Synthesis and Metabolism. Malcolm. Davies,1. Gareth. J. Thomas,. Lorna. D. Shewring, and. Roger. M. Mason. M. Davies,. G.J. Thomas,. L.D. Shewring. Institute.
Mesangial Malcolm
Cell Proteoglycans:
Davies,1
Gareth
J. Thomas,
Lorna
Synthesis
D. Shewring,
and
Roger
and
Metabolism
M. Mason
P
M. Davies, G.J. Thomas, L.D. Shewring. phrology, University of Wales College Cardiff Royal Infirmary, Cardiff, Wales R.M. Mason, Department of Cross & Westminster Hospital don, England (J. Am.
Soc.
Nephrol.
1992;
Institute of Neof Medicine,
Biochemistry, Charing Medical School, Lon-
2:S88-S94)
roteoglycans consist of a core protein to which complex linear heteropolysaccharides called glycosaminoglycans (GAG) are covabently linked. These are composed of a characteristic repeating disaccharide unit containing a hexuronic acid, either D-gbucuronic acid or L-iduronic acid, and an amino sugar, either D-glucosamine or D-galactosamine. Sugars are variably N- and 0-sulfated, which further adds to their heterogeneity. Proteoglycans may contain one or more chains of a particular GAG, e.g. chondroitin sulfate (CS), dermatan sulfate (DS), heparan sulfate (HS), heparin, and keratan sulfate. A further class of proteoglycan is recognized in which different types of GAG chains are attached to the same core protein, for example, chondroitin and heparan sulfate, as in syndecan. Hyaluronic acid (HA, also known as hyaburonan), a linear polysaccharide, is also regarded as an “honorary” proteoglycan. It differs from other GAG chains in that it is not covalently linked to protein and is the only GAG lacking sulfate groups. In addition to the GAG chains, the core proteins are substituted with N- and/or 0-linked oligosaccharides, the function of which is uncertain at present. A detailed account of the structure of proteoglycans is beyond the scope of this article, and the reader is directed to several recent comprehensive reviews (1-6). These highly pobyanionic macromolecules bind to other matrix components, such as type IV collagen, baminin, and fibronectin, and thus play an important role in the overall structure of the extracelbular matrix (ECM) and basement membranes (BM). More recently, as diverse cellular responses have been shown to be mediated by the ECM, many functions for proteoglycans have been uncovered. There is compelling evidence that proteoglycans not only participate in the formation and maintenance of the ECM and BM but also regulate cell differentiation and gene expression, migration, adhesion, and probiferation (2,3,5). Furthermore, proteoglycans, particularly HS proteoglycans (HSPG), interact with type IV collagen and laminin to provide the anionic charge barrier of BM and as such play a critical role in endowing the gbomerular basement membrane (GBM) with it properties as a selective filter (6). The importance of GBM-HSPG is further highlighted by the findings that human diabetic nephropathy, human congenital nephrotic syndrome, and models of gbmerular immune complex disease are associated with perturbations of the synthesis and/or the turnover of these molecules leading to their depletion from the GBM (7). Whether mesangial cell proteoglycans are involved in the same gbomerular disorders is as yet ,
ABSTRACT In cultures of human adult glomerular mesangial cells, large chondroitin sulfate proteoglycans (CSPG) and small dermatan sulfate proteoglycans (DSPG) are synthesized. The large CSPG has a core protein, Mr of 400,000 (major) and Mr of 500,000 (minor), and binds
to hyaluronic
The
two
bic
form
acid
to form
large
aggregates.
small DSPGs (Mr of -‘-350,000 and Mr of -‘-200,000) were related to biglycan and decorin, respectively. The majority of these proteoglycans were located in the culture medium, but a hydrophoof the
CSPG
was
extracted
from
the
cell
layer. Mesangial cells in the growing phase synthesized and secreted all three types of proteoglycans, but in cells arrested in G0 by serum deprivation the incorporation
of (35S)sulfate
in CSPG
was
drastically
reduced. In the same cells stimulated to proliferate by replacing the medium with one containing serum, the synthesis of CSPG dramatically enhanced. The synthesis
elevated
in
cells cocultured with cytokines but in contrast significantly reduced when cultured in medium taming hyperglycemic levels of glucose. Finally,
was conpre-
liminary
of CSPG
and
experiments
are
CSPG and DSPG bind vitro. These observations ized
function
of glomerular
low-density
Mesangial lipoproteins,
‘Dr.
Institute
cardiff
M. Davies,
Royal Infirmary,
was
reported
also
that
that
in cellular
processes
disease.
cell, proteoglycans, hyaluronic cell growth, cytokines
of Nephrology, Newport Road.
acid,
University of Wales College of Medicine. Cardiff. CF2 ISZ, Wales, United Kingdom.
1046-6673/021 0-0588$03.00/0 Journal of the American Society of Nephnology Copyright © 1992 by the American society of Nephrology
S88
indicate
to low-density lipoproteins in suggest a possible special-
for proteoglycans
characteristic Key Words:
DSPG
Volume
2-
Supplement
2’
1992
Davies
largely unresolved. In fact, the function of mesangiab cell-derived proteoglycans in normal renal physiology is poorly understood. To address these questions, our aim has been to define those proteoglycans synthesized by mesangial cells and to investigate their metabolism under different physiological conditions.
GLOMERULAR MESANGIAL PROTEOGLYCANS
CELL
Localization The presence of GAG in the mesangium was first reported by Kanwar et at. (8). In electron microscopy studies, these authors showed that a significant number of anionic sites present in the gbomerulus could be removed by perfusion with chondroitin ABC lyase and/or heparitinase, thus demonstrating that both HS and CS/DS GAG chains are present. The experimental data also indicated that the GBM and the mesangiab matrix are compositionally different with regard to their proteoglycan content, HS-GAG being present in both extracellular matrices, whereas CS-GAG is present exclusively in the mesangial matrix. Immunobocabization studies with both pobycbonab and monocbonal antibodies that specifically recognize the protein cores and GAG chains of different proteoglycans confirm the view that CSPG is the predominant proteoglycan within the mesangium and that very little, if any, is present in the GBM (91 1). Several studies also indicate that antibodies against BM-HSPG, although readily staining the GBM, only weakly immunobocalize within the mesangiab matrix (1 0, 1 2). The mesangiab matrix was, however, stained when rats were injected i.v. with antibody against GBM-HSPG (1 3). The immunostaining was weak and only prominent 1 day postinjection, after which it gradually disappeared. This may be taken as evidence that the mesangiab matrix HSPG is immunobogically related to BM-HSPG. On the other hand, the immunostaining could be explained by the clearance of GBM anionic sites to this location (14).
Mesangial CSPG and
Cells DSPG
Synthesize
and
Secrete
Mainly
Cultures of human gbomerular mesangiab cells synthesize and secrete several proteoglycans with CS and DS side chains (Table 1). Indeed, these chains account for the majority of the GAG extracted from the cell layer and the culture medium (15). These molecules fall in two subgroups based on their size, their ability to interact with HA, and their susceptibility to chondroitin ACII byase. High-Molecular-Weight CSPG. The first population is designated a high-molecular weight (HMW)CSPG on the grounds that it is a hydrodynamically
Journal
of the
American
Society
of
Nephrology
et al
barge molecule (Mr of 1 06) and contains GAG chains (molecular mass of 25 kDa), which are completely digested by both chondroitin ABC byase and chondroitin ACII byase. When digested with either of these enzymes, core proteins with molecular mass values of 500 (minor) and 400 kDa (major) were revealed. The basis of this protein core heterogeneity remains to be fully investigated. One explanation is that the mesangiab cell synthesizes two forms of this class of proteoglycan. The possibility that the smaller CSPG is derived from the larger species by limited proteobysis remains a second option. The HMW-CSPG was present in both the culture medium and the cell layer. The latter form was accessible to limited trypsin treatment, had the capacity to bind to octyb-Sepharose, and could be inserted into biposomes. These observations are consistent with the idea that this proteoglycan is intercalated into the cell membrane of mesangial cells. The proteoglycans in our study were extracted under dissociative conditions and are characterized as biochemical entities. However, it must be realized that proteoglycans are present in tissues in association with other components of the ECM such as laminin, fibronectin, and HA. Analysis of the radiolabeled proteoglycan in the media and cell layer under associative conditions showed that a small amount of labeled material ebuted in the excluded volume on a CL-2B column. Characterization of this material indicated the presence of both HA and proteoglycans. These results led us to investigate the mesangiab cell proteoglycans for their hyaluronate-binding capacity. This demonstrated that of the proteoglycans listed in Table 1 only the HMW-CSPG bound to HA. The amount of CSPG that bound to HA accounted for about 75% of that in the cell layer and 60% of that in the medium. Thus, it appears that not all of the mesangiab cell HMW-CSPG possess the capacity to form aggregates with HA. Large CSPG with HA-binding capacity have been isolated from several tissues including bone, skin, and aorta (for a review see reference 2) and are produced by cultured fibrobbasts (1 6) and aortic endothebial and smooth muscle cells ( 1 7). These molecules have similar but not identical biochemical, morphological, and immunological properties to a cartilage proteoglycan cabled aggrecan (18). The N-terminal end of the protein core of aggrecan contains two globular domains (01 and 02) separated by a short peptide. Although these two domains exhibit close homology, only the 01 domain binds HA. cDNA sequence analysis indicates that the 0 1 domain contains two structural motifs, one related to the immunogbobulin fold, the other being a “proteoglycan tandem repeat (PTR)” region consisting of two disulfide-bonded loops. It has been suggested that in aggrecan the PTR region contains the HA-binding re-
S89
Mesangial
TABLE
Cell
I
Proteoglycans
. Inventory
““
of human
Proteoglycan
(1)
(2)
( 1)
HSPG
(2)
b
Adapted from Size expressed
C
The distribution
Thomas as
M
,,
NA, not
(minor
analyzed;
Distribution (% Total)c
GA G Type
CM
CL
500d 400d
25 25
CS CS
7.2
3.4
200 350
45 45
25 25
DS DS
9.8 40.8
ND 5.2
1 500 450
NA NA
65 50
HS HS
2.3 19.2
1.0 4.3
proteoglycan
is expressed
The same ND, not
M, values
were
as a percentage CL, cell chondroitin
obtained
layer. ABC by similar
of the total lyase
yielded
treatment
Hydrophobicity of the CL Form
+
+
+
+
-
-
-
-
-
+
-
+
Comments
Core protein contains HABR; all of the 500 kDa bound to HA; HMW-CSPG were not detected in the CL or CM of quiescent cells labeled with (35S)sulfate The 200and 350 kDa form were identified as decorin and biglycan, respeclively
(35S)GAG two
of the
synthesized
1251-labeled HMW
pool
by the cells
proteins of CSPG
over
a 24-h
of M, of -400,000 extracted
from
cells
labeling (major
prelabeled
period band)
in and
with
(3H)
detected.
gion (HABR) of the molecule. The interaction between the HABR and HA involves a specific interaction with a decasaccharide sequence of HA. This noncovalent interaction is greatly stabilized by additional interactions with a third component, a link protein, which appears to interact with both HA and the HABR in the Gl domain. Other large CSPG, such as the fibroblast proteoglycan known as versican, bind to HA via a similar HABR, but their interaction with a link protein has not been established. In our studies, the binding of mesangial cell HMW-CSPG to HA could be competed for by hyaluronate decasaccharides or larger oligomers. Such oligosaccharide competition is unable to displace aggrecan from HA in link-stabilized native proteoglycan aggregates. The formation of HA-mesangial proteoglycan aggregates was prevented by prior reduction and alkylation of the HMWCSPG, presumably through the unfolding of the PTR region. In a dot-blot assay anti-HABR antibody (monocbonal antibody mAb 1 C6) bound to the reduced protein core of HMW-CSPG, but under the same conditions, an anti-link protein antibody (8A4) did not bind. The above evidence, although preliminary, suggests that the interaction of mesangial cell HMWCSPG with HA is not stabilized by a link protein, and in this respect, it is more similar to endothelial CSPG and versican rather than to aggrecan. Apart from aggrecan, which is viewed as a spacefilling proteoglycan with osmotic properties essential 590
HA Binding
et al. (15). of x103.
of each
band).
. .#{149},.
proteoglycans#{176}
Sizeb
medium containing 15% FOS. CM. culture medium; d Treatment of ‘251-labeled HMW-pool CSPG with 500,000 leucine.
cell
Protein Coreb
>1,000
LMW-CSPG
.
#{149}-.
mesangial
Siz&’
HMW-CSPG
“
.
for articular cartilage function, no physiological roles for the barge CSPG related to versican have been clearly defined. The sequence data for the versican core protein (1 9) indicate that it is different from the aggrecan core protein, despite homologies in the Nand C-terminal regions. Versican does not possess a G2 domain. In the C-terminal region, in addition to a lectin-like domain and a complement regulatory protein-bike region present in aggrecan, it contains two epidermal growth factor domains. Thus, it also has features rebated to the LEC-CAM group of cell adhesion molecules (20). On the basis of this homology, it has been proposed (1 9) that versican (and probably similar molecules) acts as a bridge binding HA in the ECM through the N-terminal and the cell surface via the C-terminal CAM-bike region. Such a unique interaction could play an important role in cell attachment or migration. However, it must be borne in mind that barge CSPG are known to oppose cell adhesion, probably by masking the ROD moieties of the ECM molecubes, which are normally recognized by cell surface receptors (2 1 ). To explain these two apparently contradictory views, it has been proposed that the interaction of the LEC domain of versican with the cell surface represents a weak, transient attachment that merely orientates the cell for subsequent migration (3). In mesangial cells, an interesting finding discussed below is that quiescent cells do not synthesize significant amounts of HMW-CSPG. Volume
2
‘
Supplement
2
‘
1992
#{149} *1
HMW-cSPG 0
10
0 ‘0
Cl)
5
It)
CV)
0 -0.2
0.0
0.2
0.4
0.6
0.8
1.0
Kay Figure 1. Molecular sieve profiles of 35S-labeled CSPG synthesized by human mesangial cells in culture. The total (35S)proteoglycans were extracted from the culture medium, digested with cold nitrous acid to remove all (35S)HS GAG chains, and then chromatographed on a dissociative Sepharose CL-4B analytical column (see reference 15 for full details). Journal
of the
American
Society
of
Nephrology
Cultured mesangial cells maintained for 3 days in medium supplemented with lactalbumin (LH) become arrested in a reversible manner in the G phase. Under these conditions, the synthesis of 35S-labeled proteoglycans is significantly suppressed. The effect of serum on proteoglycan synthesis was reversible. Cultures switched from medium plus 0. 1 % LH to medium with serum (5 to 20%) increased synthetic rates to serum levels, whereas cultures switched from medium containing 20 or 1 0% serum to medium alone exhibited decreased synthetic rates. These changes, whether from medium to serum or vice versa, were largely explained by the appearance (or disappearance) of the HMW-CSPG (Figure 2). The precise nature of the factor(s) in the fetal calf serum (FCS) responsible for the enhanced synthesis of this proteoglycan is not known. The effect of cytokines/growth factors and hormones on the growth of mesangial cells has been investigated extensively (25). These include insulin, parathyroid hormone, 91
Mesangial
Cell
500
Proteoglycans
mL), there was a slight decrease in [35Sjsulfate incorporation, whereas higher doses (>10 ng/mL) caused an increase (40%). TOF-fl also enhanced proteoglycan synthesis, but in our system, the response to this cytokine was lower than that recorded by other workers (24). Finally, we showed that tumor necrosis factor stimulated mesangiab cells to up-regulate their proteoglycan synthesis in a doseand time-dependent manner (26).
a
400
!
INTERACTION OF PROTEOGLYCANS
100
C
600
E a
b
I
Fcs
LH
400
‘0 Cl) It)
CV)
200
20
30
40
50
Fraction
60
70
80
No
Figure 2. The effect of serum on the synthesis of (355)proteoglycans by rat mesangial cells. Cells were labeled with (35S)sulfate for 24 h in the presence of (a) 20% FCS or (b) 0.2% lactalbumin (LH). The medium was removed, and the (35S)proteoglycans were extracted from the cell layer and chromatographed on a dissociative Sepharose CL-4B column. The arrows indicate that the profiles change from pattern (a) to (b) and vice versa depending on the culture conditions.
interleukin1 /3, tumor necrosis factor, platelet-derived growth factor, and TGF-fl, all of which have been reported to affect the growth status of mesangiab cells in culture. In addition, some of these factors have been shown to increase the synthesis of at least one component of the mesangial matrix, and thus, the presence of either one or more of them in serum may explain the above results. To examine this further, quiescent cells were incubated with serum-free medium alone, medium containing several different cytokines, or medium containing 20% serum for 24 h, and the rate of proteoglycan synthesis was determined. The inclusion of FCS stimulated proteoglycan synthesis by 200%. Interleukin1 9 caused a modest dose-dependent increase (33% at 10 ng/mL). The effect of platelet-derived growth factor was complex and possibly biphasic. At lower doses (0.1 to 1.0 ng/
S92
MESANGIAL CELL WITH PLASMA LIPOPROTEINS
Proteoglycans are involved in the pathogenesis of atherosclerosis (27). The evidence for this is based on changes In the composition and content of proteoglycans in the arterial wall with the progress of the disease, the ability of proteoglycan at physiological pH to interact with serum lipoproteins (especially bow-density lipoproteins [LDL]) in the presence of calcium to form insoluble complexes, and the isolation of lipoprotein-proteogbycan complexes from atherosclerotic lesions. Vascular smooth muscle cells are the main source of proteogbycans in the arterial wall. HMW-CSPG Isolated from human aorta, having similar physical properties to mesangial cell CSPG, form insoluble complexes with LDL in vitro (28). Because vascular smooth muscle cells and mesangiab cells are closely related and because of the many parallel pathophysiobogical mechanisms between atherosclerosis and gbomeruboscberosis (29), it is possible that mesangial proteoglycans are involved in the pathogenesis of gbomerular scarring. To explore this possibility, studies have been initiated to investigate the interaction of mesangial cell proteoglycans with human LDL. Mesangiab cell HMW-CSPG and the small DSPG (a mixture of biglycan and decorin) were extracted from the conditioned medium obtained from cultured human mesangiab cells and were purified as described by Thomas et at. (1 5). Both of these proteoglycan fractions bind LDL with high affinity and form insoluble complexes in the presence of calcium (D.C. Wheeler, R. Chana, M. Davies, unpublished data). The binding, which was maximal at 30 mM Ca2, was completely abolished by pretreatment of either proteoglycan fraction with chondroitin ABC lyase. However, we cannot rule out the possibility that the core protein of the large CSPG or either of the small interstitial DSPG is involved in the binding. Our studies comparing the mesangium-derived CSPG and DSPG indicated that the former was more efficient at binding LDL. These preliminary results suggest that the mesangial cell proteoglycans could play a pivotal role in terms of lipid accumulation in gbomeruboscberosis. Moreover, mesangial cells are stimulated to proliferate by LDL (30), and the probiferating mesangial cells produce greater amounts of HMW-CSPG than do quiescent cells.
Volume
2
-
Supplement
2
‘
1992
Davies
METABOLISM OF PROTEOGLYCANS
MESANGIAL
CELL
Animal studies as well as tissue culture experiments indicate that the degradation of proteoglycans is a normal physiological event that ensures a regular turnover of these molecules in the gbomerular ECM. Our studies with an in vivo 35S-babebing system together with an efficient extraction protocol showed that there is a complex turnover pattern of labeled proteoglycans in the gbomerubus consisting of a rapid phase followed by a slower phase (3 1 ,32). The total population of [35SJHSPG had a metabolic half-life (th) of 20 and 60 h in the first and second phases, respectiveby. The turnover of BM-HSPG was even faster and showed t,, of 5 and 20 h, respectively. Although the [35S]CSPG/DSPG molecules were metabolized mitiabby with a t,. of ‘-20 h, this pool showed very little turnover in the second phase. These results taken together with the immunobocalization studies of CSPG and the biochemical studies discussed above infer that the turnover of mesangiab cell proteoglycans differ from that of other proteoglycans in the gbomerubus. This interpretation is supported by recent in vitro pulse-chase studies with human gbmerubar cells. These studies show that the cell-associated 355-babebed proteoglycans of both mesangial and gbomerular epithebiab cells are metabolized via two separate routes: ( 1 ) by limited proteobysis of core proteins and subsequent release into the extracellubar environment where they are neither endocytosed nor metabolized further, and (2) by internalization and complete degradation. However, the overall metabobism of proteoglycans by the mesangiab cell is considerably less active compared with that observed for glomerubar epithelial cells. Thus, with mesangiab cells, 60% of the initial proteoglycans remain associated with the cell layer over a 6-h chase period. This compares with 27% in the epithelial cells.
THE EFFECT MESANGIAL
OF ELEVATED GLUCOSE CELL PROTEOGLYCAN
ON SYNTHESIS
Both the GBM and the mesangial matrix undergo changes in diabetic nephropathy. The GBM is frequently thickened, primarily as the result of the increased deposition of type IV collagen and possibly baminin, although there is also evidence for a reduction in the batter. Some evidence also suggests that the amount of HSPG in the GBM is reduced in diabetes and that this may be causally linked with the microalbuminuria, which marks the onset of nephropathy (33-35). The mesangium shows increased deposition of fibronectin, laminin, and collagens types IV, V. and VI. Recent evidence indicates that in rat mesangial cells maintained over a 4-wk period in culture medium containing 30 mM glucose exhibited enhanced synthesis of fibronectin, baminin, and
Journal
of
the
American
Society
of Nephrology
et al
type IV collagen compared with that of controls cubtured in 1 0 mM glucose (36). We have measured the synthesis of proteoglycans by human mesangial cells maintained in medium containing either 4 mM (normoglycemic) or 30 mM (hyperglycemic) glucose over 14 days. Compared with those grown in normoglycemic conditions, cells grown in high glucose exhibited a marked (40 to 50%) inhibition of proteoglycan synthesis. The synthesis of HA was also greatly reduced (27%). Analysis of the different sulfated proteoglycans indicated an equivalent reduction in the synthesis of each class rather than a specific reduction in a particular molecule. These findings are of interest because they suggest that in the presence of elevated glucose the synthesis of different mesangial matrix macromolecules is affected in different ways.
CONCLUSION
AND
PROSPECTIVES
At the present time, we have only a rough framework for understanding the biochemical events that may be important in regulating the synthesis and turnover of the constituents of gbomerular ECM and how these are perturbed in gbomeruboscberosis. The experiments outlined above describe the different types of proteoglycans synthesized by gbomerular mesangial cells in culture and demonstrate in a functionab way how the metabolism of these complex molecules is modulated by culture conditions aimed at stimulating pathophysiobogic events. Evidence has been presented that a large CSPG could play an important role in the control of mesangial cell probiferation and possibly cell migration. This proteoglycan may be related to versican or to the large CSPG isolated from aorta but further detailed analysis of the amino acid sequence of its core protein is required to confirm this. Future studies of these proteoglycans and their interaction with the other molecules of the gbomerubar ECM will bring us to a better understanding of how growth control of mesangiab cells is integrated within the gbomerulus.
REFERENCES 1 . Hascall VC, Hascall GK: Proteoglycans. In: Hay ED, ed. Cell Biology of Extraceblular Matrix. New York: Plenum; 1981:39-63. 2. Evered D, ed.: Functions of the Proteoglycans. Ciba Foundation Symposium 1 24. Chichester: John Wiley & Sons; 1986. 3. Gallagher JT: The extended family of proteoglycans: Social residents of the periceblular zone. Curr Opinion Cell Biob 1 989; 1 : 1 20 1 -1218. 4. Evered D, Whelan J, eds: The Biology of Hyaluronan. Ciba Foundation Symposium 143. Chichester: John Wiley & Sons; 1989. 5. Ruoslahti E: Structure and biology of proteoglycans. Ann Rev Cell Biol 1988;4:299-355. 6. Farquhar MG, Lemkin MC, Stow JL: Role of
S93
Mesangial
Cell Proteoglycans
proteoglycans in gbomerular function and pathobogy. In: Robinson RR, ed. Nephrobogy. Vol. 1 . New York: Springer-Verlag; 1986:580-600. 7. Maklno H, Kashihara N, Ikeda S, Lelongt B, Kanwar YS: Relevance of proteoglycans in gbmerular matrix pathology. In: Hatano M, ed. Nephrobogy. Vol. 1 . Tokyo: Springer-Verlag; 1991: 864-875. 8. Kanwar YS, Jakubowski ML, Rosenzweig LI: Distribution of sulfated glycosammnoglycans in the gbomerular basement membrane and mesangiab matrix. Eur J Cell Biol 1 983;3 1: 29-295. 9. Couchman JR, Caterson B, Christner JE, Baker JR: Mapping by monocbonab antibody detection of gbycosammnogbycans in connective tissues. Nature (Lond) 1 984;307:650-652. 1 0. Couchman JR. McCarthy KJ, Abrahamson DR, Fine JD, Parry G: Immunological and molecular approaches to the study of basement membrane proteoglycan diversity. Trans Biochem Soc 1989; 18:819-821. 1 1 . McCarthy KJ, Accavitti MA, Couchman JR: Immunological characterization of a basement membrane-specific chondroitin sulfate proteoglycan. J Cell Biob 1989;109:3187-3198. 1 2. van den Heuvel LPWJ, van den Born J, van de Velden TJAM, et al.: Isolation and partial characterisation of heparan sulphate proteoglycan from the human gbomerular basement membrane. Biochem J 1989;264:457-465. 1 3. Miettinen A, Stow JL, Mentone 5, Farquhar MG: Antibodies to basement membrane heparan sulfate proteoglycans bind to the laminae rarae of the gbomerular basement membrane (GBM) and induce subepithelial GBM thickening. J Exp Med 1986; 163:1064-1084. 14. Abrahamson DR, Perry EW: Distribution of intravenousby injected cationized ferritmn within developing gbomerular basement membranes of newborn rat kidneys. Anat Rec 1986;216: 534-543. 15. Thomas GJ, Mason RM, Davies M: Characterisation of proteoglycans synthesized by human adult glomerular mesangial cells in culture. Biochem J 1991;277:81-88. 16. Schmidtchez A, Carlstedt I, Malmstr#{246}m A, Fransson LA: Inventory of human skin fibroblast proteoglycans. Biochem J 1990;265: proteoglycans. Biochem J 1 990;265:289-300. 17. Morita H, Takeuchi T, Suzuki S, et al.: Aortic endothelial cells synthesize a large chondroitin sulphate proteoglycan capable of binding to hyaburonate. Biochem J 1 990;265:6 1-68. 1 8. Hardingham TE, Fosang AJ, Dudhea J: Domain structure in aggregating proteoglycans from cartilage. Trans Biochem Soc 1990; 18:794-796. 19. Zimmermann DR, Ruoslahti E: Multiple domains of the large fibroblast proteoglycan, versican. EMBO J 1989;8:2975-2981. 20. Coombe DR. Rider CC: Lymphocyte homing receptors cloned-a role for anionic polysaccharides in lymphocyte adhesion. Immunol Today
S94
1989; 10:289-291. 2 1 . Ruoslahti E: Proteoglycans in cell regulation. J Biol Chem 1989;264:13369-13372. 22. Fisher LW, Termine JD, Young MF: Deduced protein sequence of bone small proteoglycan I (biglycan) shows homology with proteoglycan II (decorin) and several non-connective tissue proteins in a variety of species. J Biol Chem 1989; 264:4571-4576. 23. Scott JE: Proteoglycan-fibrillar collagen interactions. Biochem J 1988;252:313-323. 24. Okuda 5, Languino LR, Ruoslahti E, Border WA: Elevated expression of transforming growth factor-3 and proteoglycan production in experimental gbomerulonephritis. Possible role in expansion of the mesangial extracellular matrix. J Clin Invest 1990;86:453-462. 25. Williams JD, Davies M. Eicosanoids and cytokines. In: Pusey CD, ed. Immunology of Renal Disease. Dordrecht: Kluwer Academic; 1991: 123- 160. 26. Shewring L, Thomas GJ, Davies M: Proteoglycan synthesis is stimulated by TNF in rat mesangial cells in vitro. Clin Sci 1989;76:S20. 27. Wight TN: Cell biology of arterial proteoglycans. Arteriosclerosis 1 989;9: 1-20. 28. Mires CS, Mourao PAS: Interaction of high mobecular weight chondroitin sulphate from human aorta with plasma low density lipoproteins. Atherosclerosis 1 988;73: 113-124. 29. Diamond JR, Karnovsky MJ: Focal and glomerubosclerosis. Analogies to atherosclerosis. Kidney Int 1988;33:917-924. 30. Coritsidis G, Rifici V, Gupta 5, et at.: Preferential binding of oxidized LDL to rat glomeruli in vivo and cultured mesangial cells in vitro. Kidney Int 1991;39:858-866. 3 1 . Beavan LA, Davies M, Mason RM: Renal gbmerular proteoglycans. An investigation of their synthesis in viva using a technique for fixation in situ. Biochem J 1988;251:41 1-418. 32. Beavan LA, Davies M, Couchman JR, Williams MA, Mason RM. In vivo turnover of the basement membrane and other heparan sulphate proteoglycans of rat gbomerulus. Arch Biochem Biophys 1 989;269:576-585. 33. Groggel GC, Stevenson J, Hovingh P, Linker A, Border WA: Changes in heparan sulfate correlate with increased glomerular permeability. Kidney Int 1988;33:517-523. 34. Deckert T, Feldt-Rasmussen B, Borch-Johnsen K, Jensen T, Kafoed-Enevoldsen A: Albuminuria reflects widespread vascular damage. Diabetobogia 1989;32:219-226. 35. Shimomura H, Spiro RG: Studies on macromolecular components of human glomerular basement membrane and alterations in diabetes: Decreased levels of heparan sulfate proteoglycan and laminin. Diabetes 1987;36:374-381. 36. Ayo SH, Radnik RA, Glass WF, et al.: Increased extracellular matrix synthesis and mRNA in mesangial cells grown in high-glucose medium. Am J Physiob 1991 ;260:F185-F191.
Volume
2’
Supplement
2’
1992
