Full text of DOI:10.1007/s001670000134

Knee Surg, Sports Traumatol, Arthrosc
(2000) 8:281–285
KNEE
DOI 10.1007/s001670000134
Type V collagen is increased during
rabbit medial collateral ligament healing
Christopher Niyibizi
Karl Kavalkovich
Tomoo Yamaji
Savio L-Y. Woo
Abstract To understand the repara-
tive process of medial collateral liga- collagen indicated that in sham con-
corresponding to types I, III, and V
Received: 1 February 2000
Accepted: 25 May 2000
ment (MCL), fibrillar collagen and
their relative ratios in healing MCL
trols types III and V collagen repre-
sented about 8% and 12%, respec-
Published online: 4 August 2000
© Springer-Verlag 2000
with anterior cruciate ligament (ACL) tively, of the type I collagen whereas
reconstruction were analyzed. Skele- the healed MCL ligaments at 6 weeks
tally mature New Zealand white rab- showed significant increase in type III
bits were subjected to a mop-end tear and V collagen to about 19% and
of MCL without repair with ACL re- 24%, respectively. By 52 weeks
construction. Rabbits were killed 6
and 52 weeks after injury. Ligamen-
tous tissues from the injury site and
sham controls were soaked in 0.5 M
acetic acid for 24 h, minced, and
type III collagen in the healed MCL
had returned to that of sham controls
while type V collagen remained ele-
vated at approximately 18%. These
data suggest that presence of type V
C. Niyibizi (✉) · K. Kavalkovich ·
T. Yamaji · S. L-Y. Woo
treated with pepsin to solubilize col- collagen in high concentration in
Collagen Biochemistry Laboratory,
Musculoskeletal Research Center,
Department of Orthopaedic Surgery,
200 Lothrop Street, Rm C313,
Presbyterian-University Hospital,
University of Pittsburgh
School of Medicine,
Pittsburgh, PA 15213, USA
e-mail: niyi@pitt.edu
lagen. Pepsin solubilized about 80%
of the total collagen as determined
by hydroxyproline analysis of the
pepsin residues. Sodium dodecyl sul- be included in the analysis when
fate polyacrylamide gel electrophore- evaluating ligament healing.
sis analysis of the solubilized colla-
gen revealed presence of fibrillar
collagen types I, III, and V. Densito- ligament · Ligament healing ·
metric scanning of the protein bands Collagen · Fibril diameter
healing ligaments may have an influ-
ence on collagen fibril diameters
seen in healed ligament and should
Keywords Medial collateral
Tel.: +1-412-648-1091
Fax: +1-412-648-8412
rameters have been examined to try to understand the
reparative process of MCL; these include the amount of
collagen present [1, 24], collagen fibril diameters [7],
Introduction
Attempts to accelerate and improve ligament repair are level of collagen cross-links [8], amount of type III colla-
currently an active area of investigation. Several studies gen present relative to type I collagen [9], and proteogly-
from different laboratories using animal models of liga- can content [11]. No correlation has been established be-
ment repair have shown that medial collateral ligament tween any of the parameters and the mechanical proper-
(MCL) has potential to heal without intervention, but that ties of the ligament [19]. The collagen fibril diameters of
the healed tissue remains mechanically, structurally, and healing ligaments have, however, been shown to remain
materially inferior to normal ligaments [6, 22, 24]. The relatively smaller than those of normal ligaments even up
failure of the ligament to heal and reach the quality of that to 2 years after injury [7]. Studies on wound healing have
of normal ligament is not clearly understood. Several pa- suggested that tensile strength, toughness, and organiza-

282
were washed in saline containing protease inhibitors, after which
tissue samples from the injury site were dissected out, washed in
distilled water, and dried. Approximately 2 mg each of the dry lig-
ament was suspended in 1 ml of cold 0.5 M acetic acid and al-
lowed to soak for 24 h. Softened tissues were minced with a
scalpel blade and resuspended in 1 ml of 0.5 M acetic acid. To this,
pepsin was added at 1:30 by weight (enzyme:tissue) and stirred at
4°C for 24 h. After 24 h the pepsin concentration was raised to
1:10, and tissue digestion with pepsin was continued further for
24 h at 4°C with stirring. Pepsin digests were clarified by centrifu-
gation in a microfuge at the maximum speed of 14,000 rpm for
10 min. The supernatants were removed, and the pepsin residues
were reextracted in 0.5 M acetic acid for 24 h and then centrifuged.
The first and second supernatants were combined and neutralized
with 0.5 M NH4HCO3. Aliquots of the samples were dried and an-
alyzed by gel electrophoresis using the methods of Laemmli [12].
To resolve type III collagen chains from α1(I) chains, interrupted
gel electrophoresis was performed as described [20]. Quantitation
of collagen bands was performed by densitometric scanning of
protein bands corresponding to α1(I), α1(III) and α1(V) chains on
a Biorad G670 imaging densitometer (Biorad Laboratories, Her-
cules, Calif., USA). To determine collagen type relative ratios the
area under the peaks for α1(III) or α1(V) chains was divided by
the area under the α1(I) peak of the same sample. The relative per-
centage was obtained by multiplying the ratio obtained by 100.
Relative ratios of α1(III) or α1(V) from each sample were pooled
with ratios obtained from different samples within the experimen-
tal or sham group. Statistical analysis was performed using Stu-
dent’s t test for paired samples with significance set at P<0.05.
tion of the repaired tissue is related to the collagen fiber
diameter [4]. These findings suggest that the healed liga-
ments with smaller collagen fibrils are mechanically weaker
than those with the normal collagen fibrils.
Collagen fibril diameters are controlled by other mole-
cules that associate with the major fibrillar forming colla-
gens. Among these type V collagen has been implicated
in the regulation of type I collagen fibril diameters [13,
15, 16]. We have previously shown that normal MCL is
composed of six genetically distinct types of collagen
[17]. Type I collagen is the major fibrillar collagen of the
MCL and types III and V collagen are quantitatively mi-
nor components. The role of types III and V collagen in
ligaments has not been clearly defined. Previous studies
on ligament healing have demonstrated that during early
phases of ligament healing type III collagen is highly ele-
vated relative to type I collagen [9]. A high concentration
of type III collagen relative to type I is suspected to result
in smaller collagen fibrils [1]. No other collagen types
have however, been examined during ligament healing.
In this study we examined changes in fibrillar collagen
ratios in a mop-end tear of MCL injury model without re-
pair so as to gain insight into the reparative process of
MCL. We report here that, in addition to type III collagen
increasing, type V collagen levels increase relative to
type I collagen following ligament injury and remain ele- Hydroxyproline analysis
vated even up to 1 year after injury. The implication of the
To determine the amount of collagen solubilized by pepsin diges-
presence of high concentration of type V collagen relative
to type I collagen is discussed in the context of the sus-
pected role for type V collagen.
tion, the residues remaining after pepsin digestion were hy-
drolyzed in 6 M HCl at 108°C for 24 h. Samples were dried, and
aliquots of the hydrolyzates were subjected to hydroxyproline
analysis as described previously [23]. The amount of total collagen
solubilized from each sample was determined by assuming that
90% of the tissue dry weight is collagen.
Materials and methods
Twenty-four skeletally mature New Zealand white rabbits were
subjected to a combined injury of the ACL with reconstruction and
a mop-end tear of the MCL without repair, as described previously
[24]. Briefly, the MCL substance was ruptured to create a mop-end
by pulling a rod medially beneath the ligament; the ends of the
ruptured MCL were opposed but not repaired [21]. The ACL was
transected at its tibial insertion, and the ligament was detached.
The ACL was reconstructed using a flexor tendon allograft which
was trimmed to approximately 4 mm in width and 6 cm in length
at the time of surgery [24]. Contralateral knees received a sham
operation by opening the skin and facia and passing a rod beneath
the MCL without rupturing. The rabbits were divided into three
groups of eight animals for each healing time interval, i.e., 6, 12,
and 52 weeks. The analyses that were performed on the MCL of
the rabbits of each group have been previously reported [23, 24].
In the present study the collagen type ratios were determined from
the MCL harvested from the rabbit knees of the 6- and 52-week
groups. Postoperatively all animals were allowed unrestricted cage
activity.
Results
Pepsin solubilized approximately 70–80% of total colla-
gen in each of the ligaments as determined from the hy-
droxyproline analysis. The collagen solubilization was
about equal for both injured and sham controls. Sodium
dodecyl sulfate polyacrylamide gel electrophoresis (SDS-
PAGE) analysis of the collagen solubilized by pepsin di-
gestion indicated the presence of fibrillar collagen types I,
III, and V.
Figure 1 shows SDS-PAGE of collagen solubilized by
pepsin from sham and ruptured MCL after 6 and 52 weeks.
After 6 weeks of healing prominent protein bands corre-
sponding to α1(III), α1(V), and α2(V) chains are evident
from the collagen solubilized from the ligament of the ex-
perimental group (lane E). The same protein bands are
less intense in the collagen solubilized from sham controls
(lane S). After 52 weeks of healing only a faint protein
band corresponding to the α1(III) chain can be seen. In
contrast, α1(V) and α2(V) chains are evident from the
collagen solubilized from the ligaments of the 52-week
experimental group. In sham controls α1(V) and α2(V)
Collagen isolation and analysis
Ligaments were dissected from rabbit knee joints following me-
chanical testing and were suspended in saline containing protease
inhibitors. Four ruptured and four sham-operated ligaments at
6 weeks, and five ruptured and five sham-operated ligaments at
52 weeks were used for collagen analysis. The harvested MCLs

283
Fig.1 SDS-PAGE of collagen
solubilized from sham (S) and
experimental (E) MCL at 6 and
52 weeks. Individual α chains
were identified based on puri-
fied collagen standards type I,
III, and V (not shown). Sam-
ples were electrophoresed un-
der reducing conditions using
the delayed reduction tech-
nique. The protein band be-
tween α1(I) and α2(I) chains is
truncated α1(I) due to exten-
sive pepsin digestion. Gels were
stained in Coomassie blue and
destained in methanol/acetic
acid/water 1:1:8
Fig.2 Densitometric scans of SDS gels of the collagen solubilized from sham ligaments at 6 weeks (6WS) and 52 weeks (52WS) and
experimental ligaments at 6 weeks (6WE) and 52 weeks (52WE). The α chains corresponding to individual peaks are identified

284
Table 1 Collagen type ratios in heating MCL
were able to solubilize about 80% of the total collagen of
the ligamentous tissues when using the pepsin solubiliza-
tion procedure described above. In addition, consecutive
pepsin extractions of the same samples showed relatively
constant ratios between α1(III) or α1(V) chains to α1(I)
chains (data not shown). We have previously demon-
strated that normal bovine MCL is comprised of fibrillar
collagen types I, III, and V [17]. The role of type III col-
lagen in ligaments or other tissues is not clear, but this
collagen may associate with type I collagen fibrils and
may regulate the fibril diameters of type I collagen [18].
Studies have, however, shown that type III collagen may
associate with large collagen fibrils as well as small colla-
gen fibrils [10]. The present investigation demonstrates
that type III collagen returns to normal levels by 52 weeks
of healing, but that type V collagen remains elevated in
relation to type I collagen over the same period.
Studies on type V collagen in chick cornea have shown
that type V collagen may be involved in the regulation of
corneal collagen fibril diameters. Type V collagen ac-
counts for about 20% of the total collagen in chick cornea;
the thin fibrils of uniform diameter present in the cornea
are believed to result from the high concentration of type V
collagen in this tissue [13].
The tissue form of type V collagen has been shown to
retain amino terminal extension peptides [5, 15]; these
peptides are believed to play a role in the regulation of
type I collagen fibrils in a heterotypic assembly. A model
of the way in which type V collagen regulates type I col-
lagen fibrils has been proposed [13]. In this model type V
collagen helical domain is buried in the interior of the het-
erotypic collagen fibrils; the amino-terminal domains re-
tained on the tissue form of type V collagen α chains pro-
ject out on the surface of the collagen fibrils, presumably
through the hole zone. In this assembly the retained
amino-terminal domains act as steric hindrance for further
addition of type I collagen molecules thereby regulating
the lateral growth of collagen fibrils. Indeed studies on in-
termolecular cross-linking of bone type V collagen have
demonstrated covalent cross-links between types I and V
collagen [15].
Further evidence for the role of type V collagen in the
regulation of collagen fibril diameters has been obtained
from studies on transgenic mice with an in-frame deletion
of exon 6 of the collagen α2(V) chain [2]. In these mice
the skin was found to be extremely fragile and to contain
disorganized fibrils of variable diameters, resembling those
seen in some forms of Ehlers-Danlos syndrome. Further-
more, linkage of the α1(V) gene to Ehlers-Danlos syn-
Type III/I
Type V/I
6 weeks
(n=4)
52 weeks
(n v=5)
6 weeks
(n=4)
52 weeks
(n=5)
Sham
7.9±1.8
7.2±2.1
8.5±3.9
0.5446
13.5±3.5
24.9±3.6
0.0105
11.1±1.5
18.1±4.9
0.0327
Experimental 19.0±5.3
0.0118
P
chains are present as faint bands. These data indicate that
type V collagen is present in higher concentration in the
ligaments of the experimental group. The protein band
migrating between α1(I) and α2(I) is truncated α1(I) gen-
erated by pepsin digestion [3].
Figure 2 shows the representative densitometric scans
of the collagen solubilized from ruptured ligaments and
sham controls after 6 and 52 weeks. The scans show the
relative intensities of each of the collagen chains. After
6 weeks of healing the α1(III), α1(V) and α2(V) chains
are evident. After 52 weeks of healing α1(V) and α2(V)
chains are evident in the experimental group while the
α1(III) chain is barely detectable, again suggesting that
type V collagen is elevated in the collagen isolated from
the ligaments of the experimental group.
Analysis of the areas under the peaks for α1(III) and
α1(V) chains from sham and injured ligaments and their
relative ratios to the area under α1(I) chains of the same
sample revealed that in control ligaments types III and V
collagen comprised about 8% and 12%, respectively, of
type I collagen (Table 1). The injured ligaments after
6 weeks showed a significant increase in type III collagen
to 19%. By 52 weeks type III collagen levels in the ex-
perimental group returned to that of the sham group or
were barely detectable on SDS-PAGE (Table 1). Simi-
larly, after 6 weeks the injured ligaments showed a signif-
icant increase in type V collagen to about 24% (Table 1).
In contrast to type III collagen, however, by 52 weeks
type V collagen in the MCL of experimental group re-
mained elevated at about 18%, which although reduced
from the 6-week group is still significantly higher than
that of the sham control. Although the data reported here
cannot be taken as absolute values, they suggest that type
V collagen remains elevated even up to 52 weeks of liga-
ment healing.
Discussion
Previous studies on MCL healing using cyanogen bro- dromes I and II has been demonstrated [14]. All together,
mide digestion have shown that type III collagen levels these data strongly demonstrate that type V collagen is in-
increase in early phases of ligament healing [6]. In the volved in the organization and regulation of type I colla-
present study we demonstrate that, in addition to type III gen fibril diameters. Although the present study was per-
collagen, type V is also increased during ligament healing formed on ligaments obtained from the combined injury
in relation to type I collagen. Although pepsin solubiliza- model, the same findings may apply to isolated MCL in-
tion of collagen from tissues is usually not efficient, we jury.

285
In view of the ascribed role for type V collagen, to- V collagen in high concentration should be considered
gether with the correlation drawn between collagen fibril among the parameters when evaluating ligament healing.
size and tissue strength in wound healing, it will be of in-
Acknowledgements This work was supported, in part, by NIH
terest to determine whether the smaller collagen fibril di-
grants R29AR42720 and AR 41820. We also thank Lou Duerring
ameters seen in healed ligament are due to the presence of
for typing the manuscript and Dr. Frank Shuler for help with the il-
high levels of type V collagen. Thus the presence of type
lustrations.
References
1.Amiel D, Frank CB, Harwood FL,
Akeson WH, Kleiner JB (1987) Colla-
gen alteration in medial collateral liga-
ment healing in a rabbit model. Con-
nect Tissue Res 16:357–366
10.Keen DR, Sakai LY, Bächinger HP,
Burgeson RE (1987) Type III collagen
can be present on banded collagen fi-
brils regardless of fibril diameter. J
Cell Biol 105:2393–2402
11.Plaas AH, Wong-Palms S, Koob T,
Hernandex D, Marchuk L, Frank CB
(2000) Proteoglycan metabolism dur-
ing repair of the ruptured medial collat-
eral ligament in skeletally mature rab-
bits. Arch Biochem Biophys 374:35-41
12.Laemmli UK (1970) Cleavage of struc-
tural proteins during assembly of the
head of bacteriophage T4. Nature 227:
680–685
13.Linsenmayer TF, Gibney E, Igoe F,
Gordon MK, Fitch JM, Fessler LI, Birk
DE (1993) Type V collagen molecular
structure and fibrillar organization of
the chicken α1 (V) NH2-terminal do-
main, a putative regulator of corneal
fibrillogenesis. J Cell Biol 121:1181–
1189
18.Romanic AM, Adachi E, Kadler KE,
Hojima Y, Prockop DJ (1991) Copoly-
merization of PN collagen-III and Col-
lagen I. PN collagen decreases the rate
of incorporation of collagen I into fi-
brils, the amount of collagen incorpo-
rated and the diameter of fibrils
formed. J Biol Chem 206:12703–
12709
19.Shrive N, Chimick D, Marchuk L, Wil-
son J, Brant R, Frank C (1995) Soft tis-
sue flaws are associated with the mate-
rial properties of the healing rabbit me-
dial collateral ligament. J Orthop Res
13:923–929
20.Sykes B, Puddle B, Francis M, Smith
R (1976) The estimation of two colla-
gens from human dermis by inter-
rupted gel electrophoresis. Biochem
Biophys Res Commun 72:1472–1480
21.Weiss J, Woo SLY, Ohland K, Horibe
S, Newton P (1991) Evaluation of a
new injury model to study medial col-
lateral ligament healing. Primary repair
versus non-operative treatment. J Or-
thop Res 9:516–528
22.Woo SLY, Inoue M, McGuire-
Burleson E, Gomez MA (1987) Treat-
ment of medial collateral ligament in-
jury. II. Structure and function of ca-
nine knees in response to differing
treatment regimens. Am J Sports Med
15:22–29
2.Andrikopoules K, Liu X, Keen DR,
Jaenisch R, Ramirez F (1995) Targeted
mutation in Col5a2 gene reveals a reg-
ulatory role for type V collagen during
matrix assembly. Nat Genet 9:31–36
3.Bätge B, Notbohm H, Diebold J,
Lehmann H, Bodo M, Deutzmann P,
Muller P (1990) A critical cross-link
region in human-bone derived collagen
type I specific cleavage site at residue
Leu 95. Eur J Biochem 192:153–159
4.Doillon C, Dunn MG, Bender E, Silver
F (1992) Collagen fiber formation in
repair tissue: development of strength
and toughness. Collagen Relat Res
5:481–492
5.Fessler JH, Fessler LI (1987) The type
V collagen. In: Mayne R, Burgesson
RE (eds) Structure and function of col- 14.Loughlin J, Irven G, Hardwick KL,
lagen types. Academic Press, Orlando,
pp 81–103
6.Frank C, Woo SLY, Amiel D, Har-
wood F, Gomez M, Akeson W (1983)
Medial collateral ligament healing: a
multidisciplinary assessment in rabbits.
Am J Sports Med 11:379–389
7.Frank C, McDonald D, Bray D, Bray
R, Rangeyyan D, Chimich D, Shrive N
(1992) Collagen fibril diameters in the
healing adult medial collateral liga-
ment. Connect Tissue Res 27:251–263
8.Frank CB, McDonald D, Wilson JE,
Eyre DR, Shrive N (1995) Rabbit me-
dial collateral ligament scar weakness
is associated with decreased collagen
pyridinoline cross-link density. J Or-
thop Res 13:157–165
9.Inoue M, Woo SLY, Gomez MA,
Amiel D, Ohland KJ, Hitabayashi LR
(1990) Effect of surgical treatment and
immobilization on the healing of the
medial collateral ligament: a long-term
multidisciplinary study. Connect Tis-
sue Res 25:13–26
Butcher S, Walsh S, Wordsworth P,
Sykes B (1995) Linkage of the gene
that encodes the α1 chain of type V
collagen (CoL5A1) to type II Ehlers-
Danlos Syndrome (EDS-II). Hum Mol
Genet 4:1649–1651
15.Niyibizi C, Eyre DR (1993) Structural
analysis of the extension peptides on
the matrix forms of type V collagen
from fetal calf bone and fetal calf skin. 23.Woo SLY, Niyibizi C, Matyas J,
Biochem Biophys Acta 1203:304–309
16.Niyibizi C, Eyre DR (1994) Structural
characteristics of cross-linking sites in
type V collagen of bone, chain speci-
ficities and heterotypic links to
Kavalkovich K, Weaver-Green C, Fox
RJ (1997) Medial collateral knee liga-
ment healing: combined medial collat-
eral and anterior cruciate ligament in-
juries studied in rabbits. Acta Orthop
Scand 68:142–148
type I collagen. Eur J Biochem
224:934–950
24.Yamaji T, Levine RE, Woo SLY,
Niyibizi C, Kavalkovich K, Weaver-
Green CM (1996) Medial collateral
ligament and anterior cruciate ligament
injury: an interdisciplinary study in
rabbits. J Orthop Res 14:223–227
17.Niyibizi C, Saggaria-Visconti C,
Kavalkovich K, Woo SLY (1995) Col-
lagens in an adult bovine medial collat-
eral ligament. Immunofluorescence lo-
calization by confocal microscopy re-
veals that type XIV collagen predomi-
nates at the ligament-bone junction.
Matrix Biol 14:743–751