Disruption o proteoglycans in neural tube fluid by B-

Disruption of Proteoglycans in Neural Tube Fluid by ␤-D-XylosideAlters Brain Enlargement in Chick Embryos M.I. ALONSO, A. GATO,*J.A. MORO, AND E. BARBOSA Departamento deAnatomı´a Humana, Facultad de Medicina, 47005 Valladolid, Spain ABSTRACT Following neurulation, the anterior end of the neural tube undergoes a dramaticincreaseinsizeduemainlytotheenlargingofthebraincavity.This cavityisfilledwithso-calledneuraltubefluid(NTF),whosepositivepressurehas been shown to play a key role in brain morphogenesis. This fluid contains a water-soluble matrix, rich in chondroitin sulfate (CS), which has been proposedasanosmotic regulator of NTF pressure genesis. The purpose of the present study is to observe the influence of CS on NTF osmolality and its relationtoNTF hydrostatic pressure and brain expansion. NTFwas obtained by means ofmicroaspirationfromthe mesencephalic cavityofchickembryos.TheosmolalityofNTFbetweenH.H.stages20and29 wasmeasuredon the basis of its cryoscopic point.CSsynthesis was disrupted by using ␤-D-xyloside and the induced variations in brain volume were measuredbymeansofmorphometry.WealsomeasuredthevariationsinNTF osmolality, hydrostatic pressure, and the concentration of CS and sodium inducedbymeans of ␤-D-xyloside. Our data reveal that, at the earliest stages of development analyzed, variations in NTF osmolality show a characteristic pattern that coincides withthedevelopmental changes in the previously described fluid pressure. Chick embryos treated with ␤-D-xyloside, a chemical that disrupts CS synthesis,displayedanotableincreaseinbrainvolumebutnootherapparent developmental alterations. Morphometric analysis revealed that this increase was due to hyperenlargement of the brain cavity. ␤-D-xyloside brings about specific changes in the biochemical composition of NTF, which entails a large increase in CS concentration, mainly in the form of free chains, and in that of sodium. As a result, the fluid’s osmolality and brain intraluminal pressure increased,whichcouldaccountfortheincreaseinsizeofthebrainanlage. These data support the hypothesis that the intraluminal pressure involved in embryonic brain enlargement is directly dependent on NTF osmolality, and that the concentrations of CS and its associated microions could play a key role in the regulation of this process. Anat. Rec. 252:499– 508, 1998. r1998 Wiley-Liss, Inc. Key words: neural tube; brain development; chondroitin sulfate; ␤-Dxyloside; chick embryo The anterior end of the neural tube, the future brain anlage,undergoesaconsiderableincreaseinvolumeatthe earliest stages of development, with a significant difference emerging between the spinal cord and the future encephalon. The mechanisms involved in this important morphogeneticprocessremain largely unknown to date. In the last decades, many efforts have been made in order to quantify the rate of brain enlargement at these particular stages of development. In chick embryos, PaGrant sponsor: Junta de Castilla y Leo´n and Iberduero, Inc.; Grantsponsor: Universidad de ValladolidtoA.Gato. *Correspondenceto:Dr.AngelGato,DepartamentodeAnatomı´a Humana, Facultad de Medicina, C/ Ramo´n y Cajal 7, 47005 Valladolid,Spain. E-mail: [email protected] Received13 June 1997;Accepted6May 1998 THE ANATOMICAL RECORD 252:499–508 (1998) r1998 WILEY-LISS, INC. checo et al. (1986) described an initial period of moderate brain growth from H.H. stages 11-18 (Hamburger and Hamilton, 1951), followed by a period of rapid growth between stages 19 and 26. In this final period, and in only 48 h, the embryonic brain undergoes a 30-fold size increase, ⬎70% of this being accounted for by growth of the brain cavity (Desmond and Jacobson, 1977). The explosive growth of the embryonic brain cavity has been directly related to the expanding force generated by the positive pressureoftheintraluminalfluiddescribedbyJelinekand Pexieder (1968). Such pressure requires the previous occlusion of the spinal cord lumen, transforming the brain cavity into a physiologically sealed system (Schoenwolf and Desmond, 1984; Desmond et al., 1993). The particular role, based on physical phenomena, that these morphogeneticmechanismscanplayinthedevelopmentalpatternof various embryonic anlages was summarized by Newman and Comper (1990). In fact, expanding forces based on intraluminalhydraulic mechanisms have been involved in the growth and development of other anlages besides the embryonic brain, such as the otic vesicle (Barbosa et al., 1986) and the eye (Coulombre, 1956). An experimental demonstration that the positive pressure exerted by NTF confined in the brain cavity is a key factor in embryonic brain expansion and morphogenesis was provided by Desmond and Jacobson (1977) and Desmond (1985). They induced an experimental reduction in this pressure in chick embryos, which led to severe dysmorphogenesis and braincollapse. The composition and function of the fluid contained in the embryonic brain cavity, and their possible influence on earlybraindevelopment,requiremoreresearch(Jacobson, 1991). In a previous report (Gato et al., 1993), we suggested that the osmotic pressure of NTF could play a key role in the genesis of this expanding force and demonstrated the presence of a water-soluble matrix rich in the proteoglycanchondroitinsulfate (CS) in this fluid. Given the special osmotic properties of the proteoglycans (PGs) (Comper and Laurent, 1978), we hypothesized that the osmotic pressure of NTF could be, at least in part, regulatedbyits CS concentration. Inordertoclarifythemechanismsleadingtothegenesis ofpressureanditsregulation, wemeasured theosmolality of NTF in chick embryos from H.H. stages 20 to 29. In ordertoanalyzetherelationbetweentheCSconcentration in NTF, the osmolality of this fluid and embryonic brain expansion, we disrupted the synthesis of CS in the NTF with ␤-D xyloside, a chemical that selectively disrupts the synthesisofsulfated PGs (Schwartz et al., 1976). ␤-D-xylosideshave beenwidely employedin thestudyof the roles played by sulfated PGs (particularly CS) at differentembryonicdevelopmentstagesintherat(MorrissKay and Crutch, 1982), chicken (Gibson et al., 1978,1979; Kanke et al., 1982; Segen and Gibson, 1982), sea urchin (Kinoshita and Saiga, 1979), and frog (Yost, 1990). ␤-Dxylosides specifically alter the synthesis of those sulfated PGs whose glycosilation begins with the sequence xylosylgalactosyl-galactosyl,includingchondroitinsulfateandits epimeres dermatan sulfate, heparan sulfate, and heparin (Schwartzet al., 1974; Robinson etal.,1975; Roden, 1980). However,p-nitrophenyl-␤-D-xylopyranosideespeciallydisruptschondroitinanddermatansulfatesynthesis.Several studieshaveshownitsincapabilityofprimingtheheparan sulfate and/or heparin synthesis in embryonic tissues (Sobue et al., 1987; Lugemwa and Esko, 1991), or in tissue cultures from different sources (Galligani et al., 1975; Spooncer et al., 1983). The data from the present study indicate that CS is the only sulfated proteoglycan in chick embryo NTF. Therefore, we can assume that the changes in NTF physicochemical properties induced by means of ␤-D-xyloside are fundamentally due to a disruption in CS synthesis. MATERIALSAND METHODS Fertile White Leghorn eggs were incubated at 38°C in a humidified atmosphere to obtain chick embryos at different developmental stages, ranging from H.H. 20 to 29 (Hamburger and Hamilton, 1951). After incubation, a small window was opened in the shell and thereafter the embryos were classified according to stage. For ␤-Dxyloside treatment, doses of 24 µl of p-nitrophenyl-␤-Dxylopyranoside4mM(Sigma,St.Louis,MO)wereinjected with a Hamilton microsyringe into the subgerminal space of H.H. stage 12–13 embryos; the control group was injected with sterile saline solution. Afterward, the opening in the shell was sealed and the eggs were reincubated for44h until the embryos reached H.H. stage 23. MorphometricAnalysis The method used to determine brain volume is very timeconsuming; therefore, the number of embryos to be measured was limited to 10 control and 10 ␤-D-xylosidetreated embryos.After fixation in Bouin’s solution for 6 h, embryos were dehydratated in graded ethanol and embedded in paraplast; 8 µm transverse serial sections were cut and stained with haematoxylin-eosin. The volume was calculated from the rostral extent of the brain until the first section in which the otic vesicles appeared, including forebrain, midbrain, and part of the hindbrain. The inner andouter limits oftheneuroepithelium were drawninone of every five sections (section factor ⫽5) using a Leitz SM Lux microscope equipped with a drawing tube (total magnification 60X), and the corresponding areas integrated with a Videoplan (Kontrol Elektronic GMBH) computerized image-analysis system, via the VIDAS 2.1. stereology program. In each embryo, the volume of the brain, of its cavity, and that of the neuroepithelial wall wereobtainedbyaddingthe correspondingsectional areas (in µm2) multiplied by 8 µm (thickness) and by 5 (section factor). According to Desmond and Jacobson (1977) and Pacheco et al. (1986), the measurement of every fifth section differs ⬍10% from the measurement of every section in H.H. stage 23 chick embryos, and this error can beregardedasacceptable.Thefinalvolumesareexpressed as the arithmetic mean of the values of the 10 embryos ⫹/- standard error. For comparison of the average volumes (total, cavity, and tissue) between experimental and control embryos, two-tailed Student’s t-test for independent samples was applied and Pvalues ⬍0.05 were interpreted asindicatingstatistical significance. Electron Microscopy After in ovo treatment, the extraembryonic membranes were excised and both control and treated embryos rinsed in saline solution. The primary brain vesicle regions were sectioned from the embryonic heads with thin tungsten needles, under dissecting microscope control, in a SilgarcoatedPetridish.Thebrainsegmentswererinsedagainin saline solution, fixed overnight in Karnovsky fixative 500 ALONSO ET AL. containing 2% tannic acid at room temperature, following the method of Singley and Solursh (1980), and then postfixed in 1% osmium tetroxide for1hatroom temperature. The samples were dehydrated in a graded acetone series and embedded in plastic (Spurr, 1969). Ultrathin sections from the different brain vesicles were cut on an AMC ultra-microtome, MT 6000 XL, stained with uranyl acetate and lead citrate (Reynolds, 1963) and examined underaJeol 1200 EX electron microscope at 80 Kv. Neural Tube FluidAcquisition NTFforosmolalitydeterminationandbiochemicalanalysis was obtained as follows. The extraembryonic membranes were removed, the embryos were rinsed twice in sterile saline solution and placed in a dry Petri dish, and the surrounding fluid was aspirated. A micropipette, connected to a microaspirator (Medical System PLI 100), was carefully placed in the middle of the mesencephalic cavity underdissectingmicroscope control, avoiding contact with the neuroepithelial wall. The NTF of several embryos was slowly aspirated, until a sufficient amount was obtained for osmolality determinations (200 µl for each H.H. stage between 20 and 29) and biochemical analysis (300 µl from both control and treated H.H. stage 23 chick embryos). NTF wasaliquotedand frozen at -40°C until it was to be used. NTF Osmolality Determinations Measurements of NTF osmolality were made on the basis of its cryoscopic point with an Osmomat 050 (Gonotec).Sincea minimum of 50 ul were necessary foraccurate assessment,apool(200µl)ofNTFwasmadewiththeNTF obtained from embryos of each H.H. stage between 20 and 28,andanotherpool(200µl)fromcontroland␤-D-xyloside treated H.H. stage 23 chick embryos. Four different determinations of NTF osmolality were performed in all cases. Mean and standard errors were calculated for each developmental stage and the nonparametric two-tailed MannWhitney U test was used to analyse the differences in osmolality among stages. Probability (P) values ⬍0.05 wereinterpretedas indicating statistical significance. NTF Pressure Determinations To measure the intralumen pressure of brain anlage, live embryos were placed in a Petri dish, as described in the section on NTF acquisition, and a micropipette of 30 µm (inner diameter) connected to a microaspirator was carefully placed in the rhombencephalic cavity. Positive pressure of the lumen caused NTF to move into the micropipette. By offsetting this pressure, using the instrument balancing pressure system, we estimated intraluminal pressure as that pressure necessary to make the NTF returntothebraincavity.Onlymeasurementsthatdidnot show NTF leakage were valued. A total of 13 measurements for control embryos and 14 for treated ones were made on H.H. 21 stage embryos. Mean and standard errors were calculated, and the two-tailed Student’s test for independent samples was applied in order to compare the average pressures of experimental embryos with those of control embryos, probability (P) values ⬍0.05 were interpretedasindicating statistical significance. GlycosaminoglycansAnalysis Isolation and purification of the NTF’s glycosaminoglycans GAGs. The method to isolate and purify the GAGs was based on that proposed by Breen et al. (1976). Aliquots of NTF from control and treated embryos were digested with protease type XIV (Sigma) at 50°C for 6 h. The GAGs were precipitated from the digest by the addition of 5 volumes of 5% potassium acetate in absoluteethanoland centrifugation at 17,500 g at 4°C. QuantitativeanalysisofGAGs.Thetotalamountof isolated GAGs was estimated by the uronic acid-carbachol reaction described by Bitter and Muir (1962). Briefly, after vigorous hydrolysis in 0.025 M sodium tetraborate in sulfuric acid, 200 µl of 0.125% carbachol (Fluka) in absolute ethanol were added. The optical density was read at 530 nm and the absorbance values were plotted against thoseobtainedwith standard solutions of CS. Qualitative analysis of GAGs. The composition of GAGsinNTFwasdeterminedbymonodimensionalelectrophoresis and specific glycosidase digestion. GAGs isolated from NTF of control and treated embryos, as well as standard solutions of GAGs (Sigma), were applied to cellulose acetate strips (Cellogel 2.5 x 17), and electrophoresedin 0.1 M formic acid-pyrydinebuffer,pH 3.1, at 270v for 1 h according to Beeley (1985). After electrophoresis, the strips were stained with 1% alcian blue 8GX (Gurr) in 0.05 M acetate buffer, pH 5.8, and ethanol (v/v). For enzymaticidentification,glycosidaseschondroitinaseABC and AC (Sigma) were used, following the digestion procedure of Breen et al. (1976). Chondroitinase ABC specifically digests chondroitin 4and 6-sulfate, and to some extent, hyaluronic acid, whereas chondroitinase AC only digests chondroitin 4and 6-sulfate (Yamagata et al., 1968). The activity of the enzymes was previously checked with standard GAGs. Afterward, aliquots of the GAGs isolated from NTF in control and treated embryos were diluted in 50 µl of 0.15 M Tris-HCL buffer containing 0.5 IU of chondroitinaseABC (pH 8), or 0.5 IU of chondroitinase AC (pH 7). The mixtures were incubated at 37°C overnight and gently shaken, and the GAGs were isolated andelectrophoresedas described below. Sodium Determination Sodium concentration (both bound and free) in NTF from control and treated embryos was measured by flame emission spectroscopy on a Variant Spectra A-80 with a Variant GTA-100 attachment. ProteinAnalysis: Western Blot The total concentration of proteins in NTF was determined by the Bio-Rad protein assay based on the Bradford (1976) dye binding procedure. For Western blot analysis, aliquots of NTF from control and treated embryos were mixed with a sample buffer containing 3.2% SDS, 16% glycerol,2.8M␤-mercaptoethanoland0.0016%bromophenol blue. Samples were applied to a 4–20% gradient SDS-PAGE gel (Bio-Rad) according to Laemly (1970), and the proteins obtained were transferred to nitrocellulose sheets, pore size 0.45µm(Bio-Rad), according to the method of Towbin et al. (1979). After incubation for2hat room temperature in the blocking solution (PBS containing 5% skimmed milk), the nitrocellulose sheets were exposedovernight,at4°C,toantichondroitinsulfatemono501CHONDROITIN SULFATE AND BRAIN ENLARGEMENT clonal antibody (CS-56 from Sigma), which recognizes the CS glycosidic fraction (Avuur and Geiger, 1984), and identified with a peroxidase-labeled mouse IgM PK 4010 Vectastain avidin biotin complex kit (Vector). The peroxidase activity was revealed with diaminobenzidine (0.5 mg/mlinPBS with 0.02% hydrogen peroxide). RESULTS Developmental Pattern of NTF Osmolality NTF osmolality was measured in the period of development (H.H. stages 20–29) in which rapid brain enlargement occurs. Osmolality determinations were not made in previous stages because of the impossibility of obtaining sufficientamounts of NTF. NTF osmolality variations followed a characteristic patternduringtheearlydevelopmentofchickembryos(Fig.1). NTF osmolality oscillated around 250 mOsm/Kg with no significant changes until H.H. stage 24 (P⬎0.05, by Mann-Whitney U test). Osmolality then dropped significantly at stage 25 but returned to 250 by stage 27. There were statistically significant differences in values of osmolalitybetweenstages24and25andbetweenstages25and 26(P⫽0,029 by Mann-Whitney U test). Effect of ␤-D-xyloside in Brain Enlargement After in ovo treatment, the brain vesicles exhibited considerable growth with respect to the rest of the neural tube and began to undergo secondary vesiculation (Fig. 2-A). Histological observation of these embryos (Fig. 2-C) reveals that the brain anlage occupies most of the embryonicheadand reveals a large internal cavity. Embryostreatedwith␤-D-xylosideexhibitedmorphological development similar to that of the control specimens, althoughthey all, to agreateror lesser degree, displayeda significantincrease in headsizewith respect tothecontrol specimens(Fig.2-B);thisseemstobeduetoahyperenlargement of the brain vesicles, whose morphology was not altered to any great extent. No morphological changes were seen in the development of other embryonic anlages with the exception of the eye, which occasionally appeared smaller. The histological sections of these embryos (Fig. 2-D)confirmed the large increase inbrainanlagesize with regardtothecontrolspecimens; thisseems tobe theresult of a global overexpansion of the brain cavity. These ␤-Dxylosideinducedchangesareunaccompaniedbyanyhistological alterations in the morphology of the brain anlage, which shows normal progress in brain vesiculation; however,some morphologicalalterations canbeappreciatedat theborderbetweenthediencephalonandthemesencephalon, such as a flattening in the diencephalo-mesencephalic fissureandin the mesencephalic floor (Fig. 2C-D). Morphometric data (Fig. 3) indicate that the size of the brain anlage of the ␤-D-xyloside-treated embryos was 37% larger than that of the control ones. Brain cavity volume increased 50% more than that of the control specimens, whereastheincreaseinthebrainneuroepithelialwallwas only 11% in the ␤-D-xyloside-treated embryos. Statistical analysis shows that the total and cavity volumes of the experimental groups were significantly greater than for the controls ( P⬍0.001), whereas differences in tissue volume were not significant ( P⬎0.05). Such data indicate that brain volume increase induced by ␤-D-xyloside is mainlydueto an overexpansion of the brain cavity. Ultrastructure of Brain Neuroblasts Our results indicate that ␤-D-xyloside-induced disruption of CS biosynthesis is accompanied by changes in the ultrastructural appearance of the apical ends of brain neuroepithelial cells. In control embryos, the apical end of the neuroblasts were joined by gap junctions, with numerous and conspicuous intercellular spaces (Fig. 4A). The apical cytoplasm of these cells showed mitochondria and intracellular membrane formations morphologically compatible with endoplasmic reticulum, Golgi cisterns and large (200–300 µm diameter) vesicular formations of a moderately electrondense content. Near the apical surface several small vesicular formations, which resemble the clathrin-coated vesicles involved in endocytosis, could be seen. The most remarkable ultrastructural changes induced by ␤-D-xyloside in neuroblasts (Fig. 4B) was a notable increase in apical intracytoplasmic organelles, such as Golgi complexes and, especially, in large vesicular formationsgathering in the vicinity of the nucleus and scattered near the apical surface. The ultrastructural appearance of these vesicles does not differ from those described in controlembryos,except in their number and greater size. ␤-D-xyloside Induces Specific Changes in NTF Biochemical Composition, Osmolality, and Hydrostatic Pressure Biochemical analysis of the NTF of control embryos (Table 1) revealed that, at H.H. stage 23, its main componentis protein (295 ug/100ul),in which the PGsrepresent a small percentage (12,5 ug/100 ul). In these embryos, the results of the qualitative analysis of the NTF (Fig.5) identify two different types of GAGs, CS and hyaluronic acid. However, the electrophoretic separation of native NTF (without prior digestion with protease) of embryos of the same stage revealed no migration of any of its components, with the GAGs (identified as positive alcian blue material) remaining at the level at which they were Fig. 1. Graph showing the pattern of NTF osmolality at early stages of chick embryo brain development. The osmolality remains stable until H.H. stage 24; however, a transitory drop can be observed at H.H. stage 25. The evolution of NTF osmolality coincides with that of brain cavity hydrostatic pressure, which has been previously described. The values are mean ⫹/- standard error (n ⫽4). * P ⫽0.0029 by Mann-Whitney U test. 502 ALONSO ET AL. deposited (Fig. 6B2). This fact suggests that the GAGs present in the NTF are probably integrated in proteoglycanmacromolecularaggregates,asoccursinotherextracellular matrices. The results of the Western blot, performed to clarify the nature of NTF chondroitin sulfate, indicate thatonly a proteinfraction,Mr197x103,waspresent (Fig.7). The results of other determinations on NTF composition, such as sodium concentration and osmolality of the control embryos, were of 2,316 ppm and 278 mOsm/Kg, respectively(Table 1). Inthe␤-D-xylosidetreatedembryos,thespecificchanges in the biochemical composition of NTF were only reflected in the PGs and ionic composition, with no alteration in its proteinconcentration(Table1).The uronicacidconcentration was 200% higher than that of the control embryos (Table 1), and the results of the electrophoretic separation plus digestion with specific glycosidases (Fig. 6A) indicated that this increase is exclusively due to CS, since the amount of hyaluronic acid remained unchanged. In the electrophoresed native NTF of the treated embryos (Fig. 6B), a band of similar mobility to that of CS appeared, which is highly suggestive of the presence of free chains of this proteoglycan. Western blots of the NTF of the treated embryos with antichondroitin sulfate CS-56 antibody showed (Fig. 7) that CS electrophoretic mobility is similar to that of the control embryos, with only a slight reduction inepitopeconcentration.Thesedatasuggestthat␤-D-xyloside does not bring about substantial variations in the molecular weight of the CS secreted into the brain cavity. Yet our measurements of Na concentration of NTF of ␤-D-xyloside embryos compared with controls showed nearly a 40% increaseinthe total sodium concentration (Table 1). Inorderto clarify the influence of the increase of CS and sodiumin theosmolality ofNTF,we measuredthisparameterinbothcontroland␤-D-xyloside-treatedembryos.The results show that ␤-D-xyloside treatment induces an 80% increase in NTF osmolality in comparison with control embryos (Table 1). Furthermore, in order to find out whether this increase in NTF osmolality was related to changes in brain intraluminal pressure during the rapid brain enlargement period, we measured the pressure of control and treated embryos at stage 21 H.H. (after 30 h of ␤-D-xyloside treatment).As Table 2 shows, the intraluminal pressure in the brain vesicles of treated embryos increased 39% in comparison with controls, and this difference is statistically significant (P⬍0.001). DISCUSSION In this report, we test whether the osmolality of NTF is directly related to hydrostatic pressure and whether experiFig. 2. ␤-D-xyloside induces overexpansion of the embryonic brain cavity in chick embryos. Although the morphologic pattern of brain vesiculation seems to be preserved, some alterations the diencephalomesencephalic borders can be observed. Macroscopic view of H.H. stage 23 control (A) and treated (B) chick embryos. Sagittal histological section of control (C) and treated (D) embryos. T, telencephalic vesicles; D, diencephalon; M, mesencephalon; R, rhombencephalon. Scale bars ⫽ 1 mm. 503CHONDROITIN SULFATE AND BRAIN ENLARGEMENT mentally induced changes in CS concentration of NTF by meansof␤-D-xylosidewouldlead tosignificantalterations in NTF osmolality and hydrostatic pressure, and how muchchangesaffect brain expansion in chickembryos. Onemajorfinding, i.e., a significant drop in the osmolality of the NTF demonstrated by us, is consistent with the drop in intraluminal pressure reported by Jelinek and Pexieder (1970). However, we found that the changes in hydrostatic pressure occur later than the osmolality changes; this could be explained by the time necessary for water interchange between NTF and embryonic tissues. Furthermore, our results show that an experimentally induced increase in the osmolality of NTF by means of ␤-D-xylosideleadstoasignificantincreaseinintraluminal pressure of the brain vesicles and to a hyperenlargement ofthebrainanlageinchickembryos.Thisrelationcouldbe explained if we bear in mind that the osmotic pressure of NTF implies an increase of water inside the brain cavity, leading to an increase in hydrostatic pressure and/or a neuroepithelial wall distension. This explanation is based ontheassumptionthattheembryonicbrainisaphysiologically sealed system (Schoenwolff and Desmond, 1984; Desmond et al., 1993). These data support the hypothesis that NTF osmolality could be involved in the control of brainexpansion. Inaddition,we have already reported the presence of an osmotically active molecule, like CS, inside the brain cavities in chick embryos (Gato et al., 1993), and we have also suggested that the osmotic pressure of NTF could be, at least in part, regulated by its CS concentration. Our study indicates that NTF contains a main protein component, two characterized PGs (CS and hyaluronic acid) and high levels of sodium, all of which could contribute to NTF osmolality; however, the special osmotic properties of CS (Comper and Laurent, 1978) make it a candidate for NTF osmotic pressure regulation. Moreover, our results show that the changes in the CS concentration of NTF, experimentally induced by means of ␤-D-xyloside (without any other associated change in the hyaluronic acid or the protein component of NTF), lead to significant modifications in NTF osmolality, hydrostatic pressure and brain enlargement. Proteoglycans have been proposed as fundamental molecules in the regulation of the water content in biological tissues. Certain major biological properties of PGs are the result of their polyanionic nature, which facilitates their interaction with water in biological tissues, hence their special osmotic properties and hydraulic conductivity. CS hasthe greatestosmotic capacity,determinedbyitsuronic acidcontentandthenatureoftheglycosidicunion(Comper, 1981; Comper and Zamparo, 1990), and relatively small variations in its concentration produce large increases in itsosmoticcapacity (Comper and Laurent, 1978). Our results show that the changes induced in CS concentrationofNTFbymeansof␤-D-xyloside,areaccompanied by a significant increase in total sodium concentration. ␤-D-xyloside is a chemical compound that specifically disrupts sulfated proteoglycan biosynthesis, but we have not found any data to justify a direct interference of this compoundwithionictransportmechanisms;therefore,the changes in sodium concentration of NTF, induced using ␤-D-xyloside, seem to be related to the CS variations. These data suggest that there is a functional relation between CS and sodium in NTF, and both molecules could be co-responsible for NTF osmolality regulation. Comper and Laurent (1978) have shown that sulfated PGs bind large amounts of inorganic ion (counterions) by electrostatic interactions so that the osmotic properties of PGs in biological solutions depend to a large extent on the actvity oftheirassociatedcounterions.Ourdatashowthatsodium isamajorioniccomponent ofchick embryoNTF,soitsions are probably the counterions associated with CS. Therefore, our data support the hypothesis that the NTF osmolality properties in chick embryos could be attributed to a coordinatedregulationin CS and sodium concentrations. Despite the fact that intraluminal PGs can regulate the sodiumconcentration in NTFbyelectrostatic interactions, there is a large sodium concentration in the NTF of chick embryos, which suggests the existence of a sodium transport mechanism across the neuroepithelium toward the brain cavity. We do not have any evidence to support this point and further research is necessary; however, Barbosa et al. (1985) have already reported the possibility of transepithelial transport of ions and water toward the inner cavity of another ectodermal derived anlage, the otic vesicle. In addition, in the neuroepithelial cells of amphibian embryos, the presence of sodium pumps, which actively transport sodium out of the neural cavity, has been describedbyHotaryandRobinson (1990)and Borgensand Shi (1995). This ionic transport generates a transneural tubepotentialthatseemstobenecessaryforthestructural integrity of the early neuroepithelium and a prerequisite for normal morphogenesis. In amphibian embryos, a transepithelial sodium transport inside the embryo has beendescribedintheepiblast(MetcalfandBorgens,1994), and it seems to be preserved after neural tube invagination and internalization during neurulation to create the Fig. 3. Comparison of brain size of control and ␤-D-xyloside treated H.H. stage 23 embryos shows a significant disruption of the brain expansion process. The total brain anlage volume of the treated embryos is 37% higher than that of the control specimens. This increase is due to brain cavity volume, which is 50% greater than that of control specimens, whereas brain tissue mass increases by only 11%. The values are mean ⫹/⫺standard error (n ⫽10).* P ⬍0.001 by Student’s t-test. 504 ALONSO ET AL. Fig. 4. ␤-D-xyloside induces ultrastructural changes in neuroepithelial cells. Images are from the apical pole of diencephalic neuroblasts of control (A) and treated (B) H.H. stage 23 chick embryos and show changes induced by ␤-D-xyloside, in particular, a surprising increase in apicalvesicular formations(V)andGolgi complexes (g).Scalebar ⫽1µm in both figures. transneural tube potential. In the epiblast of avian embryos, a similar sodium transport mechanism has been described (Stern and MacKenzie, 1983); however, the existence of a high NTF concentration of sodium after neurulation, as we have found in our work, suggests that in chick embryos the mechanism of sodium pumping from the brain cavity has not been preserved, or perhaps has been inverted in order to generate an osmotic swelling mechanism. ␤-D-xyloside induces ultrastructural changes at the apical end of brain neuroblasts, mainly the increase in the number of Golgi complexes and apical intracytoplasmic vesicles. Golgi complexes have been traditionally associated with glycosylation, fundamental for GAGs chains synthesis;similarly,vesicularformationsresemblingthose described have been related with proteoglycan synthesis and secretion out of the cell (Hay, 1991). Moreover, the presence of numerous vesicules of similar appearance and containing GAGs has been previously described in the neuroblasts’ apical cytoplasm by Mak (1978) during the fusion process of neural folds in amphibian embryos. Therefore, if we bear in mind that the cavity of the brain TABLE 1. Effects of ␤-D-Xyloside Treatment on Neural Tube Fluid Composition of H.H. Stage 23 Chick Embryos Treatment Proteins µg/100 µl Uronic acid µg/100 µl Sodium ppm Osmolality mOsm/Kg Control 295 12.5 2316 278 ␤-D-xyloside 295 37.5 3259 504 Fig. 5. Two major GAGs, hyaluronic acid and CS, were identified in NTF of H.H. stage 23 chick embryo brains, by electrophoretic separation on cellulose acetate combined with digestion by specific glycosidases. Lane 1: Standard GAGs separation pattern (HA: hyaluronic acid, KS: keratan sulfate, DS: dermatan sulfate, CS: chondroitin sulfate). Lane 2: Electrophoretic separation of GAGs extracted from NTF of H.H. stage 23 chick embryos. Lanes 3 and 4: GAGs extracted from H.H. stage 23 chick embryo after pretreatment with chondroitinase AC and ABC respectively. The identification of the two bands as hyaluronic acid (arrow) and CS (arrowhead),was based ontheirsensitivityto chondroitinaseAC(Lane 3), which specifically digests chondroitin-dermatan sulfate, and to chondroitinase ABC (Lane 4), which digests chondroitin-dermatan sulfate and hyaluronic acid. Fig. 6. Changes induced in NTF GAGs composition by means of ␤-D-xyloside valued by electrophoretic separation on cellulose acetate combined with digestion with specific glycosidases. A. Lane 1 shows the electrophoretic pattern of standard GAGs (HA: hyaluronic acid, KS: keratan sulfate, DS: dermatan sulfate, CS: chondroitin sulfate). Lane 2: Electrophoretic separation of GAGs extracted from NTF of ␤-D-xyloside treated embryos. Lane 3: The same as lane 2 after chondroitinase AC treatment. The electrophoretic separation of GAGs extracted from NTF ␤-D-xyloside treated embryos shows that most GAGs migrate to a similar levelexhibitedbythe chondroitinand dermatansulfate standards(Lane 2, arrowhead) and that they are sensitive to chondroitinase AC digestion (Lane 3). These facts suggest that the great increase in NTF uronic acid concentration induced by means of ␤-D-xyloside is mainly due to chondroitin / dermatan sulfate. B. The electrophoretic pattern of GAGs from NTF in their native state (without previous digestion with protease) from␤-D-xyloside treatedembryos,shows aband of similarmobility toCS (Lane 1, arrow), whereas GAGs from control embryos stay at the same levelat which they were deposited (Lane 2, arrow). These results suggest that in the NTF of control embryos, the CS are arranged to macromolecular aggregates of PG, and that ␤-D-xyloside primes the synthesis of CS free chains. 506 ALONSO ET AL. vesicles is delimited by the apical end of neuroblasts, we can assume that the CS present in the embryonic brain cavitycouldbeaproductofapicalsecretionoftheneuroepithelialcells. Synthesis and secretion of PGs are maximally stimulated by the presence of a ␤-D-xyloside, since this compound avoids the physiological control mechanisms (Galligani et al., 1975). The great increase in CS concentration detected in the NTF of chick embryos treated with ␤-Dxyloside suggests a capacity of PGs secretion far superior to that occurring in normal conditions, suggesting that it mustbeamechanismregulatingthesecretionofCS.Ithas been suggested (Desmond, 1989) that the expansive process of the chick embryo brain could be regulated by the secretory state of the neuroepithelium at each embryonic age. Our results suggest that this regulative mechanism could be based on the rate of CS secretion by the brain neuroblasts. In conclusion, our data indicate that NTF positive pressure, a major morphogenetic mechanism in early brain development, might be generated by an osmotic water-binding phenomenon inside a physiologically sealed braincavityandregulatedbyconcentrationsofsubstances that are osmotically active in NTF, among which CS and itsassociatedmicroion, sodium, could play a key role. ACKNOWLEDGMENTS We thank Sagrario Callejo, Marı´a La´zaro, and Pilar Martı´n for technical assistance and Dra. E. Latorre for constructivecriticismof this manuscript. LITERATURE CITED AvuurZ,GeigerB.Inmunocytochemicallocalizationofnativechondroitin-sulfate in tissues and cultured cells using specific monoclonal antibody. Cell 1984;38:811–822. Barbosa E, Giraldez F, Represa JJ Transepithelial potential and resistance in the otic vesicle of the chick embryo. J.Physiol. 1985; 365,53P. Barbosa E, Moro JA, Gato A, Represa JJ, Giraldez F. Precipitate matherial by CPC inside the otic vesicle of mouse and chick embryos.Arch Biol VII° Symp.Int. Sciences Morphologiques, 1986. BeeleyJG.GlycoproteinandProteoglycanTechniques,Vol.16.Amsterdam:Elsevier, 1985. Bitter T, Muir HM.Amodified uronic acid carbazole reaction. AnalyticalBiochem. 1962;4:330–334. Borgens R, Shi R. Uncoupling histogenesis from morphogenesis in the vertebrate embryo by collapse of the tube potential. Dev. Dyn. 1995;203:456–467. Bradford M. A rapid and sensitive method for the cuantification of microgram quantities of protein utilizing the principle of proteindyebinding.Anal. Biochem.1976; 72:248. BreenM,WeinsteinHG,BlachIJ,BorcherdingMS,SittigRAMethods Carbohidr. Chemi., 1976;7:101–115. Comper WD. Polymer Monographs, Vol.7: Heparin and Related Polysaccharides.NewYork: Gordon and Breach, 1981. Comper WD, Laurent TC. Physiological function conective tissue polysacharides.Physiol. Rev. 1978;58:255–315. Comper WD,Zamparo O. Hydrodynamic properties of connective tissuepolysaccharides. Biochem. J. 1990;269:561–564. CoulombreAJ. The role of intraocular pressure in the development of the chick eye. I. Control of eye size. J. Exp. Zool. 1956;133:221–223. Desmond ME. Reduced number of brain cells in so-called neural overgrowth.Anat. Rec. 1985;212:195–198. DesmondME. Precocious occlusion of the spinalcordin chick embryos resultsinprecociousexpansionofthebrain.Anat.Rec.1989;225:32– 33. Desmond ME, Duzy MJ, Federici BD. Second messenger regulation of occlusion of the spinal neurocoel in the chick embryo. Dev. Dyn. 1993;197:291–306. Desmond ME, Jacobson AG Embryonic brain enlargement requires cerebrospinalfluid pressure. Dev. Biol. 1977;57:188–198. Galligani L, Hopwood J, Schwartz NB, Dofman A. Stimulation of synthesis of free chondroitin sulfate chains by ␤-D-xyloside in culturedcells. J. Biol. Chem. 1975;250:5400–5406. Gato A, Moro JA, Alonso MI, Pastor JF, Represa JJ, Barbosa E. Chondroitin sulfate proteoglycan and embryonic brain enlargement inthe chick.Anat. Embryol. 1993;188:101–106. GibsonKD,DollerHJ,HoarRM.␤-D-xylosidescauseabnormalitiesof growth and development in chick embryos. Nature 1978;273:151– 157. Gibson KD, Segen BJ, Doller HJ Changes in chemical composition of chick embryos treated with a ␤-D-xyloside and Lathyrogen. Teratology1979;19:345–356. Hamburger V, Hamilton HLA series of normal stages in the developmentof the chick embryo. J. Morphol. 1951;88:49–82. Hay ED.Cell Biolgy of Extracellular Matrix, 2nd ed. Boston: Harvard MedicalSchool, 1991. Hotary KB, Robinson KR. The neural tube of the Xenopus embryo maintains a potential difference across itself. Dev. Brain Res. 1990;59:65–73. Jacobson M. Developmental Neurobiology, 3rd ed. New York: Plenum Press,1991. Jelinek R, Pexieder T. The pressure on encephalic fluid in chick embryosbetweenthe2ndand6thdayofincubation.Phisiol.Bhemo. 1961;97–305. Fig. 7. Western blot analysis of NTF of H.H. stage 23 control (CON) and ␤-Dxyloside treated (BDX) chick embryos. Monoclonal antibody CS-56 recognizes a single band, Mr197x10 3, in NTF of control embryos. In␤-Dxyloside treated embryos, thereareno apparent changes except a slight reduction in epitope concentration. TABLE 2. Results of the Intraluminal Pressure, in Control and ␤-D-Xyloside-Treated of H.H. Stage 23 Chick Embryonic Brainsa Control (hPa) ␤-D-xyloside (hPa) Pressure 37.8 ⫾1.5 (13)* 52.6 ⫾2.8 (14)* aMeasurements are mean ⫾standard error. Number of embryosmeasured are inparentheses. *P⬍0.001,by two-tailed Student’stest. 507CHONDROITIN SULFATE AND BRAIN ENLARGEMENT