Anu R. Biohe. 1995.

6:403-34
Coprigh Q 195 by Ana Revies In. Al righs resered
COLLAGENS: Molecula
Biology, Diseases, and Potentials
for Therapy
Darin J. Prockop
Dpnt of Biohemistry ad Moleula Bi ology, Jefferon Institute of
Molecular Medicine, Jefferson Medical College of Thom Jefferson Univesity,
Philadelphia, Pennsylvania 19107
Kari I. Kivirikko
Collagen Resarch Unit, Bioenter and Dpartmnt of Medical Biohemisty,
Univesity of Oulu, 90220 Oulu, Finland
KEY WORDS: fbosis, oteogenesis impefecta, transgenic mic, gene therapy, antisense
oligonucletide, antisense gene
CONTENTS
THE COLLAGEN FAMILY OF PROTEINS AND GENES...................... 4
Siructure and Funtions of the Collagen Triple Helix. . . . . . . . . . . . . . . . . . . . 4
Types ofCollagen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
BIOSyNTHESIS......................................................... 41 2
General Features . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 412
Intraellular Prcessing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . " . . . . . . . . . . . . . . 41 3
Extracellular Events . . " . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 5
Potentials for Inhibiting Fibrosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 7
MUTATIONS I N MEN AND MICE......................................... 41 9
Mutations in Patients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 419
MUiations in Trasgenic Mice. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 424
Potentials for Gene Terapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 426
SUMMARY. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 427
A
BSTRCT
Te collagen suprfamily of proteins now contains at least 19 protins formally
defned as collagens and an additional ten proteins tat have collagen-like
domains. The most abundant collagens form extrcellular fbrils or network
like structures, but the oters fulfll a variety of biological fnctions. Some of
43
06-415415/0701-0403$05.0
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
Quick links to online content
Further
ANNUAL
REVIEWS
40 PROKOP & KlVIRlKKO
the eight highly specifc post-translational enzymes involved in collagen bio
synthesis have recently ben cloned. Over 40mutations in 6 difernt colla
gens cause a variety of human diseases that include osteogeesis imperet,
chondrodysplasia, some fonns of osteoporosis, some fons of osteoarhrits,
and the renal disease known as the Alport syndrome. Many of the disese
phenotype have ben produced in tansgenic mice with mutated collagen
genes. There has been incresing interest in the possibility that th unique
post-tanslational enzymes involved in collagen biosynthesis offer attatve
targets for specifcally inhibiting excessive fibrotic reactons in a numbr of
diseases. A number of experiments suggest it may be possible to inhibit
collagen synthesis with oligo- nucleotides or antsense genes.
THE COLLAGEN FAMILY OF PROTIS AND GENES
At least 19 proteins a now known as collagens, and at least an additonal 10
proteins have collagen-like domains. Initally, collagens were defned a pro
teins of the extrcellular matix that contained large domains comprised of
repeating -Gly-X-Y- sequences and that folded into a unique tiple-helical
stcture. Screening of cDNA and genomic DNA libmrles with probes for
collagens rvealed a large number of relatd proteins with varying lengts of
repeating -Gly-X- Y- sequences. Because they were discovered in searches for
collagens, the proteins encoded were defned as collagens, even though in
some cases th tiple-helical domains were small and most of the protein
stcture was globular. Also, a few proteins studied in other contxts wer
found to contain triple-helical domains but wer not defned a collagens. The
variety in the structures of different collagens and related protins implies that
they have vastly difernt biological fnctions.
Beause of the extensive literature on collagens, in this chapter we concen
tmt primarily on rcent advances in the feld. For more detailed data, the
reader is referrd to severl prvious reviews on the stucturs and functons
of collagens (1-11), on the structures of collagen genes (12-14), on the bio
synthesis of the proteins (1, 6, 7, 15, 16), and on mutations in patients and
mice (2, 17-26). A vast literture is also available on cis-rgulatory elements
for collagen genes, and several tnscription factors have ben characterized.
The topic, however, is not reviewed here, in part beause there are conflictng
data fom similar experiments with different gene constucts in cell transfeton
assays. Also, dta fom cell tansfection assays are not always consistent wit
data fom experiments in tansgenic mice.
Structure and Functions of the Collagen Triple Heli
The collagen tiple helix is foned fom thre polypeptide chains that a each
coiled into a left-handed helix. The thre chains are then wrappe aound each
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
COLGES 405
other into a right-handed super-helix so that the fnal strcture is a rope-like
rod 0, 2, 4, 6, 7). The presence of glycine as every third amino aid in the
repeating -Gly-X-Y- sequence of each chain is essential, because a lager
amino acid will not ft in the center of the tiple helix wher the tre chains
come together. Proline is fequently in the X-positon of the -Gly-X-Y- se
quence and 4-hydroxyproline is fequently in the Y -position. These two amino
acids limit rotation of the polypeptide chains. The triple helix is frther stabi
lized by hydrogen bonds and water bridges, many of which reuire the presence
of 4-hydroxyproline. The conformation of the tiple helix places the side chains
of amino acids in the X- and Y -positons on the surface of the molecule. This
arangement explains the ability of many collagens to polymerize, since the
multple clusters of hydrophobic and charged side chains dirct self-asembly
into preisely ordered structures. The tiple helix is relatively rigid. In some
contexts, the resistance of the moleule to extension or comprssion is impor
tant for the biological fnction of the protein. In many collagens, the tiple
helix is interrupted by globular sequences that make the moleule more flex
ible, but the precise functons of the globular sequences are unknown.
Types of Collagen
For simplicity, the superfamily of collagens (6) can be divided into several
classes on the basis of the polymeric stuctures they form or related stuctural
features (see Figur 1): (a) collagens that form fbrils (types I, II, II, V, and
XI), (b) collagens that form network-like structures (te type IV family, and
types VII and X), (c) collagens that are found on the surface of collagen fbrils
and are kown as fbril-associated collagens with intruptd tiple helices
(FACITs that include types I, XI, XIV, XVI, and XIX), (d te collagen
that forms beaded flaments (tpe VI), (e) the collagen that forms anhoring
fbrils for basement membranes (tpe VI), i collagens with a transmembrne
domain (types XIII and XVII), and (g) the newly discovered typs XV and
XVIII collagens that have been only patially charactrized. An additonal
group (h) consist of protins containing tiple-helical domains that have not
been defned as collagens.
FIBRIL-FORMING COLLAGENS All these collagens (typs I-III, V, and XI) are
similar in size and in that they contain large tiple-helical domains with about
10 amino acids or 330 -Gly-X-Y- repeats pr chain. In additon, they are
also frst synthesized as larger precursors, and the precursors nee to be
proessed to collagens by cleavage of N-propeptides and C-propptdes by
spifc proteinases. Finally, they are similar in that they all assemble into
cross-stiate fbrils in which eah molecule is displaced about one-quarter of
its length relative to its nearst neighbor along the axis of the fbril (Figure 1).
Type I is te most abundant coHagen and is found in a variety of tissues. Many
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
406 PROKOP & KVIRIKKO
A 1,1I,III,V,XI
N-prpeplides
200nm
1 C-propeptides
f __ ! L (67 nm) staggered. molecues
'

e

,__

,
--

1Q<
IV
L
300nm
VIII, X C. IX, XII, XIV, XVI, XIX
monomer

100nm
D.
moDomer

VI
eUamer
to
dlmer
..
rx Xli.
LPL
I
chain
?--

I ! lil il
IifibriJ
100nm
E. VII
beaded flament

H.
m|IBOanl
1q p|0l|n
"\LO Wen agens mB>0phBg
mBnnan|n0|ng
cOI s r
p|0l|h Ong|m|n|n
J"-

OIIBgn ,,
Z0m M|I
memme
Fgur 1 Schematic for the structure of various collagens. The fgur is moifed after the fgure
presente by Hulmes (6) and reproduce here with prmission. The letlers refer to th classification
usd in the text Becaus the protein stctures ae still unkown, the sheme ds no pesent
collagen with a tranmembrane doman (type XIII and XVII) and the family of two newly
discvred cllagens (typs XV an XVIII).
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
COLLAGES 407
Table 1 Collagen types and the location of their genes on human chromosomes
Type Gene Chromosome Expression
COLlA I 17q2J.3-q22 Most connective tissues
COLlA2 7q2I.3-q22
II COL2AI 12q13-q14 Cartilage, vitreous humor
III eOL3Al 2q24.3-q3l Extensible connective tissues, e. g.
skin, lung, vascular system
IV eOL4Al 13q34 Basement membranes
COL4A2 13q34
COL4A3 2q3S-q37
COL4A4 2q3S-q37
eOL4AS Xq22
COL4A6 Xq22
V COLSAI 9q34.2-q34.3 Tissues containing collagen I, quantita-
eOL5A2 2q24.3-q3l tively minor component
COL5A3
VI COL6Al 21q22.3 Most connective tissues
eOL6A2 21q22.3
COL6A3 2q37
VII eOL7Al 3p2l Anchoring fibrils
VIII eOL8AI 3qI2-qI3.1 Many tissues, especially endothelium
eOL8A2 Ip32.3-p34.3
IX COL9AI 6q12-q14 Tissues containing collagen II
COL9A2 Ip32
COL9A3
X COLl DAI 6q21-q22 Hypertrophic cartilage
XI COLIIAI Ip21 Tissues containing collagen II
eOLl IA2 6p2L2
COL2Alb 12q13-q14
XII eOLl 2Al 6 Tissues containing collagen I
XIII COLl 3AI IDq22 Many tissues
XIV COLl 4AI Tissues containing collagen I
XV COLl SAI 9q2l-22 Many tissues
XVI COLl 6AI Ip34-35 Many tissues
XVII COLl 7AI l Oq24.3 Skin hemidcsmosomes
XVIII COLl 8AI 21q22.3 Many tissues, es pecially liver and
kidney
XIX COLl9AI 6q12-q14 Rhabdomyosarcoma cells
a For chromosome locations see References 21,44,75, 102,214.
The a3(XI) chain of type XI collagen is encoded by the same gene as the a (II) chain of type
II.
of the other fbril-foning collagens have a more selective tissue distibuton
(Table 1).
Among the new developments concering fbril-fonning collagens is the
discovery of alterative splicing of exons in te N-terminal propeptdes of tp
I (27-29) and typ XI collagens (3032). Beause of altratve splicing, the
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
48 PROKOP & KlVIRIKKO
coing sequences of an additional exon (27-29) are present in the typ I
proollagen formed in noncartilaginous tissues early in embryonic develop
mnt (33-37). Another new development concerning fibril-forming collagens
is the finding that many fbrils in vivo are composed of two or more difernt
collagen types (4, 6, 9). An additional discovery is that hybrid molecules
containing chains of both type V and typ XI collagens are present in some
tissues such as a hybrid moleule containing the a2(V) chain and the 01 (XI)
chain (4, 6, 9, 38-). Theefore, type V and type XI collagens can probably
b considerd as a single kind of collagen comprised of fve different chains
(9), i.e. al(V), a2(V), a.(V), al(XI) and a2(XI).
The gene structures of the fibril-forming collagens show a great deal of
similarity (12-14). One common feature of the genes is that the major tiple
helical domain of each chain is coded for by 42 exons. Most of the exons are
54 bp and the others are either twice 54, thre times 54, or combinations of
45 and 54 bp exons. Also, each exon begins with a complete codon for glycine,
and therfore the exon codes for a discret number of -Gly-X-y- tipeptde
units. I addition, the patter of exon sizes is similar in all the genes and has
ben highly conserved throughout evolution.
The genes for the 02(1) chain of type I collagen and al (Il ) chain of type
II collagen contain alterative promoters that code for difernt polypeptides
(41,42). The alteratve promoter of the COLlA2 gene is located witin inton
2, and the transcript contains a short open reading frame that is out of fame
with the collagen coding sequence (41). Thus, this RNA cannot encode a
collagen but may encode a noncollagen polypeptde. The transcript appars
early in embryogenesis in tissues derived fom neuroectoderm, but at later
stages of development, it is found almost exclusively in hyaline cartilage (41).
The alterative promoter of the COL3A 1 gene is located in intron 23, and the
trnscript may encode either a noncollagen polypeptide or a tuncated collagen
(42). This tanscript appears transiently in limb mesenchyme and then de
crases to low levels in intact calage (42). Te functons of these altratve
trnscripts are currently unknown.
NETWORK-FRMING COLLAGENS These collagens include the faily of type
I collagens found in basement membraes and type VIII and X collagens
(Figure 1). The collagenous domain of a type IV collagen molecule is longer
than in the fbril-forming collagens and consists of about 140 amino acids in
-Gly-X-Y- repeats that are fequently interrupted by short noncollagenous
sequences. The N-terminus of a molecule contains a small noncollagenous
domain and the C-terminus a major noncollagenous domain of about 230
amino acids (Figure I). The molecules self-assemble to form net-like strctures
in which monomers associate at the C-trmini to form dimers and at the
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
COLGES 409
N-trmini to form ttamers. In addition to these end-to-end interactions, the
tiple-helical domains intertwine to fonn supercoiled stuctures (43-45).
Although most of the type IV collagen in basement membraes consist of
a combinaton of al(IV) and a2(IV) chains, some basement membranes
contain smaller amounts of moleules of a3(IV) and a4(IV) (46-49) or of
a(IV) and a6(IV) chains (50-56 ) that a simila but not identical. Furter
variation in the stuctur of type IV collagens is caused by alterative splicing
of RNA tanscripts for the a3(IV) chain (57, 58). Also of interest is tat the
genes of type IV collagens a found in pairs with head-to-head orientatons
on different chrmosomes so that the promoter regions overlap (Table 1).
The al(I) and a2(IV) chain genes are head-to-head on chromosom 13
(43,44), the a3(IV) and a4(IV) chain genes a head-to-head on chromosome
2 (44, 47, 48, 59); and the a5(IV) and a6(IV) chain genes are head-to-head
on the X chromosome (44, 50, 5 1, 54, 55, 6 0). The stucturs of thee genes
differ distinctly fom those of the fibrillar collagens. Only a few exons are
54 or 45 bp, and many exons coding for the triple-helical domain begin wit
a split codon for glycine in which the frst G of the coon is in the preeding
exon (12-14, 4, 6 0).
The two other network-forming collagens, types VIII and X, are very dif
ferent in stucture fom type IV but similar to each other (47, 9, 6 1). The
al(VII), a2(VII), and al(X) chains all contain a collagenous sequence of
almost the same size and with eight imprfections in similar positons in the
-Gly-X- Y- sequences. Th genes for these two collagens all contain only three
exons, and almost all of the coding sequences ae found in the lage third exon
(4,12- 14,6 1). Descemet's membrane. which separates the coreal endothelial
cells fom the stoma. consists of stacks of hexagonal lattices made of type
VIII collagen (4). Type X collagen is among the most speializd of the
collagens and is synthesized primarily by hypertophic chondrocytes in the
deep-calcifying zone of cartilage (47, 6 2). The assembled form of typ X
collagen resembles the hexagonal lattice of typ VII in Decemet's membrne
(6 2).
FACIT COLLAGENS These collagens (types IX, XII, XIV, XVI, and XIX) do
not fonn fbrls themselves but are found attached to the surfaces of preexistng
fbrils of the fbril-forming collagens (3-1 1). All tese collagens are charac
terize by short triple-helical domains interupted by shor noncollagenous
sequences.
The type IX collagen molecule consists of three tple-helical domains and
four noncollagenous domains. The protein is commonly found on the surface
of fbrils of typ n collagen covalently bound to molecules of type II collagen
in antiparalel orientation (Figure 1). One unusual feature of collagen IX is
that it ofen occur as a proteoglycan in which a single glycosaminoglycan
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
410 PROKOP & KIVIRIKKO
side chain is covalently attahed to the seond noncollagenous domain of the
a2(IX) chain. In ocula and embryonic tissues, type IX collagen occurs in a
form with a shor al (IX) chain lacking nearly all of the N-terminal globula
domain. This short al (IX) chan is tascribd fm an ateratve promote
loated btween exons 6 and 7 of the al (IX) gene (63). The expression pattrs
of the long and shor forms sem to be both temprally and spatally regulated.
During avian development, the swith in expression fom the short form to the
long form occurs at the beginning of chondrogenesis during the early devel
opmnt of the vetebral column (64, 65).
Type XII and XIV collagens show several strctural similarities to type IX
collagen, paricularly in the C-terminal collagenous domains (3, 6, 8, 9,6670).
Tese two collagens also contain glycosaminoglycan side chains attahed to
the large N-terminal globular domain. The RNA transcripts for type XII and
XIV collagens undergo alteratve splicing that varies the structures of the
N-trminal globular domain. In the longest form of type XII collagen, the
N-trminal globula domain contains 18 fbronectin type II repeats and four
repeats homologous to the von Willebrnd factor A domain (8, 9). I the
longest form of type XIV collagen, te N-terminal domain contains eight and
two of these repeat respectively (69). Type XVI collagen (71, 72) and the
reently discovered tpe XIX collagen (73, 74) also show similarites in stuc
ture to the FACIT collagens and are therefor classifed into this subgroup.
BEADED FILAMENT -ORMING COLLAGEN The only collagen known to form
beaded flaments is type VI (75) (Figur 1). Each of the thre diferent chains
of the protein contains a very short triple-helical domain, and the remainder
consists of large N-terinal and C-trnal globula domains (75). Reently,
researchers found that the N-terminal globula region of the a3(VI) chain is
much larger than the same region in the other two chains (47, 75). The
N-trminal and C-terminal globular domains of all three chains contain 20
residue repats wit significant similarites to the A domains of von Willebrand
factor. The C-terinal region of the a3(VI) chain also contains thre additonal
domains that show similaities to salivar protins, to fibronetn typ II
repeats, and to Kunitz-type protease inhibitors (4-7, 9, 75). Several o2(VI)
and a3(VI) chain variants rsult fom alterative splicing of the repttve
nonollagenous subdomains (47,9,75,76).
COLLAGEN OF ANCHORING FIBRILS Type VII collagen forms anchoring fbrils
(Figure 1) that link basement membmnes to anchoring plaques of typ IV
collagen and laminin in the underlying extmcellular matix (77-82). The ti
ple-helical domain of type VII collagen, which is longer than the tiple helix
of any othe collagen, contains 1530 amino acids in -Gly-X-Y- repeats that
a interupted at 19 separate sites (82). The large N-trminal globular domain
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
COLGES
411
contains a segment homologous to cartlage matix protein. The segmnt is
followed by nine fbronectin type IIrepeats, one segment homologous t von
Willebrand factor A domain, and a segment that is cystein and proline rich.
The smaller C-terminal globular domain contains a segment homologous to
Kunitz-type protease inhibitors (82). The protein is frrst assembled into anti
parallel dimers formed by a small overlap at the C-terminal globula ends. The
C-trminal globular domains appear to be cleaved during the assembly of
dimers and the dimers are stabilized by disulfde bonds. The dimers then
assoiate laterally and in register to become the main constituents of anhoring
fibrils. Te gene for type VI collagen is about 31 kb and has 118 exons (83).
It therefor has mor exons than any oter known gene.
COLLAGENS WITH A TRANSMEMBRANE DMAIN Two recently discovered col
lagens contain a tansmembrane domain and, terfore, a probably not s
creted into the extacellular matix. Type XIII collagen ( 1 1, 8487) is found
in many tissues. In contast, type XVII collagen (80, 88-9 1) is found primarily
in the hemidesmosomes of the skin and is one of the two antigens that produce
the autoimmune disease known as bullous pemphigoid. These two collagens
ar not homologous in structure, but they both contain a single tansmembrne
N-trminal domain that is apparently cytoplasmic. The rmainder of the mole
cule is extracellular ( 1 1, 87,90,91). One of the most remarkable features of
type XIII collagen is that it undergoes extensive alterative splicing that can
generat several hundred forms of the protein (84, 92, 93). The altrative
splicing is unique among collagens in that it involves -Gly-X-Y- sequences.
How tis alteratve splicing alters the ptental of the protein for folding into
a triple helix in which all tree chains must b te same length is not clear.
FAMIL Y OF TYPES XV AND XVIII The newly discovered typ XV (9-98) and
XVIII (97, 99103) collagens have a large N-terminal globular domain, a
highly interrupted triple helix, and a large C-trminal globular doman. Both
collagens also contain several potental attachment sites for serine-linked gly
cosaminoglycans and asparagine-linked oligosaccharides, obseratons that
suggest these collagens may b extensively glycosylated. Both type XV and
type XVIII collagens are found in many tissues, but typ XVIII collagen is
expressed at much higher levels in the liver. Type XVII collagen is tanscribed
fom two alterative promoters, and the RA tanscript from one of the
promoters is frther modifed by alterative splicing witin sequences coding
for the N-trminal globular domain (103). The longest variant of type XVII
collagen has an N-terminal cysteine-rich sequence that is homologous to three
noncollagenous proteins in the family of G-protein coupled receptors ( 103).
"NONCOLLAGEN" COLLAGENS The group of proteins contining collagenous
sequences but not defned as collagens includes the subomponent C lq of
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
412 PROKOP & KlVIRlKKO
G8-O
OH

--
/
.. ....
B
..
.
J --..

J
.
.......
.
.
.....
.
..... t ,, 6" p
==
<<
===
Figure 2 Schematic for the biosynthesis of a fibril-forming collagen. (A) Intracellular events that
involve pot-translational hydroxylation an glycoylation, assoiation of plyppid chins, and
folding ofthe triple helix. (8) Extraellular events that involve cleavage of the N- and C-propptides,
self-asembly of collagen into fbils, an cros-linking of th fibrils. Reproduced with prission
from Reference 10.
complement, the tail structure of acetylcholinesterase, pulmonay surfactant
proteins SP-A ad SP-D, mannan-binding protein, conglutnin, colletn-43,
the bacterial enzyme pullanase, and type I ad tp II macrophage scavenger
reeptors (6, 11, 104-107). The type I ad II macrophage scavenger reetors
resemble type XIII and XVII collagens in that they contain a single tansmem
brane domain prceded by a short cytoplasmic N-terinal domain. Te rest
of the molecule is extracellular (11, 105).
BIOSYNTESIS
General Features
The fbril-forming collagens ar frst synthesized as larger preursor molecules
known as proollagens. The intaellula steps in the assembly of a proollagen
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
COLGES 413
a (Figure 2) cleavage of signal peptdes, hydroxylaton of Y -position proline
and lysine rsidues to 4-hydroxyproline and hydroxylysine; hydroxylation of
a few X-position proline rsidues to 3-hydroxyprolin, addition of galactose
or both galactose and glucose to some of the hydrxylysine residues, additon
of a mannose-rich oligosaccharide to one or both of the propptdes, assoiaton
of the C-terminal propeptides through a process direted by the stuctur of
these domains, and formation of bth intrchain and interchain disulfde bonds.
Aftr the C-propeptides have associated and each chain has acquired about
10 4-hydroxyproline rsidues, a nucleus of triple helix forms i the C-trminal
region, and the tiple helical confonnation is then propagated to the N-teminus
in a zipper-like manner (6, 7, 16,21, 108).
After seretion of procollagen fom fbroblasts, the N-propeptides ae
cleaved by a proollagen N-proteinase and the C-propeptdes by a separate
proollagen C-proteinase. The collagen then self-assembles into fbrils. Fi
nally, lysyl oxidase converts some lysine and hydroxylysine rsidues to alde
hyde derivatives that form a complex series of cross-links.
The assembly and proessing steps of many nonfbrillar collagens are the
sam, but there are notable exceptons. Many collagens contain N- and/or
C-teninal noncollagenous domains that are not removed and therefore not
called propeptides. Several collagens undergo N-glycosylaton. Three colla
gens (types IX, XI, and XIV) a modifed by addition of glycosaminoglycan
side chains, and two additional collagens (types XV and XVIII) have potential
attachment sites for such chains. The tiple helices of a few collagens that lack
large C-tenninal globular domains (e.g. type XI) may fold by ms of a
mechanism that does not involve formation of a nucleus in the C-terminus
(109).
Intracellular Processing
Reently, analyses of cDNAs provided the complete amino acid sequences for
the a subunit of prolyl 4-hydroxylase fom human (110), chick (111), nema
toe Caenorhabditis elegans (112), and for the subunit fom several organ
isms (113). In addition, complete amino acid seuences have ben reported
for human (114) and chick (115) lysyl hydroxylase. Prolyl 4-hydrxylase fom
vertebrates is an 2 tetamer and lysyl hydroxylase an dimer, but the
subunit stuctur of prolyl 3-hydroxylase is curntly unknown (116, 117). No
signifcant homology is found between te primary structures of lysyl hydroxy
lase and the two types of subunits of prolyl 4-hydroxylase in spite of the marked
similarities in the catalytic properties betwen these two enzymes (116, 117).
The two catalytc sites in the tetamer of prolyl 4hydroxylase are loated
in the a subunits (113, 116, 117). Even though the enzyme from all the
vertebrate sources studied is an 2 tetramer, cloning and expression of the
a subunit from C. elegans rveled that its prolyl 4-hydroxylase is an 01 dimer
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
414 PROKOP & KlVIRlKKO
(112). Also, an isoform of the a subunit of the vertebrate enzyme, defned as
a(U) subunit, was rcently discoverd (118). Expression studies have demon
stted tha the recombinat enzyme from verebrates can form both [a(I)
h
i
2
and [a(I)hi2 tetamrs (118, 119), but whether the rcombinant enzym can
also form tetmr that contain both a subunits is currently unknown. There
appear to b no major differences in the tssue distribution of the two a subunit
(118). Most of the catlytic properties of the [a(I)h and [a(II)hi2 ttamers
a highly similar, but a surprising difference is that poly(lrproline) is a very
poor inhibitor of the [a(I)hj2 enzyme (118), whereas it is a highly effetive
compttve inhibitor of the [a(I)h enzyme (117).
Nucleotde sequencing of the cDNA for the j subunit of prolyl 4hydroxy
lase indicated that this polypeptide is identical to the enzyme protein disulfde
isomerae (POI) (120). Moreover, the j subunit has POI activity even when
preent in the natve prolyl4-hydroxylase tetamr (121). POI catalyze thiol:
disulfde interhange in vitro, leading to net protein disulfde formation, re
duction, or isomrization depending on the reaction conditons. Researchers
regard it as the in vivo catalyst of disulfde bond formation in the biosyntesis
of a lage number of sereted and cell surface proteins, including collagens
(122, 123).
The POI actvity of the PDIIj subunit is not diretly involved in the hy
droxylation reaction of prolyl 4-hydroxylase. This fnding is based on rcent
data obtained by expression of a reombinant prolyl 4-hydroxylase ttamr
in insect cells (119, 12 4). The PDI1 polypeptide has two -Cys-Gly-His-Cys
sequence that rprsent two independently acting catalytc sites for the isom
erase actvity (12 4, 12 5). When both these sequences were moified to -Ser
Gly-His-Cys-, the polypeptide had no PDI activity but still associated with the
a subunits to form the i2 tetramer, and this tetamer proved to be flly active
prolyl 4-hydroxylase (12 4). Expression studies have further demonstated that
in the absence of the PDIj subunit, the a subunit forms highly insoluble
aggregates (118, 119, 126). Therefore, one function of the PDIIj subunits in
the prolyl 4-hydroxylase tetra mer is to keep the a subunits in a catalytically
active, nonaggregated conformation.
Recent reports indicate that the cellular PDI1 polypeptide may have several
additonal functons (117, 122, 123, 12 7-12 9). One is to serve as a major
cellular thyroid hormone-binding protein in th endoplasmic reticulum. A
second functon is to act as a chaperone-like polypeptide that nonspeifically
binds pptdes in the lumen of the rough endoplasmic reticulum. A third
fncton is to serve as the smler subunit of the microsomal tiglyceride
transfer protin. Further suggested functons a to serve as a dehydroascorbate
reductase (130) and to act as a developmentally regulated retnal protein termed
r-cognin (131). The PDI subunit thus appears to b an unusually versatile
polypeptde that has many biological fnctions.
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
COLGES 415
The two hydroxylysyl glycosyltansferases involved in the biosynthesis of
collagens have been extnsively charcterized (6, 7, 16), but te genes have
not yet been cloned.
Reent reports also suggest that chain assoiaton and folding of type I and
I collagens may involve a specifc moleular chaperne protin called Hsp47
or colligin (132-134). Hsp47 binds specifically to type I procollagen and to
types I and IV collagens in vito. Cellular levels of the protein parallel the
raes of synthesis of typ I or typ I collagen in many expeimental situaions.
Cross-linking studies in intact cells demonstated association of type I procol
lagen with Hsp7, and this assoiaton was incrased when cells were heat
shoked or tated with the iron chelator a,a'-dipyridyl tat effetvely inhibits
th hydoxylation of prline residues (132). Tratment of cells with antisense
oligonucleotides to Hsp47 decreased the rate of synthesis of tpe I proollagen
(134). However, Hsp47 does not bind to type m procollagen, and rsearchers
have not clearly established tat it has an essential role in proollagen biosyn
thesis.
Extracellular Events
Extrcellular collagen fibrils are formed by seretion of a soluble proollagen
that is then enzymatcally processed to an insoluble collagen. Te mechanisms
by which other collagens a incorporated into an insoluble extacellular matix
a mor obscure, since prsumably they must be soluble during intacellular
assembly. One possible mechanism is that such collagens ae secreted a
soluble proteins that bind to other macromoleules aftr serton to form
insoluble heteromolecules.
Both the N- and C-propeptdes of proollagens must be cleved by speifc
proteinases for the proteins to self-assemble into fbrils under physiological
conditions. Te N-propeptides of both types I and II procollagens ae cleaved
by te same specifc procollagen N-proteinase (6, 7, 16). The N-propptid of
type III prcollagen is probably cleaved by a different type III N-proteinase
(6, 7, 16). Whether other spcifc N-proteinases are required to cleave other
proollagens such as types V and XI is unclear. Contary to earlier repors,
type I N-protinase extractd from bovine tssues was recently shown to b
the same protein as the bettr-characterized enzyme fom chick embryos (135).
Also, rsearchers recently demonstated that if type I procollagen is aggregated
by addition of polyethylene glycol (136), the rate of cleavage by the C-prote
inase is incrased 10- to 15-fold. The rate of cleavage by te N-proteinase is
increased about fourfold. Because the turover numbrs with monomeric type
I procollagen are low, it may b that the enzymes in vivo may act on secreted
aggregates of procollagen (137).
The self-assembly of fibril-forming collagens has ben studied for many
years by warming and neutalizing solutions of the collagen extrcted fom
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
416 PROKOP & KIVIRIKKO
tissues with cold acidic buffers (137, 138). More reently, the pres was
studied in a systm in which pCcoIIagen, a soluble and partially proessed
preursor lackng te N-propptde, is cleaved to collagen by purifed prool
lagen C-proteinase in a physiological bufer and at physiological temperature
rage (137). Cleavage of pCcollagen to collagen reduces te solubility of the
protein by about l00-fold. The rsulting collagen reproducibly self-assembles
into tightly packed fbrils (137, 139).
One series of expriments was carrie out by isolating type I proollagen
and cleaving it with C-proteinase to genert type I pNcoIIagen (137). The
pNcollagen assembled into thin, shet-like stctres that were cross-stated
in longitudinal setons and of a uniform thickness. Mixtures of tpe I collagen
and pNcollagen copolymerized to form a variety of pleomorhic fbrils. The
results were consistent with the hypothesis that under some cirumstnes type
I pNcollagen has a biological role in altering the morphology of type I collagen
fbrils (137, 140).
Type IIpNcollagen also formd tue copolymer with type I collagen, and
th copolymrizaton generate fbrls that wer tinner than fbrils geneated
fom type I collagen alone (141). The results were consistnt with a model in
which type III pNcollagen can regulate the diametr of type I collagen fbrils
by coating their surface. However, the effets on fbril diamter ruired at
least a 1:1 ratio of type II pNcollagen to typ I collagen (141).
I related expriments, recombinant type II pCcollagen (142,143) was used
for fbril assembly by incubaton with C-proteinase (144). The kinetics for the
assembly of type II collagen fbrils difered markedly from those for the
assembly of type I collagen, and the critical concentation at 37C was about
50-fold geater. Also, the tpe I collagen fbrils were relatvely thin and
formed three-dimensional networks. The results indicated that th differences
in prmary stucture betwen type II and I collagens a sufcient to explain
many of the characteristc diferences in morhology betwen these two kinds
of fbrils sen in tissues (144).
The system for generating typ I collagen fibrils by enzymic cleavage of
tye I pCcollagen made it possible to follow the growth of fbrils from the
intermediate stages. The frst fbrils detetd had a blunt end and a pointed tip
(145). Initial growth of the fbril was exclusively fom the pointed tp. Later,
tips appeared on the blunt ends and the fibrils gew in both diretons. Bot
the initial tips and the later tips were nearly paaboloidal (145). Based on the
observations, a model of fbril growth was developd (146), the essential
fetures of which were a distinctive structural nucleus that formed at each end
of a growing fibril. The growth of the fbril then occurred by propagaton of
the two stuctural nuclei. The strctural nuclei had similar spiral helical con
formations, and assembly and propagaton of each structural nucleus reuired
just two kinds of specifc binding steps (146). Simila studies on fbrils of
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
COLGES 417
reombinant type I collagen demonstated that the tips were again nearly
parabloidal (147). However, the monomers in te tps of the two type were
oriented differently. In tips of type I collagen fbrils, all the monomrs were
oriented so that the N-termili pointed toward the end of the tp (145). In fbrils
of type II collagen, all the monomers wer oriented so that the C-termini
pointd towad the growing tp (147).
Reently, Fourier tansformed infred spectroscopy (FIR) was used to
study te lag period of fibrils assembled by neutalizing and warming solutons
of collagen extacted with cold acidic bufer (138). The results ae consistnt
with the conclusion that as the tempemture is raised, the tiple helix tghtens
or stiffens but then relaxes again as fibrils are formed (138. 148).
The lysyl oxidase that forms cross-links in collagen fbrils is a highly
insoluble copper-containing protein (149). Complete cDNA-derived amino
acid sequences have now ben reportd for the enzyme fom several sources
(150-153). The enzyme was identcal to a tumor suppressor protein kown as
rrg (154. 155). Erlier work had demonstate ta lysyl oxidase actvity is
markedly low in te culture medium of many malignantly trnsformd human
cell lines (156), and recently, these cells were also found to have very low
levels of lysyl oxidase mRNA (157).
Potentials for Inhibiting Fibrosis
Normal wound healing involves the formation of scars and fibrous tssue that
lagely consist of collagen fbrils. Although moderate degees of fbrous tssue
ar bnefcial in wound repair, fbrous material ofen accumulates in excessive
amounts and impairs the normal functon of the afected tissue. Such excessive
accumulation of collagen becomes an important event in scaring of the skin
following bus or traumatc injury and in fbrosis of the liver, lungs, and
kidneys following injury to these orgas. Therfore, there ha been consider
able intrst in agents that can inhibit or modulate collagen synthesis in fibrotic
diseases. Potential target sites for inhibiting collagen synthesis include tan
scripton of the genes, translation of the mRNAs, and som of the unique
post-tanslational enzymes involved in the biosynthesis of the protein.
Reent studies have demonstated that synthesis of type I collagen can be
specifically inhibited in cell culture by the use of antisense oligonucleotde
(158-16). However, the degre of inhibiton obtained is highly variable and
rarely excees 50% (158, 1 59). I related experiments. an antsense gene to
human type I collagen in which only the 3'-half was inverted was shown to
be highly effetve in inhibiting collagen synthesis in tansgenic mice express
ing an interally deleted human COLlAI minigene (161). The results raised
the possibility that chimeric gene constructs that contain intrn seuences and
in which only part of the gene is inverted may b parcularly effective as
antisense genes that can inhibit collagen synthesis in fbrotic conditons. How-
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
418 PROKOP & KlVIRIKKO
ever, both the antisense-oligonucleotide statgy and the antsense-gen stt
egy appar to prsent considerble problems in the delivery of the agents in
ways that will b effective in inhibiting fibrosis in vivo.
Severl of the post-taslational enzyms appar to be attactve trgets
for specifc inhibiton beause they ae unique to collagen biosyntesis. Tese
inlude prlyl 4-hydroxylase, procollagen C-protinase, and pehaps also
lysyl hydroxylase and lysyl oxidase. Numrous compounds are now known
that inhibit prolyl 4-hydroxylase competitively with respct to som of its
cosubstates or the peptde substate (116, 117). For example, pyrdine
2,4-dicarboxylate inhibited prolyl 4-hydroxylase with a Kj of 2 J1M. Te
problem of cell membrane permability was in part overcome by the design
of lipphilic proinhibitors that were convertd to the actve inhibitors in
tcellularly (21, 117, 162). One such derivatve inhibits hepatic collagen
accumulation in two models of liver fbrosis in rats (21, 162). Te rcent
succes in expression of an actve recombinant human prlyl 4-hydroxylase
in insect cells (119) should make it possible t defne the critcal stuctural
features of the enzym by site-directed mutagenesis (119, 163) and to produce
adequat amounts of the enzyme for crystallizaton so that more efetve
inhibitors can b designed.
Poollagen C-prteinase is another attactive target for inhibiton of fbro
sis. Most of the available evidence suggests that procollagen cannot partcipate
in fbrl assembly unless the C-propeptde is spcifcally cleaved fm the
preursor (6, 7, 16, 137). The only challenge to this prposal comes fom
experimnts in which rcombinant procollagen was synthesizd wit a muta
tion at the cleavage site so that the prtein was not cleaved by C-protinase.
Som of the protein synthesized in cell cultur was cleaved by nonspeific
proteases (164; J Bateman, personal communication). However, such nonspe
cifc cleavage is unlikely to generate collagen that is assembled into normal
fbrils. Intal studies suggested tat basic amino acids and peptdes may
speifcally inhibit C-poteinase (see 108), but the development of more ef
fetve agents such as pptdomimtcs is still in the early stages.
Severl attempts have ben made to develop inhibitors for Iysyl hydroxylase.
Minoxidil and many of its derivatives have the surprising effect of reucing
both lysyl hydroxylae actvity (165) and the mRNA (166, 167) in cultured
cells. Their mechanism of acton is unknown. Also, wheter inhibition of lysine
hydroxylation will in itself be effective in inhibiting fibrosis is unclear.
One of the frt targets explored for inhibiton of fbrosis was lysyl oxidase.
Aminopropionitile has long ben known to be a suicide inhibitor of the
enzyme (149). Recently, severl deri vati ves were developed that are even more
efective (149). However, whether an inhibition of lysyl oxidase will pevent
fibrosis is unclear, beause the cross-linking of collagens ocurs long afer
fibril assembly.
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
COLGES 419
MUTATIONS IN MEN AND MICE
Mutations in Patients
TPE I COLLAGEN Almost 20 difernt mutatons have now ben charac
terize in the eOLlAI and eOLlA2 genes that code for proal (I and pra2(1
chains of tp I proollagen (2, 17-22,25; for dtails, se 18). Most of these
mutations have ben identifed in patients with osteogeneis imperfet (01),
but they a also found in patients with related disorder (Table 2).
OJ is chaacterized by brittle bones but also involves other tissues rich in
type I col lagen so as to produce blue sclerae, abnormal teeth, thin skin, weak
tendons, and hearing loss. In th most sever forms, bones and other tssues
a so fagile that death occurs in utero or shortly after birth. In mor moderate
fors, the disease is not lethal, but the patients have repeated fracture afer.
minor tauma that may led to prent deformities of limbs and other bony
stctures.
Table 2 Diseases caused by mutations in collagen genes or deficiencies in
the activities of post-translational enzymes of collagen synthesis
Gene or enzyme
COLlAl; COLlA2
COL2 Al
COL3Al
C0L4 A3; COL4A4
COL4 AS
COL4AS and COL4 A6
COL7 Al
COL9Al
COL9A2
COLlOAI
eOLllA2
Lysyl hydroxylase
Type IN-proteinase
Lysyl oxidase
In a subset of patients.
Disease
Osteogenesis imperfecta
Osteoporosis'
Ehlers-Danlos syndrome type VIlA, VIIB
Several chondrodysplasias
Osteoarthritis'
Ehlers-Danlos syndrome type IV
Aortic aneurysms'
Alport syndrome, autosomally inherited forms
Alport syndrome, X-linked form
Alport syndrome with diffuse esophageal
leiomyomatosis, X-linked
Epidermolysis bullosa, dystrophic forms
Osteoarthritisb
Multiple epiphyseal dysplasiac
Schmid metaphyseal chondrodysplasia
Stickler syndrome, nonocular forme
Ehlers-Danlos syndrome type VI
Ehlers-Danlos syndrome type VIle
Occipital hom syndromed
Menkes syndromed
Demonstrated only in transgenic mice.
C CDOn5If3lCd OnV DV yCnCllC Ink3yC.
Secondary to an abnormality in copper metabolism.
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
420 PROKOP & KIVIRIKKO
Essentially all patients with OI have mutatons in type I collagen. Most of
th mutations are single base substtutions that convert a codon for an obligate
glycine in the repating -Gly-X-Y- sequence of the tiple helix to a codon for
an amino acid with a bulkier side chain (2, 17-22). Other mutations include
deletions, insertions, RNA splicing defects, and null alleles. The null alleles
primarily cause the mild typ I variant of OI (168). The mutations inactivatng
th alleles have ben diffcult to defne, but four paients wee recenty found
to have a pematur tanslation terminaton codon that decreased the cytoplas
mic level of the mRNA ( Korkko, P Paassilta, J Zhuang, H Kuivaniemi, L
Ala-Kokko, et al, in preparation).
Mutatons that cause synthesis of stucturally altere pro chains of type I
proollagen generlly cause mor severe phenotypes than null alleles (2, 17-
22). One of two molecular mechanisms are usually involved. Som substtu
tions for obligate glycines interupt the zippr-like folding of the tiple helix
and generate unfolded procollagen that firt accumulats in fbroblasts and is
thn degraded. The efects of the mutatons are amplifed bease both the
normal and mutated chains present in the same moleule are degraded in a
proess referd to as proollagen suicide or a dominant negative effect (1 7,
18, 108). The effets of other glycine substitutions are explained by their
consequences on the nucleated gowth of collagen fbrils. One such mutaton
was a heterozygous substtution of cysteine for glycine at position 748 of the
01(1) chain that intoduced a fexible kink into the triple helix (2, 18, 169).
Studies on fibril formation in vito (se biosynthesis) demonstated that mole
cules with the cysteine kink copolymerized into fbrils with the noral mole
cules, but the prsence of the kinked moleules delayed fbril formation,
reduced the total amount of collagen incorporated into the fbrils, and drasti
cally altered the morphology of the fibrils (18, 1 38, 1 70) (Figur 3). Other
glycine substitutons tat do not affet folding have similar effets on fbril
assembly (1 71-173).
I 01, bones are fragile in part bcause of a marked rducton in bone mass
(ostopenia) . Mutations in type I collagen have also ben found in a few
patients who have little evidence of OI but who have osteopenia and factures
characterstc of osteoporosis (18, 20, 21, 25, 174, 175). A rcent survey
suggested that 1-3% of patients with osteoporosis have a mutaton in either
the eOLl AI or eOLl A2 gene (175).
Some mutations in the type I collagen genes produce a disese known as
the type VII variant of EDS (Table 2), a syndrome characterized by joint
hyprmohility and skn abnormalities (18, 21 , 23, 25). The disese is caused
by a failure to cleave the N-propeptide from type I procollagen. The persistence
of the N-propeptide on the molecule drastcally alters fbril formation so that
the fbrils beome thin and highly irregular in cross-section. The mutatons
causing EDS VI are either RNA splicing mutations that eliminat the amino
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
Procol l agen
Sui ci de

7
r
7
7
7
) .
COLGES
421
B
A b n o r m a l
F i b r I I s

Figur 3 Schematic of how mutations such as glycin substitution alter the biosynthsis of
collagen. (A) Illustation of a glycine substitution that prevents the zippr-like folding of the tiple
helix and leads to degraation of bth normal an abnomal po chains by a proollagen suicid
or dominnt negative mechanism. (8) Illustration of a glycine substitution that do nt interfee
with th folding of the triple helix but producs a conformational change such Wa kinkin th proein.
(Moifed and rrouced with prmission from Reference 10.)
acids of the cleavage site for the N-propeptide (subtypes VIlA and VI) or
mutations that derease the activity of the cleaving enzyme, procollagen N
proteinase (subtype VIle).
TPE II COLLAGEN Over 50 different mutations in the COL2A1 gene have
ben shown to cause a heterogeneous goup of disorders of calage that are
known as chondrodysplasias and are characterize by shor-limbed dwasm
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
422 PROKOP & KIVIRIKKO
and skeletl deformites (18, 21, 24, 176182). The mutations include amino
acid substttons, deletons, insertons, RNA splicing defects, and stop codons
for premature termination of tanslaton. All fve stop codons sofar charc
terize have ben identfed in the single phenotpe of the Stckler syndrome
that involves vitous degeneration and retnal detachment in additon to de
generton of joint cartlage (21, 183-187).
Mutatons in typ II collagen (able 2) are also found in about 2% of patients
with early-onset familial osteoarhrits (177). One mutton is a substtuton of
cystine for arginine at amino acid positon 519 of the 01(11) chain, that has
now ben found in four apparently unrelated families with severe, early-onset
osteoartritis and mild chondrodysplasia (179, 188, 189). Two additional mu
tations in patents with a similar disease phenotype are a serine-for-glycine
substtuton at positon 01-274 (19) and at 01-976 (179). In additon, muta
tions in other genes for collagens such as types IX and X (Table 1) my b
found as prdisposing factors for osteathrits in som familie (se Mutatons
in Transgenic Mice, blow).
TPE III COLLAGEN About 50 diferent mutations in the COL3A 1 gene have
ben found in patients with EDS I (18, 21, 23, 191, 192), the most severe
form of EDS that can cause sudden death fom rupture of large arteies and
oter hollow organs in additon to skin and joint changes (23, 25). The muta
tions include glycine substtutions, deletons, RNA splicing defets and null
alleles (18, 21, 191, 192). Mutations in the COL3Al gene have also been found
in a subset of patients who have aterial aneurysms but who exhibit little (193)
or no evidenc (18, 21, 194-196) of other connective tssue manifestations.
Muttions in the COL3A 1 gene may also b a prdisposing facto. for intc
ranial aneurysms (18, 21, 197), but they appear to b a rar cause of this disese
(198).
TPE IV COLLAGEN No mutations have ben identifed in the genes eOUAl
and eOUA2 that encode the two major a chains of type I collagen, but
mutations have ben found in the genes coding for the minor type IV collagen
polypeptdes (able 2). The Alpor syndrome is a progrssive heritable kidney
disease characterized by hematuria caused by stuctural changes in the
glomerular basement membrane. This disease is also associatd with hearing
loss and oular lesions. The gene coding for th a(IV) chain was mapped to
th lous for te X-linked form of the Alport syndrom (50, 51), and sub
sequently, mor than 50 difernt mutations in the eOUA5 gene were found
in families with this disorder (21, 4, 199, 2(). The mutations include amino
acid substitutions, large deletions, and gee rearangements such as inversions,
insertons, and duplications. Although the Alport syndrom is primarily X
linked, autosomally inherited forms also exist, and reently, heterozygous
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
COLGES 423
mutations in both the COIA3 and COIA4 genes were chaactrized in
autosomally inheited forms of the disease (44, 201).
Deletions involving the hed-to-hed 5'-ends of both the COIA5 and
COIA6 genes have been found in several patents (54, 202) who have the
Alprt syndrom together with diffuse esophageal leiomyomatosis, a re
syndrom characterizd by proliferaton of smooth muscle cells in the esopha
gus, tahebronchial tre, and the female genital tact (4).
TYPE VII COLLAGEN About 20 mutations in the COL7Al gene have ben
found in patients with the dystophic form of epiderolysis bullosa. a disease
charactrizd by severe blistring and scarng of the skin fom minor tauma
(78-81 , 203-206). As a conseuenc of these mutations, the anchoring fbrils
that link the basement membrane to the anchoring plaues in te skin a eiter
reducd in amount or completely absent (203). Mutatons in the COL7 Al gene
wer found in both the dominantly and recessively inherited forms of the
disease (78-81). Te mutations include amino acid substitutons, an inserton
deletion, and prematur tanslation termination codons (81 , 203-20).
TYPE X COLLAGEN More than 10 different mutations have now been charac
terize in typ X collagen (207-213) in patents with Schmidt mtaphyseal
chondrodysplasia, a disease that is characterized by shorening of limbs and
bowing of legs aggravated by walking. The muttons include amino acid
substtutons, deletions, and prmature tanslation termination codons. All the
mutaions so far characterized altr the stucture ofC-terminal noncollagenous
domain of the polypptde, an observation suggesting that the mutant chains
a unable to assoiate to form tiple-helica molecules.
OTHER COLLAGENS Genetc linkage was found betwen the COL9A2 gene
lous (214) and multple epiphysel dysplasia (EDM2). Also, genetic linkage
was found betwen the COLl IA2 gene lous and a non-ocular form of the
Stickler syndrome (215).
Since correct expression of collagen genes appars to be essental for the
stctural intgrity of many tssues, mutations in more than 30 diferent col
lagen genes (Table 1 ) can probably produce disese phenotypes. Therefor,
research in this area is still at a very early stage (20). Also, similar disease
phenotypes are prbably produced by mutatons in genes for other matix
proteins, since severa diseases of catilage have ben linked to loi that do
not contain ay kown collagen genes (18, 26).
PST-TRANSLATIONAL ENZYMES A defciency of lysyl hydroxylase is found
in most but not all families with EDS VI (21 , 23, 108), a disease characterized
by hyperextensible skin and joints, scoliosis, and ocular fagility. One mutaton
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
424 PROKOP & KIVIRIKKO
in the gene for the enzyme was a homozygous duplicaton of seven exons that
appeas to b caused by a recombinaton of Alu sequences (216, 217). The
sam mutaton was found in several apparently unrelatd families (216-218).
Additional mutatons include a homozygous tanslation termination codon
(219) and several amino acid substtutons (220).
EDS vie is caused by a deficiency in type I proollagen N-protinase (221 ,
222), but no mutaions i n the genes have been chaacterized s o far.
Defciencies of Iysyl oxidae, a copper-containing protin, a sen in two
rae and severe X-linked recessive diseases, the ocipital hor syndrome and
Menkes syndrom (21, 108. 223). The diseases are caused by defets i coppe
mtbolism and lead to seondary defets in the cross-linking of collagen, but
the mehanisms producing a defcieny of Iysyl oxidase are uncler. Skin
fbroblasts fom patents with these diseases contan ad secrete rduced
amounts of the lysyl oxidase protin (21 . 224), and rcently, two (157, 224)
out of three (157. 224, 225) studies reported that these cells also contin
reduced amounts of Iysyl oxidase mNA. These observations suggest that the
abnormality in copper metabolism somehow infuences the synthesis or sta
bility of the mRNA for Iysyl oxidase.
Mutations in Transgenic Mice
Transgenic mice are particularly useful for studying matix proteins bcause
most of the proteins are large, insoluble. and difcult to test for fnction. They
a also useful for studying the consequences of disease-causing mutations in
matix genes, since the mutations affect many tissues that cannot be exained
flly in patients. Experiments have ben carried out in which mutated collagen
genes were randomly inserted in tansgenic mice to produce dominant negatve
efects, and more recently. a few expriments were carried out involving
knok-out of collagen genes. In some instances. the results fom these two
types of experiments have been complementary. In others they are apparntly
contradictory.
The frst expriments with transgenic mice used a retovirus infection of
mouse embryos. By chance. the retrovirus was inserted in the frst intron of
the COLlAI gene and prevented exprssion of the gene in most tissues
(see226). Homozygous mice died in utero with liver nerosis and bleeding.
Heterzygous mice survived and had derased collagen content, decresed
mechanical stength of bones. and hearing loss (226). No facturs wer de
tected. Subsequently. transgenic mice were prpared that expressed a mutated
COLl AI gene in which a cysteine codon was substituted for an obligate
glycine (227). Four of seven founder mice had some of the phenotpic features
of 01 such as por mineral ization in most bony structures. Again. however.
no factures were demonstated and no breeding lines of the tansgenic mice
could be developed. I contrast. extnsive factures were observed in several
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
COLGES 425
lines of tansgenic mice expressing a mini-gene version of the human COLIAI
gene in which exons 6 to 46 were deleted in-fam to cause synthesis of
shorened proal () chains (228). Mice expressing high levels of the mini-gene
protein developd a lethal phenotype of extensive fractures because the short
ened proa chains produced procollagen suicide or a dominant negative effect
through protein depletion (228). Mice expressing lower levels developd a
milder phenotype resembling human osteoporsis (229, 230). Extensive bred
ing of the tansgenic mice in the inbre line demonstatd marked phenotypic
variability and incomplete penetance that apparntly is an inhernt property
of expression of mutated collagen genes (230). In related expriments, a
naturally ocurring recessive mutation that produced bowed and britte bones
in mice was shown to be a single base deleton in the C-prpeptide that caused
synthesis of proa2 chains that could not associate with proal chains (231 ).
As a result, the only type I procollagen synthsized was homotimrs of
proal (I) chains.
In experiments with the COL2AI gene, tansgenic mice expressing an
interally deletd version of the human gene develope a phenotyp similar
to some human chondrodysplasias with dwarfsm, a short snout, a cranial
bulge, a clef palate, and delayed mineralizton of bone (232). At the same
time, similar phenotypes of severe chondrodysplasias were seen in tansgenic
mice expressing mouse COL2Al genes mutated either by substitution of a
cysteine codon for glycine at amino acid position 85 in the al (lI) chain (233)
or by a deleton that eliminated amino acids 4-1 8 of the al (II) chain (234).
In older mice fom some of the same lines, the evidence of chondrodysplasias
was less marked, and the most stiking featurs were degeneratve changes of
aricular carilage similar to osteoarthrits (235, 236). A surprising fnding was
that over-exprssion of a normal mouse COL2Al gene in transgenic mice
produced abnormally thick collagen fbrils in catilage, apparently because of
an imbalance in the amounts of different collagens being synthesized in the
tissue (237). Transgenic mice over-expressing the normal gene developed a
chondrodysplasia
Transgenic mice were prepared that exprssed a COL3A 1 gene in which a
mtionine codon was substtted for lysine at amino acid 939, the cross-link
ing site in the triple-helical region of the protein (238, 239). The phenotype
was mild, but pregnant females apparently developed uterine dysfncton
(238). The same mice had morphologic changes in healing dermal wounds,
but no drastic consequences occurrd on wound repair (239).
Transgenic mice expressing a mutated COL9Al gene (240) and tansgenic
mice in which the gene was inactivated (241) developed ealy onset osteoari
tis. The results suggested, therefore, that a subset of patients with osteoarthritis
may have mutations in type IX collagen.
In tansgenic mouse expriments with the COLIOAI gene, the results with
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
426 PROKOP & KIVIRIKKO
dominant negatve experiments and gene inactivation expriments wer dif
feent. Transgenic mice expressing constucts of the chicken gene with an
in-fae deleton in the centa triple-helical domain of the collagen (242)
developed skeletal abnormalities and a defciency of leukoyts similar t the
Schmidt type of human metphysel chondrodysplasia. The obsevations in
tnsgenic mice (242) largely provided th basis for defning mutations in the
sam gene that cause Schmidt phenotypes in patents (see above). However,
mice with homozygous inatvaton of the COLlOAI gene did not develop
any apparent phenotype (243). The inital explanaton of these conficting dt
was that mutations producing a dominant negatve effect were mor sevee in
terms of phenotype than mutations that inactivate the gene. Subseuently,
reearchers found that mice haroring a 68ano acid deleton fom the
tple-helical domain of the type X collagen gene did not develop any grss
abnormalities of the skeleton (244). At the moment, ther is no simple way to
reolve these confictng observations.
Transgenic mice expressing an intrally deletd COLl IA2 gene developed
a mild phenotype with a shor snout, prominnt forehead, shorened limbs, and
shorened tl (245). Also, DNA linkage to the lous for COLl I AI was found
in cho mice that have a naturally occurring autosomal reessive chondrodys
plasia (246). The ol(XI) chain was absent frm tissues, and the collagen fibrils
in cartlage fom the mice were unusually thick.
Potentials for Gene Therapy
Researchers have conducted prelimina experiments to test the potentals of
gene therapy for either contolling collagen deposition in fibrotic conditons
or rescuing the phenotypes produced by mutatd genes. As indicated above,
oligonucleotides partially inhibited exprssion of collagen genes in cell culture
experimnts ( 158-1 60). A related approach showe that a atisense gene to
th huma COLl AI gene is effective in tansgenic mice ( 1 61 ). The experi
mnts were carried out with mice that developed a lethal phenotype of fgile
bones beause they expressed an interlly deleted mini-gene for the prooI(1)
chain of human type I procollagen. The mice wer bred to transgenic mice
that exprssed an antisense gene that was similar to the interally deleted
human COLl AI mini-gene, but the 3'-half of the gen was inverte so as to
code for hybrid sense and antsense RNA. In mice that inherited both genes,
the incidence of lethal fagile bne phenotyp was reduced fom 92 to 27%
(1 61), and there was a corrsponding derease in the tissue levels of the human
mini-prool (I) chains. The experiment may have succeeded wher simila
experimnts with antsense cDNAs failed in the past, beause the tascript
fom the antsense gene contained invertd inton sequences and, threfor,
interacted with the sense transcript in the nucleus.
In rlated experiments, noral mice were subjeted to lethal body irrdiaton
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
COLGES 427
and then reeived stomal cells of bone marrow fm a tansgenic mouse
containing a DNA marker consistng of a human COLlAl mini-gene. Afer
five months, the donor cells were found to account for 3-5% of the cells
in bne marrow, spleen, bone, cartlage, and lung (247). Also, the huma
COLlAl mini-gene was exprssed in bone. Beause stromal cells fom bone
marrow are readily expande in culture, they may be a useful sourc of
long-lastng prcursor cells for gene therapy of bone and cartlage diseases.
Techniques were also developed for site-directd insertion of an exogenous
collagen gene into an endogenous gene locus (28). In stable tansfeton
experments with the human tumor cell line known as HT -1080, a constct
containing a short 5'-fragment from the COLlAI gene linked to 30 kb of the
COL2Al gene was found to be inserted at a high frquency into both allele
of the endogenous COLlAI gene (248). Transfected cells with the tageted
inserton expresed rlatvely high levels of the exogenous gen. The reults
suggested that it may be possible to taget inseron of exogenous gene into
preeteined loci of collagen genes. The targeted insertion has the advatge
that it provides controlled expression of te exogenous genes and avoids the
potential dangers of random insertions.
SUMMARY
Collagens a defned as proteins that: (a) contin several repeats of the amino
acid sequence -Gly-X-Y- in which the X-positon is feuently proline and the
V-positon is fquently 4-hydroxyproline and (b) have the ptental for three
chains with such repeated seuences to fold into a characteristc tiple helix.
At least 19 proteins and more than 30 gene products ar now formally defned
as collagens. An additonal 1 0 proteins have collagen-like domains. Therefor,
the number of proteins classifed within this super-family is rapidly expanding.
The most abundant collagens for extacellular fibrils, but others form net
work-like strctures. Still others bind to the surfaces of collagen fbrils, form
beaded-flaments, sere as anchoring fbrils for the skin, and a tansmem
brane proteins. The exon-inton patters of most of the collagen genes are
complex. Most of the genes are widely distributed in the genome, but three
pairs of genes are located in a unique head-to-head arangement with overlap
ping promoters.
Recently, severl of the eight highly specifc postranslatonal enzyme
involved in collagen biosynthesis wer cloned. The 1 subunit of prolyl 4-hy
droxylase is identical to th enzyme protein disulfde isomerase and appars
to have severl other distinct fnctons. Experiments with recombinant pro
collagens that ae the soluble precursors of fbril-forming collagens have
confmned previous indications that collagens self-assemble into fibrils by the
classical mechanism of nucleation ad propagaton. The results also indicated
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
428 PROKOP & KIRIKKO
that the diffeences in primary stctur betwen the two most abundant fbril
forming collagens (types I and II) are suffcient to explain many of the char
acteristic differences in morphology betwen these two kinds of fibrils.
There has been increasing interst in the possibility that the unique post
tanslatonal enzymes involved in collagen biosynthesis offe attractive targets
for spcifcally inhibitng the excesive deposition of collagen fbrils that
ocur in scar and in the fbrotc response of injury seen in most tssues.
Inhibition of prolyl 4-hydroxylase and procollagen C-proteinase appar to b
paticulaly attactive strategies.
Te imporant roles of collagens in biology have ben illustated by the
reent demonstations that over 40 mutations in 6 different collagen types
cause a varety of human diseases. Muttons in the tpe I collagen gens can
produce defets of bones and related tissues that range fom lethal ostegenesis
impefet to ostoporosis. Muttons in the gene fo typ n collagen ca
produce cartlage disorders ranging fom lethal chondrodysplasias to ealy-on
set ostearthrits. Mutations in the gene for the type X collagen, which is
expressed in hyperophic chondocytes, also prouce chondrodysplasias, and
genes for two additional collagens of catilage have ben linked to similar
phenotpes. Mutations in type III collagen produce defects of blood vessels
and other tissues that range from a severe form of the Ehlers-Danlos syndrome
to familial aortc aneurysms. Mutations in several polypptde chains of the
type IV collagens of basement membranes produc the renal disease known
as the Alport syndrome that may be associated with diffuse esophageal
leiomyomatosis. Mutations in the typ VII collagen tat forms anchoring fbrils
for basement membranes cause severe blistering and scarring of the skin.
Muttions in oter collagens are also likly to produce human diseases, but
some diseases with similar phenotypes are clearly linked to noncollagen genes.
Experiments with tansgenic mice have ben paculaly useful for defning
the structure ad function of collagens because the proteins are lage, insoluble,
and difcult to test for function. In addition, tansgenic mice have ben usefl
in demonstating the consequences of disease-causing mutatons in collagen
genes, since the mutations affect many tssues that cannot b examined fully
in patient. Experiments with mutated genes for type I collagen reproduced
the phenotyps of ostogenesis imperfecta and osteoporosis. Experiments with
mutated genes for typ II collagen reproduced the phenotpes of severe chon
drodysplasias and ostearhrits. Similar experiments with the type IIcollagen
gene produced only a mild phenotype. Experiments with the typ I gene
produced early onset osteoarthritis. Expriment with a dominant negatve
version of the type X gene produced a phenotyp of a speifc chondrodys
plasia, but a knock-out experiment with the same gene produced no apparnt
phenotype.
In exploring potentials for gene therapy, oligonucleotides were shown to
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
COLGES 429
partially inhibit expression of collagen genes in cell culture. In stable cell
tnsfection experiments, a hybrid gene containing a 5' fgment of the type I
collagen gene was found to be insertd into both alleles of the normal locus
of the tpe I gene at a high fequency. The results suggested it may b possible
to target insertion of exogenous genes into predetermined loi of collagen
genes. A phenotype of severe osteogenesis impetect in tansgenic mice
expressing a mutatd human type I collagen gene was patially rscued by
breding these mice to other tansgenic mice expressing an antsense version
of the same huma gene.
Any Annu Rekw chapter, W well a ay arUcie cit I an Annu &kw chapte,
may b purchae from the Annual Reviews Pre prints and Reprint srvice.
1-0347-07; 41529517; email: [email protected]
Literature Cited
I . Aeishmajer R, Olsen BR, Kihn K.
199. Structure, Molecular Biology, and
Patholgy of Collagen. Ann N Acad.
Sci. 580: 1-592
2. Engel J, Prokop OJ. 1 91 . Annu. Rev.
Biophys. Biophys. Cher. 20: 137-52
3. Shaw LM, Olsen BR. 1 91 . Trends Bio
cher. Sci. 16: 191-94
4. van dr Rest M, Garrone R. 1991.
FASEB 1. 5:281 4-2.1
5. Burgesn R, Nimni ME. 1 92 Clin.
Orthop. 282:25072
6. Holms DJS. 192. Essays Biochem.
27:49-67
7. Kielty CM, Hopknson I, Grant ME.
1993. In Connective Tissue and Its Heri
table Disordrs: Molecular, Gelletic
and Medical Aspcts, e. PM Royce, B
Steinmann, pp. 1 03-7. New Yok:
Wiley-Liss
8. Mayn R, Brewton RG. 1993. Curro
Opin Cell Bioi. 5:883-90
9. van dr Ret M, Garon R, Herbge
D. 1993. Adv. Mol. Cell Bioi 6: 1-7
1 0. Brewton RG, Mayne R. 1 94. In Ex
tracellular Matri Assembly and Struc
ture, es. PD Yurchenc, DE Birk, RP
Mecham, pp. 1 29-70. San Diego: Ac
demic
I I. Pihlajaniemi T, Rehn M. 1995. Prog.
Nuli Acid Res. Mol. Bioi. 5:225-2
1 2 Sandll LJ, Boyd CD. 199. In Extrael
lular Matrx Gens, ed. LJ Sanell, CD
Boyd, pp. I-5. New Yok: Academic
1 3. Yuorio E, de Crombrugghe B. 199.
Annu. Rev. Biochem. 59:837-72
14. Chu M-L, Prokop DJ. 1 93. See Ref.
7, pp. 149-5
1 5. Olsen BR. 1 91 . In Cell Biology of
Etraellular Matrx, e. ED Hay, p.
177-20. New York: Plenum. 2n e
16. Kivirikk KI. 1995. In Principles of
Medical Biology, e. EE Bilta, N Billar.
Greenwich, C: JAI. In press
1 7. Prokop OJ. 1 90. 1. Bioi. Chem. 265:
1539-52
1 8. Kuivaniemi H, Tromp G, Prokop DJ.
1991. FASEB 1. 5:252-0
19. Byers PH, Steinr RD. 1992. Annu Rev.
Med. 43:269-82
20. Prokop DJ. 1992. N. Engl. 1. Med.
326:50
21 . Kivirikk KI. 193. Ann. Med. 25: 1 13-
26
22. Byers PH. 1 93. Se Ref. 7, pp. 3 17-50
23. Steinmann B, Royce PM, Supei-Furga
A. 1993. See Ref. 7, pp. 351-0
24. Hoton WA, Hecht J. 1993. See Ref.
7, pp. 61 -75
25. Prokop DJ, Kuivaniemi H, Tromp G.
199. In Ha"ison's Prinipls ofIner
nal Medicine, ed KJ Isselbachr, E
BraunwaId, JD Willsn, 18 Martin, AS
Fauci, DL Kapr, pp. 2105-17. New
Yok: McGraw-Hill. 1 3t ed.
26. Jaenko 0, Olsen BR, Warma ML.
199. Am. 1. Hum. Genet. 54: 16
3-
27. Ryan Me, Sandell L. 1 90. 1.
Biol. Chem. 265: 1 033-39
28. Sandll U, Morris N, Robins JR, Go1d
bing MB. 1 91 . 1. Cell Bioi. 1 14: 1 307-
19
29. Nah HD, Upot W. 1991 . 1. Bioi.
Chem. 266:234652
30. Morris N, Oxfod IT, Doge K. 199.
Matrix Bioi. 14:361 (Abt.)
31 . Zhidkova NI, Justice SK, Mayn R.
199. Matrix Bioi. 1 4:365 (Abt.)
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
430 PROKOP & KIVIRIKKO
32 Tsumak N, Kimura T. 199. Matrix 57. Bernl D, Quinons S, Saus J. 1993. J.
Bioi. 14.36 (Abstr.) Bioi. Cher. 268: 1 200-94
33. Thoogo P, Bee J, von der Mark K. 58. Feng L, Xia Y, Wilson CB. 199. J.
198. Devl. BioL 1 16:497-59 Bioi. Cher. 269:2342-8
34. Hayashi M, Ninmiya Y, Hayashi K, 59. Mariyama M, Zhng K, Yag-Feng T,
Linsenmayer T, Olsn BR. 1988. De- Reeers ST. 1992. Genoics 13:80-1 3
veloprent 103:27-3 60. Zhou J, Leinonen A, Tryggvaon K.
35. Fitch 1, Mener A, Mayn R, Linen- 199. J. Bioi. Chem 269:6-14
mayer T. 1 989. Developent 105:85- 61. Muragak Y, Jacenko 0, Apte S, Mattei
89 MG, Ninomiya Y, Olsen BR. 1991 . 1.
36. Koha RA. Solursh M. 1989. Devl. Bii. Cher. 266:7721-27
Bii. 131 :558- 62. Schmid T. Cole AA. Chn Q. Boen
37. Berrd M. Yoshoka H, Rodiguez E. OK, Luchn L. LinsnmayaTF. 199.
van dr Rest M. Kimura T. Ninomiya See Ref. 10. pp. 171-20
Y. et al. 1988. 1. Bioi. Cher. 263: 63. Nishimura I, Muragaki y, Olsen BR.
1715< 198. J. Bioi. Chem 26:20331
38. Niybzi C. Eyre DR. 1989. FEBS LeIl. 6. Hayashi M. Hayashi K, Iyama K-I. Trel-
242: 314-18 stad RL. Linerunayer TF. Mayne R.
39. Kleman J-p. Hamann DJ, Ramirez F. 1992. Dev. Dyn. 194: 169-76
van der Rest M. 192. Eur. J. Biohem. 65. Swidersk R. Solursh M. 192. Dev.
21 0:32935 Dyn 194: 1 1 827
40. Mayn R, Brewton RG. Mayne PM, 6. Watt L, Lundstrum GP, McDonugh
Baker JA. 1 93. 1. BioI. Cher. 26: A, Kene DR, Burgeson R. Moris
9381-46 NP. 192. J. Bioi. Cher. 267:2093-9
41 . Bennett YD. Adams SL. 199. J. BioI. 67. Gereke DR. Foley JW. Catagnola p.
Cher. 265:2223-30 Gena M, Dublet B. Caneda R, et
42. Nah H-D, Niu Z, Adams SL. 199. 1. a. 1993. 1. Bioi. Cher. 268: 1 2177-8
Bii. Chr. 269: 1643-48 68. Oh SP. Grifith M. Hay ED. Olsen BR.
43. Timpl R. 1 989. Eur. 1. Biohern. 1 80: 1993. Dev. Dynar. 1 9:37-6
487-502 69. Wllchli C. Treb J. Kessler B. Winler-
44. Hudson BG. Reeers ST, Tryggvason hailer H, Trueb B. 1993. Eur. 1. Bio-
K. 1993. 1. BioI. Cher. 268:26033-36 chrn. 21 2:483-90
45. Yurchenco PD. 1 94. Se Ref. 1 0, pp. 70. Wllchli C. Koh M. Chiquet M. Odr-
351-88 matt BF. Trueb B. 194. 1. Cell Sci.
46. Gunwar S, Saus J, Noelken ME. Hudon 107:69-81
BG. 199. J. Bioi. Chr. 265:54669 71 . Pan TC. Zhang R-Z. Matlei M-G. Timpl
47. Morison K. Gamino GG. Reder R. Chu M-L. 1 92 Prc. Natl. Acad.
ST. 1 91 . 1. Bii. Cher. 266:34-39 Si. USA 89:6565-69
48. Kamagata Y, Mattei MG, Ninomiya Y. 72. Yamaguchi N, Kimura S, McBrde OW,
1992. 1. Bioi. Cher. 267: 23753-58 Hoi H, Yamada Y. et a. 1 992. 1.
49. Maiyama M. Kalluri R. Hudson BG. Biocher. 1 1 2:856-3
Reeers S. 1 992. 1. Bioi. Cher. 267: 73. Yohioka H. Zhag H. Ramirz F. Mat-
1253-58 tei M-G. Moradi-Ameli M. et a. 1992.
50. Myer JC. Jone T A. Pohjalainn E-R. Genrics 1 3:886
Kadri AS. Godard AD. et al. 199. 74. Myers JC. Yang H. D'Ipplito JA, Pe-
Am J. Hum Genet. 46: 1024-33 sente A. Miller MK. Dion AS. 194. 1.
51 . Hotikka SL, Eddy RL, Byers MG, Hoy- Bioi. Cher. 269: 1859-57
htya M. Shows TB, Tryggvaon K. 75. Timpl R, Chu ML. 1 94. See Ref. 10.
199. Pro. Nail. Acad. Sci. USA 87: pp. 207-2
16010 76. Zhang R-Z. Pan T-C. Timpl R. Chu
52. Pihlajaniemi T. Pohjalainen E-R. Myer M-L. 193. Bioher. 1. 291 :787-92
JC. 1 90. 1. Bioi. Cher. 265: 1 3758-6 77. Paente MG. Chung L. Ryynnn J.
53. Zho J. Hez 1M. Leinnn A. Trygg- Wooley DT, Wynn KC. et a. 1991 .
vaso K. 1992. J. Bioi. Cher. 267: Pro. Nail. Acad Si. USA 88:631-35
1 247581 78. Vitto J, Christian AM. 1 992. 1. cin.
54. Zho J. Mohizuki T. Smeets H. An- Invest. 90:67-92
tignc C. Lauriala L. et al. 1993. Sience 79. Burgesn R' 1993. 1. Invest. Dennatol.
261 : 1 1 67-9 101: 252-55
55. Oohashi T, Sugimoto M. Mattei M-G. 80. Gereke DR. Godon MK. Wagman DF.
Ninoiya Y. 194. J. Bioi. Cher. 269: Champliau MF. Burgeson R. 1994.
7520-26 See Ref. 10. pp. 41 7-39
56. Zhou J, Ding M. Zhao Z, Reeders ST. 81 . Brcknr-Tudrman L. 1994. Ann. Med.
1 994. 1. Bioi. Chern. 269: 1 3 193-9 26: 165-7 1
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
82 Christiano A. Greenpan DS. Lee S.
Uitto J. 1 94. J. Bio. Cher. 269:20256-
62
83. Christian A. Hoffman GG. Chung
Honet LC. Le S. Chng W. e al. 199.
Ge1Orics 21 : 1 6-79
84. Pihlajaniem T. Tamminn M. 1 90. J.
Bii. Chr 265: 16922-2
85. Tikka L. E10maa O. Pihlajaniemi T.
Tryggvason K. 1 91. J. Bioi. Cher.
26: 1771 319
86. Juvonen M. Pihlajaniemi T. AutioHar
mainn H. 1993. lb. Invest. 69:541-51
87. Higg P. Jaakkoa S. Pihlajaniemi T.
1995. Matri Bioi. 14:351 (Abt.)
88. Li K. Sawamua D. Giudc GJ. Dial
LA. Malei M-G. et aJ. 1991. J. Bioi.
Cher 266:240
89. Giudc GJ. Emery O. Diaz LA. 1992.
J. Invest. DermatoL 9:243-50
90. Hoknon SB. Ridelle KS. Jons JCR.
1992. J. Invst. Dennatol. 9:2670
91. Li K. Tanai K. Tan EML. Uitto I. 1993.
J. Bio. Cher 26:8825-3
92 Juvone M. Pihlajaiemi T. 1992. J.
BiL Chr 267:24639
93. Juvonen M. Sandbrg M. Pihlajaniemi
T. 1992. J. Bioi. Chem. 267:247007
94. Myer JC. Kivirikko S. Gordn MK.
Dion AS. Pihlajaniemi T. 192. Pro.
Natl. Aca. Sci. US 89: 1014-48
95. Huebn K. Canzzar LA. Jab EW.
Kivirikko S. Manzn H. et aJ. 1992.
Ge1Orics 14:220-24
96. Kivirikko S. Heinimiki p. Rehn M.
Honkanen N. Myes JC. Pihlajanemi
T. 199. J. Bioi. Cher 26:4773-79
97. Muragak Y. Ab N. Ninomiya Y. Olsen
BR. Oohima A. 1 94. J. Bioi. Cher.
269:442-6
98. Kivirikko S. Sala J. Myers JC. Au
tioHarmainen H. Pihlajaniemi T. 1995.
Subitted
99. Rehn M. Pihlajaniemi T. 1 94. Pro.
Natl. Aca. Sci. US 91 :4234-38
100. Oh SP. Kamagata Y. Muragaki Y.
Timmons S. Ooshima A. Olsen BR.
199. Proc. Natl. Acad. Si. US 91:
4229-33
101. Rehn M. Hintikka E. Pihlajaiemi T.
1994. J. Bioi. Cher. 26: 13929-35
102 OhSP. WarmanML. Seldin. MF. Cheng
S-D. Knoll IHM. e al. 1994. Genomics
19:49-99
103. Reh M. Pihlajaniemi T. 195. J. Bioi.
Chem. 270. I n press
104. Reid KBM. 1993. Bioher. Sc. Tran.
21 :468
105. Krieger M. Acto S. Ashkna I. Pear
son A. Penma M. Resnick D. 193. J.
BioL Chr. 268:456-72
106. Iensenius IC. Laursen SB. Zhng Y.
COLGES 431
Homskov U. 199. Biohem, So.
Tran. 22:95-10
107. Haagsman HP. 1 94. Biohm. So.
Trans. 22: 10
108. Pkp OJ. Kivirikko K. 198. N.
EngL J. Med. 31 1 :3766
109. Mazzoana M. Grffat H. Segeat A.
van dr Rest M. 193. J. Bioi. Cher.
268:302-32
1 10. Helaakoki T. Vuoi K. Myllyli R.
Kivirikk KI. Pihlajaniemi T. 1989.
Pro. Natl. Acad Si. US 86:4392-6
I l l . BassukJA. Kao WW-Y. HerP. Ked
ersha NL. Seyer JA. et a. 1 989. Pro.
Natl. Aca. Sci. US 86:7382-8
1 1 2. Veijola J. Koivln p. AnuneD p.
Pihlajaniemi T. Kivirikko K. 199. J.
Bii. Cher 269:2674653
1 13. Kivirikk KI. Helaaksk T. Tasanen K.
Vuri K. MyUyli R. et al. 199. An
N Acad. Sci. 580: 132-2
1 14. Hautala T. Byers MG, Eddy R, Shows
TB. Kivirikko KI. MyllyUi R. 1992.
Genris 13:62-6
1 15. Myllyli R. Pihlajaniemi T. Pajunn L.
Turenoiem-Hujanen T, Kiviriko KI.
1991. J. Bioi. Cher 26:2805-10
1 1 6. Kivirikk K. MyllyUi R, Pihlajaniemi
T. 1989. FASEB J. 3: 160-17
1 17. Kivirikk K. Myllyli R. Pihlajaniemi
T. 1992. In Post-Transltional Modf
cations ofProteins. e. J Hading. MJC
Crabe. pp. 1-51. Boa Raton: CRC
1 1 8. Helaakoki T. Annunen P. Vuoi K.
MacNeil IA. Pihlajaniemi T. Kivirikko
KI. 195. Prc. Natl. Acad. Si. USA.
In prss
1 19. Vuri K. Pihlajaniemi T, Mattila M.
Kivirikk KI. 1992. Pro. Natl. Acad.
Si. USA 89:7467-70
120. Pihlajaniemi T, Helaakosk T, Tasaen
K, Myllyli R, Huhtala M-L, et al. 1987.
EMBO J. 6:639
121. Koivu i. Myllyli R, Helaaksk T, Pi
hlajaiemi T. Taalln K. Kivirikko KI.
1987. J. Bioi. Chem 262:647-49
122. Noiva R. Lennar WJ. 192. J. Bii.
Cher. 267:3553-56
123. Fredman RB. Hirst T. Tuite M.
199. Trens Biochr. Sci. 19:331-36
124. Vuori K. Pihlajaniemi T, Myllyli. R.
Kivirikk KI. 192 EMBOJ. 1 1:4213-
17
125. Vur K. Myllyli R. Pihlajanier T.
Kivirikk KI. 192 J. Bioi. Cher 267:
721 1 -1 4
126. Jon DA, Grant M. Bulleid NJ. 1993.
EMBO J. 1 2: 1587-95
127. Wetterau IR, Comb KA. Mclean LR.
Spinnr SN. Agebc LP. 1991 . Bio
chmistr 30:9728-35
128. Nova R. Fredman RB. Lenarz. WJ.
1993. J. Bioi. Cher 26: 1921017
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
432 PROKOP & KIVIRIKKO
129. OISU M, Omura F, Yohimri T,
Kikuchi M. 1 94. J. Bioi. Chem. 26:
6874-77
1 30. Wells W, Xu DP, Yang Y, Roue
PA. 190. J. Bioi. Chm. 265: 1 5361-
1 31 . Krishna Rao ASM, Hausman R.
1 993. Pr. Natl. Acad Si US 9:
2950-54
132 Nakai A, Saloh M. Hirayashi K. Nagata
K. 1992. J. Cell BioI. 1 1 7:903-14
133. Clark EP. Jain N. Brickenden A. Lo
imer IA. Sanwal BD. 193. J. Cell Bioi.
1 21 : 193-9
1 34. Sauk JJ. Smith T. Norris K. Fereira L.
1 994. J. Bioi. Chem. 269:3941-6
135. Hojima Y. Margelin MM. Engel J. Bou
lillon M-M. van der Rest M. McKenzie
J. et a. 1994. J. Bioi. Chem. 269: 1 1381-
90
1 36. Hojima Y. Bebla B. Romanic AM.
Prokop O. 1994. Anl. Biochem. 223:
173-80
137. Prokop OJ. Hulmes OJS. 194. Se
Ref. 1 0. pp. 47-90
138. Veis A. Geoge A. 1 94. See Ref. 10.
pp. 15-5
139. Kadler K. Hojima Y. Proko O.
1 987. 1. BioI. Chem. 262:1566701
140. Romanic AM. Adachi E. Hojima Y.
Prokop OJ. 1 92. J. BioI. Chem. 267:
22265-71
141. Romanic AM. Adachi E. Kadler KE.
Hojima Y. Prokp OJ. 1 91 . J. Bioi.
Chem. 266: 1 2703-9
142 AlaKokko L. Hylan J. Smith C. Kiv
irikko KI. Jimenez SA. Prockop OJ.
1991 . J. Bioi. Chem. 26: 14175-58
143. Fertala A. Sieron A. Ganguly A. Li
SoW. Ala-Koko L. et a!. 199. Bio
chem. J. 298: 31-37
144. Fertala A. Sieron AL. Hojima Y. Gan
guly A, Pockp OJ. 1 994. J. Bioi.
Chem. 269: 1 1 584-89
145. Holms OF. Chapman JA, Prokop OJ,
Kadler KE. 192. Proc. Natl. Acad. Sci.
US 89:9855-59
146. Silver D. Miller J. Harison R, Prokop
DJ. 1992. Proc. Natl. Acad. Sci. USA
89:980-6
147. Ferlala A. Holmes OF. Kadler KE.
Prokop DJ. 1 94. Matrix Bioi. 14:359
(Abtr.)
148. George A, Veis A. 1 91 . Biohemistr
30:2372-T
149. Kagan HM, Trackan PC. 1991 . Am.
J. Respir. Cell Mol. BiO. 5:20-10
1 50. Trackan PC. Pratt AM. Wolanski A.
Tang SS, Offne GD, el al. 1 91 . Bio
chemistr 30:8282
1 51 . Hamilainen E-R, Jones TA, Sher 0,
Tasknen K. Pihlajaniemi T. Kivirikko
KI. 1991. Genomics 1 1 :508-16
152. Mariani TJ. Trackan p, Kagan HM,
Eddy RL, Shows TB, et a. 1 92. Matrix
12:242-8
153. Wu Y, Rich CB, Linceum J, Trackan
p, Kagan HM, Foster JA. 1992. J.
Bioi. Chem. 267:24199-2
1 54. Contente S, Kenyon K, Rimo1di 0,
Fredman RM. 1 90. Scine 249:79-
98
155. Kenyo K. Contente S, Trackan p,
Tang J, Kaga HM. Friedman R.
1991 . Sience 253:802
1 56. Kuivaniemi H, Kohonen R-M, Vahri
A, Kivirikk Kl. 1986. FEBS Lt. 1 95:
261-6
157. Himiliinen E-R, Kemppainen R, Kui
vaniemi H, Tromp G, Var A, et a.
199. Matrix BiO. 1 4:419 (Abtr.)
158. Colig A, Soklov BP, Nugnt p.
Basrga R, Prokop DJ. 1 91 . Biohem
istr 32:7-1 1
159. Laptev AV, Lu Z, Co1ige A, Pkp
OJ. 194. Biolmistr 33: 1 1033-39
160. Marini JC. Wang Q. 1 94. Matri Bii.
14:405
161. Khillan JS, Li SoW, Prokp O. 199.
Pro. Natl. Acad. Sci USA 91 :6298-
7302
162. Hanuske-AII HM. 1991. 1. Heptol.
I3:S8-16
163. Lamlrg A, Pihlajaniemi T, Kivirikko
KI. 1995. 1. Bioi. Chr. 270:ln press
16. Lee SoT, Kessler E, Greenpan DS.
199. J. BioI. Chen\ 265:2192-6
165. Murad S, Tennant MC, Pinnell SR.
1992. Arch Biochem Biophys. 292:
234-38
166. Hautala T. Heikkinen J, Kivirikko KI,
Myllyli R. 1992. Biohem. J. 283:51 -54
167. Yeowell HN. Ha V, WaIker LC. Murad
S. Pinnell SR. 1992. J. lnvest. Dennatol.
9:869
168. Willing MC, Pchn eJ, Atknso M,
Byers PH. 192. Am J. Hum Genet.
51 :508-1 5
169. Voel BE, Dolz R, Kadler K, Hojima
Y. Engel J. Prokp OJ. 1988. J. Bioi.
Chem. 263: 19249-55
170. Kadler K, Tore-Blanco A, Adachi E.
Voel BE, Hojima Y, Proko OJ. 1991 .
Biochemistry 30:501-8
1 71 . Torre-Blaco A, Adaci E. Roanic
AM. Prokop OJ. 192. 1. Bioi. Chem.
267:496-73
172. Lightwod SJ. Holme DF, Bras A,
Grant ME, Byers PH, Kade KE. 1992.
J. Bioi. Chem. 267:25521-28
173. Romanic AM. Sptila L, Adahi E,
Engel J, Hojima Y. Prko O. 1994.
J. Bioi. Chem. 26: 1 1614-9
1 74. Sptila LD, Constantinu CD, Scred
L. Ganguly A. Riggs BL. Pokop OJ.
1991. Pro. Natl. Acad Si US 88:
521-27
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
175. Spoila LD, Colige A, Sered L, Con
stantinou-Deltas CD, Whyte M, et al.
199. J. Bone Miner. Res. 9:923-32
176. Vikkla M, Metsaanta M, Ala-Kokko
L. 199. An. Med. 26:107-14
177. Ritvaniemi P, Korkko J, Bonaventure J,
Vikkla M, Hyland J, et a. 195. Ar
thrtis Rhu. In press.
178. Winterpcht A, Hilbrt M, Schwarze U,
Munlos S, Sprager J, Zabl BU. 1993.
Nal. Genel. 3:323-26
179. Williams CJ, Rok M, Harland L, Con
sidine E, McCar on S, et al. 199. Ma
tTi Bioi. 1 4:391 (Abl.)
1 80. Winterpcht A, Schwaze U, Menger
H, Mundo S, Spranger J, Zabl B.
199. MatTix Bioi. 14:392 (Ablr.)
181. Tiller GE, Weis MA, Polumbo PA,
Con DH, Rimoin DL, Ey DR. 199.
Matri Bioi. 14:390 (Abtr.)
182. Wilkn DJ, Bogaert R, Wilcx WR,
Rimoin DL, Eyre DR, Con DH. 199.
Matrix Bioi. 14:390 (Abtr.)
183. Alunad N, Ala-Kokko L, Knowlton
RG, Jimenez SA, Weaver EJ, et aI.
1991. Pro. Nal. Acad Si. US 88:
624-27
1 84. Brown DM, Nichls BE, Weingeist TA,
Sheffeld VC, Kimura A, Stone EM.
1992. Arch. Ophthalmol. 1 1 0.1 589-93
185. Alunad NN, McDonald-McGinn DM,
Zackai EH, Knwlton RG, LaRosa D,
et al. 193. Am. J. Hum. Genet. 52:39-
45
186. Ritvaniemi P, Hyland 1, Ignatius 1, Kiv
irikko KI, Prokop Dl, Ala-Kokko L.
1993. Genomics 17: 21 8-21
187. Williams CJ, Ganguly A, McCaron S,
Considin E, Michels V, Prokop DJ.
1994. Matrix Bioi. 14:390 (Abtr.)
188. Ala-Kokk L, Baldin C, Mokowitz
RW, Prokp DJ. 199. Pro. Natl.
Acad. Si. US 87:6565-6
189. Pun Y, Moskowitz RW, Lie S, Sund
stom W, Blok SR, et at. 199. Ar
thritis. Rheum. 37:2669
190. Sprange J, Winterpacht A, Zabl B.
199. Eu. J. Pediatr. 1 53:56-65
191. Pal G, Keek JMF, Muijs MA. 199.
Matrix BioI. 14:387 (Abstr.)
192. Goldstein lA, Schwarze V. Witz A.
Byer PH. 1994. Matrix Bioi. 14:392
(Abt.)
193. Kontusaai S, Tromp G, Kuivaniemi H,
Lada RL, Proko DJ. 1 90. Am. J.
Hum. Genet. 47: 1 12-20
194. Kontusaai S, Tromp G, Kuivaniemi H,
Romanic AM, Prckop D1. 190. J.
Clin. Invest. 86: 1465-73
195. Tromp G, Wu YL, Prcko DJ, Mad
hatheri SL, Kleinert C, et at. 1993 J.
Clin. Invest. 9 1: 2539-45
196. Andersn DW, Tromp G, Kuivaniemi
COLGES 433
H, Ricktts M, Pop P, et a. 1994.
Matrix Bii. 14:392 (Abt.)
197. Majamaa K, Savoann E-R, MyUyll
V. 1 92. Biochim. Biophys. Acta 1 1 38:
191-9
198. Kuivaniemi H, Prokp DJ, Wu Y,
Madhathe SL, Kleinrt C, et a. 1993.
Neurol. 43:2652-58
199. Barkr DF, Hostika SL, Zo J, Cho
LT, Olipant AR, et aI. 1 90. Siene
248: 1224-27
20 . Trygvason K, Zo J, Hotikka SL,
Shows TB. 1993. Kidny Int. 43:38-4
201. Mohizuk T, Lemmink HH, Mariyaa
M, Antignac C, Gubler M-, et a. 199.
Nat. Genet. 8:77-82
22. Antignac C, Zhow 1, Sank M, Coht
P, Roussel B, et al. 192. Kidny Int.
42: 1 178-83
203. Christian AM, Greenspn DS, Hoff
ma GG, Zhang X, Tani Y, e a. 1993.
Nat. Genet. 4:62-6
2. Hilal L, Rohat A, Duqesnoy P,
Blanchet-Badon C, Wecsler J, et aI.
1993. Nat. Genet. 5:287-93
205. Christian A, Ryynlnn M, Uitto 1.
199. Pro. Natl. Acad Si. US 91 :
3549-53
206. Christiano AM, AnhaIt G, Gibbn S,
Bauer EA, Uitto 1. 1 94. Genomics 21 :
160-6
207. Warma M, Abt M, Apte SS, Hef
feron T, Mcintosh I , et al. 1 93. Nat.
Gent. 5:79-2
208. McIntoh I, Abbtt MH, Warman ML,
Olsen BR, Francomao CA. 1 94. Hum.
Mol. Genet. 3:303-7
209. Wallis GA, Rash B, Swetman WA,
Thomas I, Supr M, et aI. 1 94. Am
J. Hum. Gellet. 54: 169-78
210. Dharmavaram RM, Elbrson M, Peng
M, Kirson LA, Kelley T, Jimnez SA.
199. Hu. Mol. Genet. 3:57-9
21 1 . Mcintoh I, Abbtt MH, Franomano
CA. 1994. Matrix Bioi. 14:397 (Abstr.)
21 2. Chan D, Cole WG, Rogers J, Bateman
JF. 199. Matrix Bioi. 14:396 (Abtr.)
21 3. Dharmavaram RM, Elbrson MA, Peng
M, Kirson LA, Kelly T, Jimenz SA.
199. Matri Bioi. 1 4:377 (Abt.)
214. Briggs MD, Choi HC, Warma ML,
Loughlin lA, Wordsworth P, et aI. 199.
Am. J. Hum. Genet. 55:678-8
21 5. Brunnr HG, van Berswn SEC, War
man ML, Olsn BR, Roprs H-H, Mari
man ECM. 199. Hum. Mol. Genet.
3: 1561-
216. Hautala T, Heikkinen 1, Kivirikko KI,
MyllyUi R 1993. Gellomics 15:399-4
217. Pousi B, Hautala T, Heikinen J, Pa
junen L, Kivirkko KI, Myllyll R. 1994.
Am. J. Hum. Gellet. 55:899-9
21 8. Marshall M, Walker L, Murad S, Pinnell
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.
434 PROKOP & KIVIRIKKO
S. Yeowell H. 1994. Matri Bioi. 14:398
(Abt.)
21 9. Hyland 1. Ala-Kokko L. Royce p. Ste
inmann B. Kivirikko K. MyllyHi R.
1992 Nat. Genet. 2:228-31
220. Ha V. Mashall MK. Elsa U. Pinnell
SR. Yeo well HN. 1994. J. Clin. Invest.
93: 1 71 6-21
221. Smith LT. Wereleki W, Milstone LM,
Petty EM. Seahoe MR. e a. 1992.
Am J. Hum Genet. 51 :235-4
222 Nusgens BV. Verllen-Dumoulin C.
Hennann-Le T. De Pae A. Nuytinck
L. el a. 1 92. Nal. Genel. 1 :214-17
223. Danks DM. 193. Se Ref. 7. pp. 487-
50
224. Gacer S. McGee C. Uriu-Hare JY.
Kounn T. Packan S. et al. 1993.
Arch. Biochr. Biophys. 301:32529
225. Yeowell HN, Mashall MK. Walker LC.
Ha V. Pinnell SR. 194. Arh. Biohem
Biophys. 308:299-305
226 Bonadio 1. Saundrs T. Tsai E. Gold
stein SA, Morris-Winman J. et al. 199.
Pro. Nail. Acad Si. USA 87:7145-9
227. Staey A. Bateman J. Chi T. Macaa
T. Cole W. Jaenisch R. 198. Nalur
332: 1 31-3
228. Khillan JS. Olsn AS. Kontusaari S.
Soklov B. Prokp O. 1 91 . J. Bioi.
Cher. 266:23373-79
229. Pereira R. Khillan IS. He1minen H.
Hume EL. Prokop DJ. 1993. J. Clin.
Invest. 91 :709-16
230. Pereira R. Halford K. Sokolov B. Khil
Ian JS. Proko DJ. 1 94. J. Clin. Invest.
93: 1 765-6
23 1. Chipman SD. Sweet HO. McBride
DJ, Davissn MT, Mark SC. et a.
193. Pro. NatL Aca Si. US 9:
1701-5
232 Vanenbrg p. Khilla JS. Pokp O.
Helminn H. Kontusari S. Ala-Kokko
L. 1 91 . Pro. Natl. Acad Sci. USA
88:760-4
233. Garofalo S. Vuorio E. Metwanta M.
Roati R. Toman D, et al. 1991. Pro.
Nail. Aca. Sci. US 88:968-52
234. Metsaranta M. Garofalo S. Dkr G,
Rintala M. de Crobgh B, Vuoo
E. 1992. J. Cell Bioi. 1 1 8:2312
235. Helminn HI, Kiraly K, Peluai A.
Tammi MI. Vandnbrg P, et a. 1993.
J. Cli" Invest. 92:582-95
236. Metiranta M, Siiinn A-M, Vuoo
E. 1994. Matri Bii. 14:41 1 (Abtr.)
237. Garofalo S. Metslranta M, Ellard 1,
Smith C, Horton W, et a. 1 93. Pro.
Nail. Aca. Sci. US 9:3825-2
238. Toman D, Stachr B, Masca T, Rob
brson D. Smith C, et a. 194. Malrix
Bioi. 14:413 (Abstr.)
239. Quaglino D. Toman D. de Crombggh
B. Davidsn JM. 199. Matrix Bioi.
14:412 (Abstr.)
240. Nakta K, Ono K. Miyazak I-I. Olsen
BR. Muragaki Y. et a. 1993. Pro. NaiL
Acad. Si. US 90:28(74
21. Fassler R, Scbnegelsbrg PNI, Dausman
I. Shiny T. Muragak Y, e a. 199.
Pro. Nal. Acad Si. USA 91:570-74
22. laenko 0, LuVaile P, Olsn BR. 1993.
Nature 365:56-1
243. Roati R. Hoan GSB, Pinero GJ, Garo
falo S, Keen DR, et a. 199. Nat.
Gent. 8: 1 29-35
24. Elima K, Eerola I, Makkla M.
Kannn K, Rosati R. et al. 1 94. Matrix
Bioi. 14:414 (Abstr.)
24
5
. Vanenbrg p. Khillan J. Proko DJ.
199. Matri Bio. 1 4:41 5 (Abtr.)
246. Li Y, Lacerd DA. Waan M. Beier
DR, Yoshioka H, et al. 195. Cell. In
pess
27. Pereira R. Halford KW. O'Hara M,
Leepr D, Pollad M, e a. 195. Pro.
Natl. Acad. Sci. US 92:1n prss
248. Ganguly A, Smelt S. Mewar R. Ferala
A. Siero AL. et a. 199. Pro. NaL
Acad. Si. US 9 1:735-6
A
n
n
u
.

R
e
v
.

B
i
o
c
h
e
m
.

1
9
9
5
.
6
4
:
4
0
3
-
4
3
4
.

D
o
w
n
l
o
a
d
e
d

f
r
o
m

w
w
w
.
a
n
n
u
a
l
r
e
v
i
e
w
s
.
o
r
g
b
y

H
I
N
A
R
I

o
n

0
4
/
0
9
/
1
1
.

F
o
r

p
e
r
s
o
n
a
l

u
s
e

o
n
l
y
.