Genetic Rescue of Chondrodysplasia and the Perinatal Lethal Effect ...

Genetic Rescue of Chondrodysplasia and the Perinatal Lethal Effect of Cartilage Link Protein-deficiency1

Mátyás Czipri*, Jeffrey M. Otto*, Gabriella Cs-Szabó*, Rajesh V. Kamath*, Csaba Vermes*,

Gábor Firneisz*, Kevin J. Kolman*, Hideto Watanabe†, Yefu Li

‡, Peter J. Roughley

¶,

Yoshihiko Yamada§, Björn R. Olsen,

‡ and Tibor T. Glant*

2

*Section of Biochemistry and Molecular Biology, Departments of Orthopedic Surgery and

Biochemistry, Rush University at Rush-Presbyterian-St. Luke’s Medical Center, Chicago, IL 60612,

USA; †Institute for Molecular Science of Medicine, Aichi Medical University, Nagakute, Aichi 480-

1195, Japan; ‡Department of Cell Biology, Harvard Medical School, Boston, MA 02115, USA;

¶Genetics Unit, Shriners Hospital for Crippled Children, and Department of Surgery, McGill

University, Montreal, Quebec, Canada H3G 1A6, and §Craniofacial Developmental Biology and

Regeneration Branch, National Institute of Dental and Craniofacial Research, NIH, Bethesda, MD

20982, USA; and

Running Title: Genetic Rescue of Link Protein-deficiency

Keywords: Transgenic animals, cartilage, link protein, chondrodysplasias, genetic rescue, extracellular

matrix

2To whom correspondence should be addressed/Corresponding Author:

Dr. Tibor T. Glant

Section of Biochemistry and Molecular Biology Departments of Biochemistry and Orthopedic Surgery

Rush-Presbyterian-St. Luke’s Medical Center

1653 West Congress Parkway

Chicago, IL 60612

Phone: (312) 942-5855

Fax: (312) 942-8828

E-mail: [email protected]

Copyright 2003 by The American Society for Biochemistry and Molecular Biology, Inc.

JBC Papers in Press. Published on May 5, 2003 as Manuscript M303329200 by guest on M

arch 24, 2018http://w

ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Genetic rescue of LP-deficiency - Czipri et al. 2

(SUMMARY)

The targeted disruption of cartilage link protein gene (Crtl1) in homozygous mice resulted in a

severe chondrodysplasia and perinatal lethality. This raised the question of whether the abnormalities

seen in Crtl1 null mice are all caused by the absence of link protein in cartilage, or whether the

deficiency of the protein in other tissues and organs contributed to the phenotype. To address this

question we have generated transgenic mice overexpressing cartilage link protein under the control of a

cartilage-specific promoter and then these transgenic mice were used for a genetic rescue of

abnormalities in Crtl1 null mice. While the overexpression of cartilage link protein resulted in no

abnormal phenotype, the cartilage-specific transgene expression of link protein could completely

prevent the perinatal mortality of link protein-deficient mice and, depending on the level of the link

protein expression, rescue skeletal abnormalities.

Although link protein was originally isolated from cartilage, we found and determined Crtl1

transcripts and corresponding proteins in every organ tested from mouse embryos to aging animals.

We also identified three additional members of the link protein family, all co-localized with HA-

binding proteoglycans in the mouse genome. The ubiquitous presence of link protein suggests a

general and systemic function of link protein in the organization of extracellular matrix in a number of

tissues, possibly interacting with other proteoglycans, such as versican, brevican and neurocan.

by guest on March 24, 2018

http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


(INTRODUCTION)

A highly specific macromolecular organization of cartilage extracellular matrix (ECM)3 is

required for chondrogenesis, chondrocyte differentiation, endochondral ossification and for the

maintenance of the weight-bearing function of articular cartilage. The major macromolecular

components of this ECM are the proteoglycan aggrecan, hyaluronic acid (hyaluronan, HA), link

protein (LP) and type II collagen. Hundreds of aggrecan molecules associate with a single HA filament

to form large proteoglycan aggregates. The link protein, a glycoprotein, binding simultaneously to both

HA and aggrecan, stabilizes the aggregate structure (1), and the aggregates are entrapped within the

mesh-like network of type II collagen fibrils (2-4). The negatively charged glycosaminoglycan chains

of aggrecan bind large amounts of water and are responsible for the resiliency of cartilage (5).

Mutation in a gene encoding components of cartilage ECM may result in skeletal disorders,

chondrodysplasias, in mice (6,7) and humans (8).

Cartilage LP is a member of the hyaluronectin superfamily, and shares a high degree of

sequence homology in different species, particularly in the HA binding domain (2); also known as

‘link module’ (4,9). Although LP was originally isolated from cartilage, its presence has been reported

in numerous non-cartilaginous tissues, such as aorta (10), mesonephros and other embryonic tissues

(11,12), brain (13) and sclera (14) in chicken, and the mouse cumulus-oocyte complex, without

concomitant expression of aggrecan (15,16). This wide distribution of LP suggests that it may have a

similar function in various organs to that described for cartilage by stabilizing the interactions between

HA-versican in connective tissues (17), and between HA-brevican (13,18) or HA–neurocan (19,20) in

brain. In addition, recent findings indicate that LP, or at least its N-terminal domain, may function as a

growth factor upregulating the synthesis of aggrecan and type II collagen in cartilage (21-24).

Although mutation(s) in the cartilage LP gene (CRTL1) was excluded in several human hereditable

chondrodysplasias (25,26), a recent study of mice with targeted mutation in the Crtl1 gene clearly


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


showed a prominent function of LP in chondrocyte differentiation, cartilage ECM organization and

skeletal development (27). The lack of LP in homozygous mice was perinatal lethal, accompanied by

severe skeletal abnormalities (27). To address the question of whether all the abnormalities and the

perinatal lethality are caused by the LP deficiency in cartilage, we have generated transgenic mice

overexpressing cartilage LP under the control of a cartilage-specific promoter, and then these

transgenic mice were used for genetic rescue of the phenotype in Crtl1 null mice. We isolated and

sequenced the mouse cartilage LP cDNA (Crtl1), which then was inserted into a cartilage-specific

expression vector. While the overexpression of cartilage LP has resulted in no abnormal phenotype, the

cartilage-specific transgene expression of LP could completely prevent the perinatal mortality of LP-

deficient mice and rescue skeletal abnormalities depending on the level of the LP expression.

EXPERIMENTAL PROCEDURES

Isolation and cloning of the murine LP gene - Cartilage from articular joints of neonatal (2-5-days old)

BALB/c mice were dissected, total RNA isolated and reverse transcribed (RT) as previously described

(28,29). Prior to reverse transcription, samples were digested with DNase I (Invitrogen, Carlsbad, CA)

to eliminate residual genomic DNA from the RNA sample. First-strand cDNA was synthesized from 1

g of total RNA primed by oligo-d(T) using SuperScript II reverse transcriptase (Invitrogen) as

described (28). The strategy for identification of the mouse LP gene was based upon RT-polymerase

chain reaction (PCR), first using primers homologous in various species, and then using mouse specific

primers. PCR products were directly cloned into the TA-vector (Invitrogen) and sequenced from both

directions. Overlapping clones from the 5' and 3' ends were generated, using a 5' RACE system


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


(Clontech, Palo Alto, CA) and 5’ and 3' Genome Walker (GW) systems (Clontech), and cloned into the

pT-ADV vector.

In order to identify the 5' transcription start site, primer extension was employed. A reverse

primer in exon 1 (LP1) was end-labeled and used to both reverse transcribe mRNA from mouse

chondrocytes and sequence a PCR product of genomic DNA corresponding to the LP promoter region.

Both the [33

P]-ATP labeled RT product and the sequencing reaction were run in parallel on a 7% Long

Ranger sequencing gel (FMC, Rockland, ME).

For determination of intron-exon structure, the GW kit was employed using gene-specific

nested primers located in the neighboring exons (not shown; primer sequences and PCR conditions are

available upon request). Resulting PCR products were sequenced and sequence alignments and

analysis were performed using the OmigaTM

2.0 computer package (Oxford Molecular Group, Hunt

Valley, MD) or the NIH BLAST Server. Whenever it was possible, a long range PCR using genomic

DNA template from BALB/c mice was used to determine the size of introns.

To determine tissue distribution of Crtl1, total RNA was isolated from various organs and

tissues, and reverse transcribed as described above. To identify reverse transcribed LP mRNA in

various organs, RT-PCR was employed using LP2 and LP3 or LP4 and LP5 primers (Table 1). The

same PCR primers (LP4 and LP5) were used to detect the presence of transgene in genomic DNA

samples. PCR products were separated in a 1.5% agarose gel and visualized by ethidium bromide

staining. Representative PCR products were isolated from gels (Quiaex II gel extraction kit, Qiagen

Inc., Valencia, CA) and sequenced using an ABI 310 genetic analyzer (Perkin Elmer, Branburg, NJ).

For quantitative analysis, real-time PCR was used. For Northern dot-blot analysis, total RNA was

isolated from 20.5 days old embryos (E20.5) and mRNA purified using Oligotex mRNA kit (Qiagen

Inc). Two µg of this oligo-d(T)-purified mRNA in 2 µl final volume was dotted on GeneScreen Plus

membrane (New England Nuclear, Boston, MA), baked at 80oC in a vacuum oven, prehybridized (30),


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


and then hybridized with a 32

P-dCTP-labeled 775 bp long PCR product generated by primers LP6 and

LP7 (Table 1).

Real-time quantitative PCR (RT-QT-PCR) - Crtl1 gene expression in different organs of newborn

wild type mice and in cartilage of newborn mice with different genotypes, was quantified by real-time

quantitative RT-PCR using the Smart Cycler System (Cepheid, Sunnyvale, CA), and detection was

carried out by measuring the binding of fluorescent SYBR Green-I to double stranded DNA. The PCR

reactions were carried out in microtubes in 25 l volume. cDNA template (1 l) of RT-PCR product

was added to a PCR reaction mixture which contained final concentrations of 0.5 M Crtl1 specific

forward (LP10) and reverse (LP11) primers (Table 1), 1:50,000 dilution of SYBR Green-I stock

solution (BioWhitaker Mol. Appl. Cambrex, Rockland, ME), 200 M dNTP, 1.5 mM MgCl2, and 2.5

units of Taq DNA polymerase (Promega, Madison, WI). For normalization, glyceraldehyde-

phosphate-dehydrogenase (GAPDH) cDNA was amplified with forward (GP1) and reverse (GP2)

primers (Table 1).

The fluorescence emitted by the reporter dye was detected online in real-time, and the threshold

cycle (Ct) of each sample was recorded as a quantitative measure of the amount of PCR product in the

sample. The Ct is the cycle number at which the fluorescence exceeds a fixed level above baseline. The

Crtl1 signal was normalized against the quantity of GAPDH and expressed as Ct = (CtCrtl1-CtGapdh).

The differences in Crtl1 signal were expressed as Ct = Ctorgans - Ctcartilage in the inter-organ-

comparison of Crtl1 expression, and as Ct = Ctother genotypes - Ctwild type in the Crtl1 expression

comparison between different genotypes. Relative gene expressions were than calculated as 2- Ct

. The

real-time PCR assay included two independent reverse transcribed RNA samples isolated from three

embryos, or organs and tissues of three animals.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


Crtl1-deficient and transgenic mice with tissue-specific overexpression of LP in cartilage - The Crtl1-

deficient founder male was received from the NIH/NIDCR. Gene targeting was performed by

disrupting exon 4 of the Crtl1 gene at the BsgI site upstream of the HA-binding ‘link modules’ by the

pGK-neor-poly(A) cassette in the pPNT vector (27).

In order to generate transgenic mice that over-express LP in cartilage, we designed a cartilage-

specific transgenic expression vector (31) (Fig. 1C). To generate transgenic mice that overexpress the

murine cartilage LP mRNA, a 1758 bp-long cDNA fragment containing the entire coding sequence

was blunt-end ligated into the polylinker site of the transgenic vector pSP/44-3 (Fig. 1A). After

linearization and gel purification, the transgenic construct was used for microinjection (DNX

Technologies, Princeton, NJ). Initially, ten founder mice were identified to harbor the transgene.

From these mice, four independent transgenic lines were established and then two, a low (Crtl1TgA

) and

a high (Crtl1TgC

), transgene and protein-expressing lines were used for rescue experiments. While

transgenic products were dominantly expressed in cartilaginous tissues in embryonic life, we were able

to detect transcripts eventually in all organs of Crtl1-null embryos and newborn mice carrying the LP

transgene (Fig. 1E), but only in the skeletal tissues of adult animals (data not shown). This indicates

that, despite the cartilage-specific type II collagen-driven promoter, pSP/44-3 expression vector, at

least in the embryonic life or at the perinatal age, was leaking.

Mice were genotyped by PCR using gene or neomycin (Neo)-specific primers with

genomic DNA templates (Table 1 and Fig. 1D). Genomic DNA was isolated from tails by a standard

method (32) and a primer pair of LP4 and LP5 was used to identify the presence of the transgene (Fig.

1D). Crtl1 heterozygosity was demonstrated by the presence of two PCR products (283 bp without and

1.88 Kbp with Neo, Fig. 1D lane 3) using gene-specific LP12 forward and LP13 reverse primers

(Table 1). In addition, homozygosity of the Crtl1-deficiency was confirmed by the presence of (i) Neo-


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


specific (PNeo1 and PNeo2), and (ii) gene- (LP14) and Neo-specific (PNeo3) primer-generated PCR

products, and (iii) the absence of the 283 bp PCR product by LP12 and LP13 (Table 1).

Rescue of perinatal lethality due to Crtl1-deficiency - For rescue experiments, we generated Crtl1-/-

animals that also carried the Crtl1 transgene. Initially, homozygous transgenic females of lines A and

C (Crtl1TgA

and Crtl1TgC

; F9-F10 generations) were mated with heterozygous LP-null males (Crtl1+/-

),

first to generate heterozygous Crtl1+/-

Crtl1Tg+

offspring. Progeny (N1 generation) was genotyped by

PCR, and Crtl1+/-

mice carrying the transgene (Crtl1TgA+

or Crtl1TgC+

) were selected for further

breeding. Mice heterozygous for both endogenous and transgene Crtl1 were intercrossed to generate

Crtl1-/-

Crtl1TgA+

or Crtl1-/-

Crtl1TgC+

genotypes.

Generation of polyclonal antibodies - Polyclonal antibodies were generated against murine LP by

immunizing rabbits with synthetic link protein peptides (Research Genetics, Huntsville, AL). Peptides

were designed as LPro1 142

VIEGLEDDTGV and LPro2 341

KKHKLYGVYCFRAYN based upon their

hydrophilic character (33) and predicted antigenicity (34) using methods incorporated in the Omiga 2.0

software package. Peptides were coupled to Keyhole Limpet hemocyanin (35), dissolved in PBS

(phosphate buffered saline, pH 7.4) and emulsified in 1 ml complete Freund’s adjuvant. Three rabbits

were immunized intramuscularly (3-5 mg/injection) with each peptide conjugate. Boosters were given

every 2-3 weeks with peptide conjugates emulsified in incomplete Freund’s adjuvant. Sera from

immunized rabbits were collected weekly after the 5th injection. Antibodies from rabbit sera were

purified on corresponding peptides bound to sulfolink columns (36).

While all antibodies detected LP, polyclonal antibodies LPro1-R11 and LPro2-R18 were the most

specific, showing essentially no background in Western blotting. Based upon the synthetic peptide

sequence (LPro2) used for immunization, the polyclonal antibody LPro2-R18 was not only murine LP


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


specific, but also cross-reacted with LP from human cartilage. Indeed, when LPro2-R18 was first pre-

incubated with the synthetic peptide LPro2, or pre-absorbed with either mouse LP or human cartilage

extract, it was no longer able to cross-react with either mouse or human cartilages. LPro1-R11 did not

react with human cartilage, but could be also neutralized with either synthetic peptide LPro1 or

purified mouse LP. These two antibodies were used for both Western blot analysis and

immunohistochemistry in this study.

Tissue extraction, protein purification and Western blot analysis - For quantitative analysis, cartilage

samples from 4-day-old newborn mice were pulverized under liquid nitrogen and then extracted with

4M guanidinium chloride in the presence of protease inhibitors as described (37). Extracts were

dialyzed against water, freeze-dried and normalized for protein content by Bicinchoninic Acid assay

(Pierce, Rockford, IL) and for sulfated glycosaminoglycan (GAG) content by dimethyl-methylene-blue

(DMMB) assay (38).

Crude extracts from various organs of 40 adult mice were prepared in 4M guanidinium chloride

containing protease inhibitors as described above, except that tissue samples were homogenized. As

the Crtl1 transcript was detected in many organs, but the translation product was detected only in crude

extracts from cartilaginous tissues, subsequent purification was performed. Approximately 600 mg

lyophilized tissue/organ extracts were rehydrated in 60 ml of 0.4 M guanidinium chloride in 100 mM

acetate buffer (pH 5.8) with enzyme inhibitors. Then 2 mg/ml HA (Sigma, St. Louis, MO) and 2

mg/ml highly purified rat chondrosarcoma aggrecan (D1D1)(37), free of LP, were added to each

sample. Samples were gently stirred for 24h at 4oC. The density of each sample was adjusted to

1.5g/ml by cesium chloride and centrifuged at 100,000 xg for 48 h at 10oC (37). Guanidine

concentration of the bottom 1/3 fraction was increased to 4M, stirred overnight at 4oC and the density

adjusted to 1.5 g/ml. Samples were spun as above for 60 h. The top 1/6 fraction of this dissociative


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


gradient (A1A1D6) was collected (39,40), dialyzed against water and lyophilized. This procedure

resulted in 2-8 mg lyophilized material from the 600 mg tissue extracts, which was used to detect LP in

different organs.

Crude extracts from skeletal tissue of wild type, transgenic and rescued newborn mice were

rehydrated in Tris-acetate (0.1M, pH 7.2) buffer containing 0.15 M sodium chloride. Samples

normalized to either protein or DMMB content were separated in 10% SDS-polyacrylamide slab gels

under reducing conditions. Gels were electrophoretically transferred onto nitrocellulose membranes

(BioRad, Hercules CA) and stained with rabbit antibodies to mouse LP (LPro1-R11 or LPro2-R18)

(0.05 µg/ml diluted in 1% milk/PBS) followed by affinity purified and peroxidase-conjugated goat

anti-rabbit antibody (Accurate Chemical, Westbury, NY) (28). Enhanced chemiluminesence

(Amersham) at a serial time range was used to detect immune reaction, which was then quantified

using a PDI (Huntington Station, NY) gel scanner and an integrated software package program.

Relative amounts of antibody-stained proteins were identified on the membrane expressed as pixel

densities and normalized to mouse LP (A1A1D6) purified from newborn mouse cartilage by cesium

chloride gradient centrifugation as described above.

Macroscopic and histological staining of skeletal tissues - Bones and cartilage of newborn mice were

stained with alizarin red and alcian blue in 70% ethanol as described elsewhere (41), with the

modification that the staining step was extended to 3 days for newborn mice, and 4 days for 4-day-old

mice. Specimens for histology were fixed with 10% buffered formalin, decalcified for 1-2 weeks and

then embedded in paraffin. Tissue sections (6 m) were trichrome-stained with safranin O, fast green

and iron hematoxylin.

For immunohistochemistry, mouse organs/tissues were embedded in OCT compound (Sakura

Finetec, Torrance, CA), and 6-8 µm thick frozen sections on 3-aminopropyltriethoxi-silane-coated


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


(Sigma) slides were fixed in ice-cold acetone for 5 minutes. Sections were washed with PBS, and

pretreated with protease-free chondroitinase ABC (0.5 unit/ml; Seikagaku, Japan) at 37 C for 30

minutes followed by incubation with 10 % normal goat serum in PBS. Sections were immunostained

with LPro1-R11 polyclonal antibody (1:100) diluted in PBS containing 1% normal goat serum for 1

hour. All immunostainings were carried out in room temperature. After washing, sections were treated

with rhodamine-conjugated goat anti-rabbit IgG (Santa Cruz Biotechnology, Santa Cruz, CA) diluted

1:100 in PBS for 1 hour. The slides were washed, mounted in Fluoromount (Kirkegaard and Perry

Laboratories, Gaithersburg, MD) and analyzed using a Nikon Microphot-FXA microscope or Eclipse

TE 200 confocal microscope (Nikon, Garden City, NJ).

RESULTS

Coding region and genomic structure of mouse Crtl1 gene - Using a set of primer combinations

homologous to human (42), pig (43), rat (44), bovine (45) and chick (46) LP sequences, an

approximately 650 bp long mouse LP cDNA was PCR-amplified and sequenced from a reverse-

transcribed mouse cartilage mRNA template. Subsequently, 3’ and 5’ RACE systems were used to

determine the 2780 bp murine LP sequence, which has been deposited into GeneBank-EMBL

(Accession number: AF 098460).

The coding region contained a single open reading frame of 1,065 nucleotides encoding a protein

of 355 amino acid residues, which is one residue (His) shorter than described by Deák et al. (47).

Comparative analysis demonstrated 99% homology with rat (44) and 96% homology with human

CRTL1 (43). Most of the amino acid substitutions were identified in the N-terminal region of the

protein, prior to the start of the A loop. In order to identify the 5' transcriptional start site, primer


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


extension was employed and a single extension product (5’-CTGGGCACAG…) at -368 bp position

relative to the translational start site was identified (Fig. 1B). The size of the 5'-untranslated region in

the murine Crtl1 gene is similar to that identified in human CRTL1 at -315 bp (48), but shorter than

described for the chick LP gene at -500 bp (49), and significantly longer than described for Crtl1 in rat

chondrosarcoma (48,50).

In order to determine the intron-exon structure of the murine Crtl1 gene, exon specific primers and

the GW kit were used and the sequencing results are summarized in Table 2. Because of the unusually

large size of intron 1 (43 Kbp; Table 2), initially we were unable to determine the size of the mouse

Crtl1 gene. Based upon our cDNA sequence, intron-exon borders and the genomic sequences of Celera

Discovery System (51) (htpp://www.celera.com/) and the public database Ensembl (52)

(htpp://www.ensembl.org) the mouse Crtl1 gene is 68,099 bp-long on chromosome 13 between 90.138

and 90.206 Mb region immediately upstream of the versican gene (Cspg2) in a head-to-head

orientation. Exon 1 of Crtl1 was found to be non-coding, while exon 2 contains the translational start

site, signal peptide and N-terminal sequence, exon 3 the A loop, exon 4 the B loop and exon 5 the B'

loop and 3' untranslated sequences (Fig. 1A and Table 2).

Tissue distribution of LP transcripts - Total RNA was isolated from E10.5 whole embryos, various

organs of E16.5 and E20.5 embryos, newborn (2-4 days old) and adult mice. PCR analysis

demonstrated the presence of Crtl1 transcript in every organ tested; the transcript levels were

quantified by RT-QT-PCR (Fig. 2A). Similarly, Northern dot-blot hybridization detected the Crtl1

transcript in all tissues (Fig. 2A). The presence of LP protein was also confirmed by Western blot

analysis in protein extracts from these tissues and organs of adult mice (Fig. 2A, bottom panel).

Although LP previously was identified in various organs of embryonic chicken (11,12), this is the first

systematic study to detect LP distribution in a mammalian system. These observations suggest that LP


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


is a ubiquitous protein that may function to stabilize protein-HA interactions in many organs and

tissues.

To confirm the tissue distribution of the translated Crtl1 product, frozen sections from different

organs were prepared and stained with rabbit polyclonal antibody LPro1-R11 (Fig. 2B-G). Although at

different levels, the LP was detectable in every organ tested by immunohistochemistry.

Generation of cartilage LP transgenic mice - Four LP transgenic founder mice showed stable

transmission of the transgene to progeny, and were used to establish independent transgenic lines.

Based on Southern blot and subsequent densitometric analysis, founder line C was determined to have

the highest copy number, greater than 20 copies of the LP transgene. Founders A, B and D were

determined to have less than 10 copies of the transgene (data not shown). Because founder C had the

highest copy number of the transgene, and Line A exhibited the best breeding properties, these two

lines were maintained to establish homozygous LP transgenic lines and used for rescue experiments.

Offspring of both lines exhibited normal growth without any gross- or histo-pathological

abnormalities, not even in the growth plate or articular cartilage up to 18 months of age (data not

shown). It is important to note here again that transcripts of the transgenic product, while dominantly

expressed in cartilaginous (skeletal) tissues, eventually, it could be detected in all organs of mouse

embryos and newborn mice (Fig. 1E).

Expression of transcript and protein products in wild type and genetically manipulated mice -

Because the wild type Crtl1 (LP) gene and the LP-transgene produce indistinguishable transcripts and

proteins, overexpression of mRNA was determined by normalization to GAPDH transcripts, and the

translated product by normalization to the cartilage protein decorin, since none of these products were

expected to affect, or to be affected by, the expression of the transgene. Later, it was observed that


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


decorin content determined by Western blot hybridization and the amount of sulfated

glycosaminoglycans measured by DMMB assay correlated well, and the DMMB assay was then

routinely used to normalize cartilage samples/extracts. In the earlier generations (F2-F5), there were

significant differences in both mRNA and protein levels measured in offspring (positive litter-mates)

of the same line, most likely because mice were not yet homozygous for the transgene, and there was

individual variability in the levels of integration sites and copy number seen by Southern blot analysis

(data not shown).

Although the Northern dot-blot hybridization (not shown) or real-time RT-PCR (Fig. 3A) could

not confirm a higher expression of the transgenic transcript (mRNA) in the skeletal tissue of F6-F8

generations of mice from Line A than wild type mice (Fig. 3A, lanes 1, 4), the translated product

(protein) was approximately 28-62% higher (mean 48 24 %; n = 11) in cartilage collected from 2-4-

days old newborn than wild type mice (Fig. 3B,C, lanes 1, 4). In contrast, LP mRNA level was at least

3-times higher in cartilage of transgenic Line C, which was accompanied by 153 21% (n = 9) greater

protein expression in newborn cartilage (Fig. 3A-C, lane 1,5).

Neither mRNA nor LP was detectable in homozygous newborn Crtl1-deficient mice (Fig. 3, lane

3), or in Crtl1-/-

fetuses collected at E20.5. Both LP mRNA and protein expression was significantly

less in heterozygous (Crtl1+/-

) mice (Fig. 3 Lane 2), but Crtl1+/-

mice exhibited normal skeletal growth

and breeding characteristics similar to transgenic mice which expressed higher levels of Crtl1 mRNA

and protein.

Genetic rescue from perinatal lethality - Two transgenic lines A and C with different protein

production (Fig. 3B,C) offered possibly a different outcome in genetic rescue. It was expected that

those Crtl1-/-

mice which were also carrying the transgene would survive. Indeed, while the Crtl1-/-

(LP-deficient) mice without the LP transgene died soon after birth, most of the Crtl1-/-

mice carrying


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


the LP transgene survived and many reached adult age. The Crtl1-/-

mice were born with severe

skeletal deformities and dwarfism and, in contrast to the original study (27), which described a 7%

survival rate of Crtl1-/-

mice with dwarfism and severe chondrodysplasia, we have never seen any

surviving homozygous mice deficient of wild type Crtl1.

The survival rates in N2 generations were different in the two rescue groups, and the rates directly

correlated with the levels of the transgenic protein product. In the N2 generation, the survival ratio of

Crtl1-/-

Crtl1TgA+

mice was lower (4.2 %) than in the Crtl1-/-

Crtl1TgC+

mice (14.4 %), both were lower

than the expected ratio (Table 3). The survival ratio was improved in the N3 generation, which almost

reached the expected ratio in the Crtl1-/-

mice rescued with the higher transgene-expressing line C

(Table 3). Quite surprisingly, mice with less than 20% of the wild type LP level could survive (Fig. 3C,

lane 6), although their growth was significantly retarded and they exhibited severe chondrodysplasia

(Figs. 4A-D, 5). In contrast, mice with at least 50% LP level not only survived, but also exhibited

normal growth and skeletal structure (Fig. 4E-H).

Phenotype of LP-rescued mice - There were evident and significant differences in a number of

phenotypic characteristics when Crtl1-deficient mice rescued with the low or high transgene-

expressing lines were compared. While the Crtl1-/-

Crtl1TgA+

offspring survived, eventually, all

Crtl1TgA+

-rescued mice developed more or less severe abnormalities in their skeletal structure (Fig.

4Aa-Dd). These newborn mice were smaller than their litter-mates, their snout was shortened, and the

antero-posterior diameter of the head was shorter giving a dome-like appearance to the skull (Fig.

4Aa,4Bb). Their limbs were significantly shorter due to the 5-15% shorter length of the long bones,

although the ossification centers were present in the epiphyses of the long bones (Fig. 5A,B), and both

cartilage and bone were stained with alcian blue/alizarin red (Fig. 4Aa,Bb). The growth plates of these

mice were wider and, all zones, from the resting to the hypertrophic zone, contained fewer


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


chondrocytes (Fig. 5; rescue group A). The growth plate structure was mostly disorganized, although

the columnar organization of hypertrophic chondrocytes was maintained. The lower number of

chondrocytes was associated with an extensive amount of the extracellular matrix, which was as rich in

safranin O-stained proteoglycan as the growth plate cartilage in wild type animals (Fig. 5B,C). The

hypertrophic zone of the growth plate was narrowed (Fig. 5C), and bony trabecules were thinner in the

calcification zone (Fig. 3B).

A marked hyperlordotic curvature in the upper, and a hyperkyphotic curvature in the lower thoracic

region of the spine created massive spine deformities in the sagittal plane, which were consistently

present in all mice rescued with the low amount of LP (Fig. 4Cc,Dd). The phenotypic anomalies found

in rescue group A were similar to those described in Crtl1-/-

mice (27), resembling massive

chondrodystrophic abnormalities. The skeletal deformities remained apparent during skeletal

maturation, and the dwarfism became even more evident in aging mice (Fig. 4Aa-Dd).

In contrast to the skeletal and growth abnormalities present in rescue group A, no skeletal

deformities were seen in Crtl1-/-

mice rescued by LP transgenic line C (Fig. 4Ee-Hh); except mild

phenotypic abnormalities with the smaller epiphyseal ossification centers of long bones (Fig. 5A,

rescue group C), and reduced chondrocyte numbers in the growth plate accompanied by increased

amount of ECM (Fig. 5B,C). Positive immunostaining of the cartilage in newborn mice proved the

presence of the transgene protein (LP) in the ECM of rescued mice (Fig. 5D).


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


DISCUSSION

This study summarizes the results of a partial to complete rescue of cartilage LP-deficiency-induced

skeletal abnormalities and the elimination of perinatal mortality, which is believed to be the

consequence, at least in part, of insufficient cartilage function of the upper respiratory tract (27). The

effect of the transgene, i.e. the rescue process, is dose-dependent and the results confirm that

approximately 50% reduction of LP in cartilage does not cause significant pathology.

Chondrocytes and the surrounding ECM exist in a dynamic equilibrium in terms of gene

regulation, tissue differentiation, and biochemical and biophysical characteristics of cartilage. A defect

in, or lack of, any component of this highly organized ECM can lead to a chondrodystrophic phenotype

and frequently to death. Chondrocyte ablation by diphtheria toxin A, expressed in mice by a transgenic

construct driven by a collagen type II promoter and enhancer, resulted in severe intrauterine

malformations and death (53). These transgenic embryos were small, had cleft palates and

underdeveloped limbs, generally resembling the genotype of chondrodystrophic mice (53). An

inherited 7 bp depletion generating a non-sense mutation in the mouse aggrecan gene (Agc1) led to

dwarfism, skeletal disorders and death in homozygous cmd (cartilage matrix deficiency) mutant mice

(6). Targeted inactivation of the mouse cartilage LP gene (Crtl1) also results in death and skeletal

abnormalities characteristic of human chondrodysplasias (27). In Crtl1-/-

mice the lack of LP

significantly reduces the deposition of aggrecan in ECM, and also influences the level of signaling

molecules, such as Indian hedgehog (Ihh) and parathyroid hormone receptor (PTHrP), involved in

chondrocyte differentiation (27). Thus, the mechanism which leads to skeletal developmental failure is

more complex than the loss of a structural element of the ECM.

As the decreased aggrecan level in Crtl1-/-

mice can be attributed to the lack of a stabilizing

link protein function (27), one would expect that normal levels of LP are needed to reconstitute the

aggregate structure, considering that LP and aggrecan are present in 1:1 stochiometry in the HA-


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


aggrecan-LP tertiary complex. In contrast, in our rescue experiments, significantly lower amounts of

LP were sufficient to recover a normal function of cartilage. Mice from rescue group A having only

15-20 % of wild type levels of LP in cartilage showed no differences in safranin O staining, which

indicates the proteoglycan content in tissue sections (Fig. 5B,C), yet these mice developed severe

skeletal deformities. Nevertheless, while the skeletal development was retarded, this very low amount

of LP seemed to be sufficient to form and maintain the macromolecular organization needed for a

certain level of cartilage function, thus preventing the collapse of the respiratory tract for survival.

Similarly, while the LP content was only 56% in the skeletal tissues of mice of rescue group C, when

compared to wild type littermates, the intensity of the safranin O staining in cartilage and the DMMB-

measured sulfated GAG content were highly comparable in the two groups. These ‘rescued’ mice with

approximately 50% of the wild type level of cartilage LP not only survived but grew normally without

skeletal deformities.

These findings may be related to the cartilage growth factor characteristics of LP (21-23,54). A

peptide of 16 amino acids, cleaved from the N-terminal end of cartilage LP can function as a growth

factor and is able to stimulate the expression of aggrecan (22,23) and type II collagen (54). This N-

terminal peptide of LP may be involved in the homeostatic regulation and synthesis of matrix

components during the normal turnover of cartilage ECM (22,23,54). Although these functions of the

LP initially were described in explant cultures of human and porcine cartilages, the highly conservative

sequence of this N-terminal peptide suggests a potential growth factor-like function in mice as well,

which may be critical for skeletal development. This mechanism could, at least partially, explain the

dose-dependent recovery of cartilage ECM in mice of rescue group A, where the relatively low LP

production controlled by the transgene would provide a sufficient source for N-terminal peptide to be

cleaved. This being the case, the growth hormone effect of the N-terminal LP peptide would be

restored, resulting in a more physiologic composition of the cartilage ECM.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


The null mutation of Crtl1 gene is lethal in homozygous mice (27), similar to the functional null

mutation in aggrecan in cmd embryos (6). In both cases, the deaths of newborn mice were attributed to

the collapse of the cartilage-containing upper respiratory tract resulting in respiratory failure. In

addition, in Crtl1-/-

mice a spine deformity is present, with the antero-posterior diameter of the chest

being decreased due to the hyperlordosis of the upper thoracic spine. The chest deformity evidently

compresses the heart and lungs, which may negatively influence the cardio-respiratory functions in

these chondrodystrophic mice.

In rescue group A, the 15-20% levels of LP in cartilage appeared to rescue the respiratory problems

allowing survival of Crtl1 null mice in spite of their dwarfism and relatively severe skeletal

abnormalities. We conclude therefore that restoring aggrecan levels to wild type levels should provide

tracheal cartilage with sufficient mechanical properties to allow normal respiratory function. This

raises the question of whether Crtl1 protein is, in fact, absolutely necessary for normal cartilage

structure. It is conceivable that the rescue effect on cartilage structure by 15-20 % levels of LP may be

due more to the restoration of aggrecan levels than the need for LP to be engaged in the aggregate

structure. It also raises the question of whether cartilage contains other LP-like proteins that may

stabilize HA-aggrecan complexes in the absence of sufficient amounts of Crtl1-encoded LP.

The existence of additional LP-like proteins, even in cartilage, is quite likely. All HA-binding

proteins (LP, aggrecan, versican, brevican, neurocan, CD44 and Tnfip6) have a specific sequence

motif called the “link module” (9,55,56), which is responsible for the HA binding of these molecules.

Searching for sequences resembling Crtl1 and “link modules”, we have identified at least four proteins

structurally related to LP in both the human and mouse genomes. Very unexpectedly, the ‘cartilage’

LP (Crtl1; previously named Lpr1) was paired with versican (57) on chromosome 13, whereas the

cartilage-specific large aggregating proteoglycan (aggrecan)(58) was co-localized with another LP

(Lpr3) on chromosome 7 (T. T. Glant, unpublished). In addition, the brain LP (Bral1) (59,60) was


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


paired with brevican (Bcan) (18) on chromosome 3, and the brain-specific neurocan (Cspg3)(19) with

a fourth LP gene (Lpr4) on mouse chromosome 8. The protein alignments of the four ‘link proteins’

showed 38-49 % identities, and the expression of Lpr3 mRNA was almost as high as the Crtl1 in

cartilage and brain (T. T. Glant, Unpublished). Together, the simultaneous presence and high

expression levels of Crtl1 and Lpr3 may explain the high success rate of rescue experiments even with

low levels of Crtl1 expression. Recently, the presence of these four paralogous gene pairs was also

confirmed in the human genome (61).

Although LP is abundant in cartilaginous ECM, where it simultaneously interacts with aggrecan

and HA, the Crtl1 gene encoded LP was found in a number of tissues other than cartilage. In this

study, we systematically tested many organs for the expression of Crtl1 mRNA and protein during

embryonic development and in aging animals. The ubiquitous presence of LP suggests a common

function for LP in the organization of ECM in a number of tissues, possibly by interacting with other

HA-binding proteoglycans, such as versican, brevican or neurocan. Our results suggest that LP may

have at least two functional roles in cartilage. One function is to provide structural stability of the

cartilage aggregate by interacting with aggrecan and hyaluronan. Formation of stable cartilage

aggregates is essential for the function of the growth plate, but approximately 50 % LP expression

level in a normal growth plate is sufficient for this stability. The 50% expression level is equivalent to

that we could detect in heterozygous LP-null mice where normal phenotype is observed. An

unexpected additional function is to promote cartilage development by increasing levels of

proteoglycans and perhaps other cartilage matrix proteins such as type II collagen. This function,

however, requires less amount of LP with as little as 10-15% expression level compared to the normal

LP level. Since low levels of LP expression are functional, LP may promote chondrocyte

differentiation through mediating the Ihh and PTHrP pathways (62,63).


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


Acknowledgment - Authors would like to thank Dr. Katalin Mikecz, and Dr. László Módis for helpful

suggestions, Dr. Tamás Bárdos, Jodi Gallagher, Sarah J. Rotter and Sonja Velins for technical

assistance.

Footnotes:

1This work was supported in part by the NIH (AR 47135; AR40310).

2Corresponding author:

Dr. Tibor T. Glant

Section of Biochemistry and Molecular Biology Departments of Biochemistry and Orthopedic Surgery

Rush-Presbyterian-St. Luke’s Medical Center

1653 West Congress Parkway

Chicago, IL 60612

Phone: (312) 942-5855

Fax: (312) 942-8828

E-mail: [email protected]

3Abbreviations used: Crtl1/CRTL1, cartilage link protein; HA, hyaluronic acid; ECM, extracellular

matrix; LP, link protein; GAG, glycosaminoglycan; DMMB, dimethyl-methylene-blue; RACE, rapid

amplification of cDNA ends; RT-QT-PCR, real time-quantitative-PCR; Tnfip6, tumor necrosis factor

induced protein 6; Lpr3, link protein 3; Lpr4, link protein 4; Bral1, brain link protein 1; Cspg3,

chondroitin sulfate proteoglycan 3; Bcan, brevican; Ihh, indian hedgehog; PTHrP, parathyroid

hormone related protein.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


REFERENCES

1. Hardingham, T. E. (1979) Biochem.J. 177, 237-247

2. Neame, P. J., Christner, J. E., and Baker, J. R. (1986) J.Biol.Chem. 261, 3519-3535

3. Morgelin, M., Paulsson, M., Hardingham, T. E., Heinegård, D., and Engel, J. (1988) Biochem.J. 253, 175-185

4. Neame, P. J. and Barry, F. P. (1993) J.Cell Biol. 49, 393-402

5. Wight, T. N., Heinegård, D. K., and Hascall, V. C. (1991) Proteoglycans: structure and function. In Hay, E. D.,

editor. Cell Biology of Extracellular Matrix, Plenum Press, New york, London

6. Watanabe, H., Kimata, K., Line, S., Strong, D., Gao, L.-Y., Kozak, C. A., and Yamada, Y. (1994) Nature Genet. 7,

154-157

7. Jacenko, O. and Olsen, B. R. (1995) J.Rheumatol. 22, 39-41

8. Horton, W. A. (1996) Ann.N.Y.Acad Sci 785, 150-159

9. Kohda, D., Morton, C. J., Parkar, A. A., Hatanaka, H., Inagaka, F. M., Campbell, I. D., and Day, A. J. (1996) Cell

86, 767-775

10. Gardell, S., Baker, J., Caterson, B., Heinegård, D., and Rodén, L. (1980) Biochem.Biophys.Res.Commun. 95,

1823-1831

11. Stirpe, N. S., Dickerson, K. T., and Goetinck, P. F. (1990) Dev.Biol. 137, 419-424

12. Binette, F., Cravens, J., Kahoussi, B., Haudenschild, D. R., and Goetinck, P. F. (1994) J.Biol.Chem. 269, 19116-

19122

13. Ripellino, J. A., Margolis, R. U., and Margolis, R. K. (1989) J.Cell Biol. 108, 1899-1907

14. Lohmander, L. S., DeLuca, S., Nilsson, B., Hascall, V. C., Caputo, C. B., Kimura, J. H., and Heinegård, D. (1980)

J.Biol.Chem. 255, 6084-6091

15. Sun, G. W., Kobayashi, H., and Terao, T. (1999) J.Histochem.Cytochem. 47, 1433-1442

16. Sun, G. W., Kobayashi, H., Suzuki, M., Kanayama, N., and Terao, T. (2002) Mol.Reprod.Dev. 63, 223-231

17. Yamauchi, S., Cheng, H., Caterson, B., and Yamauchi, M. (1997) J.Dent.Res. 76, 1730-1736

18. Yamada, H., Watanabe, K., Shimonaka, M., and Yamaguchi, Y. (1994) J.Biol.Chem. 269, 10119-10126

19. Rauch, U., Grimpe, B., Kulbe, G., Arnold-Ammer, I., Beier, D. R., and Fässler, R. (1995) Genomics 28, 405-410

20. Vintiner, G. M., Lo, K. K., Holder, S. E., Winter, R. M., and Malcolm, S. (1993) J.Med.Genet. 30, 773-778

21. Martin, H. and Dean, M. (1993) Eur.J.Biochem. 212, 87-94

22. McKenna, L. A., Liu, H. X., Sansom, P. A., and Dean, M. F. (1998) Arthritis Rheum. 41, 157-162

23. Liu, H., McKenna, L. A., and Dean, M. F. (1999) Biochem.Biophys.Aspects 1428, 191-200


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


24. O'Byrne, E. M., Blancuzzi, V., Wilson, D. E., Wong, M., and Jeng, A. Y. (1990) Arthritis Rheum. 33(7), 1023-

1028

25. Loughlin, J., Irven, C., and Sykes, B. (1994) Hum.Genet. 94, 698-700

26. Hecht, J. T., Blanton, S. H., Wang, Y., Daiger, S. P., Horton, W. A., Rhodes, C., Yamada, Y., and Francomano, C.

A. (1992) Am.J.Med.Genet. 44, 420-424

27. Watanabe, H. and Yamada, Y. (1999) Nature Genet. 21, 225-229

28. Cs-Szabó, G., Melching, L. I., Roughley, P. J., and Glant, T. T. (1997) Arthritis Rheum. 40, 1037-1045

29. Fülöp, C., Kamath, R. V., Li, Y., Otto, J. M., Salustri, A., Olsen, B. R., Glant, T. T., and Hascall, V. C. (1997)

Gene 202, 95-102

30. Roebuck, K. A., Vermes, C., Carpenter, L. R., Fritz, E. A., Ramesh, N., and Glant, T. T. (2001) J.Bone Miner.Res.

16, 501-510

31. Glant, T. T., Kamath, R. V., Bárdos, T., Gál, I., Szánto, S., Murad, Y. M., Sandy, J. D., Mort, J. S., Roughley, P. J.,

and Mikecz, K. (2002) Arthritis Rheum. 46, 2207-2218

32. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular cloning: a laboratory manual, Cold Spring Harbor

Laboratory Press, New York

33. Engelman, D. M., Steitz, T. A., and Goldman, A. (1986) Annu.Rev.Biophys.Biophys.Chem. 15 , 321-353

34. Hopp, T. P. and Woods, K. R. (1981) Proc Natl Acad Sci U.S.A 78, 3824-3828

35. Pierschbacher, M., Hayman, E. G., and Ruoshlati, E. (1983) Proc.Natl.Acad.Sci.USA 80, 1224-1227

36. Roughley, P. J., White, R. J., and Mort, J. S. (1996) Biochem.J. 318, 779-784

37. Glant, T. T., Mikecz, K., Roughley, P. J., Buzás, E., and Poole, A. R. (1986) Biochem.J. 236, 71-75

38. Muller, G. and Hanschke, M. (1996) Connect.Tissue Res. 33, 243-248

39. Hascall, V. C. and Heinegård, D. (1974) J.Biol.Chem. 249, 4232-4241

40. Baker, J. R. and Caterson, B. (1979) J.Biol.Chem. 254, 2387-2393

41. Kimmel, C. A. and Trammell, C. (1981) Stain Technol. 56, 271-273

42. Osborne-Lawrence, S. L., Sinclair, A. K., Hicks, R. C., Lacey, S. W., Eddy Jr., R. L., Byers, M. G., Shows, T. B.,

and Duby, A. D. (1990) Genomics 8, 562-567

43. Dudhia, J. and Hardingham, T. E. (1989) J.Mol.Biol. 206, 749-753

44. Doege, K., Hassell, J. R., Caterson, B., and Yamada, Y. (1986) Proc.Natl.Acad.Sci.USA 83, 3761-3765

45. Hering, T. M., Kollar, J., Huynh, T. D., and Sandell, L. J. (1995) Comp.Biochem.Physiol.B Biochem.Mol.Biol. 112,

197-203


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


46. Deák, F., Kiss, I., Sparks, K. J., Argraves, W. S., Hampikian, G., and Goetinck, P. F. (1986)

Proc.Natl.Acad.Sci.USA 83, 3766-3770

47. Deak, F., Mates, L., Krysan, K., Liu, Z. , Szabo, P. E., Mann, J. R., Beier, D. R., and Kiss, I. (1999) Cytogenet.Cell

Genet. 87, 75-79

48. Dudhia, J., Bayliss, M. T., and Hardingham, T. E. (1994) Biochem.J. 303, 329-333

49. Deák, F., Barta, E., Mestric, S., Biesold, M., and Kiss, I. (1991) Nucleic Acids Res. 19, 4983-4990

50. Rhodes, C., Savagner, P., Line, S., Sasaki, M., Chirigos, M., Doege, K., and Yamada, Y. (1991) Nucleic Acids Res.

19, 1933-1939

51. Kerlavage, A., Bonazzi, V., di Tommaso, M., Lawrence, C., Li, P., Mayberry, F., Mural, R., Nodell, M. , Yandell,

M., Zhang, J., and Thomas, P. (2002) Nucleic Acids Res. 30, 129-136

52. Waterston, R. H., Lindblad-Toh, K., Birney, E., Rogers, J., Abril, J. F., Agarwal, P., et al. (2002) Nature 420, 520-

562

53. Yamada, Y., Miyashita, T., Savagner, P., Horton, W., Brown, K. S., Abramczuk, J., Hou-Xiang, X., Kohno, K.,

Bolander, M., and Bruggeman, L. (1990) Ann.N.Y.Acad.Sci. 580, 81-87

54. Liu, H., McKenna, L. A., and Dean, M. F. (2000) Arch.Biochem.Biophys. 378, 116-122

55. Day, A. J. (1999) Biochem.Soc.Trans. 27, 115-121

56. Lesley, J., English, N. M., Gal, I., Mikecz, K., Day, A. J., and Hyman, R. (2002) J.Biol.Chem. 277, 26600-26608

57. Shinomura, T., Zako, M., Ito, K., Ujita, M., and Kimata, K. (1995) J.Biol.Chem. 270, 10328-10333

58. Walcz, E., Deák, F., Erhardt, P., Coulter, S. N., Fülöp, C., Horváth, P., Doege, K. J., and Glant, T. T. (1994)

Genomics 22, 364-371

59. Hirakawa, S., Oohashi, T., Su, W. D., Yoshioka, H., Murakami, T., Arata, J., and Ninomiya, Y. (2000)

Biochem.Biophys.Res.Commun. 276, 982-989

60. Oohashi, T., Hirakawa, S., Bekku, Y., Rauch, U., Zimmermann, D. R., Su, W. D., Ohtsuka, A., Murakami, T., and

Ninomiya, Y. (2002) Mol.Cell.Neurosci. 19, 43-57

61. Spicer, A.P., Joo, A., Bowling, R.A. (2003) The missing links: A hyaluronan binding link protein gene family whose

members are physically linked adjacent to chrondroitin sulfate proteoglycan core protein genes. J. Biol. Chem. PMID:

12663660

62. Arikawa-Hirasawa, E., Watanabe, H., Takami, H., Hassell, J. R., and Yamada, Y. (1999) Nat.Genet. 23, 354-358

63. Costell, M., Gustafsson, E., Aszodi, A., Morgelin, M., Bloch, W., Hunziker, E., Addicks, K., Timpl, R., and

Fassler, R. (1999) J.Cell Biol. 147, 1109-1122


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


Table I

Primers Used for Cartilage LP Cloning and Analysis

_______________________________________________________________________________________________

Primer Sequence 5' to 3' Direction Location PCR product PCR

size (bp) conditions

_______________________________________________________________________________________________

LP1 GAGTGCGGTGTGGGTGTG Rev Exon 1 Primer Extension.

Primers for RT-PCR and genotyping transgenic mice

LP2 CCTTTCAGACAGCTACACTCC For Exon 2 292 A

LP3 AAACACTCGACCTTGATAGCC Rev Exon 3

LP4 AGGAATCCACAAAATCC For Exon 3 540 B

LP5 CAGAAAACGTCATATCTGC Rev Exon 4

LP6 CTTACCCTGGAGGATTATGG For Exon 3 775 C

LP7 TTGCATGAGTTTCATATTAAA Rev Exon 5

LP8 GTGCAGTACCCAATCACCAAAC For Exon 4 319 D

LP9 GTCGAGAAATAGGATAGCGGACG Rev Exon 5

Primers for real-time PCR

LP10 TGCAGTACCCAATCACCAAACC For Exon 4 87 E

LP11 CTTGTCCCAAAACCCGTAGTTCC Rev Exon 4

GP1 CTACATGTTCCAGTATGACTCCACTCACG For GAPDH 79 F

GP2 GTTGATGACAAGCTTCCCATTCTCG Rev GAPDH

Primers for knockout genotyping

LP12 TAATGACCTTTCCTGTCTCTCC For Intron 3 283 or 1,883 D

LP13 CCCAAAACCCGTAGTTCC Rev Exon 4

PNeo1 GGATCGGCCATTGAACAAG For Neomycin 600 B

PNeo2 CACCATGATATTCGGCAAGC Rev Neomycin

LP14 GTCGGGCAGAGGCAGAGCATTTTGAG For Intron 3 2,200 G

PNeo3 GCTACCGGTGGATGTGGAATGTGTGC Rev Neomycin

_____________________________________________________________________________________________


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


Primers used for analysis of the Crtl1 (Link Protein) gene.

PCR conditions:

A: 94 C for 2 minutes, 30x (94 C for 30 seconds, 59oC for 30 seconds, 72 C for 1 minutes)

+ final extension at 72 C for 5 minutes;

B: 94 C for 2 minutes, 30x (94 C for 30 seconds, 57oC for 30 seconds, 72 C for 1 minutes)


C: 94 C for 5 minutes, 30x (94 C for 30 seconds, 59oC for 1 minutes, 72 C for 1 minutes)


D: 94oC for 2 minutes, 35x (94 C for 15 seconds, 60.5 C for 30 seconds, 72 C for 5 minutes)


E: 94oC for 2 minutes, 35x (94 C for 15 seconds, 58.6 C for 30 seconds, 72 C for 30 seconds);

F: 94oC for 2 minutes, 35x (94 C for 15 seconds, 56.0 C for 15 seconds, 72 C for 20 seconds);

G: 94 C for 3 minutes, 15x (94 C for 15 seconds, 60oC for 30 seconds, 72 C for 5 minutes),

25x (94 C for 15 seconds, 60oC for 30 seconds, 72 C for 5 minutes

+ 20 seconds increment at each cycle) + final extension at 72 C for 10 minutes.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


Table II

Exon/Intron Structure of Cartilage LP Gene (Crtl1)

______________________________________________________________________________

Exon Coding information Splice donor.................... Intron size...... Splice acceptor

size (bp) (bp)

______________________________________________________________________________

1 (1104) 5' Untranslated ACAAGgtaaaacttccatcg...43,122....atgtgatccttacagCAGAG

2 (130) Leader/N-terminus TCAAGgtaagggaacacaag...16,787....cccttctcattatagCAGAA

3 (372) A Loop ACAAGgtaggtatttataaa....3,381....ctctcctttgaatagGTGTG

4 (303) B Loop CAACGgtaagacagcagctg....2,360....tccttctccccacagGCCGA

5 (540) B' Loop/3' Untranslated

_________________________________________________________________

Capital and lower case letters represent exon and intron sequences respectively. The consensus splice

junctions are in boldface and underlined. The Crtl1 gene spans 68,099 bp and resides on chromosome

13.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


Table III

Survival Rates of Rescued Mice at Eight Weeks of Age

Parents Generation Genotypes Mendelian Expected Observed survival

ratio survival ratio ratios

(%) (%) line A line C

Crtl1+/- X Crtl1Tg+/+ N1 All genotypes 100 100 100 (n=67) 100 (n=45)

Crtl1+/+Crtl1Tg+/- 50 50 53.7 (n=36) 46.7 (n=21)

Crtl1+/-Crtl1Tg+/- 50 50 46.3 (n=31) 53.3 (n=24)

Crtl1+/-Crtl1Tg+/- N2 All genotypes 100 100 100 (n=71) 100 (n=118)

Crtl1-/-Crtl1Tg- 6.25 0 0 0

Crtl1-/-Crtl1Tg+/- or +/+ 18.75 20 4.2 (n=3) 14.4 (n=17)

Crtl1+/- or -/-Crtl1Tg-/- or +/- or +/+ 75 80 95.8 (n=68) 85.6 (n=101)

Crtl1+/-Crtl1Tg+/- or +/+ N3 All genotypes 100 100 100 (n=77) 100 (n=52)

Crtl1-/-Crtl1Tg- 2.25 0 0 0

Crtl1-/-Crtl1Tg+/- or +/+ 22.75 29.3 12.9 (n=10) 21.1 (n=11)

Crtl1+/- or -/-Crtl1Tg-/- or +/- or +/+ 75 70.7 87.1 (n=67) 78.9 (n=41)

Ratios are expressed in percentage of the total number of mice.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


FIGURE LEGENDS

FIG. 1. Schematic representation of the murine LP (Crtl1) gene (A), the transcription site at

-365 bp position (B), transgenic construct (C), and PCR samples of genotyping results (D,E).

Panel A indicates the assembled Crtl1 gene structure. The LP gene is shown as a continuous box with

shaded ends (non-coding sequences) and a white center (coding sequence). The scale in Kbp is given

above the sequenced cDNA, whereas the locations of introns are indicated by the designation of IVS1,

IVS2, etc. Important landmarks are identified above the gene box. START indicates the transcriptional

start site with the corresponding primer extension reaction shown in panel B. The lane labeled RT

denotes the reverse transcription reaction, while the arrow indicates the predominant product. Lanes

labeled A, C, G, T indicate the corresponding dideoxynucleotide used in sequence analysis. The

asterisk indicates the nucleotide that corresponds to the RT product, thus indicating the initiation site

for transcription. ATG marks the translational start site, TGA indicates the termination codon, and

AATAAA indicates the first polyadenylation signal, which was incorporated into the transgenic

construct. ‘Neo’ in box shows the position in exon 4 (at the BsgI site) where the NeoR casette was

inserted to disrupt the LP gene (27), just before the HA-binding motif of the B loop (Table 2).

Schematic structure of the LP transgenic construct is shown in panel C. The transgenic construct used

for driving cartilage-specific over-expression consists of: (i) a 1.8 Kb type II collagen promoter,

followed by (ii) a multiple cloning site for insertion of the LP transgene, (iii) a 0.5 Kb fragment

consisting of the SV40 promoter and polyadenylation site, and finally, (iv) a 1.5 kb type II collagen

enhancer from the 5' end of intron I. Panel D summarizes the PCR genotyping results. Lane 1 shows

the presence of the 540 bp transgenic construct using LP4 and LP5 primers (Table 1). Lane 2 is the

product of the wild type gene with LP12 (in intron 3) and LP13 (in exon 4) primers (Table 1). Using

the same primers (LP12/LP13), Lane 3 shows the PCR results of a heterozygous Crtl1 knockout with

(1883 bp) or without (283 bp) the Neo-gene. Lane 4 presents a homozygous mouse with Neo-disrupted


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


Crtl1 gene. Lane 5 shows the genomic PCR results of a mouse with targeted disruption of the Crtl1

gene (homozygous for wild type Crtl1 gene), but also carrying the transgene (540 bp). This mouse

survived with normal growth, and exhibited a normal skeleton structure. Panel E shows 319 bp RT-

PCR products of the transgenic transcript in various organs of a Crtl1-/-

Crtl1TgC+

E20.5 fetus using

primer combinations of P8 and P9 (Table 1). Lane “+/+” represents positive wild type (Crtl1-/-

) and

“-/-“ negative (Crtl1-/-

) controls. Lane numberings of Panel E are: lane 1: skeletal tissue; lane 2 brain;

lane 3: liver; lane 4: lungs; lane 5: heart; lane 6: kidney; lane 7: spleen; lane 8: intestines.

FIG. 2. Mouse Crtl1 gene expression and immunolocalization of cartilage LP in wild type mouse

organs of different ages. Panel A summarizes the results of Crtl1 expression in different organs at

different ages as indicated. RT-PCRs using primer pair of LP2 and LP3 (Table 1) were performed as

described in methods. Total RNA was isolated from the whole 10.5-day old embryo E10.5, whereas

the numbers above the other RT-PCR products represent the following organs: lane 1: skeletal tissue;

lane 2 brain; lane 3: liver; lane 4: lungs; lane 5: heart; lane 6: kidney; lane 7: spleen; lane 8: intestines;

lane 9: testis; lane 10: eye; lane 11: ovarium and lane 12: uterus. Under the RT-PCR panel of E20.5, a

Northern dot-blot shows the Crtl1 expression in the corresponding organs using 2 g of mRNA from

each organ, except the first dot having only 0.5 µg mRNA from skeletal tissue. Quantitative

differences measured by Real Time PCR (RT-QT-PCR) and the expression level of Crtl1 gene in

different organs were first normalized to GAPDH, and then to the expression level in skeletal tissue

shown on lane 1. Results of newborn (2-days-old) mice are shown. Western blotting under the PCR

panel of adult samples shows the presence of LP in various organs. LP was purified by cesium chloride

gradient centrifugation as described in Methods. A total of 10 µg protein of the top 1/6th

fraction of the

dissociative gradient (A1A1D6) was loaded on each lane and stained with rabbit LPro1-R18 antibody.

Immunohistochemistry panels (B-G) show the presence of LP in frozen sections of an E13.5 mouse


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


embryo (D-F), an E19.5 fetal kidney (G) and in wrist of a newborn mouse (B and C), all stained with

rabbit polyclonal anti-LPro1-R11 antibody, except panel C (wrist) where the preimmune serum of the

same rabbit was used as a negative control. Panels show LP accumulation in the cartilaginous carpal

bones (B), primordial cartilaginous vertebrae and baso-occipital bone (D), the dermis of the skin (E),

the aorta (F) and fetal kidney (G). Bars: B-D, 200 µm; E-G, 50 µm.

FIG. 3. Quantitative analysis of genotyping and phenotyping results of newborn (2-days-old)

mice with various genetic combinations used in this study. Panel A summarizes results of Real-time

PCR (RT-QT-PCR) (n = 9; 3 mice from 3 different litters), and Panel C shows the quantitative results

of Western blot analysis (n = 9-11 mice of each column from 4-5 independent litters). In both cases,

expression levels are compared to the mRNA and protein expressions measured in wild type mice (first

column), and these are indicated by white numbers inside each column. Panel B is a representative

Western blot stained with LPro1-R11 rabbit antibody. Columns/lane 1: wild type (Crtl1+/+

); lane 2:

heterozygous LP-knockout (Crtl1+/-

); lane 3: homozygous LP-deficient (Crtl-/-

); lane 4: homozygous

LP transgenic (Crtl1+/+

Crtl1TgA+/+

) mice; lane 5: homozygous LP transgenic Crtl1+/+

Crtl1TgC+/+

; lane

6: Crtl1-/-

mice rescued with LP transgene (Crtl1-/-

Crtl1TgA+/- or +/+

), and lane 7: rescued Crtl1-/-

Crtl1TgC+/- or +/+

mice. Asterisks above columns (panel C) indicate significant differences (*p<0.01) in

Crtl expressions measured in wild type samples (first column) relative to other samples. There are also

significant differences between rescue groups Crtl1-/-

Crtl1TgA+/+

(column 4) and Crtl1-/-

Crtl1TgC+/+

(column 5).

FIG. 4. Wild type and LP-deficient (Crtl1-/-

) mice rescued with LP-transgene as either ‘Rescue

Group A’: Crtl1-/-

CrtlTgA+

or ‘Rescue Group C’: Crtl1-/-

Crtl1TgC+

of different ages. In each paired

panel, animal in the upper position is a normal wild type compared to the corresponding rescued litter-


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


mate of the F2 generation. Gross phenotype are shown on the left side panels of both sets (A-H),

whereas skeletons stained with either alcian blue/alizarin red (a, b, e and f) or shown as whole body

radiographs (c, d, g and h) are on the right side panels. Newborn (Aa and Ee; less than 1-day old), 4-

days old (Bb and Ff), three-weeks old (Cc and Gg), and adult mice (Dd and Hh) are shown. Note the

shortened snout, fore- and hind limbs, and dorsal kyphosis of newborn mice in Rescue Group A (Aa-

Dd), in contrast to the normal appearance of newborn rescued mice in Rescue Group C (Ee-Hh). The

skeletal growth of Rescued Group A is delayed, the long bones are shorter, the skull is dome-like and

there is a significant hyperlordotic curvature in the upper and a hyperkyphotic curvature in the lower

thoracic spine (Cc-Dd). Dwarfism is becoming even more apparent with time in Rescue Group A (Bb-

Dd), while the body size of mice in Rescue Group C remains comparable to the age-matched wild type

controls (Ff-Hh).

FIG. 5. Radiographs (A), safranin O/fast green-stained micrographs of knee joints of 2-weeks-

old (B and C), and LPro1-R11 antibody-immunostained metacarpal bones of newborn (D) wild

type and rescued Crtl1-/-

Crtl1TgA+

and Crtl-/-

Crtl1TgC+

transgenic mice. Arrows are pointing to the

epiphyseal ossification center of the distal femur and proximal tibia (A and B). Bars: B, 1 mm; C, 200

µm; D, 500 µm.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

20001500

750500

250

1883

540

283

1 2 3 4 5D

*

RTACGT

B

((

2

( ()

)

(Sn

a

RB

Sm

aI

Bam

HI

Xba

I

Eco

RI

Xba

I

Hin

dII

I

Hin

dII

I

SP6Promoter

1.8 Kb

Sp

eI

Sst

IS

al

I

No

tI

Ba

mH

I)X

ba

I

1.5 Kb0.5 KbA

at

IS

ph

IM

luI

Sp

lI I)

Kp

nI

Rsr

II

Pst

I

Eco

I)K

pn

I

Sse

8387

I

Transgene

SP/44-3pPoly A EnhancerSV40L PCol2A1 Prom.

C

A

‘Neo’

Kbp

1 1.50.5 2 2.5

IVS 1 IVS 2

Coding 3' Flanking

IVS 3 IVS4

ATG TGA AATAAASTART

5' Flanking

1 2 3 4 5 6 7 8

500

250319

+/++/+ -/--/-E

Czipri . Fig. 1et al


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Czipri . Fig 2et al

E

GF

CB

D

A1

0.2

9

0.0

9

0.4

8

0.2

8

0.2

2

0.0

7

0.1

3New-born

RT-QT-PCR(Relative to the expression in skeletal tissue)

1 2 3 4 5 6 7 8

E10.5 E16.5 1 2 3 5

Northern

1 2 3 4 5 6 7 8E20.5

Adult

46 Kda

1 2 3 4 5 6 7 8 9 10 11 12


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Czipri . Fig. 3et al

46

kDa

1 2 3 4 5 6 7

B

Western blotC

1

0.560

0.140.54

1.48

2.53

0

0.5

1

1.5

2

2.5

3

1 2 3 4 5 6 7genotypes

rela

tive

pro

tein

exp

ressio

n

RT-QT-PCRA

10.78

0.460

3.26

0.97

0.39

0

0.5

1

1.5

2

2.5

3

3.5

1 2 3 4 5 6 7genotypes

rela

tive

mR

NA

exp

ressio

n


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Czipri Fig.4.et al.

Rescue Group A ( )Crtl1 Crtl1-/- TgA+

Rescue Group C ( )Crtl1 Crtl1-/- TgC+

rescue

WT

rescue

WT

rescue

WT

rescue

WT

A

B

C

D

E

F

G

H

a e

b f

rescue

WTWT

rescue

c g

rescue

WTWT

rescue

d h


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Czipri Fig. 5.et al.

Wild type Rescue Group A( )Crtl1 Crtl1

-/- TgA+

Rescue Group C( )Crtl1 Crtl1

-/- TgC+

A

B

C

D


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Yamada, Bjorn R. Olsen and Tibor T. GlantGabor Firneisz, Kevin J. Kolman, Hideto Watanabe, Yefu Li, Peter J. Roughley, Yoshihiko

Matyas Czipri, Jeffrey M. Otto, Gabriella Cs-Szabo, Rajesh V. Kamath, Csaba Vermes,protein-deficiency

Genetic rescue of chondrodysplasia and the perinatal lethal effect of cartilage link

published online May 5, 2003J. Biol. Chem.

10.1074/jbc.M303329200Access the most updated version of this article at doi:

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/lookup/doi/10.1074/jbc.M303329200

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;M303329200v1&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/early/2003/05/05/jbc.M303329200.citation

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=early/2003/05/05/jbc&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/early/2003/05/05/jbc.M303329200.citation

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/

Genetic Rescue of Chondrodysplasia and the Perinatal Lethal Effect ...

Documents

Transcript of Genetic Rescue of Chondrodysplasia and the Perinatal Lethal Effect ...