- Email: [email protected]
THE PROTEOGLYCAN DSD-1-PG OCCURS IN PERINEURONAL NETS AROUND PARVALBUMIN-IMMUNOREACTIVE INTERNEURONS OF THE RAT CEREBRAL CORTEX
CDPergamon
Int.J. DevlNeuroscience, Vol. 14,No. 3, pp. 249–255,1996 CopyrightG 1996ISDN Publishedby ElsevierScienceLtd. Printed in Great Britain PII: S0736-5748(96)00011-1 073$5748/96 $15.00+0.00
THE PROTEOGLYCAN DSD-1-PG OCCURS IN PERINEURONAL NETS AROUND PARVALBUMIN-IMMUNOREACTIVE INTERNEURONSOF THE RAT CEREBRAL CORTEX EVA SABINE WINTERGERST,*7 ANDREAS FAISSNERJ and MARCO R. CELI07 TInstitute of Histology and General Embryology, University of Fribourg, CH-1705 Fribourg, Switzerland; $Department of Neurobiology, University of Heidelberg, D-69120 Heidelberg, Germany (Received 10 August 1995;revised9 November 1995;accepted 30 November 1995) Abstract—Proteoglycans involved in the shaping of the developing brain are often preserved in the adult brain in more restricted locations. We have studied the fate of DSD-1-PG, a chondroitin sulfate proteoglycan containing the hybrid epitope DSD-1. DSD-1-PG exerts neurite outgrowth promoting activity and has been shown to occur in the developing brain during late brain development and into adulthood. In the adult rat brain, monoclinal and polyclonal antibodies against DSD-1-PG labelled only the circumference of a selected subpopulation of neurons. These nerve cells invariably expressed the calciumbinding protein parvalbumin. The label occupied the extracellular space in close vicinity to the cell body, surrounding axon terminals and glial end feet, but was absent from synaptic clefts. DSD-1-PG is thus shown to be an additional representative of the growing list of substances found in perineuronal locations in the adult mammalian brain. Copyright ~ 1996 ISDN. Published by Elsevier Scienee Ltd. Key words: proteoglycan, DSD-1-PG, perineuronal net, parvalburnin, interneurons, rat, cerebral cortex.
An increasing number of observations indicates that hyaluronan, some proteoglycans as well as glycoproteins,expressedduring the developmentof the nervous system,persist in the adult brain. Examples of this phenomenon are the occurrence of the glycoproteinstenascin-C14and tenascinearly during development and in the fully developed brain. R 16as well as proteoglycans’>638> 17s2’s22’25
Whereas these substances occur widespread in the brain during ontogeny, their location becomes more restricted during postnatal life. From infiltratingthe intersticesbetween cellsand processesin the wholeneuraxis,hyaluronan, proteoglycansand glycoproteinsare later enrichedin aperineuronal location. It is a surprising,and as yet unexplained,observation that they preferentiallyaccumulate These nerve cellsbelong mainly around a small subset of neurons in the adult nervous system.3’4’5’3s to the interneuronal subpopulation39 and often express the calcium-binding protein Parvalbumin T2A.J3,W5,J6 The aim of this study was to test the distribution of DSD-1-PG in the adult brain. DSD-1-PG is predominantly a chondroitin sulfateproteoglycan.It has been definedwith the help of MAB 473HD which reacts with the epitope DSD-1.M In the developing brain DSD-1-pG found in discrete patterns, e.g. in barrel field boundaries.” Immunofluorescencedouble-labelling studies performed on mouse cerebella cultures have established that DSD-1-PG is detectable on the surface of immature astrocytes and oligodendrocytesand becomes down-regulated with maturation. DSD-1-PG was shown to exert neurite outgrowth promoting activity towards embryonic mesencephalic and hippocampal neurons.19 we found that in adults DSD-l-PG-immunoreactivityis restricted to the perimeter of the perikarya of parvalbumin-positiveinterneurons.
EXPERIMENTAL PROCEDURES Animals
ZUR-Siv rats were used for all experiments. All animals were bred at the Institut fur Labortierkunde at the University of Ztirich. *To whom correspondence should be addressed. 249
250
E. S. Wintergerst
et al.
Antibodies
Polyclonaland monoclinal antibodies to the DSD-1-proteoglycan (PDSD-1-PG; rat IgM MAb 473HD), and monoclinal antibodies recognizingthe calcium-bindingprotein parvalbumin (mouse MAb PV235)were used. Production, characterization and specificityof pDSD-1-PG, MAb473HD and parvalbumin antibodies have been described previously.15>18 In brief, polyclonal anti-DSD-l PG antibodies react with the central nervous systemchondroitin sulfate proteoglycan DSD-1-PG, but do not react with other glycoproteins of the extracellular matrix, i.e. tenascin, laminin or fibronectinin enzyme-linkedimmunosorbentassay(ELISA). The PDSD-1-PG showa higheraffinity towards the core glycoproteinof DSD-1-PG as compared to the intact molecule.MAb473HD reacts with the chondroitin sulfate/dermatan sulfate hybrid epitope DSD-1 expressed on DSD-1-PG. Primary antibodies were visualizedby fluoresceinisothiocyanate-conjugatedgoat anti-rabbit IgG antibodies (Molecular Probes), or after incubation with biotinylated anti-mouse antibodies (Vector Laboratories, Burlingham,CA, U.S.A.) with Avidin-TexasRed (Molecular Probes). Tissue preparation
For immunolight-and immunoelectronmicroscopyrats were asphyxiatedwith COZand perfused through the left ventriclewith 0.1 M phosphate buffer, pH 7.4, containing 4!4. paraformaldehyde (w/v) for 15 min. The brains were removed and 40-pm-thick frontal sections were prepared for conventional immunohistochemistrywith a vibratome. For indirect double immunofluorescence staining, rat brains were rinsed in phosphate buffer and cryoprotected in 18°i0sucrose (w/v) in phosphate buffer. Cryostat sections, 10-#m-thick,were thaw-mounted on to slides coated with gelatinechrome alum. Immunohistochemistry
The sectionswereincubated free-floatingon a shaker at 4°Cfor 72hr, usingthe primary antibodies pDSD-1-PG or MAb473HD diluted in Tris-bufferedsaline(TBS), pH 7.3, containing 10?4obovine serum. The sections were then rinsed six times for 5 min in TBS, and further processed with the avidin–biotin complex using the vectastain elite kit (Vector Laboratories) according to the manufacturer’s recommendations. The antibody-enzyme complex was visualizedusing a solution containing 0.020/.3,3’-diaminobenzidineand 0.006°i0hydrogen peroxide in 0.05M TBS, pH 7.3. Sectionswere either mounted on gelatinechrome alum-coated slides,dehydrated in a graded series of ethanol, cleared in XY101, and coverslippedin Eukitt (Gribi AG, Belp, Switzerland),or processed for ultrastructural analysis.Therefore, sectionswere postfixedinitiallyin 2.5% (v/v)glutaraldehyde and then in 1YO(w/v) osmiumtetroxide(both in phosphate buffer) at 4“C, each for a period of 45 min. The tissue was stained with 10/0(w/v) uranyl acetate in 70°/0(v/v) ethanol, dehydrated in a graded seriesof increasing ethanol concentrations and embedded in Epon 812. Ultrathin sections were cut, mounted on copper grids and examined in a Zeiss EM 10. Double immunoj?uorescence staining
Cryosections were incubated with TBS containing IYo bovine serum albumin (BSA), for 30 min at room temperature. Slideswere then incubated concomitantly with a monoclinal antibody recognizing parvalbumin15and an antiserum directed against the DSD-1-proteoglycan.18The location of the DSD-1-PG was revealed by incubating sections for 3 hr with fluorescein isothiocyanate-conjugatedgoat anti-rabbit IgG diluted in phosphate buffer; parvalbumin antibodies were labelledfor 2 hr with a secondarybiotinylatedanti-mouseantibody diluted in phosphate buffer and then exposed to Avidin-TexasRed for 3 hr. RESULTS Light microscopic localization of DSD-I-PG in the neocortex of the developing and adult rat
To analysethe spatiotemporal expressionpattern of the DSD-1-PG in the rat cerebral cortex, we investigatedby immunohistochemistrya developmentalseriesof frontal sectionsthrough the cerebral cortex. Polyclonal as well as monoclinal antibodies directed against the DSD-1-PG appeared to stain
Perineuronalnets
251
perineuronal nets localizedaround nonpyramidal neurons in the rat neocortex from the late second week of postnatal development(Fig. 1)onwards into adulthood (i.e. older than 6 weeks;Figs 1 and 2). The immunostained neurons were differentin sizeand shape. The distribution of DSD-1-PG in the cerebral cortex varied considerably throughout the depth of the cortex (not shown). Neurons enwrapped by DSD-l-PG-positive perineuronal nets were mainly located in the cortical layers II and IV, but were also present in the other layers, with the exception of layer I. Clearly outlined perineuronal nets were the predominant feature of DSD-1-PG staining. However, there was considerable variability in staining intensity and penetration of the immunoreactivematerial into the perineuronal space.At a highermagnificationthe stainingwasclearlyreticulated(Fig. lbf,). Double lmmunofluorescencestaining revealed that DSD-I -PG-immunoreactivity preferentially accumulated around neurons expressingparvalbumin at any stage tested (Fig. 3). The fully developednets
Fig. 1. Localization of DSD-1-PG in frontal vibratome sections of the adult rat neocortex as achieved by the application of PDSD-1-PG. Deep layer III and layer IV contain numerous nonpyramidal neurons ensheathed by perineuronal nets (a). The appearance of the perineuronal nets is variable with regard to their staining intensity and the penetration of the immunoreactive material into the surrounding neuropil (b-f). Weakly stained and rather confined neuropil occurs as well as an intensely stained perineuronal environment showing a lattice-like pattern of immunoreactivity in regions far from the neuronal cell body (compareb, c, anddwitheand O. Note that dendritic stems areensheathedby sharply continedperineuronal nets (a–f). In (e) a tangentially cut cell surface shows the interlacing net of DSD-1-PG-positive material. Bar in (f)= 75 pm for (a); 15pm for (b}(f).
252
E. S.Wintergerstet al.
Fig. 2. Irnmunoelectron microscopic localization of DSD-1-PG in frontal sections of the neocortex of an adult rat using pDSD-1-PG, showing a perineuronal distribution of Iabelled extracellular material (a). The perikaryon of a neuron with a lobular nucleus (neuron A) is surrounded by a neuropil which is interspersed with immunoreactive extracellular material. Note that the neighbourhood of the neuron with a round nucleus (neuron B) remained unstained (a). The immunopositive signal is confined to a limited region of the neuropil around the cell body as seen at higher magnification (b, c), thereby surrounding glial structures (arrowhead in c) but excluding the synaptic cleft (arrow in b). Bar in (c)= 3.5 pm for (a); 1.5pm for (b) and (c).
appeared for the first time at postnatal day 19(Fig. 3a4) and persistedinto adulthood (Fig. 1, Fig. 3e and f). Electron microscopic localization of DSD-I-PG in the neocortex of the adult rat
In regions containing intensively stained perineuronal nets as evidenced in light microscopic investigations,nets could also be identifiedusing the electron microscope. At the ultrastructural levelDSD-1-PG immunoreactivitywas found to infiltrate the extracellular space to a distance of a few micrometers from the cell surface of some neurons (Fig. 2). It surrounded neuronal and glial
Perineuromdnets
253
Fig. 3, Double immunofluorescence staining for DSD-1-PG in perineuronal nets (b, d, f) applying pDSD1-PG, and parvalbumin applying MAb 235 (a, c, e) of frontal cryosections, of the neocortex of 3-week-old (a+) and adult (e and f) rats. Numerous parvalbumin-immunoreactive neurons localized in deep layer III and layer IV (a, c, e) are enwrapped by perineuronal nets as revealed by DSD-1-PG immunoreactivity (b, d, f; marked with arrows in b). Allocation of parvalburnin-positive neurons (c, e) with perineuronal nets (d, f) at a higher magnification in the cortex of a 3-week-old rat (c, d) and in the cortex of an adult rat (e, f). Bar in (f)=80 pm for (a) and (b); 20 pm for (c~f).
structures, such as synaptic terminals and glial processes(Fig. 2c), but was absent from synaptic clefts (Fig. 2b). The labelling was confined to the extracellular space but was associated with membranes. Neurons showinga pericellularzone of labelledextracellularmatrix were often in the vicinityof neurons with a non-labelledextracellularspace (Fig. 2; compare neurons A and B). DISCUSSION Applying monospecificpolyclonal and monoclinal antibodies against DSD-1-PG to the adult rat brain, we found an exclusivelocation of this hybrid chondroitin sulfate/derrnatane sulfate proteoglycan around parvalbumin-immunoreactiveneurons. The antibodies against DSD-1-PG havebeen characterizedextensivelyin previouspublicationsand satisfyvariouscriteria of specificity. In particular, they do not cross-react with versican, which has been shown by others to occur in perineuronal Focalization.b
254
E. S. Wintergerstet al.
At the light microscopiclevelthe staining with antibodies against DSD-1-PG did not differfrom that seen with a variety of lectins24’38’@ and antibodies directed against knownlGand unknown antigens.26,28 . A1lthese tools la~l the perikaryal surface and the proximal dendrites of the same subpopulation of interneurons in the brain in a reticulated manner. This labellingpattern has been called the “perineuronal net of extracellular matrix” (see Celio and Bli,imcke13 for review). As pre-embedding staining with diaminobenzidine/Hz02as a substrate was used, the ultrastructural localization of the DSD-1-PG antigen cannot be accurate.23Nevertheless, the primary site of occurrenceis undoubtedly the extracellular space. The validity of the association of label with the outer leaflet of membranes of glial processes32and nerve terminals will await confirmation by immunogold techniques. As for the other componentsof perineuronalnets, the preferentialassociation of DSD-1-PG with the surface of parvalbumin-immunoreactiveneurons in the adult animal is of unknown functional significance.Parvalbumin neurons have indented nuclei, as do most interneurons, belong to the and a special membrane cYtoGABA-subpopulation of interneurons,10have a highmetabolism37 skeleton,]zare rich in certain isoformsof K+-channels42and are characterized neurophysiologically The enwrapping of their cell bodies and proximal dendrites with the by a high firing rate. Wg,so,ql perineuronal net of extracellular matrix may relate to their peculiar chemical constitution and electrophysiologicalcharacteristics.Procedures that could reproduce and disrupt perineuronal nets in organotypic cultures of brain slicesmay shed light on their roles in situ. Heparan sulfates and chondroitin sulfate proteoglycans have already been recognized in perineuronal nets by using various monoclinal antibodies (seeTable I in Ref. 13for review).With the addition of DSD-1-PG a dermatan sulfate~chondroitinsulfate hybrid, as a new component, perineuronal nets of extracellular matrix are being shown to be even more complex structures, composed of a wide variety of different molecules. Some of the molecules are known to have repulsiveand/or permissiveeffectson the growth of axons during development(for reviewsee Ref. 19),but their roles in the adult brain remain unknown. The DSD-1-epitope has been shown to be involved in neurite outgrowth promotion and could serve a similar function in the perineuronal nets. Other components may concentrate growth factors around neurons20and still others are thought to stabilizesynapseson the surface of neurons.2’27 Thus, perineuronal nets of extracellular matrix are composed of a mesh of different molecules that creates a very peculiar milieu around parvalbumin neurons which could play a role in shaping their functional characteristics. Acknowledgements—The authors wish to thank C. Dumas, D. Uldry and C. Mandl for excellent technical assistance. This work was supported by the Swiss National Foundation, grant 31.0364.93 and Deutsche Forschungsgemeinschaft, grant DFG, Fa 159/5-1,2.
REFERENCES Bertolotto A.,RoccaG.,Canavese G.,Migheli A.andSchiffer D. (1991) Chondroitin-sulfate proteoglycan surrounds a subset of human and rat CNS neurons. J. Neurosci. Res. 29, 225–234. 2. Bertolotto A., Rocca G. and Schiffer D. (1990) Chondroitin 4-sulfate proteoglycan forms an extracellular network in human and rat central nervous system. J. Neurol. Sci. 100,113-123. 3. Bignami A., Asher R. and Perides G. (1992) Co-localization of hyaluronic acid and the chondroitin sulfate proteoglycan in rat cerebral cortex. Brain Res. 579, 173–177. 4. Bignami A., Asher R., Perides G. and Rahemtulla F?.(1992) The extracellular matrix of cerebral gray matter: Golgi’s pericellular nets and Nissl’s “nervoses Grau” revisited. Znt.J. deul Neurosci. 10,291-299. 5. Bignami A. and Dahl D. (1988) Expression of brain-specific hyaluronectin (BHN), a hyaluronate-binding protein, in dog postnatal development. Expl Neurol. 99, 107–117. 6, Bignami A., Perides G. and Rahemtulla F. (1993) Versican, a hyahrronate-binding proteoglycan of embryonal precartilaginous mesenchyma, is mainly expressed postnatally in rat brain. J. Neurosci. Res. 34, 97–106. 7. Brtickner G., Brauer K., Hartig W., Wolff J. R., Rickmann M. J., Derouiche A., Delpech B., Girard N., Oertel W. H. and Reichenbach A. (1993) Perineuronal nets provide a polyanionic, glia-associated form of microenvironment around certain neurons in many parts of the rat brain. Glia 8, 183–200. 8. Brttckner G., Delpech B., Delpech A. and Girard N. (1990) Concentration of hyalunorectin and anionic glycoconjugates in perineuronal glial cell processes at GABA-ergic synapses of rat cerebellum. Acta Histochem. Suppl. 38, 161–165. 9. Celio M. R. (1984) Parvalbumin as a marker of fast-tiring neurons. Neurosci. Lett. Suppl. 18,332. 10,Celio M. R. (1986) Parvalbumin in most garnma-aminobutyric acid-containing neurons of the rat cerebral cortex. Science 231,995-997. 11, Celio M. R. (1990) Calbindin D-28k and parvalbumin in the rat brain. Neuroscience 35,375475. 1.
Perineuronal
nets
255
12. Celio M. R. (1993) Perineuronal nets of extracellular matrix around parvalbumin-containing neurons of the hippocampus. Hippocampus 3, 55+1. 13. Celio M. R. and Blitmcke I. (1994) Perineuronal nets. A s~ecialized form of extracellular matrix in the adult nervous system. Brain Res. Rev. 19, 128–145. 14 Celio M. R., Chiquet-Ehrismann R. (1993) “Perineuronal nets” around cortical interneurons expressing parvalbumin are rich in tenascin. Neurosci. Lett. 162, 137–140. 15 Celio M. R., Baier W., Scharer L,, De Viragh P. and Gerday Ch. (1988) Monoclinal antibodies directed against the calcium binding protein parvalbumin. Cell Calcium 9, 81–86. 16 Celio M. R. and Rathjen F. (1993) Restrictingoccurs in “perineuronal nets” of the adult rat brain. Soc, Neurosci. Abstr. 19,689. 17 Delpech B., Bertrand P,, Hermelin B,, Delpech A., Girard N., Halkin E. and Chauzy C, (1986) Hyaluronectin. In F’ronfiersin Matrix Biology, Vol. 11 (eds Labat-Robert J., Timpl R. and Robert L,), pp. 78–89, Karger, Basel, 18 Faissner A., Clement A., Lochter A., Streit A., Mandl C. and Schachner M. (1994) Isolation of a neural chondroitin sulfate proteoglycan with neurite outgrowth promoting properties. J. Cell Bio/. 126, 783–799. 19, Faissner A., Scholze A. and Gotz B. (1994) Tenascin glycoproteins in developing neural tissue: only decoration? Devl Neurobiol. 2, 53–66. 20 Flaumenhaft R. and Rifkin D. B. (1991) Extracelhrlar matrix regulation of growth factor and protease activity. Curr. Opin. Cell Biol, 3,817-823. 21. Fujita S, C., Tada Y., Murakami F., Hayashi M. and Matsumura M. (1989) Glycosaminoglycan-related epitopes surrounding different subsets of mammalian central neurons. Neurosci. Res. 7, 117–130. 22 Girard N., Courel M. N,, Delpech A., Bruckner G. and Delpech B. (1992) Staining ofhyahrronan in rat cerebellum with a hyahrronectin-antihyaluronectin immune complex. Histochem. J. 24, 21–24. 23. Goldfischer S. (1981) Effusion artifact: a chronic problem in DAB-cytochemistry, J. Histochem. Cytochem. 98, 1360, 24. Hartig W., Brauer K. and Bruckner G. (1992) Wisteria jioribunda agglutinin-labelled nets surround parvalbumincontaining neurons. NeuroReport 3, 869–872. 25. Herndon M. E. and Lander A. (1990) A diverse set of developmentally regulated proteoglycans is expressed in the rat central nervous system. Neuron 4, 949–961. 26. Hinton D. R., Henderson V. W., Blanks J. C., Rudnicka M. and Miller C. A. (1988) Monoclinal antibodies react with neuronal subpopulations in the human nervous system. J. conzp.Neurol. 267, 398%408. 27. Hockfield S., Kalb R. G., Zaremba S. and Fryer H. (1990) Expression of neural proteoglycans correlates with the acquisition of mature neuronal properties in the mammalian brain. Cold Spring Harb. Symp. quant. Biol. 55, 505–514. 28. Hockfield S. and McKay R. D.G. (1983) A surface antigen expressed by a subset of neurons in the vertebrate central nervous system. Proc. natn. Acad. Sci. U.S.A. 80, 5758–5761. 29. Kawaguchi Y, (1993) Physiological, morphological and histochemical characterization of three classes of interneurons in rat neostriatum. J. Neurosci, 13, 4908J1923. 30, Kawaguchi Y., Katsumaru H., Kosaka T., Heizmann C, W. and Hama K. (1987) Fast spiking cells in rat hippocampus (CA1 region) contain the calcium-binding protein parvalbumin. Brain Res. 416,369-374. 31. Kawaguchi Y. and Kubota Y. (1993) Correlation of physiological subgroupings ofnonpyramidal cells with parvalbuminimmunoreactive and calbindin D-28k-immunoreactive neurons in layer V of rat frontal cortex. .J. Neurophysiol. 70, 387– 396. 32,Kosaka T. and Hama K. (1986) Three-dimensional structure of astrocytes in the rat dentate gyrus. J. comp. Neurol. 249, 242-260. 33.Kosaka T. and Heizmann C. W. (1989) Selective staining of a population of parvalbmnin-containing GABAergic neurons in the rat cerebral cortex by lectins with specific affinity for terminal N-acetylgalactosamine. Brain Res. 483, 158-163. 34.Kosaka T., Heizmann C. W. and Barnstable C. J. (1989) Monoclinal antibody VCI.1 selectively stains a population of GABAergic neurons containing the calcium-binding protein parvalbumin in the rat cerebral cortex. E.xplBrain Res. 78, 43-50, 35. Kosaka T., Heizmann C. W. and Fujita S. C. (1992) Monoclinal antibody 473 selectively stains a population of GABAergic neurons containing the calcium-binding protein parvalbumin in the rat cerebral cortex. Expl Brain Res. 89, 109– 114. 36. Kosaka T., Isogai K., Barnstable C. J. and Heizmann C. W. (1990) Monoclinal antibody HNK-1 selectively stains a subpopulation of GABAergic neurons containing the calcium-binding protein parvalbumin in the rat cerebral cortex. Expl Brain Res. 82,566-574. 37.McCasland J. S., Hibbard L. S. and Woolsey T. A. (1993) Calcium-binding protein phenotype and function of GABAergic neurons in barrel cortex. Soc. Neurosci. Abstr. (Vol. 2), 19, 1567. 38.Nakagawa F., Schulte B. A. and Spicer S. S. (1986) Selective cytochemical demonstration of glycoconjugate-containing terminal N-acetylgalactosamine on some brain neurons. J. camp. Neurol. 243, 28&290. 39.Nagakawa F., Schulte B. A., Wu J. Y. and Spicer S. S. (1986) GABA-ergic neurons of the rodent brain correspond partially with those staining for glycoconjugate with N-acetylgalactosamine. J. Neurocytol. 15,389-396. 40. Nagakawa F., Schulte B. A., Wu J. Y. and Spicer S. S. (1987) Postnatal appearance of glycoconjugate with terminal Nacetylgalactosamine on the surface of selected neurons in mouse brain. Deul Neurosci. 9, 53–60. 41. Steindler D. A., Settles D., Laywell E. D., Yooshiki A., Faissner A., Erickson H. P. and Kusakabe M. (1995) Tenascinknockout mice: barrels, boundary molecules, and glial scars. J. Neurosci. 15, 1971-1983. 42, Weiser M., Vega-Saenz de Micra E., Kentros C., Moreno H., Franzen L., Hillman D., Baker H. and Rudy B. (1994) Differential expression of shaw-related K+ channels in the rat central nervous system. J. Neurosci. 14, 949–972.
Int.J. DevlNeuroscience, Vol. 14,No. 3, pp. 249–255,1996 CopyrightG 1996ISDN Publishedby ElsevierScienceLtd. Printed in Great Britain PII: S0736-5748(96)00011-1 073$5748/96 $15.00+0.00
THE PROTEOGLYCAN DSD-1-PG OCCURS IN PERINEURONAL NETS AROUND PARVALBUMIN-IMMUNOREACTIVE INTERNEURONSOF THE RAT CEREBRAL CORTEX EVA SABINE WINTERGERST,*7 ANDREAS FAISSNERJ and MARCO R. CELI07 TInstitute of Histology and General Embryology, University of Fribourg, CH-1705 Fribourg, Switzerland; $Department of Neurobiology, University of Heidelberg, D-69120 Heidelberg, Germany (Received 10 August 1995;revised9 November 1995;accepted 30 November 1995) Abstract—Proteoglycans involved in the shaping of the developing brain are often preserved in the adult brain in more restricted locations. We have studied the fate of DSD-1-PG, a chondroitin sulfate proteoglycan containing the hybrid epitope DSD-1. DSD-1-PG exerts neurite outgrowth promoting activity and has been shown to occur in the developing brain during late brain development and into adulthood. In the adult rat brain, monoclinal and polyclonal antibodies against DSD-1-PG labelled only the circumference of a selected subpopulation of neurons. These nerve cells invariably expressed the calciumbinding protein parvalbumin. The label occupied the extracellular space in close vicinity to the cell body, surrounding axon terminals and glial end feet, but was absent from synaptic clefts. DSD-1-PG is thus shown to be an additional representative of the growing list of substances found in perineuronal locations in the adult mammalian brain. Copyright ~ 1996 ISDN. Published by Elsevier Scienee Ltd. Key words: proteoglycan, DSD-1-PG, perineuronal net, parvalburnin, interneurons, rat, cerebral cortex.
An increasing number of observations indicates that hyaluronan, some proteoglycans as well as glycoproteins,expressedduring the developmentof the nervous system,persist in the adult brain. Examples of this phenomenon are the occurrence of the glycoproteinstenascin-C14and tenascinearly during development and in the fully developed brain. R 16as well as proteoglycans’>638> 17s2’s22’25
Whereas these substances occur widespread in the brain during ontogeny, their location becomes more restricted during postnatal life. From infiltratingthe intersticesbetween cellsand processesin the wholeneuraxis,hyaluronan, proteoglycansand glycoproteinsare later enrichedin aperineuronal location. It is a surprising,and as yet unexplained,observation that they preferentiallyaccumulate These nerve cellsbelong mainly around a small subset of neurons in the adult nervous system.3’4’5’3s to the interneuronal subpopulation39 and often express the calcium-binding protein Parvalbumin T2A.J3,W5,J6 The aim of this study was to test the distribution of DSD-1-PG in the adult brain. DSD-1-PG is predominantly a chondroitin sulfateproteoglycan.It has been definedwith the help of MAB 473HD which reacts with the epitope DSD-1.M In the developing brain DSD-1-pG found in discrete patterns, e.g. in barrel field boundaries.” Immunofluorescencedouble-labelling studies performed on mouse cerebella cultures have established that DSD-1-PG is detectable on the surface of immature astrocytes and oligodendrocytesand becomes down-regulated with maturation. DSD-1-PG was shown to exert neurite outgrowth promoting activity towards embryonic mesencephalic and hippocampal neurons.19 we found that in adults DSD-l-PG-immunoreactivityis restricted to the perimeter of the perikarya of parvalbumin-positiveinterneurons.
EXPERIMENTAL PROCEDURES Animals
ZUR-Siv rats were used for all experiments. All animals were bred at the Institut fur Labortierkunde at the University of Ztirich. *To whom correspondence should be addressed. 249
250
E. S. Wintergerst
et al.
Antibodies
Polyclonaland monoclinal antibodies to the DSD-1-proteoglycan (PDSD-1-PG; rat IgM MAb 473HD), and monoclinal antibodies recognizingthe calcium-bindingprotein parvalbumin (mouse MAb PV235)were used. Production, characterization and specificityof pDSD-1-PG, MAb473HD and parvalbumin antibodies have been described previously.15>18 In brief, polyclonal anti-DSD-l PG antibodies react with the central nervous systemchondroitin sulfate proteoglycan DSD-1-PG, but do not react with other glycoproteins of the extracellular matrix, i.e. tenascin, laminin or fibronectinin enzyme-linkedimmunosorbentassay(ELISA). The PDSD-1-PG showa higheraffinity towards the core glycoproteinof DSD-1-PG as compared to the intact molecule.MAb473HD reacts with the chondroitin sulfate/dermatan sulfate hybrid epitope DSD-1 expressed on DSD-1-PG. Primary antibodies were visualizedby fluoresceinisothiocyanate-conjugatedgoat anti-rabbit IgG antibodies (Molecular Probes), or after incubation with biotinylated anti-mouse antibodies (Vector Laboratories, Burlingham,CA, U.S.A.) with Avidin-TexasRed (Molecular Probes). Tissue preparation
For immunolight-and immunoelectronmicroscopyrats were asphyxiatedwith COZand perfused through the left ventriclewith 0.1 M phosphate buffer, pH 7.4, containing 4!4. paraformaldehyde (w/v) for 15 min. The brains were removed and 40-pm-thick frontal sections were prepared for conventional immunohistochemistrywith a vibratome. For indirect double immunofluorescence staining, rat brains were rinsed in phosphate buffer and cryoprotected in 18°i0sucrose (w/v) in phosphate buffer. Cryostat sections, 10-#m-thick,were thaw-mounted on to slides coated with gelatinechrome alum. Immunohistochemistry
The sectionswereincubated free-floatingon a shaker at 4°Cfor 72hr, usingthe primary antibodies pDSD-1-PG or MAb473HD diluted in Tris-bufferedsaline(TBS), pH 7.3, containing 10?4obovine serum. The sections were then rinsed six times for 5 min in TBS, and further processed with the avidin–biotin complex using the vectastain elite kit (Vector Laboratories) according to the manufacturer’s recommendations. The antibody-enzyme complex was visualizedusing a solution containing 0.020/.3,3’-diaminobenzidineand 0.006°i0hydrogen peroxide in 0.05M TBS, pH 7.3. Sectionswere either mounted on gelatinechrome alum-coated slides,dehydrated in a graded series of ethanol, cleared in XY101, and coverslippedin Eukitt (Gribi AG, Belp, Switzerland),or processed for ultrastructural analysis.Therefore, sectionswere postfixedinitiallyin 2.5% (v/v)glutaraldehyde and then in 1YO(w/v) osmiumtetroxide(both in phosphate buffer) at 4“C, each for a period of 45 min. The tissue was stained with 10/0(w/v) uranyl acetate in 70°/0(v/v) ethanol, dehydrated in a graded seriesof increasing ethanol concentrations and embedded in Epon 812. Ultrathin sections were cut, mounted on copper grids and examined in a Zeiss EM 10. Double immunoj?uorescence staining
Cryosections were incubated with TBS containing IYo bovine serum albumin (BSA), for 30 min at room temperature. Slideswere then incubated concomitantly with a monoclinal antibody recognizing parvalbumin15and an antiserum directed against the DSD-1-proteoglycan.18The location of the DSD-1-PG was revealed by incubating sections for 3 hr with fluorescein isothiocyanate-conjugatedgoat anti-rabbit IgG diluted in phosphate buffer; parvalbumin antibodies were labelledfor 2 hr with a secondarybiotinylatedanti-mouseantibody diluted in phosphate buffer and then exposed to Avidin-TexasRed for 3 hr. RESULTS Light microscopic localization of DSD-I-PG in the neocortex of the developing and adult rat
To analysethe spatiotemporal expressionpattern of the DSD-1-PG in the rat cerebral cortex, we investigatedby immunohistochemistrya developmentalseriesof frontal sectionsthrough the cerebral cortex. Polyclonal as well as monoclinal antibodies directed against the DSD-1-PG appeared to stain
Perineuronalnets
251
perineuronal nets localizedaround nonpyramidal neurons in the rat neocortex from the late second week of postnatal development(Fig. 1)onwards into adulthood (i.e. older than 6 weeks;Figs 1 and 2). The immunostained neurons were differentin sizeand shape. The distribution of DSD-1-PG in the cerebral cortex varied considerably throughout the depth of the cortex (not shown). Neurons enwrapped by DSD-l-PG-positive perineuronal nets were mainly located in the cortical layers II and IV, but were also present in the other layers, with the exception of layer I. Clearly outlined perineuronal nets were the predominant feature of DSD-1-PG staining. However, there was considerable variability in staining intensity and penetration of the immunoreactivematerial into the perineuronal space.At a highermagnificationthe stainingwasclearlyreticulated(Fig. lbf,). Double lmmunofluorescencestaining revealed that DSD-I -PG-immunoreactivity preferentially accumulated around neurons expressingparvalbumin at any stage tested (Fig. 3). The fully developednets
Fig. 1. Localization of DSD-1-PG in frontal vibratome sections of the adult rat neocortex as achieved by the application of PDSD-1-PG. Deep layer III and layer IV contain numerous nonpyramidal neurons ensheathed by perineuronal nets (a). The appearance of the perineuronal nets is variable with regard to their staining intensity and the penetration of the immunoreactive material into the surrounding neuropil (b-f). Weakly stained and rather confined neuropil occurs as well as an intensely stained perineuronal environment showing a lattice-like pattern of immunoreactivity in regions far from the neuronal cell body (compareb, c, anddwitheand O. Note that dendritic stems areensheathedby sharply continedperineuronal nets (a–f). In (e) a tangentially cut cell surface shows the interlacing net of DSD-1-PG-positive material. Bar in (f)= 75 pm for (a); 15pm for (b}(f).
252
E. S.Wintergerstet al.
Fig. 2. Irnmunoelectron microscopic localization of DSD-1-PG in frontal sections of the neocortex of an adult rat using pDSD-1-PG, showing a perineuronal distribution of Iabelled extracellular material (a). The perikaryon of a neuron with a lobular nucleus (neuron A) is surrounded by a neuropil which is interspersed with immunoreactive extracellular material. Note that the neighbourhood of the neuron with a round nucleus (neuron B) remained unstained (a). The immunopositive signal is confined to a limited region of the neuropil around the cell body as seen at higher magnification (b, c), thereby surrounding glial structures (arrowhead in c) but excluding the synaptic cleft (arrow in b). Bar in (c)= 3.5 pm for (a); 1.5pm for (b) and (c).
appeared for the first time at postnatal day 19(Fig. 3a4) and persistedinto adulthood (Fig. 1, Fig. 3e and f). Electron microscopic localization of DSD-I-PG in the neocortex of the adult rat
In regions containing intensively stained perineuronal nets as evidenced in light microscopic investigations,nets could also be identifiedusing the electron microscope. At the ultrastructural levelDSD-1-PG immunoreactivitywas found to infiltrate the extracellular space to a distance of a few micrometers from the cell surface of some neurons (Fig. 2). It surrounded neuronal and glial
Perineuromdnets
253
Fig. 3, Double immunofluorescence staining for DSD-1-PG in perineuronal nets (b, d, f) applying pDSD1-PG, and parvalbumin applying MAb 235 (a, c, e) of frontal cryosections, of the neocortex of 3-week-old (a+) and adult (e and f) rats. Numerous parvalbumin-immunoreactive neurons localized in deep layer III and layer IV (a, c, e) are enwrapped by perineuronal nets as revealed by DSD-1-PG immunoreactivity (b, d, f; marked with arrows in b). Allocation of parvalburnin-positive neurons (c, e) with perineuronal nets (d, f) at a higher magnification in the cortex of a 3-week-old rat (c, d) and in the cortex of an adult rat (e, f). Bar in (f)=80 pm for (a) and (b); 20 pm for (c~f).
structures, such as synaptic terminals and glial processes(Fig. 2c), but was absent from synaptic clefts (Fig. 2b). The labelling was confined to the extracellular space but was associated with membranes. Neurons showinga pericellularzone of labelledextracellularmatrix were often in the vicinityof neurons with a non-labelledextracellularspace (Fig. 2; compare neurons A and B). DISCUSSION Applying monospecificpolyclonal and monoclinal antibodies against DSD-1-PG to the adult rat brain, we found an exclusivelocation of this hybrid chondroitin sulfate/derrnatane sulfate proteoglycan around parvalbumin-immunoreactiveneurons. The antibodies against DSD-1-PG havebeen characterizedextensivelyin previouspublicationsand satisfyvariouscriteria of specificity. In particular, they do not cross-react with versican, which has been shown by others to occur in perineuronal Focalization.b
254
E. S. Wintergerstet al.
At the light microscopiclevelthe staining with antibodies against DSD-1-PG did not differfrom that seen with a variety of lectins24’38’@ and antibodies directed against knownlGand unknown antigens.26,28 . A1lthese tools la~l the perikaryal surface and the proximal dendrites of the same subpopulation of interneurons in the brain in a reticulated manner. This labellingpattern has been called the “perineuronal net of extracellular matrix” (see Celio and Bli,imcke13 for review). As pre-embedding staining with diaminobenzidine/Hz02as a substrate was used, the ultrastructural localization of the DSD-1-PG antigen cannot be accurate.23Nevertheless, the primary site of occurrenceis undoubtedly the extracellular space. The validity of the association of label with the outer leaflet of membranes of glial processes32and nerve terminals will await confirmation by immunogold techniques. As for the other componentsof perineuronalnets, the preferentialassociation of DSD-1-PG with the surface of parvalbumin-immunoreactiveneurons in the adult animal is of unknown functional significance.Parvalbumin neurons have indented nuclei, as do most interneurons, belong to the and a special membrane cYtoGABA-subpopulation of interneurons,10have a highmetabolism37 skeleton,]zare rich in certain isoformsof K+-channels42and are characterized neurophysiologically The enwrapping of their cell bodies and proximal dendrites with the by a high firing rate. Wg,so,ql perineuronal net of extracellular matrix may relate to their peculiar chemical constitution and electrophysiologicalcharacteristics.Procedures that could reproduce and disrupt perineuronal nets in organotypic cultures of brain slicesmay shed light on their roles in situ. Heparan sulfates and chondroitin sulfate proteoglycans have already been recognized in perineuronal nets by using various monoclinal antibodies (seeTable I in Ref. 13for review).With the addition of DSD-1-PG a dermatan sulfate~chondroitinsulfate hybrid, as a new component, perineuronal nets of extracellular matrix are being shown to be even more complex structures, composed of a wide variety of different molecules. Some of the molecules are known to have repulsiveand/or permissiveeffectson the growth of axons during development(for reviewsee Ref. 19),but their roles in the adult brain remain unknown. The DSD-1-epitope has been shown to be involved in neurite outgrowth promotion and could serve a similar function in the perineuronal nets. Other components may concentrate growth factors around neurons20and still others are thought to stabilizesynapseson the surface of neurons.2’27 Thus, perineuronal nets of extracellular matrix are composed of a mesh of different molecules that creates a very peculiar milieu around parvalbumin neurons which could play a role in shaping their functional characteristics. Acknowledgements—The authors wish to thank C. Dumas, D. Uldry and C. Mandl for excellent technical assistance. This work was supported by the Swiss National Foundation, grant 31.0364.93 and Deutsche Forschungsgemeinschaft, grant DFG, Fa 159/5-1,2.
REFERENCES Bertolotto A.,RoccaG.,Canavese G.,Migheli A.andSchiffer D. (1991) Chondroitin-sulfate proteoglycan surrounds a subset of human and rat CNS neurons. J. Neurosci. Res. 29, 225–234. 2. Bertolotto A., Rocca G. and Schiffer D. (1990) Chondroitin 4-sulfate proteoglycan forms an extracellular network in human and rat central nervous system. J. Neurol. Sci. 100,113-123. 3. Bignami A., Asher R. and Perides G. (1992) Co-localization of hyaluronic acid and the chondroitin sulfate proteoglycan in rat cerebral cortex. Brain Res. 579, 173–177. 4. Bignami A., Asher R., Perides G. and Rahemtulla F?.(1992) The extracellular matrix of cerebral gray matter: Golgi’s pericellular nets and Nissl’s “nervoses Grau” revisited. Znt.J. deul Neurosci. 10,291-299. 5. Bignami A. and Dahl D. (1988) Expression of brain-specific hyaluronectin (BHN), a hyaluronate-binding protein, in dog postnatal development. Expl Neurol. 99, 107–117. 6, Bignami A., Perides G. and Rahemtulla F. (1993) Versican, a hyahrronate-binding proteoglycan of embryonal precartilaginous mesenchyma, is mainly expressed postnatally in rat brain. J. Neurosci. Res. 34, 97–106. 7. Brtickner G., Brauer K., Hartig W., Wolff J. R., Rickmann M. J., Derouiche A., Delpech B., Girard N., Oertel W. H. and Reichenbach A. (1993) Perineuronal nets provide a polyanionic, glia-associated form of microenvironment around certain neurons in many parts of the rat brain. Glia 8, 183–200. 8. Brttckner G., Delpech B., Delpech A. and Girard N. (1990) Concentration of hyalunorectin and anionic glycoconjugates in perineuronal glial cell processes at GABA-ergic synapses of rat cerebellum. Acta Histochem. Suppl. 38, 161–165. 9. Celio M. R. (1984) Parvalbumin as a marker of fast-tiring neurons. Neurosci. Lett. Suppl. 18,332. 10,Celio M. R. (1986) Parvalbumin in most garnma-aminobutyric acid-containing neurons of the rat cerebral cortex. Science 231,995-997. 11, Celio M. R. (1990) Calbindin D-28k and parvalbumin in the rat brain. Neuroscience 35,375475. 1.
Perineuronal
nets
255
12. Celio M. R. (1993) Perineuronal nets of extracellular matrix around parvalbumin-containing neurons of the hippocampus. Hippocampus 3, 55+1. 13. Celio M. R. and Blitmcke I. (1994) Perineuronal nets. A s~ecialized form of extracellular matrix in the adult nervous system. Brain Res. Rev. 19, 128–145. 14 Celio M. R., Chiquet-Ehrismann R. (1993) “Perineuronal nets” around cortical interneurons expressing parvalbumin are rich in tenascin. Neurosci. Lett. 162, 137–140. 15 Celio M. R., Baier W., Scharer L,, De Viragh P. and Gerday Ch. (1988) Monoclinal antibodies directed against the calcium binding protein parvalbumin. Cell Calcium 9, 81–86. 16 Celio M. R. and Rathjen F. (1993) Restrictingoccurs in “perineuronal nets” of the adult rat brain. Soc, Neurosci. Abstr. 19,689. 17 Delpech B., Bertrand P,, Hermelin B,, Delpech A., Girard N., Halkin E. and Chauzy C, (1986) Hyaluronectin. In F’ronfiersin Matrix Biology, Vol. 11 (eds Labat-Robert J., Timpl R. and Robert L,), pp. 78–89, Karger, Basel, 18 Faissner A., Clement A., Lochter A., Streit A., Mandl C. and Schachner M. (1994) Isolation of a neural chondroitin sulfate proteoglycan with neurite outgrowth promoting properties. J. Cell Bio/. 126, 783–799. 19, Faissner A., Scholze A. and Gotz B. (1994) Tenascin glycoproteins in developing neural tissue: only decoration? Devl Neurobiol. 2, 53–66. 20 Flaumenhaft R. and Rifkin D. B. (1991) Extracelhrlar matrix regulation of growth factor and protease activity. Curr. Opin. Cell Biol, 3,817-823. 21. Fujita S, C., Tada Y., Murakami F., Hayashi M. and Matsumura M. (1989) Glycosaminoglycan-related epitopes surrounding different subsets of mammalian central neurons. Neurosci. Res. 7, 117–130. 22 Girard N., Courel M. N,, Delpech A., Bruckner G. and Delpech B. (1992) Staining ofhyahrronan in rat cerebellum with a hyahrronectin-antihyaluronectin immune complex. Histochem. J. 24, 21–24. 23. Goldfischer S. (1981) Effusion artifact: a chronic problem in DAB-cytochemistry, J. Histochem. Cytochem. 98, 1360, 24. Hartig W., Brauer K. and Bruckner G. (1992) Wisteria jioribunda agglutinin-labelled nets surround parvalbumincontaining neurons. NeuroReport 3, 869–872. 25. Herndon M. E. and Lander A. (1990) A diverse set of developmentally regulated proteoglycans is expressed in the rat central nervous system. Neuron 4, 949–961. 26. Hinton D. R., Henderson V. W., Blanks J. C., Rudnicka M. and Miller C. A. (1988) Monoclinal antibodies react with neuronal subpopulations in the human nervous system. J. conzp.Neurol. 267, 398%408. 27. Hockfield S., Kalb R. G., Zaremba S. and Fryer H. (1990) Expression of neural proteoglycans correlates with the acquisition of mature neuronal properties in the mammalian brain. Cold Spring Harb. Symp. quant. Biol. 55, 505–514. 28. Hockfield S. and McKay R. D.G. (1983) A surface antigen expressed by a subset of neurons in the vertebrate central nervous system. Proc. natn. Acad. Sci. U.S.A. 80, 5758–5761. 29. Kawaguchi Y, (1993) Physiological, morphological and histochemical characterization of three classes of interneurons in rat neostriatum. J. Neurosci, 13, 4908J1923. 30, Kawaguchi Y., Katsumaru H., Kosaka T., Heizmann C, W. and Hama K. (1987) Fast spiking cells in rat hippocampus (CA1 region) contain the calcium-binding protein parvalbumin. Brain Res. 416,369-374. 31. Kawaguchi Y. and Kubota Y. (1993) Correlation of physiological subgroupings ofnonpyramidal cells with parvalbuminimmunoreactive and calbindin D-28k-immunoreactive neurons in layer V of rat frontal cortex. .J. Neurophysiol. 70, 387– 396. 32,Kosaka T. and Hama K. (1986) Three-dimensional structure of astrocytes in the rat dentate gyrus. J. comp. Neurol. 249, 242-260. 33.Kosaka T. and Heizmann C. W. (1989) Selective staining of a population of parvalbmnin-containing GABAergic neurons in the rat cerebral cortex by lectins with specific affinity for terminal N-acetylgalactosamine. Brain Res. 483, 158-163. 34.Kosaka T., Heizmann C. W. and Barnstable C. J. (1989) Monoclinal antibody VCI.1 selectively stains a population of GABAergic neurons containing the calcium-binding protein parvalbumin in the rat cerebral cortex. E.xplBrain Res. 78, 43-50, 35. Kosaka T., Heizmann C. W. and Fujita S. C. (1992) Monoclinal antibody 473 selectively stains a population of GABAergic neurons containing the calcium-binding protein parvalbumin in the rat cerebral cortex. Expl Brain Res. 89, 109– 114. 36. Kosaka T., Isogai K., Barnstable C. J. and Heizmann C. W. (1990) Monoclinal antibody HNK-1 selectively stains a subpopulation of GABAergic neurons containing the calcium-binding protein parvalbumin in the rat cerebral cortex. Expl Brain Res. 82,566-574. 37.McCasland J. S., Hibbard L. S. and Woolsey T. A. (1993) Calcium-binding protein phenotype and function of GABAergic neurons in barrel cortex. Soc. Neurosci. Abstr. (Vol. 2), 19, 1567. 38.Nakagawa F., Schulte B. A. and Spicer S. S. (1986) Selective cytochemical demonstration of glycoconjugate-containing terminal N-acetylgalactosamine on some brain neurons. J. camp. Neurol. 243, 28&290. 39.Nagakawa F., Schulte B. A., Wu J. Y. and Spicer S. S. (1986) GABA-ergic neurons of the rodent brain correspond partially with those staining for glycoconjugate with N-acetylgalactosamine. J. Neurocytol. 15,389-396. 40. Nagakawa F., Schulte B. A., Wu J. Y. and Spicer S. S. (1987) Postnatal appearance of glycoconjugate with terminal Nacetylgalactosamine on the surface of selected neurons in mouse brain. Deul Neurosci. 9, 53–60. 41. Steindler D. A., Settles D., Laywell E. D., Yooshiki A., Faissner A., Erickson H. P. and Kusakabe M. (1995) Tenascinknockout mice: barrels, boundary molecules, and glial scars. J. Neurosci. 15, 1971-1983. 42, Weiser M., Vega-Saenz de Micra E., Kentros C., Moreno H., Franzen L., Hillman D., Baker H. and Rudy B. (1994) Differential expression of shaw-related K+ channels in the rat central nervous system. J. Neurosci. 14, 949–972.