorphiclarva processed with the 142 antibody, showing intense immunoreactive fiber bundles between periventricular cellular layers (arrows), as well asstrong diffuse labeling in the dorsolateral tectal region (white asterisks). (D) High magnification of (C) showing the strong reelin immunoreactivityin fiber bundles between periventricular cell layers (black arrows), as well as intense diffuse staining (white asterisk) and some reelinimmunopositive neuronal somata in lateral regions (open arrows). Scale bars: 100 �m in (C); 50 �m in (A); 25 �m in (D); 10 �m in (B).

Rouvroit et al., 1999a), or even by a self-degradationactivity of the reelin protein (Quattrocchi et al., 2001).In addition, the existence of alternative splicing ofreelin has been recently pointed out (Lambert de Rou-vroit et al., 1999b). The existence of the two bands (400and 180 kDa) in the lamprey brain, demonstrates thatboth of them were present early in vertebrate evolution.

In addition, the existence of bands of similar weightto those shown in this work was previously described inmurine brain protein extracts (Lambert de Rouvroit etal., 1999a).

4.2. Immunohistochemical localization of Reelin

4.2.1. Methodological considerationsBoth G10 and 142 anti-reelin antibodies showed

similar results when used in immunohistochemical la-beling of larval lamprey brain sections, similarly towhat happened in adult lamprey brain samples in West-ern blot experiments, indicating the validity of using

these antibodies to study the distribution of reelinexpression in the lamprey brain.

In preliminary experiments in adult lamprey brainsections, we were unable to detect consistent reelinimmunoreactivity with the conditions used in this work(Section 3), although we obtained a good signal inWestern blotting assays. This might be due to a lowexpression of the protein in the adult brain. This is inagreement with the fact that a relative high concentra-tion of anti-reelin antibody (compare the dilution usedin Western blotting with the dilution used in immuno-histochemistry in the larval brain, taking into accountthat antibodies in Western blot are usually employed50–100 times more diluted than in immunohistochem-istry), and a very high concentration of lamprey braintotal protein were necessary (Section 2) to detect thereelin expression in Western blot assays.

In several brain regions reelin positive neuronal so-mata were observed surrounded by a punctate diffuselabeling, similar to the appearance of reelin immunore-

E. Perez-Costas et al. / Journal of Chemical Neuroanatomy 23 (2002) 211–221 219

activity in developing rat brains (see for comparisonFig. 2F–G). Similar results were also described in othervertebrates (Impagnatiello et al., 1998; Pesold et al.,1998, 1999; Goffinet et al., 1999; Meyer and Wahle,1999; Perez-Garcıa et al., 2001).

In addition to neuronal somata, some (but not all)fiber tracts also appeared as reelin positive, and incontrols in which the primary antibody was omitted noimmunostaining was observed, both data indicatingthat reelin labeling in fiber tracts was not related tobackground. It is of interest to note that in lampreybrain all tracts are non-myelinated, and due to this,fibers that compose them are large-bore to avoid a lowconductivity, and therefore easily accessible to labelingwith an antibody. One should also account that in theembryonic rat brain some developing unmyelinatedfiber tracts were also reelin immunopositive (see i.e. thecomparison of Fig. 2D–E).

In the adult mammalian brain many fiber tracts aremyelinated and difficult to stain with immunohisto-chemical methods for light microscopy. However, thereare many brain regions containing reelin positive pro-jection neurons that show an intense diffuse labeling inthe areas where these neurons stablish synaptic con-tacts, as it occurs i.e. in the olfactory bulb mitral cellsor in central nervous system dopaminergic neurons(Caruncho et al., 1999; Pesold et al., 2001). Moreover,it has been recently reported an axonal secretion ofreelin by Cajal-Retzius cells into the cortical marginalzone in mice (Derer et al., 2001).

4.2.2. Reelin distribution and possible rolesLampreys show a laminar-type brain (term referring

to the periventricular lamina where most neuronal so-mata remain). Among the lamprey brain areas wherethere is also a number of neurons migrated away fromthe ventricle are the olfactory bulb, pallium, habenulaand optic tectum (see as a review Nieuwenhuys andNicholson, 1998). Interestingly, these are the regionsthat show more reelin immunopositivity in lampreys(with stronger labeling in premetamorphic larvalbrains). This may suggest that reelin may be alreadyinvolved in some regulation of neuronal migration inlamprey brain, although these animals do not showclearly layered brain regions (even though in the adultoptic tectum some plexiform and cellular layers may befound). It should be pointed out that these regions havebeen shown to contain reelin expressing neurons infishes, reptiles, birds and mammals (Impagnatiello etal., 1998; Meyer and Goffinet, 1998; Pesold et al., 1998,1999; Bernier et al., 1999, 2000; Goffinet et al., 1999;Meyer and Wahle, 1999; Perez-Garcıa et al., 2001).Furthermore, lampreys also show reelin immunolabel-ing in the hypothalamus, similarly to what has beenshown in adult murine brains (Smalheiser et al., 2000).

Most reelin positive regions in the lamprey brainshow a diffuse labeling where sometimes is difficult todiscern immunopositive somata from the surroundingextracellular matrix and neuropil. Furthermore, someof these regions, such as the habenula and optic tectum,receive afferents that also are reelin positive and cancontribute to the diffuse appearance of the labeling.However, high magnification micrographs allow inmany cases the identification of neuronal somataclearly immunopositive (see i.e. Fig. 2F, Fig. 3B, Fig.3D, Fig. 4E inset). In addition to this diffuse labelingthere is also a more intense labeling restricted to specificaxons i.e. those of the stria medullaris and optic tracts.This finding suggest that, in addition to its constitutiverelease from neuronal somata and proximal dendrites,reelin can also be transported through axons, as it wasdemonstrated in the central nervous system of mam-mals for dopaminergic neurons, cerebellar granule cells,and Cajal-Retzius cells (Pesold et al., 1998, 2001; Dereret al., 2001). Interestingly, the lamprey tracts im-munopositive for reelin show an intense labeling re-stricted to the fibers but not to the surrounding regions,and a diffuse labeling in the areas where these fibersmake synaptic contacts, indicating that reelin wouldprobably be released from axon terminals.

Up to now reelin has been shown to be involved inthe regulation of neuronal migration towards layeredcortical regions, controlling the radial migration ofpyramidal cells (see as reviews Frotscher, 1997, 1999;Curran and D’Arcangelo, 1998; Lambert de Rouvroitand Goffinet, 1998a,b, 2001), and possibly also thetangential migration of GABAergic neurons from theganglionic eminences (Caruncho et al., unpublished re-sults). In addition, reelin can also be involved in axonpathfinding and synaptogenesis in hippocampus (DelRıo et al., 1997; Borrell et al., 1999), in controllingmigration and dendritic arborization pattern in Purk-inje cells (Miyata et al., 1996, 1997), and in regulatingactivity in dendritic spines in pyramidal cells (Liu et al.,2001). The present report shows that reelin immunore-activity is observed in some significative fiber tracts inspecific larval stages (i.e. optic tract and optic chiasma,which will acquire their functional significance in thepremetamorphic larval stage) suggesting that reelin inlampreys may be already having a role in regulatingaxon pathfinding and synaptogenesis. Interestingly,reelin appeared in the few laminated areas of the lam-prey brain in the moment of organization of theselaminated structures (i.e. in the optic tectum), thismight indicate that a possible role of reelin in regulat-ing neuronal migration should also exist in lampreys.

It is of interest to note that our results confirm thatreelin immunostaining increases in intensity during lar-val brain development in all of the reelin immunoposi-tive areas, reaching its pinacle in premetamorphiclarvae. This finding of highest reelin levels just when the

E. Perez-Costas et al. / Journal of Chemical Neuroanatomy 23 (2002) 211–221220

animal enters a period of intense changes (which is veryevident in regions related to the visual system forexample), is in agreement with the idea that reelin mayplay an important role in neuronal plasticity (Pesold etal., 1998, 1999; Guidotti et al., 2000; Rodrıguez et al.,2000).

In summary, the expression of a reelin like protein inthe lamprey brain may suggest different possible rolesfor this protein in this primitive vertebrate, includingsome of the previously described roles in other species.

The reelin signal transduction pathway appears to bemediated by integrins including the �3-subunit (Antonet al., 1999; Dulabon et al., 2000; Rodrıguez et al.,2000), cadherin-like receptors (Senzaki et al., 1999),and/or by the apolipoprotein E receptor type 2 and thevery low density lipoprotein receptor (D’Arcangelo etal., 1999; Hiesberger et al., 1999; Tromssdorf et al.,1999), as well as by the cytosolic adaptor proteindisabled-1 (Howell et al., 1997, 1999; Sheldon et al.,1997; Rice et al., 1998; Cooper and Howell, 1999). Weare already starting the experiments to study theirpossible expression in lamprey in order to get a betterknowledge of reelin functions in primitive vertebrateswith laminar brains. The reelin expression observed inlamprey brain suggests that molecules similar to theseshould also be found there.

Acknowledgements

The authors wish to thank Dr Andre Goffinet (Uni-versity of Namur, Belgium) for the generous gift of the142 and G10 antibodies. Supported by a grant from theXunta de Galicia (PGIDT99PXI20003B), and a Pre-doctoral fellowship to E. P.-C. (University of Santiagode Compostela).

References

Alcantara, S., Ruiz, M., D’Arcangelo, G., Ezan, F., de Lecea, L.,Curran, T., Sotelo, C., Soriano, E., 1998. Regional and cellularpatterns of reelin mRNA expression in the forebrain of thedeveloping and adult mouse. J. Neurosci. 17, 7779–7799.

Anton, E.S., Kreindberg, J.A., Rakic, P., 1999. Distinc functions ofalpha 3 and alpha (v) integrin receptors in neuronal migration andlaminar organization of the cerebral cortex. Neuron 22, 277–289.

Bar, I., Lambert de Rouvroit, I., Royaux, I., Krizman, D.B., Dernon-court, C., Ruelle, D., Beckers, M.C., Goffinet, A.M., 1995. AYAC contig containing the reeler locus with preliminary charac-terization of candidate gene fragments. Genomics 26, 543–549.

Bernier, B., Bar, I., Pieau, C., Lambert de Rouvroit, C., Goffinet,A.M., 1999. Reelin mRNA expression during embryonic braindevelopment in the turtle Emys orbicularis. J. Comp. Neurol. 413,463–479.

Bernier, B., Bar, I., D’Arcangelo, G., Curran, T., Goffinet, A.M.,2000. Reelin mRNA expression during embryonic brain develop-ment in chick. J. Comp. Neurol. 422, 448–463.

Borrell, V., Del Rıo, J.A., Alcantara, S., Derer, M., Martınez, A.,D’Arcangelo, G., Nakajima, K., Mikoshiba, K., Derer, P., Cur-ran, T., Soriano, E., 1999. Reelin regulates the development andsynaptogenesis of the layer-specific entorhino-hippocampal con-nections. J. Neurosci. 19, 1345–1358.

Caruncho, H.J., Rodrıguez, M.A., Liu, W.S., Manev, H., Guidotti,A., Costa, E., Pesold, C., 1999. Reelin is expressed by bothinterneurons and projection neurons in the adult rat brain. Soc.Neurosci. Abstr. 25, 1293.

Cooper, J.A., Howell, B.W., 1999. Lipoprotein receptors: signalingfunctions in the brain. Cell 97, 671–674.

Curran, T., D’Arcangelo, G., 1998. Role of reelin in the control ofbrain development. Brain Res. Rev. 26, 285–294.

D’Arcangelo, G., Miao, G.G., Chen, S.C., Soares, H.D., Morgan,J.I., Curran, T., 1995. A protein related to extracellular matrixproteins deleted in the mouse mutant reeler. Nature 374, 719–723.

D’Arcangelo, G., Homayouni, R., Keshvara, L., Rice, D.S., Sheldon,M., Curran, T., 1999. Reelin is a ligand for lipoprotein receptors.Neuron 24, 471–479.

De Bergeyck, V., Naerhuyzen, B., Goffinet, A.M., Lambert de Rou-vroit, C., 1998. A panel of monoclonal antibodies against reelin,the extracellular matrix protein defective in reeler mutant mice. J.Neurosci. Meth. 82, 17–24.

Del Rıo, J.A., Heimrich, B., Borrell, V., Forster, E., Drakew, A.,Alcantara, S., Nakajima, K., Miyata, T., Ogawa, M., Mikoshiba,K., Derer, P., Frotscher, M., Soriano, E., 1997. A role forCajal-Retzius cells and reelin in the development of hippocampalconnections. Nature 385, 70–74.

Derer, P., Derer, M., Goffinet, A., 2001. Axonal secretion of reelin byCajal-Retzius cells: evidence from comparison of normal andrelnorl mutant mice. J. Comp. Neurol. 440, 136–143.

Dulabon, L., Olson, E.C., Taglienti, M.G., Eisnehuth, S., McGrath,B., Walsch, C.A., Kreidberg, J.A., Anton, E.S., 2000. Reelinbinds �3�1 integrin and inhibits neuronal migration. Neuron 27,33–44.

Frotscher, M., 1997. Dual role of Cajal-Retzius cells and reelin incortical development. Cell Tissue Res. 290, 315–322.

Frotscher, M., 1999. Cajal-Retzius cells, reelin, and the formation oflayers. Curr. Opin. Neurobiol. 8, 570–575.

Gilmore, E.C., Herrup, K., 2000. Receiving reelin. Curr. Biol. 10,R162–R166.

Goffinet, A.M., Bar, I., Bernier, B., Trujillo, C., Raynaud, A., Meyer,G., 1999. Reelin expression during embryonic brain developmentin lacertidian lizards. J. Comp. Neurol. 414, 533–550.

Guidotti, A., Pesold, C., Costa, E., 2000. New neurochemical mark-ers of psychosis: a working hypothesis of their operation. Neu-rochem. Res. 25, 1207–1218.

Hiesberger, T., Tromsdorf, M., Howell, B.W., Goffinet, A., Mumby,M.C., Cooper, J.A., Herz, J., 1999. Direct binding of reelin toVLDL receptor and ApoE receptor 2 induces tyrosine phosphory-lation of disabled-1 and modulates tau phosphorilation. Neuron24, 481–489.

Hirotsune, S., Takahara, N., Sasaki, N., Hirose, K., Yoshiki, A.,Ohashi, T., Kusakabe, M., Murakami, Y., Muramatsu, M.,Watanabe, S., Nakao, K., Katsuki, M., Hayashizaki, Y., 1995.The reeler gene encodes a protein with and EGF-like motifexpressed by pioneer neurons. Nat. Genet. 10, 77–83.

Howell, B.W., Hawkes, R., Soriano, P., Cooper, J.A., 1997. Neuronalposition in the developing brain regulated by mouse disabled-1.Nature 389, 733–737.

Howell, B.W., Herrick, T.M., Cooper, J.A., 1999. Reelin inducedtyrosine phosphorilation of disabled-1 during neuronal position-ing. Genes Dev. 13, 643–648.

Ikeda, Y., Terashima, T., 1997. Expression of reelin, the gene respon-sible for the reeler mutation, in embryonic development andadulthood in the mouse. Dev. Dyn. 210, 157–172.

E. Perez-Costas et al. / Journal of Chemical Neuroanatomy 23 (2002) 211–221 221

Impagnatiello, F., Guidotti, A.R., Pesold, C., Dwivedi, Y., Carun-cho, H., Pisu, M.G., Uzunov, D.P., Smalheiser, N.R., Davis,J.M., Pandey, G.N., Pappas, G.D., Tueting, P., Sharma, R.P.,Costa, E., 1998. A decrease of reelin expression as a putativevulnerability factor in schizophrenia. Proc. Natl Acad. Sci. USA95, 15718–15723.

Lacor, P., Grayson, D.R., Auta, J., Sugaya, I., Costa, E., Guidotti,A., 2000. Reelin secretion from glutamatergic neurons in culture isindependent of neurotransmitter regulation. Proc. Natl Acad. Sci.USA 97, 3556–3561.

Lambert de Rouvroit, C., de Bergeyck, V., Cortvrindt, C., Bar, I.,Eeckhout, Y., Goffinet, A.M., 1999a. Reelin, the extracellularmatrix protein defficient in reeler mutant mice, is processed by ametalloproteinase. Exp. Neurol. 156, 214–217.

Lambert de Rouvroit, C., Bernier, B., Royaux, I., de Begeyck, V.,Goffinet, A., 1999b. Evolutionary conserved, alternative splicingof reelin during brain development. Exp. Neurol. 156, 229–238.

Lambert de Rouvroit, C., Goffinet, A.M., 1998a. The reeler mouse asa model of brain development. Adv. Anat. Embryol. 150, 1–106.

Lambert de Rouvroit, C., Goffinet, A.M., 1998b. A new view of earlycortical development. Biochem. Pharmacol. 56, 1402–1409.

Lambert de Rouvroit, C., Goffinet, A.M., 2001. Neuronal migration.Mech. Dev. 105, 47–56.

Liu, W.S., Pesold, C., Rodrıguez, M.A., Carboni, G., Auta, J., Lacor,P., Larson, J., Condie, B.G., Guidotti, A., Costa, E., 2001.Down-regulation of dendritic spine and glutamic acid decarboxy-lase 67 expressions in the reelin haploinsufficient heterozygousreeler mouse. Proc. Natl Acad. Sci. USA 98, 3477–3482.

Meyer, G., Goffinet, A.M., 1998. Prenatal development of reelin-im-munoreactive neurons in the human neocortex. J. Comp. Neurol.397, 29–40.

Meyer, G., Wahle, P., 1999. The paleocortical ventricle is the originof reelin-expressing neurons in the marginal zone of the foetalhuman neocortex. Eur. J. Neurosci. 11, 3937–3944.

Miyata, T., Nakajima, K., Aruga, J., Takahashi, S., Ikenaka, K.,Mikoshiba, K., Ogawa, M., 1996. Distribution of a reeler gene-re-lated antigen in the developing cerebellum: an immunohistochem-ical study with an allogenic antibody CR-50 on normal and reelermice. J. Comp. Neurol. 372, 215–228.

Miyata, T., Nakajima, K., Mikoshiba, K., Ogawa, M., 1997. Regula-tion of Purkinje cell alignment by reelin as revealed with CR-50antibody. J. Neurosci. 17, 3599–3609.

Nieuwenhuys, R., Nicholson, C., 1998. Lampreys, Petromyzontoidea.In: Nieuwenhuys, R., Ten Donkelaar, H.J., Nicholson, C. (Eds.),The Central Nervous System of Vertebrates, vol. 1. Springer,Berlin, pp. 397–495.

Ogawa, M., Miyata, T., Nakajima, K., Yagyu, K., Seike, M., Ike-naka, K., Yamamoto, H., Mikoshiba, K., 1995. The reeler gene-associated antigen on Cajal-Retzius neurons is a crucial moleculefor laminar organization of cortical neurons. Neuron 14, 899–912.

Perez-Garcıa, C.G., Gonzalez-Delgado, F.J., Suarez-Sola, M.L.,Castro-Fuentes, R., Martın-Trujillo, J.M., Ferres-Torres, R.,Meyer, G., 2001. Reelin-immunoreactive neurons in the adultvertebrate pallium. J. Chem. Neuroanat. 21, 41–51.

Pesold, C., Impagnatiello, F., Pisu, M.G., Uzunov, D.P., Costa, E.,Guidotti, A., Caruncho, H.J., 1998. Reelin is preferentially ex-pressed in neurons synthesizing �-aminobutyric acid in cortex andhippocampus of adult rats. Proc. Natl Acad. Sci. USA 95, 3221–3226.

Pesold, C., Liu, W.S., Guidotti, A., Costa, E., Caruncho, H.J., 1999.Cortical bitufted, horizontal, and Martinotti cells preferentiallyexpress and secrete reelin into perineuronal nets, nonsynapticallymodulating gene expression. Proc. Natl Acad. Sci. USA 96,3217–3222.

Pesold, C., Rodrıguez, M.A., Liu, W.S., Caruncho, H.J., Costa, E.,Guidotti, A., 2001. Reelin is expressed in dopaminergic systems ofadult primate brain. Soc. Neurosci. Abstr. 27, 1426.

Pombal, M.A., Puelles, L., 1999. Prosomeric map of the lampreyforebrain based on calretinin immunocytochemistry, Nissl stain,and ancillary markers. J. Comp. Neurol. 414, 391–422.

Quattrocchi, C.C., Wannenes, F., Persico, A.M., Ciafre, S.A., D’Ar-cangelo, G., Farace, M.G., Keller, F., 2001. Reelin is a serinprotease of the extracellular matrix. J. Biol. Chem. 277, 303–309.

Rakic, P., Caviness, V.S. Jr., 1995. Cortical development: view fromneurological mutants two decades later. Neuron 14, 1101–1104.

Rice, D.S., Curran, T., 1999a. Mutant mice with scrambled brains:understanding the signaling pathways that control cell positioningin the CNS. Genes Dev. 13, 2758–2773.

Rice, D.S., Curran, T., 1999b. Role of reelin and disabled-1 in thedevelopment of the visual system. Soc. Neurosci. Abstr. 25, 2263.

Rice, D.S., Sheldon, M., D’Arcangelo, G., Nakajima, K., Goldowitz,D., Curran, T., 1998. Disabled-1 acts downstream of Reelin as asignaling pathway that controls laminar organization of the mam-malian brain. Development 125, 3719–3729.

Rodrıguez, M.A., Pesold, C., Liu, W.S., Kriho, V., Guidotti, A.,Pappas, G.D., Costa, E., 2000. Colocalization of integrin recep-tors and reelin in dendritic spine postsynaptic densities of adultnonhuman primate cortex. Proc. Natl Acad. Sci. USA 97, 3550–3555.

Schiffmann, S.N., Bernier, B., Goffinet, A.M., 1997. Reelin mRNAexpression during mouse brain development. Eur. J. Neurosci. 9,1055–1071.

Senzaki, K., Ogawa, M., Yagi, T., 1999. Proteins of the CNR familyare multiple receptors of reelin. Cell 99, 635–647.

Sheldon, M., Rice, D.S., D’Arcangelo, G., Yoneshima, H., Naka-jima, K., Mikoshiba, K., Howell, B., Cooper, J.A., Goldowitz,D., Curran, T., 1997. Scrambler and yotari disrupt the disabledgene and produce a reeler-like phenotype in mice. Nature 389,730–733.

Smalheiser, N.R., Costa, E., Guidotti, A., Impagnatiello, F., Auta, J.,Lacor, P., Kriho, V., Pappas, G.D., 2000. Expression of reelin inadult mammalian blood, liver, pituitary pars intermedia, andadrenal cromaffin cells. Proc. Natl Acad. Sci. USA 97, 1281–1286.

Tromssdorf, F., Gotthardt, M., Hiesberger, T., Shelton, J.,Stockinger, W., Nimpf, J., Hammer, R.E., Richardson, J.A.,Herz, J., 1999. Reeler/disabled-like disruption of neuronal migra-tion in knockout mice lacking the VLDL receptor and ApoEreceptor 2. Cell 97, 689–701.