University of Southern California Dissertations and Theses

The inability of collagen cross-linking to influence the insolubilization of proteodermatan sulfate and a sialo-glycopeptide

(USC Thesis Other)

The inability of collagen cross-linking to influence the insolubilization of proteodermatan sulfate and a sialo-glycopeptide

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content THE INABILITY OF COLLAGEN CROSS-LINKING
TO INFLUENCE THE INSOLUBILIZATION OF
PROTEODERMATAN SULFATE AND A SIALO-GLYCOPEPTIDE
by
Paul David Benya
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Biochemistry)
June 1973
IN F O R M A T IO N TO USERS
This dissertation was produced from a microfilm copy of the original document.
While the most advanced technological means to photograph and reproduce this
document have been used, the quality is heavily dependent upon the quality of
the original submitted.
The following explanation o f techniques is provided to help you understand
markings or patterns which may appear on this reproduction.
1. The sign or "target" for pages apparently lacking from the document
photographed is "Missing Page{s)'\ If it was possible to obtain the
missing page(s) or section, they are spliced into the film along with
adjacent pages. This may have necessitated cutting thru an image and
duplicating adjacent pages to insure you complete continuity.
2. When an image on the film is obliterated with a large round black
mark, it is an indication that the photographer suspected that the
copy may have moved during exposure and thus cause a blurred
image. You will find a good image of the page in the adjacent frame.
3. When a map, drawing or chart, etc., was part of the material being
‘ photographed the photographer followed a definite method in
"sectioning" the material. It is customary to begin photoing at the
upper left hand corner of a large sheet and to continue photoing from
left to right in equal sections with a small overlap. If necessary,
sectioning is continued again — beginning below the first row and
continuing on until complete.
4. The majority of users indicate that the textual content is of greatest
value, however, a somewhat higher quality reproduction could be
made from "photographs" if essential to the understanding of the
dissertation. Silver prints of "photographs" may be ordered at
additional charge by writing the Order Department, giving the catalog
number, title, author and specific pages you wish reproduced.
University Microfilms
300 North Zeeb Road
Ann Arbor, Michigan 48106
A Xerox Education Company
(f'V i
fc '/ !
¥;■
| 73-31,631
| BENYA, Paul David, 1945-
THE INABILITY OF COLLAGEN CROSS-LINKING
TO INFLUENCE THE INSOLUBILIZATION OF
PROTEODERMATAN SULFATE AND A SIALO-
| GLYCOPEPTIDE.
University of Southern California,
Ph.D., 1973
Biochemistry
University Microfilms, A XEROX Company, Ann Arbor, Michigan
^Copyright by
Paul David Benya
1973
THIS DISSERTATION HAS BEEN M ICROFILMED EXACTLY AS RECEIVED.
UNIVERSITY OF SO U TH ERN CALIFORNIA
T H E G R A D U A T E SCHO O L.
U N IV E R S IT Y PARK
LO S A N G E L E S , C A L IF O R N IA 9 0 0 0 7
This dissertation, written by
PauJ__Day id Benya..................................................
under the direction of h.).s... Dissertation Com
mittee, and approved by all its members, has
been presented to and accepted by The Graduate
School, in partial fulfillm ent of requirements of
the degree of
D O C T O R O F P H IL O S O P H Y
Dean
Date.
DISSERTATION COMMITTEE
ACKNOWLEDGMENT
This dissertation is dedicated to my parents,
Margaret and Stephen Benya, fox their love and continued
support of my academic education, and to my grandmother,
Louetta Borrmann, for her early support of my scientific
curiosity.
My intellectual, emotional and professional devel
opment has been greatly enhanced by many conversations
with Paul Barnes, Ken Berger, Richard Croissant, Susan
Croker, Brent Feldman, Don Furuto, Mel Golditch, Mike
Schneir, and Hal Slavkin. I would especially like to
thank Mike Schneir for permitting me the freedom to
follow my own lead in developing this dissertation
research, Hal Slavkin and Mike Schneir for their con
tinued effort to provide travel to scientific meetings
and to provide exposure to the scientific community, and
Don Furuto for his interest in my work.
I would also like to thank the entire staff of the
Department of Biochemistry, School of Dentistry, Univer
sity of Southern California for the concern and under
standing they expressed on my behalf. And I gratefully
acknowledge the use of the complete facilities of this
department.
ii
Financial Support for my graduate training was
provided by a Traineeship from the National Institutes
of Health (Department of Biochemistry, School of Medicine,
University of Southern California) and a Pre-doctoral
Fellowship from the National Institute of Dental Research
(Grant #DE-4286l-03).
iii
TABLE OF CONTENTS
Chapter
I. INTRODUCTION ............................. 1
Proteoglycans ................................ 2
Glycoproteins ................................ 6
The Problem ................................ 7
Specific Introduction ...................... 9
II. MATERIALS AND M E T H O D S ..................... . . 12
Materials .................................. 12
Analytical Methods ........................... 13
Experimental Procedures ................... 18
III. RESULTS 31
Sialo-glycopeptide (SGPep) ................. 38
Proteodermatan Sulfate (PDS ) ............... 44
Isotope Incorporation ...................... 50
Collagen Solubility Transitions .......... 52
PDS Insolubilization and Degradation . . . 58
SGPep Insolubilization ...................... 60
IV. DISCUSSION ............................. 62
SGPep ..................................... 65
PDS ..................................... 72
Conclusion..................................... 76
REFERENCES ..................................... 78
iv
LIST OF TABLES
Table Page
I. Distribution of Collagen and Sialic Acid . . 34
II. Comparison of "Heteropolysaccharides" . . 42
III. Characteristics of CTAB-MgCl2 Peaks . . 48
IV. Proteodermatan Sulfate: Total Protein and
ProteiniUronate Ratio .................... 49
V. Hydroxyproline Specific Activity ............. 53
14
VI. Distribution of Total C-Hydroxyproline
(Percent) 54
VII. Insolubilization of SGPep and PDS; Isotope
Incorporation into C E ................. 59
v
LIST OF ILLUSTRATIONS
Figure Page
1. Experimental Protocol ........................... 20
2. General Extraction Scheme ....................... 22
3. Processing CE 25
4. Collagen Extraction Kinetics .................. 33
5. Sephadex G-75 Chromatography of Clarified CE 37
6. Sephadex G-50 Chromatography of the G-75
Retarded Fraction; Characterization of
SGPep 40
7. MgClp Elution of the CTAB-Precipitated G-75
Void Volume Fraction; Isolation of PDS . . 46
vi
SELECTED ABBREVIATIONS
BM Basement membrane.
CTAB Cetyl-trimethylammonium bromide.
DS Dermatan sulfate.
PCS Proteochondroitin sulfate.
PDS Proteodermatan sulfate.
sGP Structural glycoprotein.
SGPep Sialo-glycopeptide; specifically as isolated
and characterized in this dissertation.
vii
CHAPTER I
INTRODUCTION
Extreme variation in the architecture of extra
cellular collagen matrices from tissue to tissue has
fascinated many connective tissue researchers. The sig
nificance and complexity of how a biological system con
structs such a variety of matrices is apparent by compar
ing the collagen structure and the functional aspects of
several connective tissues: the large, wandering fiber
bundles of skin, with its tough, elastic containment and
protection functions; the small, tightly packed fibers
oriented parallel to a line of stress, the resistance to
which represents the function of tendon; and the slowly
rotating series of orthogonal arrays of fine fibers that
constitute the matrix of cornea, with its functional trans
parency. The early assumption that all tissue collagens
were essentially the same stimulated research concerning
the noncollagen molecules of different connective tissues.
It was hoped that the variation of these molecules might
be responsible for the diversity of the collagen matrices.
If this hypothesis was valid, and the mechanisms involved
understood, it might be possible to alleviate or prevent
the dysfunction (scarring, corneal opacities) generated
during "repair" by the deposition of an inappropriate
2
matrix. In addition, insight might he gained into the
concept of biological self-assembly.
Two types of noncollagen connective tissue compo
nents, easily distinguished from collagen and possessing
theoretical capacity to interact with collagen, are the
glycoproteins ("mucoproteins" ) and the glycosaminoglycans
or their native covalent complexes with protein, the pro
teoglycans. Their correlations with connective tissue
fiber architecture are exemplified by the models of Meyer,
for glycosaminoglycans (l), and Robert, for glycoproteins
(2). Although useful, these models are based on indirect
evidence, and the function of these molecules in terms of
collagen architecture has not yet been demonstrated. In
deed, most suggestions of collagen-association stem from
parallel tissue extractability and in vitro recombination
experiments. Both approaches offer very little specific
ity for determining the in vivo situation. Direct demon
stration of association with collagen by electron micro
scopic localization may not be possible or specific enough
in some cases. And unless a covalent linkage with collagen
is involved, one may have to rely on the summation of in
creasingly defined, indirect experimentation.
Proteoglycans
The literature prior to 1968 concerning glycosamino-
glycan-collagen interactions has been admirably reviewed
"by Jackson and Bentley (3). In general, the pertinent
literature can he divided into two groups: references
dealing with the extractibility of proteoglycans or
glycosaminoglycans, reviewed by Hallen (4); and those
concerning in vitro associations between isolated glyco
saminoglycans or proteoglycans and collagen (3)* The
conclusion of the extraction literature is that the more
severe the conditions required for extraction of a par
ticular glycosaminoglycan. or proteoglycan the more
likely that it is associated with the insoluble collagen
matrix, Following this logic for skin, there is some
possibility that proteochondroitin sulfate (PCS) is
associated with collagen since most of it was only ex
tracted with 0.5 or 1.0 M salt (5,6,7) and some remained
in the residue (8). Proteodermatan sulfate (PDS), how
ever, is almost certainly associated with collagen since
it was only slightly extracted by 1.0 M salt or even 6 M
urea at 4°. The combined extraction power of 6 M urea
and 60° was required to bring PDS into solution by de
naturing the salt-residue (presumably the insoluble
collagen)(8 ).
The early in vitro studies of glycosaminoglycan-
collagen interactions do not provide significant informa
tion about the in vivo situation because they were not
performed under physiological conditions of pH and ionic
strength, or they were performed with glycosaminoglycans,
a degraded form of the in vivo proteoglycan. More recent
studies of in vitro collagen fiber formation at 37° in
the presence of proteoglycans must be interpreted in the
light of the following observations:
1. Collagen will form native fibrils under physi
ological conditions of pH and ionic strength in the absence
of other macromolecules (9,10).
2. The collagen-collagen interactions involved in
in vitro fiber formation are very sensitive to changes in
pH, ionic strength, and temperature (11).
3. The rapid growth phase of fibers in vitro is
preceded by a lag phase during which collagen nuclei are
formed (11).
4. A nuclei-forming fraction of collagen appears
to exist (8,11,12).
5. Compared to normal salt- or acid-soluble col
lagen, increased aggregation properties have been observed
for a collagen of high molecular weight (13).
The interactions of PCS and PDS with collagen in
vitro are nearly the same when similar experiments are
performed (8,12,14). Each was capable of forming an
amorphous precipitate with 50-60 percent of the collagen
present when incubated at 4°. A few 550-650 2. banded
fibers were observed in this precipitate. Instantaneous
precipitation with collagen was also observed at 37° and
resulted in prolongation of the lag phase. It appears
that both the initial 37° precipitate and the 4° precip
itate contain the nuclei-forming collagen fraction. Both
initial precipitates and the whole starting material form
640 5£-banded fibers upon incubation at 37°. These fibers
are more stable to temperature reversal (37° to 4°) than
fibers formed in the absence of PDS. The effect of in
creased ionic strength on the stability of collagen-
pro'teoglyc an fibers formed at 37° has not been determined.
However, the lack of influence of 0.2 M ionic strength on
PCS inhibition of fiber growth (14) suggests that the
interaction between PCS and collagen is stable under these
conditions, although the 4° precipitate is not. It should
be noted that the associations between protein-free chond-
roitin sulfate or DS and intact, insoluble collagen, pres
ent under physiological conditions, were eliminated com
pletely by solutions of 0.4 M ionic strength (15). On the
basis of this and similar evidence, Jackson and Bentley
(3)> Mathews (15), Obrink (16), and Toole (S,l4) have con
cluded that interaction between proteoglycans and collagen
is ionic in nature and that proteoglycans should be influ
ential during in vivo fiber formation. More recent evidence
suggests a minor role for PCS during in vivo fiber forma
tion and provides, for the present, a distinction between
PCS and PDS in this regard. Specifically, the presumed
in vivo form of PCS, the complex of proteoglycan subunits
with the glycoprotein link fraction was shown to have no
effect on in vitro fiber formation, while proteoglycan
subunit alone appeared slightly more inhibitory to nucle-
ation than earlier preparations of PCS (17).
Glycoproteins
The earlier literature concerning "mucoprotein"-
collagen interactions has been reviewed by Jackson and
Bentley (3), and may be summarized as follows: (l) Asso
ciation with collagen was implied by discrepancies between
the amino acid and carbohydrate compositions of soluble
collagen and those of the insoluble residue; (2) "Muco-
proteins" could be extracted from the insoluble residues
of several connective tissues; (3) Insoluble collagen
could be dispersed or solubilized by "carbohydrate-spe-
cific" agents. In the latter group, either the studies
were not reproducible or the specificity of the agent was
not accurately determined.
More recently, a large group of studies has been
published that describe a class of insoluble molecules,
the "structural glycoproteins" (sGP)(18,19,20, review 21).
These glycoproteins have been operationally defined as
being insoluble after removal of the collagenous matrix
by collagenase or treatment with hot TCA and subsequently
soluble in urea or urea-mercaptoethanol. In some cases
no attempt was made to solubilize without degradation, so
the data concern the nature of the glycopeptides obtained
by proteolytic digestion (22,23). It is unclear whether
the authors intend that sGP be considered associated with
collagen (influence the matrix in some fashion (2,22)) or
with themselves (to form an independent structural matrix
(18,20 )).
The most direct suggestion of glycoprotein associ
ation with the insoluble collagen matrix is provided by
the work of Adelmann (24). In this case, material was
solubilized from a 2.0 M NaCl-extracted residue of rat
skin with the aid of a highly-purified preparation of
bacterial collagenase. The major component of this extract
was an "acid-protein" that was immuno-electrophoretically
distinct from rat serum and was present in smaller amounts
in each of the previous, lower ionic strength, salt-
extracts. This acid-protein was also reported to have
a promoting influence on collagen fiber formation.
The Problem
In general, the literature suggests that a proteo
glycan or glycoprotein is associated with collagen if it:
(l) remains insoluble after all of the soluble forms of
collagen have been extracted; (2) can be shown to bind to
collagen or influence fiber formation in vitro; (3) can
be solubilized by specific collagenase digestion of the
insoluble collagen matrix. It is curious that no concern
has been expressed regarding the lack of demonstrable
0.5 M salt-insoluble in vitro aggregates of proteoglycan
or glycoprotein and collagen. How is the insolubility of
molecules, presumably associated with collagen in vivo,
to be explained if interactions between PCS and insoluble
collagen are destroyed by 0.35 M ionic strength (15), and
even interactions between tropocollagen molecules (unless
mediated by covalent cross-links) are destroyed by 0.5 M
salt?
One explanation for the observed insolubility of
noncollagen molecules has been suggested by Jackson and
Bentley (25) in regard to ,lmucoproteins, , :
It would appear possible that, since the collagen
is laid down in a matrix containing these pro
teins, that they become 'trapped' in the collagen
fiber lattice as this increases in diameter. . . .
The conditions necessary to extract them completely
therefore would be those necessary to take the
collagen into solution. . . .
If collagen insolubilized by intermolecular covalent cross
links (0.5 M salt-insoluble) is included in the definition
of "collagen," this concept of entrapment could explain the
insolubility of collagen-associated molecules.
The exciting aspect of entrapment is that molecules
only weakly associated with collagen in vivo could ulti
mately and indirectly become insoluble in 0.5 M salt.
Each molecule possessing an affinity for collagen would
9
be actively involved in its own insolubilization, while
its final insolubility would depend on the formation of
intermolecular collagen cross-links. Such cross-linking-
dependent insolubilization of noneollagen molecules,
coupled with collagenase-dependent solubilization, ap
peared to be a stringent criterion for collagen-associated
molecules.
The goal of this dissertation was to test the hypoth
esis that the insolubility of presumed collagen-associated
molecules is due to entrapment. This was accomplished by
evaluating the rates of insolubilization of such molecules
when collagen cross-linking had been inhibited.
Specific Introduction
The effect of collagen cross-linking on the insolu
bilization of presumed collagen-associated molecules was
studied by following the appearance of radioactive sialo-
glycopeptide (SGPep) and dermatan sulfate proteoglycan
(PDS) in the collagenase-digested insoluble collagen ma
trix of rat skin; with and without pretreatment with the
lathyrogen, S-aminoproprionitrile (BAPN).
The presumption of collagen-association for PDS was
based on its insolubility (6 M urea, 60° required for
solubilization (8)), increased content with age (26,27),
sensitivity to BAPN (7), solubilization by protease-free
bacterial collagenase, and effect on in vitro collagen
10
fiber formation (8). In the case of SGPep, the presump
tion was based on its salt-insolubility, age-related
increase paralleling that of insoluble collagen (28),
and solubilization by bacterial collagenase.
BAPN is a specific inhibitor of lysyl-oxidase, the
enzyme that generates the collagen aldehydes required for
the formation of covalent cross-links (29). The effect
of BAPN on collagen is experimentally observed as an
increase in 0.5 M salt-soluble collagen caused by the
absence of cross-linking-dependent insolubilization
(Review 30,31). Some of the observed effects of BAPN
(9-37 days of treatment) on the isotope incorporation,
chemical content, and distribution of noncollagen mole
cules may actually be secondary degenerative effects due
to loss of structural integrity in collagen fibers (7,
32,33). This possibility is supported by the small
effects reported (compared to those on collagen) (7,32,33)#
the presence of tissue lesions and "high-dose" toxicity
(34)# and the fact that the effects of BAPN on collagen
have been observed after much shorter treatment (30#3l).
Since the insolubilization of collagen depends on alde
hyde-derived cross-links, so should the insolubiliza
tion of entrapped, associated noncollagen molecules.
Consequently, the effects of BAPN should be observed
simultaneously on both collagen and collagen-associated
11
macromolecules. This permitted the present study to be
performed after "low-dose" BAPN treatment for only 3
days, when collagen is affected and secondary degener
ative effects should be minimal.
Insolubilization of recently synthesized PDS and
SGPep was determined by measuring the specific or total
radioactivity of these molecules isolated from the 0.5 M
salt-residue by collagenase digestion. The isotopic
approach was taken to provide the specificity (recently
synthesized) and the sensitivity required to detect the
appearance of a few molecules in a large pool.
glucosamine was used as a precursor for both PDS and
SGPep because it is not incorporated into rat skin col-
14
lagen. C-proline was simultaneously injected as both
a collagen and noncollagen protein precursor. Evalu
ation of collagen solubility-transitions was then pos
sible in the same animals used to study PDS and SGPep
insolubilization.
Protease-free collagenase was used in the absence
of endogenous proteolytic activity (35) to provide spe
cific degradation of the insoluble collagen matrix. This
limits the complexity of the collagenase extract (CE),
facilitating the purification of SGPep and PDS, and
generates small, easily removed collagen peptides.
CHAPTER II
MATERIALS AND METHODS
Materials
All chemicals were analytical reagent grade.
Carbazole, obtained from Matheson Coleman and Bell, was
recrystalized from a 3?°, saturated acetone solution.
Cetyl-trimethylammonium bromide (CTAB) was obtained from
J. T. Baker Chemical Company, Phillipsburg, New Jersey,
and recrystalized twice from a 3 percent deionized water
solution in order to remove a 260 nm-absorbing contaminant.
Whatman standard grade cellulose powder (30g) was washed
in series with 1 L each of 0.5 percent CTAB, 4 M NaCl,
deionized water, and then lyophilized before use. AG 50W -
X12, H+-form, was obtained from BioRad Laboratories (Rich
mond, California), Sephadex gel-filtration materials from
Pharmacia (Uppsala, Sweden), and ultrafiltration membranes
and apparatus from Amicon Corporation (Lexington, Massa
chusetts ).
Clostridium histolyticum collagenase (CLSPA), ob
tained from Worthington Biochemicals (Freehold, N. J.),
was purified before use by Sephadex G-200 chromatography
(36) and subsequently treated with dithiothreitol and
p-tosyl-lysine ohloromethylketone as previously described
12
13
(35). The dialyzed preparation of collagenase is free
of noncollagenase protease activity, mucopolysaecharidase
activity, and unhound inhibitors. The absence of muco
polysaccharidase in G-200 purified collagenase was demon
strated by the absence of hyaluronic acid and chondroitin
sulfate C depolymerization when exposed to 1 mg of col
lagenase for 18 hr at 37° in collagenase buffer (CB: 50
mM Tris-HCl, pH 7.6, 5 mM CaCl2 > 0.02# NaN^) and then
chromatographed on Sephadex G-75. The activity of testi
cular hyaluronidase, in the presence or absence of col
lagenase, was easily detected with this technique. Pro-
nase (CB grade) was obtained from Calbiochem and incubated
in CB at 60° for 30 min prior to use in order to destroy
other contaminating enzymatic activities.
Analytical Methods
Sialic acid. Sialic acid was assayed by the method
of Warren (37) using N-acetyl-neuraminic acid (A grade,
Calbiochem) and malonaldehyde dimethoxyacetal (l,l,3>3-
tetramethoxypropane) (Aldrich Chemical Co.) as standards
for dichromatic corrections; extinction coefficients were
determined in each assay. The original assay was modified
to yield a 4-fold increase in sensitivity while maintaining
a 0.2 ml sample volume. To 0.2 ml of sample, 20 pi of
16.5 N HoPO, was added (1.5 N H_P0, final concentration)
3 4 3 4
and the sample hydrolyzed for one hour at 80°. Samples
14
were cooled, 0,1 ml of 0.05 M NalO^ in 9 M H^PO^ was
added, the samples mixed and allowed to stand for 20 min
at room temperature. Periodation was terminated "by the
addition of 0.1 ml of 2.0 N sodium arsenite in 1.4 M
Na2S0^ and 0.1 N ^SO^, followed by mixing until the brown
color disappeared. 2-Thiobarbituric acid (0.6# in 0.5
M NagSO^) was then added (0.8 ml), the tubes capped,
boiled for 15 min, and cooled to room temperature.
Cyclohexanone (0.8 ml) was added, mixed, centrifuged,
and its absorbance at 532 and 549 nm determined.
The distribution of sialic acid (Table I) was de
termined by analysis after each sample was dialyzed (except
the starting tissue) versus deionized water (3 X 20 L, 4°),
maintained at 75° for 30 min, lyophilized, suspended in
1.0 ml of CB, digested for 8 hr at 37° with two 50 pi
additions of pronase (10 mg/ml CB), lyophilized, and dis
solved in 1.0 ml of deionized water.
Hexosamine. Hexosamine was determined by the
Johnson (38) modification of the Elson-Morgan (39) method.
All reagent volumes were reduced 10-fold. Samples were
hydrolyzed (6 N HC1, 22 hr, 100°) in sealed evacuated
ampules (40) and the acid removed under vacuum, over
NaOH pellets.
15
For the determination of hexosamine specific ac
tivity, dry hydrolysates were dissolved in 2.0 ml of
0.01 N HC1 and loaded on 2.0 ml (bed volume) columns of
AG 50W - X12 (200-400 mesh, H+ form). The columns were
washed with 3 X 2.0 ml of 0.01 N HC1, eluted with 5 X 2.0
ml of 0.5 N HC1 and finally with 3 X 2.0 ml of 3.0 N HC1.
The second through fourth 0.5 N HC1 fractions were pooled,
dried (as above), and dissolved in 2,5 ml of 0.15 M
^200^. Aliquots were assayed for hexosamine and counted
in 15 ml of Aquasol (Packard Inst. Co.) with double-label
( C, H) settings (see below). The column procedure de
scribed yields almost complete separation of labeled
proline (containing 20 mg carrier-proline) from labeled
glucosamine (0.5# of proline in the glucosamine fraction)
and provides 93 percent recovery of 160 ug glucosamine by
colorimetric assay.
Separation of glucosamine from galactosamine was
accomplished using an 8.5 cm short column on a Beckman
120-C amino acid analyzer eluted with 0.1 N sodium citrate
buffer, pH 5.24 at 60°. Quantification was performed by
the half-height, dot-counting method. The 5-fold sensi
tivity module was used for all chromatograms.
Uronic acid. Uronic acid was assayed using the
method of Bitter and Muir (41) modified by the use of a
combined carbazole reaction-acid hydrolysis of 15 min at
16
100°. Glucuronolactone was used as a standard. Uronic
acid ratios were determined on samples and National Heart
Institute standard glycosaminoglycans (a generous gift
£
from Dr. M. B. Mathews) by comparing the absorbance ob
tained with the modified Bitter-Muir assay to that with
the Dische carbazole assay (42). All standards and samples
were analyzed as their CTAB precipitates (see below).
Dermatan sulfate. The dermatan sulfate (DS) assay
was performed as described by Di Ferrante (43) except that
reaction with Schiff's reagent was continued until maximum
color development was observed in DS standards (ca. 40
min). Only semiquantitative visual observations were
possible for experimental samples because the assay was
performed on CTAB precipitated material. National Heart
Institute standard glycosaminoglycan were tested both in
solution and as CTAB precipitates.
Hydroxyproline. Hydroxyproline was measured with
the method of Stegemann (44) as modified by Grant (45) for
the Technicon Autoanalyzer. Care was taken to avoid or
correct for the inhibition of this assay caused by Tris.
o
Extracts were hydrolyzed in 6 N HC1 for 90 min at 131
without prior dialysis. The equivalence ug collagen =
^Department of Pediatrics, The University of
Chicago, Chicago, Illinois.
17
7.46 X ug hydroxyproline has been used throughout this
study. The specific activity of hydroxyproline was
determined by the method of Juva and Prockop (46) modified
by a 4-fold reduction in the reagent volumes. All ex
tracts were dialyzed versus deionized water (3 X 20 L) for
36 hr at 4° (except collagenase extract), evaporated to
dryness, and hydrolyzed as above.
Protein. Protein was measured using the method of
Lowry (47) modified by a 5-fold reduction in reagent
volumes and the inclusion of a 10 min., 50° heating step
after the addition of the first reagent (48). Bovine
serum albumin was used as a protein standard.
Total hexose. Hexose was measured using the or-
cinol method described by Winzler (49). Glucose:
galactose, 1:1 by weight, was used as a standard. The
sample and all reagent volumes were reduced 10-fold.
Liquid scintillation counting. A Packard Liquid
Scintillation Spectrophotometer (3320) and low background
polyethylene counting vials were used throughout this
14
study. Single-label settings counted unquenched C with
88.5 percent efficiency and -^H with 41.6 percent effici-
14
ency. Double-label settings counted C with 58 percent
3
efficiency and H with 34.7 percent efficiency. There was
3 14
less than 0.01 percent H spill-over into the C channel
18
14 3
and 15.6 percent spill-over of C into the H channel
(unquenched). Quench and spill-over correction curves
were based on automatic external standard ratios. All
samples were counted in 15 ml of Aquasol (Packard Inst.
Co.) except those from the hydroxyproline specific activ
ity determinations, which were counted in Omnifluor (New
England Nuclear) as described for this assay.
Calculations. Calculation of data as amount/g dry
weight starting tissue was based on the starting wet weight
of each sample (22g), the wet:dry weight ratio determined
for each sample (three determinations), and the fraction
of the starting material used in each assay. All color
imetric determinations (except hydroxyproline and those
performed on the G-50 fractions) were performed in du
plicate and the average used for subsequent calculations.
Experimental Procedures
Animals and diet. White, male Sprague-Dawley rats
(weighing 135J5 g) were used in this study. All rats were
individually caged and fed ad libitum a control diet of
ground Purina Lab Chow for 2 days prior to the onset of
the experimental protocol. The amount of food consumed
and weight gained by each rat was determined each day be
tween 3 and 5 p.m. Isotope injection and animal sacrifice
19
(see below) were also performed at this time of day.
Experimental protocol. Sixteen rats were fed ad
libitum the control diet supplemented with 0.5 percent
beta-aminoproprionitrile-fumarate (BAPN). Beginning the
following day each BAPN-fed rat was paired with a control
rat and the control was fed only the amount consumed by
its pair. Pair-feeding was used to assure equivalent
caloric intake for the BAPN and control rats, and was
continued until the termination of the experiment. All
pairs were conserved during the formation of sample groups.
Three days after beginning BAPN treatment, both BAPN (B)
and control (C) rats received 0.1 ml of phosphate buf
fered saline (Grand Island Biological Co., Huntington
14-
Beach, Calif.) containing 10 yCi L-proline-C (U) (233
mCi/mMole) and 25 yCi D-glucosamine-6-H^(N) (3.6 Ci/mMole)
(both obtained from New England Nuclear) by ventral, sub
cutaneous injection. One day later, four groups of 4
rats each (A^gj, an(i their corresponding pair-fed
controls A(q ), b(c )) were sacrificed by Nembutal anes
thesia and decapitation. The next day groups c(b )> °(B)'
C(c j, and q ^ were sacrificed. This protocol is summar
ized in Figure 1.
Additional rats (nonradioactive and fed the control
diet) were divided into groups of 4 rats each and processed
as described above and below. Extracts obtained from these
A(B) A(C)
B(B) B(C)
C(B) c(c)
D(B) D(C)
BAPN
treat pair-
ment feeding
begins begins
J_________________ L
DAYS
FIGURE 1. Experimental protocol.
label
inj ec-
tion
j______
3
Labeled
1 day
Sacrifice
A(B) A(C)
B(B) B(C)
I ______
4
Labeled
2 days
Sacrifice
C(B) C(C)
D(B) D(C)
______I ____
5
For details, see "Methods."
21
animals were used only to obtain representative chroma
tographic profiles.
Tissue processing. Whole skins (except from the
head and limbs) were removed and stored frozen after most
of the hair was removed with electric clippers. The skins
were then thawed and the fat and fascia removed by scraping
with a scalpel. After freezing on a surface cooled by dry
ice, the last traces of hair were removed by scraping with
a scalpel. The skins were then minced, pulverized under
liquid nitrogen in a Spex Mill 6700 (Spex Industries,
Metuchen, N, Y.), pooled and mixed while frozen, and
stored at -30° until used.
Extractions. Extractions were performed in 250 ml
polycarbonate screw-capped centrifuge bottles, each con
taining 22 g wet weight pulverized tissue (80-95# of the
tissue from 4 rats) and 80 ml (unless otherwise indicated)
of extraction buffer. The pH of each buffer was adjusted
so that it was 7.4-7.6 at the temperature employed. Se
quential extractions were performed by shaking tissues
(120 cyeles/min, reciprocally) in the following buffers
for the indicated times and at the indicated temperatures.
Figure 2 summarizes this procedure. All buffers contained
0.02 percent NaN^ as an antibacterial agent. 0.15 M salt:
0.1 M NaCl, 50 mM Tris-HCl, (3 X 3 hr, 1 X 15 min), 4°;
22
Pulverized Rat Skin
0.15 M Salt I 3 X 3 hr, 4°
i: X 15 min, 4°
0.50 M Salt I 4 X 24 hr, 4°
I X 15 min, 4°
CB I 3 X 15 min, 4°
I Maintained at 75-80°
- Jr 15 min
Collagenase ^ 18 hr, 37°
1
Adjusted to 0.5 M NaCl,
15 mM EDTA by addition
of solid reagents.
Combine as CE i Extract removed after
t________•1' 15 min at 25°
__________CW|2 X 15 min, 25°
Residue
FIGURE 2. General extraction scheme. Extraction
was performed by shaking 22 g wet weight of pul
verized rat skin with 80 ml solvent at 4° in
250 ml centrifuge bottles. Extracts were removed
from tissue residues by centrifugation at 10,000
X g for 10 or 30 min. See "Methods" for further
details.
23
0.5 M salt; 0.45 M NaCl, 50 mM Tris-HCl, (4 X 24 hr,
1 X 15 min), 4°; CB: 50 mM Tris-HCl, 5 mM CaCl2, (3 X 15
min), 4°; Collagenase: CB containing 1 mg collagenase
per g dry weight starting tissue (l X 18 hr), 3?°j without
centrifugation, solid NaCl and EDTA (tetrasodium salt)
were added to final concentrations of 0.5 M and 15 mM,
respectively, (l X 15 min), 25°; Collagenase Wash, CW:
40 ml of 0.5 M NaCl, 15 mM EDTA, 1 mM K„HP0,, (2 X 15
2 4
min), 25°. Each extract was removed from its residue
by a 10,000 X g centrifugation for 10 min, except for
collagenase and CW extracts which were centrifuged for
30 min. The volume of each extract was measured and the
extract centrifuged at 105,000 X g (RCF ) for 30 min.
max
This high-speed centrifugation caused most of the opaque
lipid-lipoprotein material present in the extracts to
float and aggregate, and permitted its removal from the
0.15 M salt, 0.5 M salt, and CB extracts by gentle decanta-
tion and subsequent filtration through Whatman GFB glass
fiber filters. This technique was not as effective with
collagenase and CW extracts; consequently, the floating
opaque material was gently dispersed in the 105,000 X g
supernatant and removed subsequently by NaBr buoyant
density centrifugation.
Because of the high viscosity of the first three
0.5 M salt extracts and the consequent looseness of the
10.000 X g pellets (especially for BAPN samples), the
105.000 X g pellets from these extracts were added back
to the ongoing extraction of the 10,000 X g pellets. This
was done to maximize the amount of insoluble collagen
matrix exposed to collagenase digestion. For the collage
nase and CW extracts, the 105,000 X g pellets were combined
with the 10,000 X g pellets to yield the tissue residue.
Colorimetric assays for sialic acid and hydroxy
proline and the determination of hydroxyproline specific
activity were performed using the 105,000 X g supernatants
from each extract. The lipoprotein-containing superna
tants from collagenase and CW extracts were used to deter
mine the collagen extraction kinetics and then they were
combined as CE (collagenase extract) and used for the
other chemical and isotopic determinations, and the isola
tion of PDS and SGPep.
Processing collagenase extract. The majority of
CE was used for the isolation of PDS and SGPep. The pro
cedures involved in these isolations and the stages used
to obtain the various data presented are diagrammatically
summarized in Figure 3.
NaBr buoyant density centrifugation. CE from BAPN
and control tissues were separately dialyzed versus
3 X 20 L of deionized water for a total of 96 hr. and then
CE (105,000 X g supernatant) 25
I
Dialyzed vs. water
Lyophilized
NaBr Buoyant Density Centrifugation
I < J r I
floating
lipoprotein
layer
(not used)
"clarified CE"
1
gelatinous
pellet
(discard)
G-75 Chromatography (Figure 5)
f
A
void volume fraction
I Dialyzed vs. water
v Lyophilized
—►Analyzed for sialic acid
^ and hydroxyproline
CTAB-MgCl2 Chromatography (Fig. 7)
retarded fraction
Concentrated on
UM-2 ultrafilters
Desalted, G-10
Lyophilized__________
G-50 Chromatography
T
T
IV
(Fig. 6, Table II) J
II III
(PDS) (PCS)
— ► Analyzed for
total protein (Table IV)
230:280 nm ratio (Table III)
Dialyzed vs. water
Dialyzed vs. EDTA
^|r Dialyzed vs. water
•Analyzed for
uronate, %, 14C (Tables III, VII)
protein:uronate ratio (Table IV)
Dialyzed vs. 1 M NH^Ac
Dialyzed vs. water
'r Lyophilized
— ►Analyzed for sialic acid
I and hydroxyproline
y Hexosamine isolation (AG 50W-X12)
—►Analyzed for hexosamine total
specific activity (Table VII)
FIGURE 3. Processing CE.
26
lyophilized. Each sample was dissolved in approximately
30 ml of NaBr solution containing 0.26 g NaBr per ml
final volume of 0.5 M NaCl, 1 mM EDTA (tetrasodium salt),
50 mM NagHPO^, pH 7.4, p = 1.22 g/cc. Samples were trans
ferred to 3 X 1" cellulose nitrate centrifuge tubes and
centrifuged in a SW 27.1 rotor at 4° and 24,000 rpm
(RCF = 107,000 X g) for 4 hr. Clarified CW was re
max °
moved from the bottom of the tube by puncture with a
double-ended needle. Care was taken not to disturb the
small gelatinous pellet or the floating opaque layer
formed during centrifugation. Clarified CE from BAPN and
control tissues were concentrated by ultrafiltration using
Amicon UM-2 ultrafilters (2,000 M.W. cut-off, 43 mm diam
eter). Ultrafiltrates were saved and subsequently com
bined with the corresponding G-75 retarded fraction. The
concentrated samples contained a variable amount of aggre
gated material which was removed by centrifuge-filtration
through Whatman GFC grade glass fiber filters. Each
filter was then exposed to two 0.5 ml aliquots of 4 M
guanidine-HCl in 5 mM Na2HP0^, pH 7.4, and then cleared
by centrifugation. These guanidine washes were added
back to their respective concentrated, clarified CE prior
to G-75 chromatography.
Sephadex G-75 chromatography. Concentrated, clari
fied CE (less than 10 ml) was fractionated on a 2.5 X 39.5
27
cm column of Sephadex G-75 fine, eluted at 90 ml/hr with
0.5 M NaCl, 1 mM EDTA, 5 mM Na2H0P^, pH 7.4, and monitored
continuously at 280 nm. Fractions (3.0 ml) were collected
for subsequent analysis, and experimental samples were
pooled as void volume or retarded volume fractions.
CE from nonradioactive rats was divided so that a
dilute sample could be fractionated and analyzed as an
example of the experimental profiles obtained. An aliquot,
2.0 ml, of the total concentrate (10.85 ml) of clarified
CE was diluted to 10 ml with NaBr solution and chromo-
tographed as described.
G-75 retarded fraction. The retarded fraction from
G-75 chromatography was combined with its corresponding
NaBr ultrafiltrate, concentrated by ultrafiltration on
UM-2 ultrafilters, desalted on a 2.5 X 24 cm column of
Sephadex G-10 (eluted at 100 ml/hr), lyophilized, and
weighed. Radioactive samples were dissolved in 4 ml of
1 M NH^Ac, pH 7.0, and dialyzed versus this solution for
24 hr at 4°, then versus deionized water for 4 hr at 4°,
lyophilized, and weighed. This material was dissolved
in 2.5 ml of deionized water and 2 ml was taken for the
determination of hexosamine specific activity.
Analytical G-50 chromatography. Desalted, lyo
philized G-75 retarded fraction from nonradioactive rats
was dissolved in 3.5 ml of 0.2 M NH^Ac, pH 7.0 and
chromatographed on a 1.5 X 88 cm column of Sephadex G-50
fine, eluted with 0.2 M NH^Ac, pH 7.0, at 17 ml/hr and
monitored continuously at 280 nm. Fractions (1.96 ml)
were collected and aliquots lyophilized three times and
assayed for hexosamine, hexose, hydroxyproline, and sialic
acid.
G-75 void volume fraction. G-75 void volume mate
rial was dialyzed versus deionized water, lyophilized and
dissolved in deionized water. The samples were then ad
justed in 0.1 M NaCl "by the addition of 4 M NaCl. Aliquots
were taken for colorimetric assays to determine the total
content of the void volume fraction. The majority of this
•material was used for CTAB-MgCl2 chromatography.
CTAB-MgClp chromatography. An aliquot (usually 3.5
ml) of the G-75 void volume fraction was precipitated in
a 25 ml disposable plastic beaker with the following se
quential additions performed with constant stirring:
sample (3.5 ml); phosphate buffer, 5 mM NagHPO^, 0.1
M NaCl, pH 7.4 (4.2 ml); 1 percent CTAB in phosphate buffer
(1.05 ml); CTAB-cellulose, 2 g cellulose in 20 ml of 0.01
percent CTAB in phosphate buffer (1.8 ml). The resultant
suspension was used to partially fill a 0.9 X 1.5 cm
flow-adapted column. Additional CTAB-cellulose was added
to completely fill the column which was then eluted with
10 ml of 0.01 percent CTAB in phosphate buffer, and 15 ml
of deionized water. The absorbance at 280 nm was moni
tored continuously with a flow-cell until the water
effluent contained no material absorbing at 280 nm.
The effluent from the column-packing process, and the
CTAB and water washes were combined as CTAB-soluble mate
rial; this material was not analyzed further. The CTAB-
precipitable material was eluted from the column by a
0-1.5 M MgClg linear gradient (35 ml) and 4 M MgClg at
8 ml/hr. Four peaks, I-IV, were obtained and the frac
tions pooled as described in Figure 6. The volume of each
pooled peak was measured and the absorbance at 230 and 280
nm was determined. After the addition of 0.1 ml of 1
percent CTAB (in phosphate buffer) to the material from
each peak, all samples were dialyzed at 25° versus
2 X 20 L deionized water for 12 hr in Spectropore 1.
dialysis tubing (6,000-8,000 M.W. cut-off, National
Scientific Co.); then versus 0.1 M EDTA (tetrasodium
salt), 5 mM Na2HP0^, pH 7.4, 1 X 4 L for 4 hr; and finally
versus deionized water as above. The resultant CTAB-
precipitated material was transferred to polyethylene
centrifuge tubes and collected by centrifugation for 15
min at high speed in a clinical centrifuge (swinging
bucket rotor). The pellets were dissolved in 1 or 2 ml
(peak II) of 60 percent n-propanol. All aliquots for
colorimetric assay or liquid scintillation counting were
lyophilized in the appropriate tubes to remove the pro
panol. CTAB-precipitates were soluble in Aquasol.
CHAPTER III
RESULTS
Experimental animals were fed 3-aminoproprionitril'e
(BAPN) or control diets, injected with H-glucosamine and
14 J
C-proline, and sacrificed according to the protocol de
scribed in Figure 1. Each control rat was paired with a
BAPN-treated rat and fed only the amount of food consumed
by its pair. This was done to avoid differences in growth
rate due to the decreased food intake usually observed
with BAPN-treated animals. The growth rate was reduced
from 5-6g/day to l-3g/day by BAPN treatment (ca. 45 mg
BAPN/lOOg body wt./day) and pair-feeding. No significant
difference in weight existed between the BAPN and control
groups at the time of sacrifice (ca. I62±7g).
Whole skins were minced, pulverized, pooled, and
processed according to the extraction scheme briefly de
scribed in Figure 2. Each solvent was employed to accom
plish a generalized goal: 0.15 M salt, to remove in vivo-
soluble components of the tissue such as serum proteins,
hyaluronate, newly synthesized collagen, etc.; 0.5 M salt,
to provide the solubility distinction between intermole-
cularly cross-linked (insoluble) and noncross-linked or
3!
32
intramolecularly cross-linked (soluble) collagen so that
the effects of BAPN could be observedj CB, to prepare the
0.5 M salt-residue for collagenase digestion; 80° treat
ment , to inactivate the observed endogenous protease ac
tivity, since more selective methods were ineffective
(32); collagenase digestion, to selectively degrade the
collagen matrix; CW, to eliminate ionic and calcium-
dependent coupling of "collagen-associated" molecules to
the collagenase-residue and to provide a stabilizing en
vironment for the large amount of opaque, lipoprotein
like material released during collagenase digestion.
The conditions of extraction (time, repetition)
were determined in previous experiments which demonstrated
that the extraction of collagen and noncollagen protein
are similar with the solvents employed in the present
study. Complete extraction of collagen with each solvent
was achieved for both control and BAPN samples (Figure 4),
even though the latter contained larger amounts of soluble
collagen. Consequently, each combined extract represents
a rigidly defined fraction of connective tissue compo
nents .
The distribution of sialic acid (a marker for
sialo-glycoproteins) and collagen in the various extracts
is presented in Table I. No significant difference be
tween the sialic acid distributions of the BAPN and control
*
o
o
E
txo
200
100
20
1
J I
0.12 M SALT 0.2 N SAU
CE
FIGURE A. Collagen extraction kinetics. The data are presented as
means + range, n=2. In some cases the range is hot presented because
it was too small to be visualized oh this scale. Filled bars repre
sent control samples; open bars represent BAPN samples. Groups C and
D were used to obtain these data.
VjJ
VjO
TABLE I
DISTRIBUTION OF COLLAGEN AND SIALIC ACID
SIALIC ACID8
BAPN CONTROL
COLLAGEN13
BAPN........... CONTROL
Starting tissue 587.±6. 566.±24. 356.±30. 332+22.
0.15 M salt 243.±11.
221.±20.
32.7±5.3 10.3±1.1
0.5 M salt 161.±13. 161.±20. 45.6+4.8 23.9±2.6
CB 2.38+.60 1.72+.78
0.76±.13 0.85±.19
CE 72.5+8.7 (12) 77.0±7.9 (14) 215.±37. 261.±17.
Residue 42.8+4.4
39.7±4.3
0.566+.012 0.611±.094
Total recovered 522. (89) 500. (88) 295. (83) 297. (89)
aThe data are presented as yg sialic acid per g dry wt. of starting tissue,
± 1 S.D., n = 4 (groups A,B,C,D)j parentheses indicate the percentage of
the starting tissue content. All samples were treated with pronase after
dialysis and before hydrolysis, as described in "Methods."
^The data are presented as mg collagen per g dry wt. of starting tissue,
± 1 S.D., n = 4. Extracts were not dialyzed prior to hydrolysis.
35
groups was observed, although the mean value for CE was
slightly lower for the BAPN samples. The distribution
of collagen, however, demonstrated marked differences
between control and BAPN tissues. As expected, the 0.15
and 0.5 M salt-extracts of BAPN samples contained in
creased amounts of collagen (3-fold and 2-fold, respec
tively) when'compared to control samples. BAPN also
caused a decrease in the collagen content of the insoluble
collagen fraction, CE.
The majority of CE was used for the isolation of
PDS and SGPep. The procedures involved in these isolations
and the stages used to obtain the various data presented
are diagramatically summarized in Figure 3. The removal
of the opaque lipoprotein material by NaBr buoyant density
centrifugation was the required first step, otherwise the
further fractionation of CE could not be followed using
u.v. absorption. Clarified CE was then fractionated on
Sephadex G-75 (Figure 5) to separate SGPep from PDS, and
PDS from collagen peptides. The void volume fraction,
containing PDS, contained only 6.84 yg hyp/g dry wt.
starting tissue, or approximately 0.02 percent of the
amount originally present in CE.
Although all eight radioactive sample G-75 profiles
were qualitatively identical, the void volume fraction and
the retarded fraction were separately tested for complete-
FIGURE 5. Chromatography of concentrated,
clarified CE on Sephadex G-75. An aliquot
of CE from nonradioactive rats (2.0 of 10.85
ml) was diluted to 10 ml with elution "buffer
(0.5 M NaCl, 1 mM EDTA, 5 mM Na2HP0,, pH 7.4)
and loaded on a 2.5 X 39.5 cm column of Sephadex
G-75 eluted at 90 ml/hr. The effluent was
monitored continuously at 280 nm; 3.0 ml
fractions were collected and analyzed for
absorbance at 230 and 280 nm and hydroxyproline
content. This u.v. profile is representative
of those obtained for each radioactive sample.
Each radioactive sample was pooled as a void
volume fraction (V ) or a retarded volume
fraction (RTD). Molecular weight marker
arrows indicate the center of the marker's
elution profile when chromatographed on the
same column.
36
2J>|»
A .
>•
“ u
O
S
o
8
«i
u
z
i
a
<
RTO
LY90ZYME
14^500 M W
HEMOSLOBN
64,000 M W
J6
2J0 A
Z
30 40 50 <0 70 60
FRACTION NO.
FIGURE 5. Sephadex G-75 chromatography of clarified CE.
ABSORBANCE 230 nm
38
ness of collagenase digestion by exposure to a second
collagenase digestion. After incubation for 18 hours in
the presence or absence of collagenase, aliquots of the
void volume fraction were chromatographed on G-75, and
aliquots of the retarded volume fraction were chromato
graphed on G-50. No collagenase-dependent changes in the
u.v. profiles (G-50, G-75) or sialic acid distribution
(G-50) were observed.
Sialo-glyeopeptide (SGPep)
All G-75 retarded fractions were concentrated by
ultrafiltration (UM-2, M.W. cut-off = 2,000), desalted on
a preparative G-10 column, and lyophilized. These pro
cedures were selected for their capacity to retain low
molecular weight material. Subsequently, the entire
retarded fraction from a group of nonradioactive rats was
fractionated on Sephadex G-50 to partially characterize
the sialic acid-containing molecules of this fraction.
The elution profiles of sialic acid, hexose, hexosamine,
hydroxyproline, Aoc}ri , and A-„rt (not shown) were
280 nm 230 nm
determined and are presented in Figure 6.
The region of Figure 6 between fractions 28 and 48
contained all of the sialic acid and is referred to as
SGPep. The hexose, hexosamine, and sialic acid profiles
in this region are all aligned with regard to a major peak
that reached a maximum in fraction 4l and a minor peak
FIGURE 6. Sephadex G-50 chromatography of the
G-75 retarded fraction. The entire retarded
fraction from a group of nonradioactive rats
was loaded in 3.5 ml of elution buffer on a
1.5 X 88 cm column of Sephadex G-50 and eluted
with 0.2 M NH^Ac, pH 7.0 at 17 ml/hr. The
effluent was monitored continuously at 280 nm
and 1.96 ml fractions were collected. Appro
priate aliq.uots were taken from every other
fraction for chemical analysis. The region
of the hexosamine profile from fraction #4-6
to #54- was obtained by subtracting the absorbance
of a contaminating chromaphore and represents an
estimate of the actual value.
39
(MICROMOLES) H Y P (o— o)
t o -
7.5 -
5.0 -
2.5 ■
BLUE LYSOZYME
DEXTRAN 14,500 M .W .
E
o 20
co
M
I aJ
o
z_ 1.5
<
CO
e
o
CO
5 to
0.5
• o'
* . o
70 30 40 50 60
12
10 I
l i l
m
o
x
8 “
N
I
o
CO
tu
_ l
o
Z
o
z
<
3.0
2.5
I
l
u
o =
< <
2. 0 o g
S1 x
< tu
CO X
1 .5
1 . 0
0.5
M
I
o
CO
UJ
-I
o
z
o
z
<
FRACTION NUMBER
FIGURE 6. Sephadex G-50 chromatography of the G-75 retarded fraction
characterization SGPep.
that reaches a maximum in fraction 33. The only exception
is the hexose profile which is slightly skewed to the
right due to the spill-over of hexose from the major
hexose-hydroxyproline peak that reaches a maximum in
fraction 49* The hexosamine profile from fraction 46 to
54 is an estimate, corrected for the presence of an orange
chromophore hy comparing the absorption spectra of gluco
samine and the orange chromophore (also present in frac
tions 60-64). This contaminant may be the hexose-amino
acid chromophore previously reported (50). The carbo
hydrate data from this figure were used to estimate the
general carbohydrate compositions of the major and minor
SGPep peaks (Table II). Fractions 40 and 32, respec
tively, were used because they contained the least con
tamination from neighboring peaks. No corrections for
spill-over were made.
The A2gQ nm profile of Figure 6 shows no indication
of the major SGPep peak and only a slight indication of
the minor SGPep peak. The same is true for the Ag^g ^
profile (not shown). These profiles suggest a low peptide
content for SGPep and, at the same time, demonstrate the
presence of significant peptide contamination. These
peptides, however, do not contain hydroxyproline, since,
with sensitivity much higher than that presented in
Figure 6 (2 yg hydroxyproline), the hydroxyproline
42
TABLE II
COMPARISON OF "HETEROPOLYSACCHARIDES"a
Collagenase
+
Collagenase Pronase
#32 #40 DBMb BBMC SGPA'
Hexose 2.2 2.8 2.7 1.4 1.4
Hexosamine 1.0 1.0 1.0 1.0 1.0
Sialic acid 0.43 0.31
0.7 0.6 0.3
a
The carbohydrate compositions of several "heteropoly-
saccharides" were recalculated to fit the categories
listed above, and are presented as molar ratios, hexo-
samine = unity. #32 and #40 represent the data of
Figure 6, fractions 32 and 40. "Heteropolysaccharides"
were obtained by the various investigators from a salt-
residue or isolated basement membrane (BM) after col
lagenase or collagenase-pronase digestion.
bDBM = dog glomerular basement membrane; Kefalides (51),
peak E from G-200 chromatography.
CBBM = bovine glomerular basement membrane; Spiro (52),
purified by G-50 chromatography.
SGPA - salt-residue of aorta, represents structural
glycoprotein of porcine aorta; Moczar (23), purified
by G-50 chromatography.
profile showed no indication of a SGPep peak.
Fractions 31 and 43 from Figure 6 were lyophilized,
hydrolyzed, and their content of glucosamine, galacto-
samine, and basic amino acids determined by ion-exchange
chromatography. The ratio of glucosamine to galacto-
samine increased from 3 in the minor peak to 18 in the
major SGPep peak (data not shown), while the sialic acid
to hexosamine ratio decreased and the hexose to hexo-
samine ratio increased (Table II). Analysis of the basic
amino acids of fractions 31 and 43 demonstrated the pres
ence of lysine and hydroxylysine, in addition to the other
basic amino acids. It is not possible, however, to decide
whether the hydroxylysine was derived from SGPep or the
contaminating peptides also present in fractions 31 and 43.
The similarity of the estimated width of the major
and minor SGPep peaks to the width of the lysozyme stand
ard peak (not shown) suggests that each SGPep fraction is
relatively homogeneous with regard to molecular weight.
The absolute value of the molecular weight indicated by
cochromatography with the 14,500 molecular weight lyso
zyme standard is probably not an accurate estimate.
Comparing the elution position of SGPep to that of a
similar glycopeptide (see Discussion) suggests a lower
limit of about 8,000 for the molecular weight of the
major SGPep peak.
44
' j
The use of ■'H-glucosamine as a precursor to SGPep
permitted the specific activity of SGPep to be determined
without interference from the hexose-hydroxyproline frac
tion, regardless of whether these two fractions could be
physically separated from one another. Consequently, all
radioactive retarded fractions were dialyzed versus 1 M
NH^Ac and then deionized water, and lyophilized. This
treatment was intended to remove as much peptide material
as possible. It reduced the hydroxyproline content to 56
Ug/g dry wt. starting tissue, and the dry weight 95 per
cent; the sialic acid content decreased only 18 percent.
Correcting only for losses at this step, SGPep contains
70 percent of the sialic acid of clarified CE (100# =
43.5 yg/g dry wt, starting tissue). Following lyophiliza-
tion, the majority of each retarded fraction was hydro
lyzed and the hexosamines separated from proline by ion-
exchange chromatography on small columns of AG 50W - X12.
Double-label liquid scintillation counting and hexosamine
analysis were used at this step to determine the total and
specific activities of SGPep.
Proteodermatan Sulfate (PDS)
PDS was isolated from the G-75 void volume by CTAB
precipitation and solubilization of the precipitate (held in
a cellulose column) by elution with a linear gradient of
MgCl2 (Figure 7). Fractions were pooled as indicated to
FIGURE 7. McG12 elution of the CTAB-
precipitated G-75 void volume fraction;
isolation of PDS. An aliquot (3.0 ml of
5.0) of the void volume fraction from A(B)
in 0.1 M NaCl was diluted to 7.7 ml with 0.1
M NaCl, 5 mM Na2HP0 4 , pH 7.4 and then pre
cipitated by the addition of 1.05 ml of 1%
CTAB in the same buffer. After 10 min at
room temperature a cellulose suspension
was added and the mixture transferred to a
flow-adapted column (0.9 X 1.5cm). The
column was washed with buffer containing
0.01$ CTAB and then with deionized water
until the effluent was free of 280 nm-
absorbing material. The column was then
eluted with a 35 ml linear gradient of
unbuffered MgCl2 (0^1.5 M) and finally
with 4 M MgClo. Elution was performed
at 8 ml/hr ana the effluent monitored at
280 nm; 0.4 ml fractions were eollected-
and pooled as indica ted by the overlying
bars. The CTAB-soluble fraction is not
shown.
45
4.0
TZ m.
E
c
125
ZD
K
O
,75
.25
E
c
o
w
o
z
<
a
c
s
a
05
20 30 40 50 80 SO 90
FRACTION NO.
FIGURE 7. MgCl2 elution of the CTAB-precipitated G-75 void volume
fraction; isolation of PDS.
-r~
o
( n ' t m« c i a ( •
47
give peaks I-IV. This material was dialyzed against deion
ized water and EDTA in order to reprecipitate CTAB com
plexes. The CTAB complexes were dissolved in 60 percent
n-propanol and aliquots assayed for uronate (Dische and
Bitter-Muir), dermatan sulfate, protein, H, and C
(Tables III and IV).
The uronic acid ratio (Dische/Bitter-Muir) identi
fied the glycosaminoglycan component of peak II as dermatan
sulfate. This was supported by the reactivity of peak II
in the dermatan sulfate assay of Di Ferrante (43). The
proteoglycan nature of this peak was demonstrated by its
230/280 ratio and the significant amount of absolute ab
sorbance at 280 nm. Consistent with identification as
14 ■ ?
PDS was the incorporation of both C-proline and ^H-
hexosamine.
It is apparent that the modified Bitter-Muir uronate
assay produced misleading results for peaks I and IV since
the Dische assay detects no uronate (ratio goes to zero)
in these peaks. The material in peaks I and IV was also
reactive in the dermatan sulfate assay. This material
has been tentatively identified as RNA on the basis of the
following observations: a 230/280 nm ratio near one, u.v.
profiles similar to nucleic acid, the similarity of the
conditions of the dermatan sulfate assay to those of a
micro-assay for RNA (53), the requirement of repetitive
negative charge for CTAB precipitation, and the absence
TABLE III
CHARACTERISTICS OF CTAB-MgCl2 PEAKS
I II III IV
DS CSC HEP
UA ratio8- 0 0.44 1.11 0 0.46 1.03 1.14
UA (55)13 14.5 52.0 5.0 28.0
230/280° 1.43 4.65 3-89 1.15
d
DS assay
+ ++++
-
+ + + - -
3h , 1^C CPMe
+ ++++ - -
aThe uronic acid (UA) content of aliquots was determined using glucuronolactone as
a standard and both the Dische and the modified Bitter-Muir assay. The results are
expressed as the ratio, Dische/Bitter-Muir. National Heart Institute standard
glycosaminoglycans were assayed identically and the values are presented for compari
b
The Bitter-Muir-positive material in peaks I-IV was summed and the distribution is
expressed as percentages of this value.
cThe ratio of absorbance at 230 and 280 nm of each peak was determined immediately
after elution and pooling, but before reprecipitation. The data represent means,
n=8.
d
The dermatan sulfate assay of Di Ferrante was performed on CTAB-precipitated mate
rial, consequently, only semiquantitative visual observations were possible.National
Heart Institute standard glycosaminoglycans were assayed both in solution and as
CTAB-precipitates.
The data represent all radioactive samples.
49
TABLE IV
PROTEODERMATAN SULFATE:
TOTAL PROTEIN AND PROTEIN:URONATE RATIO
Labeled BAPN
1 DAY Control
Total Proteina Protein:Uronate Ratio*3
1.03+.04
1.48+.09
1.061+.077
1.387+.030
Labeled BAPN
2 DAYS Control
0.96+.12
1.50+.03
0.773+.035
1.519+.021
Total protein in PDS (isolated, pooled, peak II)
is expressed as total absorbance at 280 nm (k ^
X vol.) per g. dry weight starting tissue.
The data are presented as means+range, n=2.
Reprecipitated peak II (PDS) was assayed for
protein by the modified Lowry method and uronate
by the modified Bitter-Muir method. The data are
presented as weight ratios, means + range, n=2.
*3 lA
of significant or C incorporation.
Peak III has been tentatively identified as PCS.
Its 230/280 nm ratio, absorbance at 280 nm, uronic acid
ratio, and absence of reactivity in the dermatan sulfate
assay are consistent with this identification. The ab-
3 14
sence of significant H and C incorporation might be
explained by a low level of PCS synthesis or insolubil
ization in rats of this age.
The total content of PDS obtained from CE was es
timated by measuring the absorbance at 280 nm and the
volume of pooled peak II before reprecipitation. Thus,
the stage of highest purity and least experimental vari
ability was utilized for this measurement. On the basis
of these data (Table IV), it appears that BAPN caused a
33 percent decrease in CE PDS. However, the progressive
decrease in protein content of PDS indicated by the pro-
teimuronate ratio of reprecipitated PDS (Table IV) demon
strates that the protein portion of PDS has decreased.
Consequently, the amount of DS in CE is relatively un
changed by BAPN treatment.
Isotope Incorporation
The solubility-transitions of collagen and the in
solubilization of SGPep and PDS were monitored by observing
the changes in total or specific radioactivity of these
molecules. Radioactive precursors were used in order to
51
increase sensitivity and focus attention primarily on
those molecules synthesized and processed under the present
experimental conditions. The homogeneity of the prepara
tions tested was, therefore, important.
Collagen soluhility-transitions were easily fol
lowed without isolating collagen itself. The specific
or total activity of isolated hydroxyproline was used
instead, since this amino acid is only found in collagen
and elastin, and only a small amount of the latter is found
in rat skin. The homogeneity of the PDS preparation is
indicated by the virtual absence of hydroxyproline, separ
ation from a very similar proteoglycan (PCS), and the re
quirements of the isolation procedure itself. Kobayashi
(54) has found similar CTAB methodology superior to den
sity gradient centrifugation in the presence of guanidine
for removing noncovalent contaminants from preparations
of PCS. SGPep, on the other hand, was not isolated as an
intact macromolecule. Although SGPep represents two dif
ferent molecular weight components, their well-defined
and low molecular weights after isolation in the absence
of noncollagenase proteolytic activity suggest that they
were derived from a larger molecule by the specific action
of collagenase. Since sensitivity to collagenase has not
been demonstrated for any serum protein (24) and only
recently for the collagen-like basement membrane (51,52
52
and Discussion), SGPep probably represents the cleavage
products of a single collagen-like molecule or the pro
ducts of two very similar such molecules. In either case,
the specific activity of the SGPep fraction should provide
some information about the insolubilization of its pre
cursor macromoleeule.
In order to determine the rate of insolubilization
of a protein on the basis of its specific activity in CE,
the time when the specific activity of the proline pre
cursor decreases must be known. This information was ob
tained by using the most recently synthesized collagen
(0.15 M salt-soluble (25)) as an internal marker for the
decline of precursor-proline specific activity. Since
this collagen fraction demonstrated a lower specific
activity on day 2 compared to day 1 (Table V), a signif
icant loss of precursor specific activity must have oc
curred near the onset of day 2.
Collagen Solubility Transitions
The solubility-transitions of collagen are easily
followed in terms of the percent distribution of total
hydroxyproline radioactivity (Table VI). This method of
presenting the,data is possible because the total hydro
xyproline radioactivity (the sum of the values for 0.15 M
salt, 0.5 M salt and CE) demonstrated no consistent in
crease from day 1 to day 2, and no consistent difference
TABLE V
HYDROXYPROLINE SPECIFIC ACTIVITY*
(DPM/ymole)
SAMPLE
LABELED 1 DAY LABELED 2 DAYS
BAPN CONTROL BAPN CONTROL
0.15 4l8±75 423±30 332+27 280+4
0.50 291+45 58l±58 280+18 376±31
CE 3.76±1.04 17.6±2.6 3.851.52 48.8*17.4
The data are presented as means+range, n=2.
54
DISTRIBUTION
TABLE VI
OF TOTAL l4C'
(PERCENT)*
-HYDROXYPROLINE
LABELED 1 DAY LABELED 2 DAYS
SAMPLE BAPN CONTROL BAPN CONTROL
0.15 46+4.4 19+2.0 47+2.1 13+2.5
o
•
o
51±3.7 62+0.4 50±1.8 37+5.5
CE 3±0.7 20±1.6 3+0.3 50+7.6
*The DPM in isolated hydroxyproline from each
extract listed above were summed to give the
total l^C-hydroxyproline for each sample; the
DPM in each extract was expressed as a percent
age of this value. Total i^Q-hydroxyproline
incorporated = 25,423.±1,728. DPM/g dry wt.
starting tissue (mean+1 S.D., n=8). The data
are presented as means+range, n=2.
55
between BAPN and control rats (data not presented). The
lack of increasing total radioactivity with time is
consistent with the duration of the precursor-proline
pulse indicated by the specific activity of the 0.15 M
salt-soluble collagen. By presenting the data as a
percentage of the total hydroxyproline radioactivity
observed, variation in the amount of label taken up by
the skins of different groups of rats is eliminated and
a more defined picture of the distribution of recently
synthesized collagen molecules is possible.
In control rats (Table VI), collagen moved from the
0.15 M salt-soluble pool to the 0.5 M salt-soluble pool,
and finally into the insoluble pool, CE. Since the in
solubilization of PDS and SGPep is evaluated solely on
the basis of the appearance of label in CE, it is impor
tant to note that the majority of the collagen synthe
sized in day 1 did not appear in CE, but remained in the
soluble pool until day 2. This is seen as a steady in
crease in the hydroxyproline radioactivity of CE over the
two-day period (Table VI). These data are consistent
with those reported by Nimni (55) and Jackson and Bentley
(25).
The effectiveness of BAPN is clearly demonstrated
by the large increases in collagen content of the 0.15 and
0.5 M salt-extracts (Table I, and Figure 3). Knowing the
control hydroxyproline DPM that arrived in CE on the
second day of labeling (Table VI; 25,423 DPM (0.5 - 0.2) =
9,627 DPM) and dividing this by the hydroxyproline spe
cific activity on day 1 of the 0.5 M salt-soluble collagen
(the immediate precursor of CE collagen), provides an es
timate of the number of ymoles of hydroxyproline that were
insolubilized as collagen in one day (9,627 DPM * 581
DPM/ymole hydroxyproline = 16.5 ymoles hydroxyproline or
16.1 mg collagen). Consequently, BAPN must have caused
nearly complete inhibition of collagen insolubilization
for three days prior to animal sacrifice in order to ac
count for the 43 mg increase in soluble collagen and the
46 mg decrease in CE collagen (Table I),
In terms of the distribution of labeled collagen
(Table VI), BAPN caused an increase in 0.15 and 0.5 M
14
salt-soluble C-hydroxyproline on the first day of label
ing. No shift in the distribution of total label was ob
served from day 1 to day 2. Only 3 percent of the total
label was present in CE. Since the insolubilization of
PDS and SGPep is evaluated on the basis of their CE spe
cific activities, it is important that the lack of collagen
insolubilization is easily detected in terms of the spe
cific activity of CE hydroxyproline (Table V). On the
basis of specific activity data for the second day of
labeling, collagen insolubilization was inhibited 92
57
percent by BAPN treatment ((1 - 3.85/48.8) X 100 = 92%).
Consequently, few if any (see below), collagen molecules
were available to carry PDS and SGPep into the insoluble
fraction.
The presence of 3 percent of the total hydroxy
proline label in BAPN CE (Table VI) may be explained in
two ways. First, this label may represent a leak in the
BAPN-block of fibrous collagen insolubilization. Although
this seems unlikely because of the duration of effective
inhibition of insolubilization (see above), it is possible
if the leak was stopped completely sometime during the
first day of labeling; otherwise BAPN CE label would in
crease in the same fashion and for the same reasons as
control CE hydroxyproline label (Table VI). Second, this
label may represent hydroxyproline-containing molecules
("collagen") insolubilized by a mechanism other than
aldehyde-dependent cross-linking. Finally, it is unlikely
that this label represents the insolubilization of elastin,
since aldehyde-deficient elastin is either salt-soluble
(56) or partially cross-linked (57) and elastin is not
solubilized by collagenase (58,59).
The data of Tables V and VI require a qualifica
tion of the concept of a BAPN-insensitive "collagen."
Such a molecule could not have a significant soluble pre
cursor pool, as demonstrated for collagen, since this would
cause BAPN CE hydroxyproline label to increase from day 1
to day 2 as in control CE but at a lower level. This was
not observed. The alternative is a "collagen" that is
rapidly insolubilized after synthesis. In this case no
increase in specific activity on day 2 would be expected
because of the loss of precursor-proline specific activity
by day 2. The continued insolubilization of unlabeled
molecules on day 2 would not cause a significant decrease
in the hydroxyproline specific activity of BAPN CE because
the amount of BAPN-insensitive "collagen" insolubilized
per day contains only a small fraction of the total hydro
xyproline of CE.
PDS Insolubilization and Degradation
Incorporation of ^H-glucosamine into PDS in CE is
presented in Table VII. Since no major change in the DS
pool of CE was detected (Table IV) and DS from BAPN and
control samples had identical specific activities (^H DPM/
nanomole uronate) on both the first and second day of
labeling, no inhibition of PDS insolubilization was caused
3
by BAPN. The continued increase in H-PDS specific activ
ity from the first to the second day prevents any deter
mination of the half-life of the DS portion of PDS and
suggests that the cellular pool of galactosamine precursors
of DS is relatively stable compared to the proline pool
and the pool of glucosamine precursors of SGPep.
TABLE VII
INSOLUBILIZATION OF SGPep AND PDS;
Isotope Incorporation into CE
SGPepa PDSb
3h-dpm 3h-dpm 3h-dpm 14c-dpm
nanomole HexNH? g dry wt. nanomole UA nanomole UA
Labeled BAPN 3 • 66±. 14 722±63 14.31.77 2.721.06
1 DAY Control 3.02+.09 748i65 13.8i.15 2.82±.13
Labeled BAPN 3.491.47 679+26 22.3+1.0 1.93±.15
2 DAYS Control 3.021.02 646+52 23.3l.62 2.44+.04
Representing SGPep, the G-75 retarded fraction was dialyzed, lyophilized,
and the hexosamines separated from proline on ionic exchange columns of
AG 50W - X12. R-DPM/g dry wt. starting tissue was calculated from the
amount of isolated hexosamine present and the specific activity. The data
are presented as means+range, n=2.
bPDS was isolated by CTAB-MgCl2 chromatography (peak II). Colorimetric
uronic acid (UA) determinations and double-label scintillation counting
of peak II from each sample were used to obtain the data. The data are
presented as means±range, n=2.
60
1A
The incorporation of C-proline into the protein
core of PDS is also presented in Table VII. The decrease
in PDS proline specific activity from the first to the
second day of labeling indicates that all of the PDS syn
thesized during the pulse of labeled precursor-proline
on the first day was also insolubilized (i.e., appeared
in CE) before the end of the day. Therefore, PDS either
does not have a very large soluble precursor pool, or its
transition through this pool is more rapid than collagen's
transition through its 0.15 and 0.5 M salt-soluble pre
cursor pools (Table VI). If maximum incorporation of
1 /
C-proline is assumed to have occurred by the end of
day 1, a half-life of A.75 days can be calculated for
control PDS protein core by plotting the data of Table VII
on semi-log paper. Since only two points are available
for each plot, the values obtained may not be accurate;
however, the comparison between BAPN and control values
(obtained in the same way) seems valid. Thus, a 2.3-fold
increase in the rate of degradation of protein core was
observed for BAPN samples (t| = 2 days). This rapid
degradation is consistent with the decreased protein:
uronate ratios of BAPN PDS.
SGPep Insolubilization
3
Only the incorporation of H-glucosamine into SGPep
was studied. Proline specific activity was not determined
because of the probability of proline-containing peptide
contaminants. The hexosamine total and specific activities
for SGPep are presented in Table VII. The similarity be
tween control and BAPN total activities for SGPep indi
cates that no inhibition of insolubilization was caused
by BAPN. The elevated BAPN specific activity of SGPep
and its apparent decrease when compared to the unchanged
controls suggest a decreased pool and a slightly higher
rate of degradation for the BAPN samples.
CHAPTER IV
DISCUSSION
The results of labeling PDS and SGPep in CE demon
strate that these molecules were insolubilized in the
absence of aldehyde-dependent collagen cross-linking,
i.e., when collagen insolubilization was inhibited more
than 90 percent. This means that entrapment due to cross-
linking no longer can be considered the mechanism of 0.5 M
salt-insolubility for these molecules and, consequently,
if an association with collagen does exist, the associ
ative bond itself must be stable under these conditions.
The effects of BAPN on the noncollagen carbohydrate-
containing molecules of rat skin have previously been
studied by measuring the variations of hexosamine (7) and
hexosamine, uronate, hexose, and sialic acid (60) in
various extracts. Orlowski's data (7) obtained after 21
days of BAPN treatment, suggest an increase of DS in the
0.5 M salt-extract and a decrease in the insoluble frac
tion. Wagh's data (60) obtained after 37 days of BAPN
treatment, demonstrate a similar increase in uronate,
hexosamine, and sialic acid, although the hexosamines
were not fractionated and no direct measurement of the
insoluble fraction was made. These data are consistent
62
with inhibited insolubilization of proteoglycans and
glycoproteins similar to, and caused by the lack of
aldehyde-dependent collagen insolubilization. However,
the absence of any effect of BAPN on the insolubilization
(measured directly) of PDS or SGPep, a decreased protein
content of PDS, and an increased rate of degradation of
PDS protein core have been demonstrated in the present
study. These results suggest that the data of Orlowski
and Wagh could be more easily explained by a continuation
of the catabolic process detected in the present, short
duration study.
The half-life of PDS protein core was estimated to
be 4.75 days on the basis of the data in Table VII. This
is slightly shorter than the 6-day value estimated from
Bentley's data (32) for DS from the skins of rats of
approximately the same age. This discrepancy may be
explained by the small number of points used in the
present study to calculate the half-life or by a biphasic
degradation curve, i.e., more rapid degradation just after
synthesis, followed by a slower rate. Evidence for the
latter has been presented by Bentley (32) on the basis
of total DPM in DS/aorta.
Treatment with BAPN caused a 2.3-fold increase in
the rate of degradation of PDS protein core. BAPN did not
significantly effect the DS pool size in the present study
or the degradation of DS in Bentley's study (32). This
suggests that the carbohydrate portion of PDS is independ
ently degraded and that the protein degradation may be due
to a general protease rather than one with specificity
for PDS protein core. PDS with little (8) or no (15)
protein content would be expected in the 0.5 M salt-extract.
The lack of change in both the DS pool size and the rate
of insolubilization, therefore, suggests that up to 50
percent of the protein core of PDS can be removed without
influencing the insolubility of this molecule. The rate
at which the protein content decreased (Table IV) suggests
that this degradation did not begin with the onset of BAPN
effects on collagen.
The presence of labeled hydroxyproline in CE from
BAPN samples has been used to suggest the existence of a
BAPN-insensitive "collagen." Support for such a "collagen"
is also provided by Nimni's data (55) concerning the in
hibition of collagen insolubilization in rat skin by treat
ment with the lathyrogen, penicillamine. In this case,
samples were taken over a much wider range of times (2.5 hr
to 12 days) and with a higher frequency than in the present
study. In the presence of penicillamine after 9 days of
treatment, the salt- and citrate-insoluble collagen was
only slightly labeled but reached its maximum activity 10
hours after label injection and remained relatively
65
constant until the termination of the experiment. Two
explanations for these experimental observations seem
plausible. First, the radioactivity could represent
collagen insolubilized by association with noncollagen
molecules. In this case the collagen insolubilized must
have been recently synthesized. Second, the label could
be in a hydroxyproline-containing, collagen-like molecule
that is insolubilized by some other mechanism than alde
hyde-dependent cross-linking. Basement membrane is an
example of this possibility. Both possibilities will gain
further support from the following discussion.
In order to gain insight into the location and
function of PDS and SGPep, and to create a framework for
the present experimental observations, greater emphasis
will now be placed on the information from the literature.
SGPep
Information concerning the isolation and character
ization of SGPep has been presented in "Results." SGPep
from rat skin has not been previously reported in the
literature even though it contains more than half of the
insoluble sialic acid of this tissue. There are several
similarities between the characteristics and the methods
of preparation of SGPep and some fractions of basement
membrane. These will become apparent during the following
selected review of basement membrane biochemistry.
66
Basement membranes (BM) from both vascular and
avascular tissue sources possess strikingly similar carbo
hydrate compositions (Spiro, Review (52), (61)). The
carbohydrate is divided, almost equally by weight, between
two components, a disaccharide unit and a "heteropoly
saccharide" unit. While the disaccharide unit, glucose-
galactose-hydroxylysine, is identical to that found in
both soluble and insoluble collagen (62), the heteropoly
saccharide has not been detected in soluble collagen. The
branched heteropolysaccharide contains approximately 4
residues of galactose, 3 of mannose, 5 of hexosamines, 3
of sialic acid, 1 of fucose, and no glucose (63). The
hexosamine is present as both glucosamine and galacto-
samine in approximately an 6:1 ratio (64). This carbo
hydrate unit is very similar to those of thyroglobulin,
a^-acid glycoprotein, fetuin, and ovalbumin (63). The
molecular weight of the heteropolysaccharide is 3500 based
on the molecular weight (5590 by equilibrium sedimenta
tion) of the collagenase-pronase derived glycopeptide and
its carbohydrate composition (63). Despite its actual
molecular weight, glycopeptide elutes in the void volume
of Sephadex G-25, and just begins to elute slightly behind
the void volume of G-50, producing a broad peak twice the
width of the void volume peak (63).
67
A most important characteristic of glomerular "base
ment membrane is its sensitivity to purified collagenase
alone. Kefalides (51) has demonstrated that treatment of
isolated, intact dog glomerular BM with collagenase pro
duces a small glycopeptide that just precedes a hydroxy-
proline-hexose peak in the low molecular weight region of
the G-200 profile. This material contained hydroxylysine,
but not hydroxyproline. The carbohydrate composition was
very similar to that of isolated heteropolysaccharide
(above, and Table II). Spiro (52) has performed almost
the identical experiment on bovine BM followed by chroma
tography on Bio-gel A-0.5 m. Again collagenase digestion
produced a low molecular weight, sialic acid-containing
glycopeptide which preceded a major hexose peak. In both
cases a small amount of collagenase-residue was observed.
Kefalides (51) characterized the carbohydrate composition
of this material and found it more like that of hetero-
polysaccharide than whole BM. In addition, it could be
solubilized by reduction and alkylation in 8 M urea, and
it contained only trace amounts of hydroxylysine and
hydroxyproline.
BM is very insoluble due to the presence of both
hydrogen-hydrophobic bonds and disulfide cross-links
(other kinds of cross-links are possible). However, 80 per
cent of the weight of the membrane can be solubilized by
68
reduction and alkylation in 8 M urea, generally performed
at 40°. Limited solubility in 8 M urea alone has also
been observed (51,65). Limited TCA-hydrolysis (4$, at
90°) of BM produced hexosamine-containing fragments devoid
of hydroxyproline (66). This treatment is normally used
to cleave the peptide backbone of insoluble fibrous col
lagen which solubilizes this material.
It is unclear whether BM contains a significant
amount of collagen structure. Although a collagen-like
fragment (ca. 8% by weight) has been isolated from whole
BM by mild pronase treatment (67), and incorporation of
labeled precursors into a procollagen-like molecule by
embryonic chick lens cells has been reported (68), the
amount of such material in intact BM seems small (52).
On the contrary, most of BM appears to be composed of a
collagen-like glycoprotein. Support for this conclusion
is derived from the following observations: No heteropoly-
saccharide-free fractions are obtained by agarose gel-
filtration of reduced and alkylated (in 8 M urea) BM. No
fractions free of heteropolysaccharide are observed (69)
after DEAE-Sephadex chromatography of the same material in
8 M urea. Both high and low molecular weight fragments
containing heteropolysaccharide are produced by collagenase
digestion of BM. The small to medium fragments are devoid
of hydroxyproline and, in some cases, hydroxylysinej the
69
large fragments contain these amino acids (51). These
data demonstrate that heteropolysaccharide can be found
both within and without collagen-like (collagenase-sensi-
tive) amino acid sequences. The same conclusion can also
be reached for disaccharide using similar logic and data.
The antigenicity of BM has been described by
Kefalides (51):
The cross-reactivity observed between glomerular,
alveolar, and lens basement membranes indicates
that identical antigenic components reside in
homologous basement membranes whether they are
found in vascular or avascular tissue.
I would like to suggest that SGPep is derived from
the epithelial, endothelial, and hair follicle basement
membranes or similar "collagen-like" glycoproteins. This
suggestion is consistent with the BAPN-insensitive insol
ubilization of both hydroxyproline and SGPep, the produc
tion of SGPep from the salt-residue by the action of bac
terial collagenase, and the available characteristics of
SGPep (Table II, and Results).
The concept of "similar collagen-like glycopro
teins" is derived from the similarities between structural
glycoproteins (sGP) and fractions of BM obtained by simi
lar methods (Table II). Structural glycoproteins have
always been prepared from insoluble residues after complete
removal of hydroxyproline (presumed to be fibrous collagen)
by treatment with either collagenase or hot TCA. The
70
residue that remains was then extracted with urea or urea-
mercaptoethanol in order to obtain the sGP fraction (21).
The sGP from both skin (18) and the media of aorta
(21) have amino acid compositions characterized by the
absence of hydroxyproline and hydroxylysine, the presence
of cysteine, and elevations in leucine, glutamic and as
partic acids. The detailed carbohydrate composition of
aorta sGP (23) and the general compositions of both aorta
and skin (18) sGP (Table II) are very similar to those of
BM heteropolysaccharide, except for slightly decreased
levels of fucose and sialic acid. It is significant that
the amino acid and carbohydrate compositions of these sGP
are completely different from those of whole BM. However,
treatment of BM with collagenase generates (51) two medium
molecular weight (ca. 100,000 and 60,000) fragments and a
residue (solubilized by urea-mercaptoethanol) consistent
with all of the described characteristics of these sGP.
In addition, compositionally similar fragments of BM are
obtained by hot TCA extraction (66).
Evidence of a hybrid primary structure of sGP
similar to that proposed for BM (above) is available.
Some preparation of sGP contain the disaccharide hydroxy-
lysine-gal-glc (21). Immunological cross-reactivity is
observed between sGP and its preceding TCA-extract. TCA-
extracts of skin, tendon, and aorta contain heteropoly-
71
saccharide "tightly" bound to "collagen" (71). Collagenase
digestion of the salt-insoluble residue of rat skin caused
82 percent of the noncollagen nitrogen released to be
dialyzable (24). Some or all of this effect, however,
may have been due to the presence of endogenous protease.
Immunological cross-reactivity also has been ob
served between sGP and BM. Fluorescein-labeled antibodies
directed against sGP localize over both "BM" and fibrillar
structures (71). In calf skin (72), labeled antibody to
corneal sGP localizes over vascular structures, hair
follicles, and lamellar structures.
It is not intended that BM and sGP be taken as the
same molecule or structure, although this may be the case.
I have intended to show that in the absence of destructive
methods of isolation, they may represent a class of struc
tural molecules with a hybrid collagen-glycoprotein struc
ture. Such a class of molecules would be likely candidates
for the precursor of SGPep. Since the function of such
molecules would probably be intermediate between that of
fibrillar collagen and the globular, hydrophobic glyco
proteins of cell membranes, and would probably be depend
ent on both elements of the structure (certainly for BM),
the importance of demonstrating and retaining this struc
ture in future studies is apparent. This would be especi
ally true for sGP from connective tissues devoid of BM.
72
Experimental methods and approaches are already available
to test every aspect of this suggestion.
PDS
Initially the most tempting role to postulate for
PDS is that of a nucleating agent in collagen fibrillo-
genesis. However, several lines of reasoning deny this
contention.
PDS has been shown to precipitate the nucleating
fraction of collagen instantaneously at either 4° or 37°
(8). This precipitate is amorphous, as are the collagen
nuclei formed during the lag phase of fiber formation (11).
It cannot, however, represent "true nuclei" since the lag
phase would decrease and fiber growth would occur at an
earlier time. Just the opposite was observed (11). This
suggests that PDS actually inhibits collagen nucleation.
PDS could produce the larger fibers (fewer in number) ob
served by decreasing the availability of nucleating collagen
rather than by increasing the dimensions of the nuclei,
as has been proposed (11,16) for glycosaminoglycan-collagen
interactions. The lack of stability, demonstrated by the
sensitivity of the initial precipitate to small increases
in ionic strength (14) and its subsequent rearrangement at
37° to form 640 ft banded collagen fibers (8), also supports
the contention that "true nuclei" are not present. Also,
73
PDS has no effect on the much more rapid growth phase (8).
More speculatively, the formation of microfibrils
may preclude PDS-collagen interactions except at their
surface. This possibility gains support from the micro
fibrillar subunit model of fiber formation proposed by
Veis (73) and modified by Smith (74). Veis has shown that
native fibers can be reduced to these subunits (75) and
that a high molecular weight collagen fraction will form
similar fibrils in vitro (13).
Although the metabolic turnover of DS has been used
to suggest collagen-association (4), it really seems a poor
criterion and certainly provides no positive support for
a nucleation function (7). DS from rat skin has a half-
life of 5i-6 days (32,76) while chondroitin sulfate,
heparin, and hyaluronic acid from the same source (76)
have half-lives of 8, 8, and 3 days, respectively. It has
been implied that collagen would protect a proteoglycan
associated with it and, thus, increase the proteoglycans
half-life in relation to other proteoglycans (4). Chon
droitin sulfate and heparin have not been seriously con
sidered associated with collagen in rat skin, and yet they
possess half-lives longer than DS. More important, how
ever, is the half-life comparison between DS and insoluble
collagen (t^ greater than 28 days (55)). It seems reason-
■ a
able that PDS at the core of cross-linked collagen fibrils
should have a half-life similar to that of the surrounding
74
collagen and significantly different from that of unpro
tected proteoglycan. This is not observed. This logic
has been used by Orlowski (7) to suggest a transient role
for proteoglycan in collagen fiber formation, a conclusion
consistent with the turnover of the protein core described
in this study, and the organizing role in fiber formation
proposed by Jackson and Bentley (3) and Mathews (77).
A positive function for PDS at the microfibrillar
level is suggested by the macroscopic fibrous form of PDS-
collagen gels formed in vitro at 37°. Gels formed in the
absence of PDS were homogeneous in appearance (8). The
PDS-dependent nature and the increased stability of this
architecture were demonstrated by the resistance of the
PDS-collagen gels to dissolution at 4° (78). Two biolog
ical systems are also consistent with this function.
Altered fibril and fiber architecture has been observed in
scleroderma (79,80) and corneal opacities (81), coincident
with increased amounts or the appearance of PDS.
The results of the present study ask one major
question of any description of the in vivo location or
function of PDS. What type of interaction between PDS and
other connective tissue components leads to the insolu
bility of PDS in 0.5 M salt, in the absence of cross-linking-
dependent entrapment? In the case of interaction with
collagen, it has been shown (15) that PCS, an ionic analog
to PDS, no longer ie bound at ionic strength 0.35. The
other major type of interaction possible is through
hydrogen-hydrophobic bonds. The demonstration of revers
ible aggregation between proteoglycan subunit and gly
coprotein link (82,83), and proteoglycan subunit-core
and glyboprotein link (84,85) after dissociation with
2 M guanidine-HCl (both dependent on intramolecular dis
ulfide bonds (protein conformation)) proves that such
interactions are possible for proteoglycan in the presence
or absence of the high degree of steric hindrance created
by extended glycosaminoglycan chains. The nonionic na
ture of the aggregates is supported by the observation
of Mathews (86) that a preparation of PCS which contained
aggregates produced lower molecular weight compounds after
treatment with 8 M urea than after treatment with hyal-
uronidase. Unfortunately, no such elegant experiments
have been performed with PDS. However, the fact that PDS
consists of 50 percent protein (8), compared to 7 percent
for PCS (82), would appear to increase the potential of
PDS for hydrogen-hydrophobic bonding. The salt-solubility
of the PCS-collagen precipitate formed at 4° under physi
ological conditions, and the stability of the PCS-collagen
association at ionic strength 0.2 (14) appear consistent
with the concept of hydrogen-hydrophobic interactions be
tween proteoglycan and collagen, but inconsistent with
76
insolubilization of proteoglycan or collagen by this
mechanism.
Conclusion
The picture that emerges from the present data and
literature is unclear but more amenable to experimental
test than before the present study was performed. PDS
seems to be associated with collagen under physiological
conditions, probably at the fibrillar level of organiza
tion. This association probably involves both salt and
hydrogen-hydrophobic bonds, but might not be responsible
for the insolubility of PDS. SGPep appears to be derived
from a hybrid collagen-glycoprotein molecule located in
the basement membranes (not associated with fibrous col
lagen) or in the fibrillar components of the tissue
(probably associated with fibrous collagen). In either
case, the characteristics of the hybrid precursor insure
insolubilization (by hydrogen-hydrophobic and disulfide
bonds) even at the level of self-aggregation. It is
tempting to speculate that an association between PDS
and the precursor of SGPep also exists in the fibrillar
matrix. This association would provide an additional
opportunity for the SGPep precursor to influence collagen
architecture and would also provide a cross-linking-
independent mechanism for PDS insolubilization.
77
BAPN has been effectively used to demonstrate that
collagen cross-linfcing-dependent entrapment of PDS and
SGPep is not the mechanism of insolubilization for these
molecules. Thus, the major impact of the present study
has heen to focus attention on alternative mechanisms of
insolubilization. Viable alternatives for PDS exist, but
they have little or no experimental support, while the
characteristics of SGPep have suggested an identity for
its precursor and, consequently, a mechanism for its
insolubilization.
a
REFERENCES
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Meyer, K., Harvey Lectures 51, 88 (1957).
Robert, L. and Robert, B., in "Protides of Biological
Fluids," ed. by H. Peeters, Vol. 15, p. 143.
Elsevier, Amsterdam (1967).
Jackson, D. S., and Bentley, J. P., in "Treatise on
Collagen," ed. by B. S. Gould, Vol. 2, Part A,
p. 189. Academic Press, New York (1968).
Hallen, A. A., in "Chemistry and Molecular Biology of
the Intercellular Matrix," ed. by E. A. Balazs,
Vol. 2, p. 903. Academic Press, New York (1970).
Sweeney, P. R., Pearce, R. H., and Vance, H. G.,
Canad. J. Biochem. 4l_, 2307 (1963 ).
Lowther, D. A., Toole, B. P., and Meyer, F. A., Arch.
Biochem. Biophys. 118, 1 (1967).
Orlowski, W. A., and Orbison, L. J., Lab. Invest.
23, 246 (1970).
Toole, B. P., and Lowther, D. A., Arch. Biochem.
Biophys. 128, 567 (1968).
Gross, J., Highberger, J. H. and Schmitt, F. 0.,
Proc. Nat. Acad. Sci. (USA) 41, 1 (1955).
Jackson, D. S., and Fessler, J. H ., Nature (London)
176, 69 (1955).
79
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
80
Wood, G. C. and Keech, M. K. , Biochem. J. 75.* 588
(1960).
Lowther, D. A., Toole, B. P., and Herrington, A. C.,
in "Chemistry and Molecular Biology of the Inter
cellular Matrix," ed. by E. A. Balazs, Vol. 2,
p. 1135. Academic Press, New York (1970).
Clark, C. C., and Veis, A., Biochem. 11_, 494 (1972).
Toole, B. P., and Lowther, D. A., Biochem. J. 109,
857 (1968).
Mathews, M. B., in "Chemistry and Molecular Biology
of the Intercellular Matrix," ed. by E. A. Balazs,
Vol. 2, p. 1155. Academic Press, New York (1970).
Obrink, 0., in "Chemistry and Molecular Biology of
the Intercellular Matrix," ed. by E. A. Balazs,
Vol. 2, p. 1171. Academic Press, New York (1970).
Lowther, D. A., and Natarajan, M., Biochem. J.
127, 607 (1972).
Timpl, R., Wolff, I., and Weiser, M., Biochim.
Biophys. Acta, 194, 112 (1969).
Furthmayr, H., and Timpl, R., Biochim. Biophys. Acta,
221, 396 (1970).
Wolff, I., Fuchswans, W., Weiser, M., Furthmayr, H.,
and Timpl, R., Eur. J. Biochem. 20, 426 (1971).
Robert, A. M., Robert, B., and Robert, L., in "Chem
istry and Molecular Biology of the Intercellular
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
81
Matrix," ed. by E. A. Balazs, Vol. 1, p. 237.
Academic Press, New York (1970).
Moczar, E., and Moczar, M. in "Chemistry and Molecular
Biology of the Intercellular Matrix," ed. by E. A.
Balazs, Vol. 1, p. 243. Academic Press, New York
(1970).
Moczar, M., Moczar, E., and Robert, L., Athero
sclerosis, 12, 31 (1970).
Adelmann, B., Marquardt, H., and Kuhn, K., Biochem.
Z. 346, 282 (1966).
Jackson, D. S., and Bentley, J. P., J. Biophysic, and
Biochem. Cytol. £, 37 (i960).
Loewi, G., and Meyer, K., Biochim. Biophys. Acta
27, 453 (1958).
Kaplan, D., and Meyer, K., Proc. Soc. Exp. Biol, and
Med. 105, 78 (i960).
Furuto, D., Barnes, P., Benya, P., and Schneir, M.,
International Association Dental Research Abstracts,
Abstract #475, p. 177 (1973).
Narayanan, A. S., Siegel, R. C., and Martin, G. R.,
Biochem. Biophys. Res. Cummun. 46, 745 (1972).
Tanzer, M. L., Int. Rev. conn. Tissue Res. 3_, 91
(1965).
Wirtschafter, Z. T., and Bentley, J. P., Lab. Invest.
11, 365 (1962).
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
82
Bentley, J. P., Wuthrich, R. C., and Van Bueren,
A. M., Atherosclerosis, 12, 159 (1970).
Orhison, J. L,, and McCrary, C., Lab. Invest. 25,
309 (1971).
Haney, A. F., Peacock, E. E., and Madden, J. W.,
Proc. Soc. Expt'l Biol. Med. 142, 289 (1973).
Benya, P. D., Berger, K., Golditch, M., and Schneir,
M., Anal. Biochem. (in press.)
Peterkofsky, B., and Diegelmann, R., Biochem. 10,
988 (1971).
Warren, L., J. Biol. Chem. 234, 1971 (1959).
Johnson, A. R., Anal. Biochem. 4£, 628 (1971).
Elson, L. A., and Morgan, W. B. J. , Biochem. J. 27.,
1824 (1933).
Mashburn, T. A., Jr., and Hoffman, P., Anal. Biochem.
36, 213 (1970).
Bitter, T., and Muir, H., Anal. Biochem. 4, 330
(1962).
Dische, Z., J. Biol. Chem. 167, 189 (1947).
Di Ferrante, N., Donnelly, P. V., and Berglund, R. K.,
Biochem. J. 124, 549 (1971).
Stegemann, H., Z. Physiol. Chem. 311, 41 (1958).
Grant, R. A., J. Clin. Path. 17, 685 (1964).
Juva, K., and Prockop, D. J., Anal. Biochem. 15,
77 (1966).
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
83
Lowry, 0. H., Rosebrough, N. J., Farr, A. L., Randall,
R. J., J. Biol. Chem. 193, 265 (1951).
Hartree, E. F. , Anal. Biochem. 48_, 422 (1972).
Winzler, R. J., in "Methods of Biochemical Analysis,"
ed. by D. Glick, Vol. 2, p. 290, Interscience, New
York (1955).
Gardell, S., in "Methods of Biochemical Analysis,"
ed. by D. Glick, Vol. 6, p. 289. Interscience,
New York (1958).
Kefalides, N. A., Conn. Tiss. Res. 1, 3 (1972).
Spiro, R. G., in "Chemistry and Molecular Biology
of the Intercellular Matrix," ed. by E. A. Balzs,
Vol. 1, p. 195. Academic Press, New York (1970).
Randerath, K., and Randerath, E., Anal. Biochem. 28,
110 (1969).
Kobayashi, T. K., and Pedrini, V., Abstra. of Orthoped.
Res. Soc. 18th Annual Meeting, p. 45, Washington,
D. C. (1972).
Nimni, M. E., J. Biol. Chem. 243, 1457 (1968).
Sandberg, L. B., Hackett, T. N., and Carnes, W. H.,
Biochim. Biophys. Acta, 181, 201 (1969).
Partridge, S. M., in "Chemistry and Molecular Biology
of the Intercellular Matrix," ed. by E. A. Balazs,
Vol. 1, p. 593* Academic Press, New York (1970).
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
84
Ross, R., and Bornstein, P., in "Chemistry and Mole
cular Biology of the Intercellular Matrix," ed. by
E. A. Balazs, Vol. 1, p. 64l. Academic Press, New
York (1970).
John, R., and Thomas, J., Biochem. J. 127, 261 (1972).
Wagh, P. V., Sun, C. N., White, H. J., Read, R. C.,
Proceedings International Congress on Gerontology,
Abstract #157, p. 59 (1972).
Kefalides, N. A., and Denduchis, B., Biochem. 8,
4613 (1969).
Cunningham, L. W., and Ford, J. D., J. Biol. Chem.
243, 2390 (1968).
Spiro, R. G., J. Biol. Chem. 242, 1923 (1967).
Spiro, R. G., J. Biol. Chem. 242, 1915 (1967).
Hudson, B. G., and Spiro, R. G., J. Biol. Chem. 247,
42 39 (1972).
Dische, R. M., Pappas, G. D., Grauer, A., and Dische,
Z., Biochem. Biophys. Res. Cummun. 20, 63 (1965).
Kefalides, N. A., Biochem. 7., 3103 (1968).
Grant, M. E., Kefalides, N. A., and Prockop, D. J.,
J. Biol. Chem. 247, 3539 (1972).
Hudson, B. G., and Spiro, R. G,, J. Biol. Chem. 247,
4239 (1972).
Moczar, M., and Robert, L., Atherosclerosis, 11,
7 (1970).
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
85
Robert, L., Darrell, R. W., Robert, B., in "Chemistry
and Molecular Biology of the Intercellular Matrix,"
ed. by E. A. Balazs, Vol. 3> p. 1591. Academic
Press, New York (1970).
Robert, L., Parlebas, J., Oudea, P., Zweibaum, A., and
Robert, B., in "Protides of Biological Fluids," ed.
by Peeters, Vol. 12, p. 110. Elsevier, Amsterdam
(1965).
Veis, A., in "Treatise on Collagen," ed. by G. N.
Ramachandran, Vol. 1, p. 367. Academic Press,
London (1967).
Smith, J. W., Nature (London), 219, 157 (1968).
Veis, A., Bhatnagar, R. S., Shuttleworth, C. A., and
Mussell, S., Biochim. Biophys. Acta, 200, 97
(1970).
Kofoed, J. A., Bozzini, C. E., and Tocci, A. A.,
J. Dent. Res. 50, 171 (1971).
Mathews, M. B., Biochem. J. 96, 710 (1965).
Toole, B. P., Nature (London), 222, 872 (1969).
Fleischmajer, R., Damiano, V., and Nedwich, A.,
Science, 171, 1019 (1971).
Fleishmajer, R., and Perlish, J. S., J. Invest. Derm.,
58, 129 (1972).
Anseth, A., and Fransson, L. A., Exp. Eye Res. £,
302 (1969).
82. Sajdera, S. W., and Haseall, V. C., J. Biol. Chem.
244, 77 (1969).
83. Haseall, V. C., and Sajdera, S. W., J. Biol. Chem.
244, 2384 (1969).
84. Haseall, V. C., Riolo, R. L., Hayward, J. Jr., and
Reynolds, C. C., J. Biol. Chem. 247, 4521 (1972).
85. Gregory, J. D., and Haseall, V. C., Fed. Proc. 31,
433 (1972).
86. Mathews, M. B., and Lozaityte, I., Areh. Biochem.
Biophys. 74, 158 (1958).

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Physicochemical Studies Of High-Density Lipoproteins Of Bovine Plasma

PDF

Steric effects in the acid, amine, and glyceraldehyde-3-phosphate dehydrogenase catalyzed hydrolysis of acyl phosphates

PDF

The RNA polymerases of Gymodinium breve and soybean hypocotyls

PDF

The Effect Of Polyols On Placental Estradiol Dehydrogenase

PDF

Changes in the rate of net synthesis of oral tissue collagen with age and wound healing

PDF

Crosslink Precursors In Fetal Calf Skin Soluble Collagens

PDF

The ferric complexes of citrate, ETDA and NTA: their nature and the ligand exchange reaction with human transferrin

PDF

Protein Kinase Activity In Human Lymphocyte Nuclei: Effect Of Cyclic Nucleotides And Phytohemagglutinin

PDF

The Relationship Of Macrophage Antigen Processing To Recognition Of Immunogenicity

PDF

Microelectrophoresis Studies On Cancer Cells

PDF

Studies On A Modified Fat

PDF

A History Of The Canadian Radio League: 1930-1936

PDF

Attitudes Of Academic Librarians In The Pacific Coast States Toward Library Technicians

PDF

Studies on the pulmonary surface-active lipoprotein: isolation and characterization of its peptide moiety

PDF

Fluorescence Quenching By The Stable Free-Radical Di-Tert-Butylnitroxide And Thermal Rearrangement And Cleavage Reactions Of Alpha-Iminooxetanes

PDF

Changes In Leucine Transfer Ribonucleic Acid And Leucine Transfer Ribonucleic Acid Synthetase During Cotyledon Senescence

PDF

Microheterogeneity And Crosslinking Of Salt Soluble Collagen

PDF

Studies On Antigens Associated With Wilm'S Tumor (Nephroblastoma)

PDF

Studies On The Enzymatic Formation Of Amino Acid - Ribonucleic Acid Compounds

PDF

Isolation And Characterization Of Hageman Factor And Plasma Thromboplastin Antecedent

Asset Metadata

Creator Benya, Paul David (author)

Core Title The inability of collagen cross-linking to influence the insolubilization of proteodermatan sulfate and a sialo-glycopeptide

Degree Doctor of Philosophy

Degree Program Biochemistry

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag chemistry, biochemistry,OAI-PMH Harvest

Format dissertations (aat)

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Schneir, Michael L. (committee chair), Allerton, Samuel E. (committee member), O'Brien, Richard L. (committee member), Petruska, John A. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-860766

Unique identifier UC11364489

Identifier 7331631.pdf (filename),usctheses-c18-860766 (legacy record id)

Legacy Identifier 7331631

Dmrecord 860766

Document Type Dissertation

Format dissertations (aat)

Rights Benya, Paul David

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA