David A. Carrino

Sulfated proteoglycans in astroglial barriers inhibit neurite outgrowth in vitro

In vivo studies of the roof plate of the spinal cord and midline optic tectum in rodent and the developing subplate in the telencephalon of the chick showed that two glycosaminoglycans, keratin sulfate and chondroitin sulfate, possibly in the proteoglycan form (KS-PG, CS-PG, or KS/CS-PG), were present at times when axons approach closely but do not invade these territories. To address the question of whether KS/CS-PG actively inhibits growth cone elongation and to determine which component(s) of the proteoglycan may be critical to this phenomenon, we used a technique employing nitrocellulose-coated petri dishes onto which stripes of various purified macromolecules were attached. Isolated E9 chick dorsal root ganglia were grown on lanes of KS/CS-PG in alteration with lanes of the growth-promoting molecule laminin (LN). Neurite outgrowth was abundant along stripes of LN. In contrast, upon encountering a stripe containing KS/CS-PG, neurites either stopped abruptly or turned and traveled along the KS/CS-PG stripe border. The effect was dependent upon the concentration of the proteoglycan with intermediate concentrations producing intermittent patterns of crossing. We mixed LN with the KS/CS-PG, where the LN was in concentrations which alone support outgrowth, and observed that the KS/CS-PG was still inhibitory when such a growth-promoting molecule was present. A 10-fold higher concentration of LN was able to overcome the inhibitory effect of the KS/CS-PG. These results suggest that the interaction of inhibitory and growth-promoting molecules can interact to produce a wide spectrum of neurite patterns ranging from complete inhibition to totally unimpeded outgrowth. Selective enzymatic removal of the KS or CS from the KS/CS-PG permitted various degrees of neurite outgrowth to occur across the previously inhibitory lanes, and digestion of both glycoaminoglycan moieties, leaving only the protein core of the molecule, resulted in a complete lack of inhibition. These assays demonstrated that KS/CS-PG is inhibitory to embryonic dorsal root ganglia neurites in vitro and that complete inhibition requires contributions from both KS and CS moieties.

Immunochemical and Mechanical Characterization of Cartilage Subtypes in Rabbit

Cartilage is categorized into three general subgroups, hyaline, elastic, and fibrocartilage, based primarily on morphologic criteria and secondarily on collagen (Types I and II) and elastin content. To more precisely define the different cartilage subtypes, rabbit cartilage isolated from joint, nose, auricle, epiglottis, and meniscus was characterized by immunohistochemical (IHC) localization of elastin and of collagen Types I, II, V, VI, and X, by biochemical analysis of total glycosaminoglycan (GAG) content, and by biomechanical indentation assay. Toluidine blue staining and safranin-O staining were used for morphological assessment of the cartilage subtypes. IHC staining of the cartilage samples showed a characteristic pattern of staining for the collagen antibodies that varied in both location and intensity. Auricular cartilage is discriminated from other subtypes by interterritorial elastin staining and no staining for Type VI collagen. Epiglottal cartilage is characterized by positive elastin staining and intense staining for Type VI collagen. The unique pattern for nasal cartilage is intense staining for Type V collagen and collagen X, whereas articular cartilage is negative for elastin (interterritorially) and only weakly positive for collagen Types V and VI. Meniscal cartilage shows the greatest intensity of staining for Type I collagen, weak staining for collagens V and VI, and no staining with antibody to collagen Type X. Matching cartilage samples were categorized by total GAG content, which showed increasing total GAG content from elastic cartilage (auricle, epiglottis) to fibrocartilage (meniscus) to hyaline cartilage (nose, knee joint). Analysis of aggregate modulus showed nasal and auricular cartilage to have the greatest stiffness, epiglottal and meniscal tissue the lowest, and articular cartilage intermediate. This study illustrates the differences and identifies unique characteristics of the different cartilage subtypes in rabbits. The results provide a baseline of data for generating and evaluating engineered repair cartilage tissue synthesized in vitro or for post-implantation analysis.

Substrate-bonded hyaluronic acid exhibits a size-dependent stimulation of chondrogenic differentiation of stage 24 limb mesenchymal cells in culture☆

Extracellular matrix molecules including glycosaminoglycans have been implicated in several differentiative and morphogenetic processes including cell aggregation and migration. Previous reports have shown that plating of stage 24 limb mesenchyme cells onto hyaluronic acid (HA) bonded to the culture substrate causes an increase in the number of cells exhibiting chondrogenesis. This increased chondrogenesis is now shown to be dependent upon the source of the HA. When limb mesenchymal cells are plated onto HA from bovine vitreous humor, human umbilical cord, or large molecular weight HA (Healon), increased chondrogenesis is observed only on the bovine vitreous humor HA. Unsulfated chondroitin, which has a structure and charge density similar to those of HA, is capable of enhancing chondrogenesis, while cells plated onto sulfated glycosaminoglycan substrates are indistinguishable from controls. The evidence in this report suggests that the differentiation response is related to the molecular size of the HA bound to the culture substrate. Healon and human umbilical cord HA are ineffective because their molecular weight is too large, while smaller HA derived from these larger molecules or normally present in bovine vitreous humor preparations stimulates the chondrogenic differentiation of stage 24 limb mesenchymal cells in culture. The most active size class of HA elutes from a Sepharose CL-2B column with a Kav between 0.6 and 0.7 and, thus, has a molecular weight of approximately 200,000-400,000. These observations reinforce the hypothesis that local cues have an informational effect on the differentiation of chick limb mesenchymal cells.

Partial biochemical and immunochemical characterization of avian eggshell extracellular matrices

There is evidence to suggest that extracellular matrix molecules, such as proteoglycans, are involved in the regulation of mineral deposition in calcifying tissues. One mineralizing system which is characterized by extremely rapid mineralization is the hen eggshell. This eggshell consists of a pair of nonmineralized eggshell membranes subjacent to the calcified eggshell proper; the eggshell proper is organized into palisades (columns) of mineralized matrix separated by pores. Between the membranes and the shell proper are compacted foci of tissue called mammillary knobs, which are thought to be sites where mineralization is initiated. Previous work from this laboratory has shown the presence of types I, V, and X collagen in the shell membranes. To address the question of the possible role of proteoglycans and glycosaminoglycans in mineralization of the eggshell, two approaches were used. First, immunohistochemistry was performed with monoclonal antibodies to various proteoglycan and glycosaminoglycan epitopes. This analysis indicates that different glycosaminoglycans are localized to discrete regions within the eggshell. Dermatan sulfate is present within the matrix of the shell proper and, to a lesser extent, the mammillary knobs and the outer portion of the shell membranes. In contrast, keratan sulfate is found in the shell membranes and prominently in the mammillary knobs. Interestingly, different keratan sulfate antibodies immunostain distinct regions of the eggshell, which suggests that various types of keratan sulfate are distributed differently. The second approach utilized was to extract the eggshell membranes and recover anionic molecules by anion-exchange chromatography. This resulted in the extraction of material which was recognized by antibodies to keratan sulfate, but not to chondroitin sulfate. This material was very large, as evidenced by its elution in the void volume of a Sepharose CL-2B column. The large size may be due to the extensive cross-links known to occur in the eggshell. If eggshell membranes are extracted at elevated temperature, the material recovered is of much smaller size. These results indicate that molecules recognized by antibodies to glycosaminoglycans are present in the eggshell, and their localized distribution relative to the calcified matrix suggests that they may be involved in the regulation of mineral deposition.

Dermatan Sulfate Proteoglycans from the Mineralized Matrix of the Avian Eggshell

The eggshell of the chicken is a useful model to study matrix components which affect biomineralization. As an extension of our previous immunohistochemical work which suggested the presence of dermatan sulfate proteoglycans in the mineralized region of the eggshell, a study was undertaken to characterize these molecules biochemically. After demineralization with HCl and extraction with 4 M guanidinium chloride containing protease inhibitors, the extract was partitioned by anion exchange chromatography. Step elution with 0.25 M and 1.0 M sodium chloride resulted in the generation of two fractions, both of which contain chondroitinase-sensitive proteoglycans with molecular weights estimated at 200,000 by gel electrophoresis. The proteoglycans in each fraction have core proteins with molecular weights of approximately 120,000 and glycosaminoglycans with average molecular weights of 22,000. Based on differential sensitivity to chondroitinase ABC and AC II, these glycosaminoglycans contain a small proportion of dermatan sulfate. The disaccharide compositions of these glycosaminoglycans differ for the proteoglycans eluted with 0.25 M and 1.0 M sodium chloride. Those eluted with lower sodium chloride are enriched in unsulfated chondroitin and have much more 4-sulfated than 6-sulfated disaccharides; those eluted with 1.0 M sodium chloride contain primarily 4-sulfated disaccharides, a small amount of 6-sulfated disaccharides, and less unsulfated disaccharides than the proteoglycans eluted with 0.25 M sodium chloride. The large difference in the proportions of unsulfated chondroitin may be the reason for the elution at different sodium chloride concentrations. Both of the anion exchange column fractions contain other proteins in addition to the proteoglycans. These proteins are not separated from the proteoglycans by a second anion exchange column or by molecular sieve chromatography under dissociative conditions. Of particular interest is the observation that the eggshell proteoglycans and their core proteins are recognized by a monoclonal antibody which recognizes an epitope on the core protein of avian versican. This suggests that, in spite of the large differences in the sizes of the core proteins of versican and the eggshell proteoglycans, these core proteins share some homology. Because anionic molecules are thought to be important regulators of biomineralization, and because preparations like those analyzed in this study have been shown to influence in vitro calcium carbonate crystallization, the eggshell proteoglycans may play a role in eggshell mineralization.

Pericellular Versican Regulates the Fibroblast-Myofibroblast Transition: A ROLE FOR ADAMTS5 PROTEASE-MEDIATED PROTEOLYSIS*

The cell and its glycosaminoglycan-rich pericellular matrix (PCM) comprise a functional unit. Because modification of PCM influences cell behavior, we investigated molecular mechanisms that regulate PCM volume and composition. In fibroblasts and other cells, aggregates of hyaluronan and versican are found in the PCM. Dermal fibroblasts from Adamts5−/− mice, which lack a versican-degrading protease, ADAMTS5, had reduced versican proteolysis, increased PCM, altered cell shape, enhanced α-smooth muscle actin (SMA) expression and increased contractility within three-dimensional collagen gels. The myofibroblast-like phenotype was associated with activation of TGFβ signaling. We tested the hypothesis that fibroblast-myofibroblast transition in Adamts5−/− cells resulted from versican accumulation in PCM. First, we noted that versican overexpression in human dermal fibroblasts led to increased SMA expression, enhanced contractility, and increased Smad2 phosphorylation. In contrast, dermal fibroblasts from Vcan haploinsufficient (Vcanhdf/+) mice had reduced contractility relative to wild type fibroblasts. Using a genetic approach to directly test if myofibroblast transition in Adamts5−/− cells resulted from increased PCM versican content, we generated Adamts5−/−;Vcanhdf/+ mice and isolated their dermal fibroblasts for comparison with dermal fibroblasts from Adamts5−/− mice. In Adamts5−/− fibroblasts, Vcan haploinsufficiency or exogenous ADAMTS5 restored normal fibroblast contractility. These findings demonstrate that altering PCM versican content through proteolytic activity of ADAMTS5 profoundly influenced the dermal fibroblast phenotype and may regulate a phenotypic continuum between the fibroblast and its alter ego, the myofibroblast. We propose that a physiological function of ADAMTS5 in dermal fibroblasts is to maintain optimal versican content and PCM volume by continually trimming versican in hyaluronan-versican aggregates.

Versican in human fetal skin development

 The extracellular matrix of human fetal skin differs substantially from that of adult skin. Fetal skin contains sparse amounts of fibrillar collagen enmeshed in a highly hydrated amorphous matrix composed of hyaluronan and sulfated proteoglycans. Both fetal and adult skin contain two major interstitial proteoglycans that are extracted by chaotrophic agents and detergents. These are the large chondroitin sulfate proteoglycan versican and the small dermatan sulfate proteoglycan decorin. For this study, proteoglycans extracted from fetal and adult skin were compared on Western blots to determine the relative amounts of versican. Decorin present in the same samples provided an internal standard for these studies. Fetal skin differed from adult skin in that it contained a significantly higher proportion of versican than did adult skin. Immunohistochemical studies compared early-fetal with mid-fetal skin and found that versican was a significant component of the interstitial extracellular matrix at both of these stages of skin development. However, by the mid-fetal period, interstitial versican became restricted to the upper half of the dermis, although versican also continued to be highly expressed around hair follicles, glands, and vasculature in the lower half of the dermis. Fetal skin extracts differed from an adult skin extract by the presence of a 66-kDa protein immunologically related to versican and by the absence of a 17-kDa core protein of a proteoglycan related to decorin. Both of these molecular species may represent degradation products of their respective proteoglycans. Monoclonal antibodies which detect epitopes in native chondroitin sulfate glycosaminoglycan chains recognized versican extracted from fetal skin. However, the tissue distribution of these antigens did not entirely conform to that for versican core protein, suggesting that versican in different regions of the skin may be substituted with glycosaminoglycan chains with different microchemistries. The results of these studies indicate that human fetal skin is structurally different from adult skin in terms of both the distribution and the composition of the large, aggregating chondroitin sulfate proteoglycan versican.

Structural Domains in Chondroitin Sulfate Identified by Anti-Chondroitin Sulfate Monoclonal Antibodies. Immunosequencing of Chondroitin Sulfates

Monoclonal antibodies have been developed that recognize epitopes in native chondroitin sulfate chains. One of these antibodies, CS-56, reportedly recognizes chondroitin 4- and 6-sulfates. However, this antibody, and four other anti-chondroitin sulfate antibodies, 4C3, 4D3, 6C3 and 7D4, do not recognize epitopes in chondroitin sulfate chains from Swarm rat chondrosarcoma proteoglycan, an indication that native chondroitin sulfate epitopes are more structurally complex than the standard 0-, 4-, and 6-sulfated disaccharide repeats that constitute the backbone of chondroitin sulfate chains. A series of limited chondroitinase digestions was performed on the large aggregating proteoglycan monomer extracted from embryonic chick chondrocyte cultures to identify the digestion parameters required to release the different native chondroitin sulfate epitopes. Some epitopes were more accessible to enzymatic digestion than other epitopes. The approximate location of epitopes was determined by measuring the size of undigested oligosaccharides retained on the core protein following a limited digestion, and correlating this with the level of immunoreactivity for the different antibodies. These analyses identified the locations of three different antigenic domains. Domain 1 resides at the linkage region and contains epitopes for two of the five antibodies, and a portion of the epitopes for a third antibody. Domain 2 lies in the interior of the chain and contains epitopes for three of the five antibodies. Domain 3 resides at the non-reducing terminus and does not contain epitopes for any of the anti-chondroitin sulfate antibodies used in this study. These results indicate that specific native chondroitin sulfate epitopes are non-randomly distributed within the linear framework of chondroitin sulfate chains.

The Avian Eggshell Extracellular Matrix as a Model for Biomineralization

The avian eggshell is a complex, extracellularly assembled structure which contains both mineralized and non-mineralized regions. The composition of the hen eggshell organic matrix was examined by immunohistochemistry with antibodies to different extracellular matrix molecules. Type I collagen is found in the shell membranes, but only after treatment of the tissue sections with pepsin. When incomplete eggshells are removed from the oviduct and immunostained, type I collagen can be detected in the shell membranes without pepsin treatment. The shell membranes, which are non-mineralized, also contain type X collagen, and this immunostaining does not require pepsin treatment. The occurrence of type X collagen in the shell membranes is surprising, since this collagen has not been found in any tissue other than hypertrophic cartilage. Immunostaining for various glycosaminoglycans shows the presence of keratan sulfate and dermatan sulfate. Several different antibodies to keratan sulfate stain different regions of the eggshell; one keratan sulfate epitope is prominent in the calcium reserve assemblies. Dermatan sulfate staining is very intense in the palisade region. Demineralized matrix from the palisade region was extracted with guanidine and fractionated by ion exchange chromatography. A approximately 200-kDa dermatan sulfate proteoglycan is found in these extracts, along with a number of protein components. This preparation was tested for its ability to affect calcium carbonate crystal formation in vitro. Pieces of demineralized shell membranes were used as a substrate for crystal formation and various amounts of the palisade matrix dermatan sulfate proteoglycan preparation were added to the solution from which the crystals were formed. This material causes a concentration-dependent change in crystal morphology to one in which the crystals are smaller and more rounded, which more closely approximates the crystals normally observed in eggshells. These results suggest that the dermatan sulfate proteoglycans may be important in modulating crystal morphology in the hen eggshell and correlate with mineralization-modulating biomolecules from other calcified tissue, which are generally anionic.

Generation of a monoclonal antibody against avian small dermatan sulfate proteoglycan : immunolocalization and tissue distribution of PG-II (decorin) in embryonic tissues

Chick embryonic skeletal muscle synthesizes three major types of proteoglycans: large chondroitin sulfate proteoglycans, small dermatan sulfate proteoglycans and small heparan sulfate proteoglycans. A monoclonal antibody has been raised which recognizes the small dermatan sulfate proteoglycan. Immunoblot analysis of a partially purified preparation of skeletal muscle proteoglycans indicates that the antibody reacts with a molecule which migrates with an estimated Mr of 100,000. Prior treatment of the proteoglycans with chondroitinase results in immunostaining of a species of estimated Mr 45,000. These values for the intact proteoglycan and its core protein suggest that the antibody is directed against a proteoglycan of the PG-II or decorin class. Immunohistochemistry indicates a widespread distribution of the proteoglycan, which is localized in connective tissue septa of skeletal and cardiac muscle, dermis, tendon, bone, perichondrium and cornea. Immunoblot analysis of the proteoglycan core proteins from these tissues demonstrates that the antibody recognizes the same 45,000-dalton band in each tissue. The widespread tissue distribution is also consistent with the antibody being directed against an epitope of PG-II. Neither the glycosaminoglycan chains nor N-linked oligosaccharides are required for reactivity and the antibody cross-reacts with other avian material, but not mammalian. This antibody, which has been designated CB-1, reveals developmental stage-specific changes in the deposition of PG-II in embryonic limb bud and skeletal muscle.