BIOLOGY OF REPRODUCTION 84, 1053–1062 (2011)Published online before print 19 January 2011.DOI 10.1095/biolreprod.110.089599

Cervical Softening During Pregnancy: Regulated Changes in Collagen Cross-Linkingand Composition of Matricellular Proteins in the Mouse1

Meredith L. Akins,3,4 Katherine Luby-Phelps,5 Ruud A. Bank,6 and Mala Mahendroo2,3,4

Department of Obstetrics and Gynecology,3 The Cecil H. and Ida Green Center for Reproductive Biology Sciences,4

and Department of Cell Biology,5 University of Texas Southwestern Medical Center, Dallas, TexasDepartment of Medical Biology,6 University Medical Center Groningen, Groningen, The Netherlands

ABSTRACT

A greater understanding of the parturition process is essentialin the prevention of preterm birth, which occurs in 12.7% ofinfants born in the United States annually. Cervical remodelingis a critical component of this process. Beginning early inpregnancy, remodeling requires cumulative, progressive changesin the cervical extracellular matrix (ECM) that result inreorganization of collagen fibril structure with a gradual lossof tensile strength. In the current study, we undertook a detailedbiochemical analysis of factors in the cervix that modulatecollagen structure during early mouse pregnancy, includingexpression of proteins involved in processing of procollagen,assembly of collagen fibrils, cross-link formation, and depositionof collagen in the ECM. Changes in these factors correlated withchanges in the types of collagen cross-links formed and packingof collagen fibrils as measured by electron microscopy. Early inpregnancy there is a decline in expression of two matricellularproteins, thrombospondin 2 and tenascin C, as well as a declinein expression of lysyl hydroxylase, which is involved in cross-linkformation. These changes are accompanied by a decline in bothHP and LP cross-links by gestation Days 12 and 14, respectively,as well as a progressive increase in collagen fibril diameter. Incontrast, collagen abundance remains constant over the courseof pregnancy. We conclude that early changes in tensile strengthduring cervical softening result in part from changes in thenumber and type of collagen cross-links and are associated witha decline in expression of two matricellular proteins thrombo-spondin 2 and tenascin C.

cervix, cervical remodeling, cervical softening, collagen cross-linking, parturition

INTRODUCTION

The composition and structure of the cervical extracellularmatrix (ECM) regulates the ability of the cervix to remainclosed and firm during pregnancy and to open and distend atthe time of parturition. Greater understanding of the molecularchanges within the cervical ECM during pregnancy andparturition will enhance our understanding of this criticalphysiological process as well as identify mechanismsunderlying aberrant remodeling that accompany prematurebirth. The cervical ECM is secreted by the epithelia andfibroblasts within the cervical tissue and is comprised of five

major components: fibrillar collagens, elastin fibers, proteo-glycans, hyaluronan, and matricellular proteins [1, 2]. Earlybiochemical modifications result in palpable changes in tissuecompliance by the first trimester of pregnancy in women. Inmouse studies, quantifiable changes in tissue biomechanicsare detectable by mid-pregnancy [2, 3]. This early phase oftissue remodeling, termed cervical softening, is characterizedby an increased percentage of soluble collagen and quanti-fiable changes in collagen fiber microstructure while collagencontent remains constant [3, 4]. Molecular processes thataccount for changes in collagen solubility and microstructureduring cervical softening are not well characterized and arethe focus of this study.

Fibrillar collagen types I and III are the main structuralproteins of the cervix [1]. While the collagen family includesboth fibrillar and nonfibrillar collagens, fibrillar collagens arethe primary source of the tensile strength of tissue [5]. Theload-bearing capacity of a tissue is in part determined bycollagen type and abundance, posttranslational processing ofcollagen, assembly of collagen into fibrils and fibers,deposition of collagen in the ECM, and collagen degradation[6]. Collagen synthesis is complex. Fibrillar collagen issynthesized in the endoplasmic reticulum (ER) in a pro-formthat is folded and assembled into a triple helix, initiated at the Cterminus, with the aid of chaperone proteins [7, 8]. While stillin the ER, lysine residues in the procollagen chain arehydroxylated by the enzyme lysyl hydroxylase. Once theprocollagen trimer is secreted into the extracellular space, theN- and C-terminal propeptides are cleaved, and collagenspontaneously aggregates into fibrils, which subsequentlyassemble into fibers [9, 10]. Assembly of fibrils from maturecollagen molecules involves intermolecular cross-links be-tween lysine residues on adjacent collagen molecules. Somelysine residues are hydroxylated in the ER by the enzyme lysylhydroxylase, and the degree of hydroxylation determines thetype of cross-link formed. In turn, the type of cross-link formeddetermines strength and mechanical stability of the resultingtissue [11, 12]. Collagen cross-linking is catalyzed by lysyloxidase (LOX) and occurs between hydroxylated or non-hydroxylated lysine residues in collagen, resulting in pyridino-line or nonpyridinoline cross-links, respectively.

The importance of regulated changes in collagen processingduring parturition is supported by numerous reports thatpreterm birth due to cervical insufficiency and pretermpremature rupture of membranes are increased in women withinherited defects in collagen and elastin synthesis or assembly(e.g., Ehlers-Danlos and Marfan syndromes) [13]. Recentstudies suggest polymorphisms associated with genes impor-tant for connective tissue synthesis and metabolism maypredispose women to preterm birth due to cervical incompe-tence and preterm premature rupture of membranes [13].Previous studies report that activity of the cross-link-forming

1Supported by National Institutes of Health grant R01 HD043154.2Correspondence: FAX: 214 648 9242;e-mail: mala.mahendroo@utsouthwestern.edu

Received: 5 November 2010.First decision: 22 November 2010.Accepted: 10 January 2011.� 2011 by the Society for the Study of Reproduction, Inc.eISSN: 1529-7268 http://www.biolreprod.orgISSN: 0006-3363

1053

Dow

nloaded from w

ww

.biolreprod.org.

enzyme, LOX, is reduced in the cervix during mousepregnancy, and Lox gene expression is regulated in thepregnant mouse cervix [14, 15]. These data suggest a declinein the number of collagen cross-links in the mouse cervix overthe course of pregnancy.

In addition to collagen synthesis and cross-link formation,the type of fibrillar collagen can affect ECM composition andstrength. Studies in collagen III-deficient mice indicate thatcollagen III is essential for normal collagen I fibrillogenesisand that changes in the ratio of collagen I:III can alter themechanical properties of tissue [16, 17]. The ratio of type I-to-III collagen in the human nonpregnant cervix is reported to be70% and 30%, respectively, but it has not been determined if

changes in this ratio contribute to changes in cervical tissuecompliance during pregnancy [18].

ECM architecture and overall tissue strength are determinedby a network of interactions among collagens, proteoglycans,and matricellular proteins. Noncollagenous proteins in thecervical ECM, such as proteoglycans and matricellularproteins, can affect matrix organization and consequentlytissue strength. Proteoglycans organize the ECM by controllingthe size and packing of collagen fibrils [19]. The glycosami-noglycan chains that attach to the protein core can affect thehydration or equilibrium tension of the tissue. Severalproteoglycans are known to be expressed in the pregnantcervix, including versican, decorin, biglycan, asporin, andfibromodulin [3, 20–23]. Matricellular proteins, such as

FIG. 1. Cervical collagen Ia1 and IIIa1levels remain constant during pregnancy. A)Collagen Ia1 (Col1a1) and (B) collagen IIIa1(Col3a1) mRNA levels were measured viaQPCR using cervices at NP metestrus andDay 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, and18, and 2 h postpartum and 24 h postpar-tum and normalized to Day 18. For NP, n ¼15 cervices; for pregnant and postpartum, n¼ 4–6 cervices per time point. *, Signifi-cance at P , 0.05. C) The amount ofcollagen in the cervix was measuredthrough hydroxyproline content. Hydroxy-proline assay was carried out in NP estrus,gestation Day 5, 8, 10, 12, 14, 16, and 18,2-h postpartum, 10-h postpartum, and 24-hpostpartum cervices and normalized to dryweight; n ¼ 5 cervices per time point. D)Dot blot analysis of collagen I and collagenIII shows soluble protein levels at NP estrusand gestation Days 6, 8, 10, 12, 15, and 18.E and F) Optical density analysis is shown of(E) collagen I and (F) collagen III dot blot; n¼ 3 cervices per time point. G and H)Immunofluorescence of (G) collagen I and(H) collagen III is shown in gestation Day 6,12, 15, and 18 cervical sections.

1054 AKINS ET AL.

Dow

nloaded from w

ww

.biolreprod.org.

tenascins, thrombospondins, and the secreted protein acidic andrich in cysteine (SPARC) protein, are not structural proteinswithin the ECM, but they modulate the functions of structuralproteins such as collagen as well as cell functions throughinteractions with cell surface receptors, proteases, and growthfactors [24]. Targeted gene loss of specific matricellularproteins results in aberrant matrix organization and remodeling[25–27]. In this study, we assess changes in relative abundanceof fibril collagens I and III, collagen processing, collagen cross-links, proteoglycans, matricellular proteins, and collagenultrastructure in order to identify early molecular eventsleading to increased cervical compliance and collagensolubility required for successful birth.

MATERIALS AND METHODS

Animal and Tissue Collection

Nonpregnant (NP) mice. To assess stage of cycle, vaginal smears weretaken and analyzed for characteristic cell structure and immune cell presencespecific for each phase of the cycle [28]. Mice were observed through at leastone full cycle in order to insure proper cycling prior to tissue collection. In thisstudy the term ‘‘NP’’ refers to data pooled from an equal number of cervicescollected in proestrus, estrus, and metestrus. ‘‘NP estrus’’ or ‘‘NP metestrus’’refers to cervices assessed at a single stage of the estrus cycle.

Pregnant and postpartum mice. Females were housed overnight withmales and separated the following morning. Vaginal plugs were checked at thetime of separation; mice with plugs were considered to be at Day 0 of theirpregnancy, with birth occurring early on gestation Day 19. In general, cerviceswere collected at noon for all time points from gestation Day 8–18, except forDay 18, the day on which cervices were collected between 1800 and 1900 h inorder to collect cervices a few hours prior to onset of labor. Postpartum cervicescollected after vaginal birth on gestation Day 19 were obtained exactly 2 or 4 hafter delivery of the first pup or approximately 10–12 h, 24 h, or 48 h afterdelivery of the first pup.

Mice used for these studies were of C57B6/129Sv mixed strain. Cervicaltissue was dissected from the reproductive tract, and vaginal tissue wascarefully removed from cervical tissue. Cervices were flash-frozen in liquidnitrogen immediately following their extraction. All studies were conducted inaccordance with the standards of humane animal care described in the NationalInstitutes of Health Guide for the Care and Use of Laboratory Animals, using

protocols approved by an institutional animal care and research advisorycommittee.

Quantitative Real-Time PCR (QPCR)

Total RNA was extracted from frozen cervical tissue in RNA Stat 60 (Tel-Test Inc., Friendswood, TX). RNA was then treated with DNase (New EnglandBiolabs, lpswich, MA) to remove genomic contaminants. cDNA wassynthesized, and RT-PCR was carried out using the SYBR Green detectionsystem (Applied BioSystems, Carlsbad, CA) or using the TaqMan Probessystem (Applied BioSystems). All gene expression is relative to gestation Day18 levels.

Immunofluorescence

Cervical tissue was collected on Days 6, 12, 15, and 18 of gestation andfrozen in OCT embedding compound medium (Sakura Finetek USA, Torrance,CA). Five-micrometer cervical sections were cut from tissue blocks. Sectionswere fixed for 10 min in acetone at �208C. Tissue was rehydrated in PBS,blocked in 10% normal goat serum (catalog no. 01–6201; Invitrogen, Carlsbad,CA), and incubated with collagen I or collagen III rabbit polyclonal antibodies(codes ab34710 and ab7778; Abcam, Cambridge, UK). Sections were thenwashed in PBS and incubated with goat a-rabbit immunoglobulin G antibodiescoupled to Alexa 488 (product no. A11008; Molecular Probes, Invitrogen).Slides were viewed on a Zeiss LSM510 Meta NLO confocal microscope usingan Acroplan 403/0.8 W objective lens. Fluorescence signal intensity wasmeasured using ImageJ 1.41k software (http://rsbweb.nih.gov/ij/).

Immunoblotting

Collagen was extracted with 7 M urea (Sigma, St. Louis, MO), 0.1 Msodium phosphate with 1% protease inhibitor (Sigma), overnight at 48C. Theprotein concentration was determined by a Bradford protein assay (Pierce;Thermo Scientific, Rockford, IL). Ten micrograms of protein were loaded on a4%–20% Tris-HCl polyacrylamide gel and electrophoresed at 100 V. Afterovernight transfer to nitrocellulose membrane and Ponceu S (Sigma) staining toassess equal loading of protein, immunoblotting was performed using rabbitpolyclonal anti-mouse collagen I (catalog no. MD20151; MD Biosciences, St.Paul, MN).

Protein used in dot blot analysis was extracted as described forimmunoblotting. Two and one-half micrograms of protein were spotted ontoa nitrocellulose membrane and probed with rabbit polyclonal anti-collagen I or

FIG. 2. mRNA expression of collagenprocessing proteins is shown. A and B)mRNA expression of C-propeptide collagenprocessing enzymes Bmp1 (A) and Tll1 (B)were evaluated at NP, gestation Day 8, 9,10, 11, 12, 13, 14, 15, 16, 17, and 18, and 2h postpartum and 24-h postpartum andnormalized to Day 18. For NP, n ¼ 15cervices; for pregnant and postpartum, n ¼4–6 cervices per time point. C) C-propep-tide was evaluated via Western blotting inNP, gestation Day 6, 8, 9, 10, 11, 12, 13,14, 15, 16, 17, and 18, and 4-h postpartum,12-h postpartum, 24-h postpartum, and 48-h postpartum cervices. A 30-kDa bandcorresponding to the collagen C-propeptideis present. D) The N-propeptide collagenprocessing enzyme Adamts2 mRNA ex-pression was evaluated at NP, and gestationDay 10, 12, 15, and 18, and 24-h postpar-tum and normalized to Day 18. For NP, n ¼15 cervices; for pregnant and postpartum, n¼ 4–5 cervices per time point. E) The of N-propeptide collagen processing enzymeAdamts14 mRNA expression was evaluatedat the same time points as described in thelegend to panel D. For NP, n ¼ 15 cervices;for pregnant and postpartum, n ¼ 4–5cervices per time point. *, Significance at P, 0.05.

CHANGES IN CERVICAL COLLAGEN DURING PREGNANCY 1055

Dow

nloaded from w

ww

.biolreprod.org.

rabbit polyclonal anti-collagen III primary antibody (codes ab34710 andab7778; Abcam). Secondary antibody for both Western and dot blots wasdonkey anti-rabbit horseradish peroxidase (catalog no. 711036152; JacksonImmunoResearch, Westgrove, PA). Chemiluminescence was visualized usingECL (GE Healthcare, Buchinghamshire, UK). Digital images of the blots wereanalyzed quantitatively using Multi Gauge software (Fuji Film, Tokyo, Japan).

Collagen Content and Cross-Links

Frozen cervical tissue was lyophilized, weighed, and then hydrolyzed in 6M HCl at 1008C for 20–22 h. Collagen cross-links were measured via HPLC asdescribed previously [29, 30]. In brief, cervical samples were lyophilized andthen hydrolyzed in 6 M HCl at 1008C for 20–22 h. The samples were then driedand dissolved into 10 lM pyridoxine and 2.40 mM homo-arginine in water.Samples were then diluted 1:4 in 0.5% heptafluorobutyric acid in 10%acetonitrile. Cross-links were analyzed by reversed-phase HPLC and calculatedbased on internal pyridoxine standards. Hydroxyproline assays were carried outas previously described [31]. Total collagen was calculated using a ratio of 300hydroxyproline residues per triple helix [30].

Electron Microscopy

Nonpregnant mice in metestrus or pregnant mice at gestation Days 6, 12,15, and 18 were perfused with 1% glutaraldehyde, 2% paraformaldehydefixative in 0.1 M phosphate buffer. Cervical and uterine tissue was removed

and fixed in 2.5% glutaraldehyde in 0.1 M cacodylate buffer containing 0.1%ruthenium red overnight at 48C. Tissue was then cut into 250 lm longitudinalsections using a Vibratome Series 3000 sectioning system (Vibratome Co., St.Louis, MO). Vaginal and uterine tissue was removed from the cervix, and thecervix was rinsed for 30 min with cacodylate buffer. The tissue was postfixedwith 1% osmium in 0.1 M cacodylate buffer containing 0.1% ruthenium red for90 min. Specimens were dehydrated through a graded series of ethanols (50%,70%, 95%, and 100%) followed by propylene oxide and embedded in Epon(EMbed-812; Electron Microscopy Sciences, Hatfield, PA). Thin sections werestained with uranyl acetate and lead citrate. Sections were viewed bytransmission electron microscopy (TEM) with a Tecnai microscope (FEICompany, Hillsboro, OR), and images were captured at 42003 or 20 5003magnification with a Morada 11 1-megapixel charge-coupled device camera.

Collagen Fibril Measurements

Electron microscope images of transversely sectioned collagen fibrilstaken at 20 5003 magnification were analyzed with Image J 1.41k software(http://rsb.info.nih.gov/ij/). An intensity threshold was interactively deter-mined for optimal segmentation of fibril from background. Segmentedimages were converted to a binary mask, and erode, open, dilation, and fillholes binary operations were used to separate merged fibrils. Particles wereanalyzed using the Particle Analysis function of Image J, with parametersset for size, 1000–10 000 nm2, and circularity, 0–1. Ellipses were fit to allparticles, and the minor angle of the ellipse was taken as the fibril diameter.

FIG. 3. Decline in Plod2 mRNA expression correlates to a reduction in pyridinoline cross-links during pregnancy and an increase in soluble collagen. A)mRNA levels of the Plod family were evaluated in NP, gestation Days 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, and 18 and 2 h postpartum and 24 h postpartumand normalized to Day 18. For NP, n¼15 cervices; for pregnant and postpartum, n¼4–6 cervices per time point. B) HPLC analysis of HP cross-links in NPestrus, gestation Day 5, 8, 10, 12, 14, 16, and 18 and 2-h postpartum, 10-h postpartum, and 24-h postpartum cervices; n¼5 cervices per time point. C) LPcross-links were evaluated in the same time-points shown in panel B; n¼ 5 cervices per time-point. D) Extractable collagen was assessed in NP, gestationDay 6, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, and 18 and 4-h postpartum, 12-h postpartum, 24-h postpartum, and 48-h postpartum cervices. Blots wereprobed with rabbit anti-collagen I. Bands are present at ;250 kDa and 140 kDa (indicated as Kd in figure), representing procollagen and mature collagen,respectively. An unidentified 50-kDa band is also present in NP and postpartum tissues. *, Significance at P , 0.05.

1056 AKINS ET AL.

Dow

nloaded from w

ww

.biolreprod.org.

Ten images from three animals were analyzed for each time point, resultingin n ¼ 8197 fibrils/Day 6, 12 958 fibrils/Day 12, 8362 fibrils/Day 15, and12 161 fibrils/Day 18. Due to low contrast ratio of fibrils versusbackground in NP samples, 516 fibrils were taken from three animals,one image each.

Statistics

Statistics were performed using Prism software version 5.0b (GraphPadSoftware, La Jolla, CA). For comparison of QPCR, collagen content, andcollagen cross-links data, one-way ANOVA was used, followed by a Dunnetcomparison test using NP or Day 8 as the control. Fibril diameter data wereanalyzed using a one-way ANOVA followed by a Dunn multiple comparisontest.

RESULTS

Expression of Type I and III Fibril Collagen

To observe whether changes in the relative amounts of type Iand III fibril collagens could account for the progressive declinein tissue compliance through gestation, mRNA levels for thealpha-1 chain of collagen I and collagen III were determined byQPCR. Collagen I alpha-1 mRNA levels in the first half ofpregnancy were similar to NP but significantly increased in thelatter days of gestation (P , 0.01). Levels declined to NP levelsby 24 h postpartum (Fig. 1A). Collagen III alpha-1 mRNA levelswere similar to NP throughout gestation. While there was atransient and significant increase in mRNA levels 2 hpostpartum, levels quickly declined 24 h after delivery (Fig.1B). Total cervical collagen protein was estimated by hydroxy-proline assay of acid-hydrolyzed tissue. Collagen levelsthroughout pregnancy were found to be at similar or elevatedlevels compared to NP estrus cervix (Fig. 1C). Collagen levelswere increased significantly compared to NP estrus cervix ongestation Days 5, 8, 16, and 18 and 24 h postpartum whennormalized to dry weight (P , 0.0001) (Fig. 1C).

Two methods were used to estimate the relative abundanceand ratio of extractable type I and type III fibril collagen. Dotblots were performed on cervical tissue homogenates extracted

in 7 M urea and spotted on a nitrocellulose membrane at equalprotein loads (Fig. 1D). Conventional Western immunoblottingcould not be used for this assay because the antibodies availableto distinguish type I from type III do not work on denaturedsamples. Extractable collagen I increased approximately twofoldfrom NP estrus to early pregnancy (Day 6), and levels remainedconstant for the remainder of pregnancy (Fig. 1E), while therewere no significant differences in the extractable collagen IIIduring gestation compared to NP estrus (Fig. 1F). Immunoflu-orescent staining for collagen I and III on cervical tissue showedno change in signal intensity throughout gestation. Overall signalintensity for collagen I was greater than collagen III duringpregnancy. The collagen I:collagen III ratio was a factor of 2.4 atall time points (Fig. 1, G and H). Taken together, these resultsindicate that relative to the nonpregnant cervix, collagen levelsare constant or elevated during pregnancy and that the relativeabundance of fibrillar type I-to-III collagen remains unchangedthroughout pregnancy.

Intracellular trafficking and folding of procollagen chainsrequire interaction with ER chaperone proteins, such as heatshock protein 47 (Hsp47, official symbol SERPINH1) andprotein disulfide isomerase (PDI), as targeted deletion of thesegenes results in loss of tissue collagen [32]. Consistent withconstant or elevated collagen content through pregnancy (asseen in Fig. 1), we observed that protein expression ofSERPINH1 and PDI in pregnant and postpartum cervixappears similar or slightly elevated during pregnancy (Supple-mental Fig. S1, available online at www.biolreprod.org).

Collagen Processing

Once secreted from the cell, C- or N-terminal propeptidesare cleaved to form mature collagen. Collagen molecules inwhich the C-propeptide is not cleaved are unable to formfibrils, while the inability to cleave the N-propeptide results information of irregularly shaped fibrils [33, 34]. Both scenariosresult in collagen with reduced tensile strength. To examine thepossibility that C- or N-terminal propeptide processing may be

FIG. 4. Cervical proteoglycan mRNA expression is shown during pregnancy and parturition. mRNA levels of (A) decorin, (B) biglycan, (C) osteomodulin,(D) asporin, and (E) fibromodulin were determined via QPCR on gestation Days 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, and 18 and 2 h postpartum and 24 hpostpartum and normalized to Day 18; n¼ 3–6 cervices per time point. *Significance at P , 0.05.

CHANGES IN CERVICAL COLLAGEN DURING PREGNANCY 1057

Dow

nloaded from w

ww

.biolreprod.org.

altered during pregnancy, we determined mRNA levels ofenzymes responsible for cleavage of N- and C-propeptidesfrom procollagen. The expression of genes encoding bonemorphogenetic protein-1 (Bmp1) and tolloid-like 1 (Tll1),which cleave the C-propeptide from the procollagen molecule,was analyzed via QPCR. mRNA levels for Bmp1 expressionwere abundant and constant throughout pregnancy, with smallbut significant elevations observed on gestation Day 18 andduring postpartum. There was a trend for reduced Tll1expression levels during pregnancy, though not all time pointsachieved significance (Fig. 2, A and B). The procollagen C-endopeptidase enhancer family (PCOLCE1 and �2), reportedto enhance the C-proteinase activity of TLL1 and BMP1, wasalso evaluated, and mRNA levels of Pcolce1 and -2 remainedconstant throughout gestation compared to NP (data notshown). Despite the decrease in Tll1 expression, immunoblotsshowed increased abundance of the 30- kDa C-propeptideduring pregnancy compared to the NP cervix, suggesting C-terminal processing occurs normally (Fig. 2C).

Three reported ADAMTS proteases (a disintegrin-like andmetalloprotease domain with thrombospondin type I motifs),ADAMTS2, ADAMTS3, and ADAMTS14, cleave the N-propeptide from the procollagen molecule. Using standardPCR, Adamts2 and -14 were found to be present in the cervix,while Adamts3 was undetectable (data not shown). QPCRanalysis of Adamts2 revealed stable expression throughoutpregnancy and postpartum (Fig. 2D). Adamts14 expressionincreased significantly on Day 10 of gestation and remainedelevated through Day 15 (P , 0.01). Levels dropped on Day18 and remained at NP levels throughout postpartum period(Fig. 2E). While an antibody that recognizes the N-propeptideof mouse collagen is available, we were unable to visualize N-propeptide in our system.

Collagen Cross-Links

Changes in the type or number of collagen cross-links canaffect the tensile strength and solubility of collagen. Hydrox-ylation of lysine residues by the enzyme lysyl hydroxylaseregulates the type of collagen cross-links formed. Expression ofthree genes that encode lysyl hydroxylase (Plod1, -2, and -3)was measured by QPCR (Fig. 3A). No appreciable change wasseen in Plod1 or Plod3 mRNA expression during gestation. Incontrast, compared to the NP, Plod2 mRNA levels appearedsuppressed throughout gestation and reached significance ongestation Days 9–13. Levels increased twofold at 2 hpostpartum and returned to NP levels by 24 h postpartum.

Previous reports of reduced activity of the cross-link-forming enzyme LOX along with our observation that Plod2 issuppressed significantly led us to evaluate the type and degreeof lysine cross-links during gestation. Lysylpyridinolines (LP)cross-links derived from one lysine and two hydroxylysinesand hydroxylysylpyridinolines (HP) cross-links derived fromthree hydroxylysines are the stronger pyridinoline cross-linksthat predominate in rigid connective tissues such as bone,cartilage, and tendon. LP and HP cross-links are relatively lowin the more flexible connective tissues such as skin, wherecross-links between lysine predominate [11]. Thus, a reductionin LP and HP cross-linking could contribute to increasedcompliance of the cervix during ripening.

HP and LP cross-links in cervical collagen at time pointsthroughout pregnancy were measured by HPLC. HP cross-links declined significantly from NP estrus levels on Day 12 ofgestation and remained low throughout gestation and duringpostpartum repair (Fig. 3B). LP cross-links also declined by

Day 14 of gestation and remained low until Day 18, increasingafter birth (Fig. 3C).

The progressive decline in LP and HP cross-links as birthapproaches likely contributes toward increased solubility ofcollagen. Protein was extracted in 7 M urea from NP, gestationDays 6 and 8–18, and 4-, 12-, 24-, and 48-h postpartumcervices, and immunoblotting of equal protein loads were used

FIG. 5. Cervical matricellular protein expression is shown duringpregnancy and parturition. mRNA levels of (A) thrombospondin 2, (B)Sparc, and (C) tenascin C were analyzed via QPCR at NP, gestation Days8, 9, 10, 11, 12, 13, 14, 15, 16, 17, and 18 and 2-h postpartum and 24-hpostpartum and normalized to Day 18. For NP, n ¼ 11 cervices forpregnant, and postpartum n¼5–7 cervices per time point. *Significance atP , 0.05.

1058 AKINS ET AL.

Dow

nloaded from w

ww

.biolreprod.org.

to detect collagen I. Because cross-linked collagen is notreadily soluble, these blots cannot assess changes in totalcollagen but rather give an estimate of extractable collagen I.Three bands were visible on the Western blot: a broad band at;250 kDa that we attribute to procollagen dimers and trimers,a mature collagen monomer band at 140 kDa, and anunidentified band at 50 kDa that appears in NP and postpartumsamples (Fig. 3D). The 140-kDa mature collagen band is notvisible in NP samples. During pregnancy this band increased indensity, suggesting that mature collagen is more extractable(soluble) during pregnancy. The band declines in density by 4h postpartum and is not seen in later postpartum samples.Increased extractability of collagen during pregnancy isconsistent with our previous observations of increased collagensolubility as determined by extraction in acetic acid and pepsinand with current data showing a decline in HP and LP collagencross-links [3]. The 250-kDa band appeared by Day 11 ofpregnancy and remained visible throughout gestation butdisappeared postpartum. The presence of this procollagen bandsuggests an increase in the proportion of newly synthesized butunprocessed collagen.

Proteoglycans

Given the ability of small leucine-rich proteoglycans tomodulate collagen structure, the mRNA expression of decorin,biglycan, fibromodulin, asporin, and osteomodulin wasquantified by QPCR (Fig. 4). Decorin, biglycan, andosteomodulin mRNA levels showed no significant changesthroughout pregnancy (Fig. 4, A–C). Asporin levels remainconstant throughout pregnancy, with a transient upregulation 2h postpartum (Fig. 4D). Fibromodulin levels remained constantthroughout gestation and were significantly downregulatedpostpartum (Fig. 4E).

Matricellular Proteins

Expression of mRNAs encoding the matricellular proteinsThbs2, Tnc, and Sparc, were obtained via QPCR (Fig. 5).Thbs2 mRNA expression dropped from NP levels by Day 8 ofgestation and remained low until 2 h after birth, returning to NPlevels by 24 h postpartum (Fig. 5A). The decline in expressionon Days 8, 10, and 11 did not achieve significance. Similar toThbs2 expression, that of tenascin C mRNA declinedsignificantly on Day 8 of gestation and remained downregu-lated for the remainder of pregnancy. Tnc mRNA increasedsixfold by 2 h postpartum and returned to NP levels by 24 hpostpartum (Fig. 5B). In contrast, there was a trend forincreased Sparc expression from gestation Day 10 through Day17, although significance was only achieved at gestation Days14 and 17. Relative expression returned to NP levels on Day 18of gestation throughout postpartum (Fig. 5C).

Collagen Ultrastructure

The observed changes in collagen solubility, type, anddegree of cross-links as well as matricellular protein compo-sition can affect collagen organization and ultrastructure.Electron micrographs of cervical tissue at Days 6 and 18 weretaken at 42003 magnification (Fig. 6). Cellular componentsand electron-dense components of the ECM appeared to be inclose proximity during early pregnancy (Fig. 6A). In lategestation, collagen fibers were more dispersed and notassociated with cellular components of the tissue (Fig. 6B),and generally there was a dramatic loss of electron-dense ECM.To evaluate changes in collagen ultrastructure, collagen fibrildiameters were measured from TEMs of cervical tissue fromNP metestrus and gestation Days 6, 12, 15, and 18 as describedin Materials and Methods (Fig. 7A–E). The mean fibrildiameter significantly increased from early to late pregnancyalong with a shift in the distribution toward a higher frequencyof fibrils with a larger diameter. As early as Day 6, there was anincrease in fibril diameter of 12.1 nm compared to NPmetestrus, with an additional 20-nm increase from Day 6 toDay 18 (Fig. 7, F–J).

DISCUSSION

This study identifies changes in the cervical ECM early inpregnancy that contribute cumulatively to the progressivedecline in tensile strength during the cervical softening phase.Most notable are changes in the degree and types of collagencross-links and a decline in expression of two matricellularproteins, thrombospondin 2 and tenascin C. The decline incollagen cross-links, as well as decline in expression of THBS2and TNC, must independently or cumulatively result in anincreased solubility of collagen during pregnancy as well as thestriking changes in collagen fibril ultrastructure.

Based on these observations, we propose a model in whichmature collagen in the cervical ECM is gradually replaced withless cross-linked collagen beginning early in gestation. As theless cross-linked collagen becomes more abundant in the ECM,tissue compliance begins to increase, eventually reaching athreshold of measurable change. This threshold defines thebeginning of cervical softening, which, in the mouse, ismeasurable by gestation Day 12, and tissue stiffness declinesprogressively thereafter [3]. The decline in LP and HP cross-links correlates with the decline in mRNA expression of Plod2as well as the reported decline in activity of LOX in the mousecervix [14]. While the other two lysyl hydroxylase genes(Plod1 and Plod3) were expressed at normal levels in themouse cervix, they did not compensate for the decline in Plod2expression. It has been suggested that PLOD2 may havegreater specificity for hydroxylation of the telopeptide regionof collagen, important in formation of pyridinoline cross-links,while the other lysyl hydroxylases may preferentially hydrox-

FIG. 6. Cervical ECM becomes dispersedthroughout pregnancy. Electron micro-graphs of cervical ECM taken at a magnifi-cation of 34200 on (A) Day 6 and (B) Day18 of pregnancy. Bar ¼ 1000 nm.

CHANGES IN CERVICAL COLLAGEN DURING PREGNANCY 1059

Dow

nloaded from w

ww

.biolreprod.org.

FIG

.7.

Coll

agen

fibri

lsin

crea

sein

size

duri

ng

pre

gnan

cy.El

ectr

on

mic

rogr

aphs

wer

eta

ken

ata

mag

nif

icat

ion

of3

20

500

of(A

)N

Pm

etes

trus,

(B)D

ay6,(C

)D

ay12,(D

)D

ay15,an

d(E

)D

ay18.B

ar¼

1000

nm

.A

nal

ysis

of

freq

uen

cyof

fibri

ldia

met

erof

(F)

NP

met

estr

us,

(G)

Day

6,

(H)

Day

12,

(I)

Day

15,

and

(J)

Day

18

issh

ow

n;

n¼

8260–1

3030

fibri

ls.

1060 AKINS ET AL.

Dow

nloaded from w

ww

.biolreprod.org.

ylate lysine residues in the triple helical regions of variouscollagens [35]. Mutations in the Plod2 gene and a reduction inpyridinoline cross-links have been described in patients withBruck syndrome, which is characterized by fragile bones,scoliosis, and osteoporosis [36, 37].

In addition to a decline in collagen cross-linking, alteredlevels of matricellular proteins likely contribute to earlychanges in tissue compliance as well. In mice, THBS2 appearsto regulate cell-matrix adhesion, inhibit angiogenesis, andregulate collagen fibril assembly [25, 38]. Loss of THBS2results in defective cell adhesion of fibroblasts and increasedcollagen solubility in skin, and collagen fibril size is increasedwhile tensile strength is decreased. Wound healing isaccelerated in these mice, along with increased tissuevascularization [39]. In addition, THBS2 null mice have areported acceleration of cervical softening without prematurebirth [40]. Both THBS2 and tenascin C play an importantfunction in promoting cell migration during wound healingafter injury and, consistent with this function, are upregulatedseveralfold in the cervix at the time of birth and postpartum[41]. Future studies are required to understand the mechanismsby which THBS2 and tenascin C contribute to collagen fibrilassembly as well as to a potential role in tissue vascularization,which is increased in late pregnancy [42].

The early, progressive and cumulative changes in thecervical ECM are supported by the observed increase incollagen fibril diameter as measured in electron micrographs.Increased fibril diameter between gestation Days 12, 15, and 18coincide with biomechanical changes that occur in latepregnancy (Fig. 7) [43, 44]. These changes also correlate withquantifiable changes in cervical collagen fiber morphology, asdetermined by second harmonic generation microscopy at thesame time points in mouse pregnancy [4]. The increasing fibrildiameter may result from reduced packing of fibrils due to boththe decline in HP and LP cross-links as well as the decline inTHBS2 and/or tenascin C expression. The net result is a loss oftensile strength. This is supported by the observation that areduction in pyridinoline cross-links, or loss of THBS2 inknockout mouse models leads to an increase in collagen fibrildiameter as measured by electron micrographs [25, 45].

A number of genes/proteins important in synthesis,trafficking, or processing of collagen were expressed at levelssimilar to or slightly more elevated than those in nonpregnantcervix (Figs. 2 and 3). The resulting continued production ofnewly synthesized collagen might ensure that collagen ismaintained at a constant level, yet it allows for a gradualturnover of well-cross-linked collagen with poorly cross-linkedcollagen. Consistent with this hypothesis is the observation thatSparc transcripts (Fig. 5B) and protein (data not shown) areelevated during pregnancy and decline to NP levels bygestation Day 18. Sparc expression is frequently associatedwith tissues in which there is a high rate of collagen turnover,and Sparc is required for appropriate procollagen processingand deposition of collagen in the ECM. Mice deficient in Sparchave reduced collagen content, and collagen is tethered to thecell surface, with reduced fibril aggregation in the ECM [46,47].

Proteoglycans also regulate and affect collagen fibrillogen-esis, as both decorin knockout and the fibromodulin/biglycandouble-knockout mice exhibit alterations in collagen ultra-structure in skin [48, 49]. Both the protein core and the GAGchain can influence ECM function [50]. Given the lack oftranscriptional regulation of genes encoding proteoglycans inthe human cervix [23] and our studies in the mouse, furtherinvestigations are required to evaluate postranslational regula-tion of the protein core as well as regulation of glycosamino-

glycan synthesis, chain length, and degree of sulfation. A rolefor small proteoglycans, such as decorin, in modulation ofcervical collagen structure is also supported by studies in thepregnant rat [51–53].

This work has identified early pregnancy changes incollagen processing and ECM composition that are likelyresponsible for the initial increase in tissue compliance duringcervical softening. In contrast to the accelerated changes thatoccur during cervical ripening and dilation at the end ofpregnancy, these early changes occur in an environment rich inprogesterone and relatively low estrogen. Future studies arerequired to identify steroid and peptide hormones that mayregulate collagen cross-link formation and Thbs2 and Tncexpression. These studies not only enhance our understandingof the progressive physiological changes in normal cervicalsoftening, but they also identify specific genes/proteins inwhich mutations or misregulation may account for clinicalcomplications such as cervical insufficiency or result inpremature cervical shortening of the cervix, which is a riskfactor for preterm birth [54, 55]. Future studies will addressthese important questions and provide necessary understandingrequired for development of therapies to prevent preterm birth,the leading cause of infant death in the first year of life.

ACKNOWLEDGMENTS

We would like to thank Dr. Larry Fisher at the NIH for use of the C-propeptide antibody. We thank Dr. Christopher Gilpin, Dr. Xinran Liu, andthe UTSW Molecular and Cellular Imaging Center for help with tissuepreparation and visualization for TEM. Finally, we extend our appreciationto Dr. Brenda Timmons for assistance with data analysis.

REFERENCES

1. Leppert PC. Anatomy and physiology of cervical ripening. Clin ObstetGynecol 1995; 38:267–279.

2. Word RA, Li XH, Hnat M, Carrick K. Dynamics of cervical remodelingduring pregnancy and parturition: mechanisms and current concepts.Semin Reprod Med 2007; 25:69–79.

3. Read CP, Word RA, Ruscheinsky MA, Timmons BC, Mahendroo MS.Cervical remodeling during pregnancy and parturition: molecularcharacterization of the softening phase in mice. Reproduction 2007;134:327–340.

4. Akins ML, Luby-Phelps K, Mahendroo M. Second harmonic generationimaging as a potential tool for staging pregnancy and predicting pretermbirth. J Biomed Opt 2010; 15:026020-1–026020-10.

5. Kadler KE, Holmes DF, Trotter JA, Chapman JA. Collagen fibrilformation. Biochem J 1996; 316(Pt 1):1–11.

6. Myers KM, Socrate S, Paskaleva A, House M. A study of the anisotropyand tension/compression behavior of human cervical tissue. J BiomechEng 2010; 132:021003-1–021003-15.

7. Lamande SR, Bateman JF. Procollagen folding and assembly: the role ofendoplasmic reticulum enzymes and molecular chaperones. Semin CellDev Biol 1999; 10:455–464.

8. Hendershot LM, Bulleid NJ. Protein-specific chaperones: the role of hsp47begins to gel. Curr Biol 2000; 10:R912–R915.

9. Leung MK, Fessler LI, Greenberg DB, Fessler JH. Separate amino andcarboxyl procollagen peptidases in chick embryo tendon. J Biol Chem1979; 254:224–232.

10. Kadler KE, Hulmes DJ, Hojima Y, Prockop DJ. Assembly of type Icollagen fibrils de novo by the specific enzymic cleavage of pC collagen.The fibrils formed at about 37 degrees C are similar in diameter,roundness, and apparent flexibility to the collagen fibrils seen inconnective tissue. Ann N Y Acad Sci 1990; 580:214–224.

11. Eyre DR, Paz MA, Gallop PM. Cross-linking in collagen and elastin.Annu Rev Biochem 1984; 53:717–748.

12. Canty EG, Kadler KE. Procollagen trafficking, processing and fibrillo-genesis. J Cell Sci 2005; 118:1341–1353.

13. Anum EA, Hill LD, Pandya A, Strauss JF III. Connective tissue andrelated disorders and preterm birth: clues to genes contributing toprematurity. Placenta 2009; 30:207–215.

14. Ozasa H, Tominaga T, Nishimura T, Takeda T. Lysyl oxidase activity in

CHANGES IN CERVICAL COLLAGEN DURING PREGNANCY 1061

Dow

nloaded from w

ww

.biolreprod.org.

the mouse uterine cervix is physiologically regulated by estrogen.Endocrinology 1981; 109:618–621.

15. Drewes PG, Yanagisawa H, Starcher B, Hornstra I, Csiszar K, Marinis SI,Keller P, Word RA. Pelvic organ prolapse in fibulin-5 knockout mice:pregnancy-induced changes in elastic fiber homeostasis in mouse vagina.Am J Pathol 2007; 170:578–589.

16. Lui PP, Chan LS, Lee YW, Fu SC, Chan KM. Sustained expression ofproteoglycans and collagen type III/type I ratio in a calcified tendinopathymodel. Rheumatology (Oxford) 2010; 49:231–239.

17. Liu X, Wu H, Byrne M, Krane S, Jaenisch R. Type III collagen is crucialfor collagen I fibrillogenesis and for normal cardiovascular development.Proc Natl Acad Sci U S A 1997; 94:1852–1856.

18. Maillot KV, Zimmermann BK. The solubility of collagen of the uterinecervix during pregnancy and labour. Arch Gynakol 1976; 220:275–280.

19. Kalamajski S, Oldberg A. The role of small leucine-rich proteoglycans incollagen fibrillogenesis. Matrix Biol 2010; 29:248–253.

20. Danforth DN, Veis A, Breen M, Weinstein HG, Buckingham JC, ManaloP. The effect of pregnancy and labor on the human cervix: changes incollagen, glycoproteins, and glycosaminoglycans. Am J Obstet Gynecol1974; 120:641–651.

21. Norman M, Ekman G, Ulmsten U, Barchan K, Malmstrom A.Proteoglycan metabolism in the connective tissue of pregnant and non-pregnant human cervix. An in vitro study. Biochem J 1991; 275(Pt 2):515–520.

22. Osmers R, Rath W, Pflanz MA, Kuhn W, Stuhlsatz HW, Szeverenyi M.Glycosaminoglycans in cervical connective tissue during pregnancy andparturition. Obstet Gynecol 1993; 81:88–92.

23. Westergren-Thorsson G, Norman M, Bjornsson S, Endresen U, Stjern-holm Y, Ekman G, Malmstrom A. Differential expressions of mRNA forproteoglycans, collagens and transforming growth factor-beta in thehuman cervix during pregnancy and involution. Biochim Biophys Acta1998; 1406:203–213.

24. Bornstein P, Sage EH. Matricellular proteins: extracellular modulators ofcell function. Curr Opin Cell Biol 2002; 14:608–616.

25. Kyriakides TR, Zhu YH, Smith LT, Bain SD, Yang Z, Lin MT, DanielsonKG, Iozzo RV, LaMarca M, McKinney CE, Ginns EI, Bornstein P. Micethat lack thrombospondin 2 display connective tissue abnormalities thatare associated with disordered collagen fibrillogenesis, an increasedvascular density, and a bleeding diathesis. J Cell Biol 1998; 140:419–430.

26. Jones PL, Jones FS. Tenascin-C in development and disease: generegulation and cell function. Matrix Biol 2000; 19:581–596.

27. Yan Q, Sage EH. SPARC, a matricellular glycoprotein with importantbiological functions. J Histochem Cytochem 1999; 47:1495–1506.

28. Taylor P. Practical Teratology. London: Academic Press; 1986:3–8.29. Bank RA, Beekman B, Verzijl N, de Roos JA, Sakkee AN, TeKoppele

JM. Sensitive fluorimetric quantitation of pyridinium and pentosidinecrosslinks in biological samples in a single high-performance liquidchromatographic run. J Chromatogr B Biomed Sci Appl 1997; 703:37–44.

30. Breeveld-Dwarkasing VN, te Koppele JM, Bank RA, van der WeijdenGC, Taverne MA, van Dissel-Emiliani FM. Changes in water content,collagen degradation, collagen content, and concentration in repeatedbiopsies of the cervix of pregnant cows. Biol Reprod 2003; 69:1608–1614.

31. Stegemann H, Stalder K. Determination of hydroxyproline. Clin ChimActa 1967; 18:267–273.

32. Nagai N, Hosokawa M, Itohara S, Adachi E, Matsushita T, Hosokawa N,Nagata K. Embryonic lethality of molecular chaperone hsp47 knockoutmice is associated with defects in collagen biosynthesis. J Cell Biol 2000;150:1499–1506.

33. Prockop DJ, Kivirikko KI. Collagens: molecular biology, diseases, andpotentials for therapy. Annu Rev Biochem 1995; 64:403–434.

34. Watson RB, Holmes DF, Graham HK, Nusgens BV, Kadler KE. Surfacelocated procollagen N-propeptides on dermatosparactic collagen fibrils arenot cleaved by procollagen N-proteinase and do not inhibit binding ofdecorin to the fibril surface. J Mol Biol 1998; 278:195–204.

35. van der Slot AJ, Zuurmond AM, Bardoel AF, Wijmenga C, Pruijs HE,Sillence DO, Brinckmann J, Abraham DJ, Black CM, Verzijl N, DeGrootJ, Hanemaaijer R, et al. Identification of PLOD2 as telopeptide lysylhydroxylase, an important enzyme in fibrosis. J Biol Chem 2003; 278:40967–40972.

36. Ha-Vinh R, Alanay Y, Bank RA, Campos-Xavier AB, Zankl A, Superti-Furga A, Bonafe L. Phenotypic and molecular characterization of Brucksyndrome (osteogenesis imperfecta with contractures of the large joints)

caused by a recessive mutation in PLOD2. Am J Med Genet A 2004; 131:115–120.

37. Hyry M, Lantto J, Myllyharju J. Missense mutations that cause Brucksyndrome affect enzymatic activity, folding, and oligomerization of lysylhydroxylase 2. J Biol Chem 2009; 284:30917–30924.

38. Yang Z, Kyriakides TR, Bornstein P. Matricellular proteins as modulatorsof cell-matrix interactions: adhesive defect in thrombospondin 2-nullfibroblasts is a consequence of increased levels of matrix metal-loproteinase-2. Mol Biol Cell 2000; 11:3353–3364.

39. Kyriakides TR, Leach KJ, Hoffman AS, Ratner BD, Bornstein P. Micethat lack the angiogenesis inhibitor, thrombospondin 2, mount an alteredforeign body reaction characterized by increased vascularity. Proc NatlAcad Sci U S A 1999; 96:4449–4454.

40. Kokenyesi R, Armstrong LC, Agah A, Artal R, Bornstein P. Thrombo-spondin 2 deficiency in pregnant mice results in premature softening of theuterine cervix. Biol Reprod 2004; 70:385–390.

41. Timmons BC, Mahendroo M. Processes regulating cervical ripening differfrom cervical dilation and postpartum repair: insights from geneexpression studies. Reprod Sci 2007; 14:53–62.

42. Mowa CN, Jesmin S, Sakuma I, Usip S, Togashi H, Yoshioka M, HattoriY, Papka R. Characterization of vascular endothelial growth factor(VEGF) in the uterine cervix over pregnancy: effects of denervation andimplications for cervical ripening. J Histochem Cytochem 2004; 52:1665–1674.

43. Word RA, Landrum CP, Timmons BC, Young SG, Mahendroo MS.Transgene insertion on mouse chromosome 6 impairs function of theuterine cervix and causes failure of parturition. Biol Reprod 2005; 73:1046–1056.

44. Mahendroo MS, Cala KM, Landrum DP, Russell DW. Fetal death in micelacking 5alpha-reductase type 1 caused by estrogen excess. MolEndocrinol 1997; 11:917–927.

45. Takaluoma K, Hyry M, Lantto J, Sormunen R, Bank RA, Kivirikko KI,Myllyharju J, Soininen R. Tissue-specific changes in the hydroxylysinecontent and cross-links of collagens and alterations in fibril morphology inlysyl hydroxylase 1 knock-out mice. J Biol Chem 2007; 282:6588–6596.

46. Rentz TJ, Poobalarahi F, Bornstein P, Sage EH, Bradshaw AD. SPARCregulates processing of procollagen I and collagen fibrillogenesis indermal fibroblasts. J Biol Chem 2007; 282:22062–22071.

47. Bradshaw AD, Puolakkainen P, Dasgupta J, Davidson JM, Wight TN,Helene Sage E. SPARC-null mice display abnormalities in the dermischaracterized by decreased collagen fibril diameter and reduced tensilestrength. J Invest Dermatol 2003; 120:949–955.

48. Danielson KG, Baribault H, Holmes DF, Graham H, Kadler KE, IozzoRV. Targeted disruption of decorin leads to abnormal collagen fibrilmorphology and skin fragility. J Cell Biol 1997; 136:729–743.

49. Ameye L, Aria D, Jepsen K, Oldberg A, Xu T, Young MF. Abnormalcollagen fibrils in tendons of biglycan/fibromodulin-deficient mice lead togait impairment, ectopic ossification, and osteoarthritis. FASEB J 2002;16:673–680.

50. Ruhland C, Schonherr E, Robenek H, Hansen U, Iozzo RV, Bruckner P,Seidler DG. The glycosaminoglycan chain of decorin plays an importantrole in collagen fibril formation at the early stages of fibrillogenesis. FEBSJ 2007; 274:4246–4255.

51. Kokenyesi R, Woessner JF Jr. Relationship between dilatation of the ratuterine cervix and a small dermatan sulfate proteoglycan. Biol Reprod1990; 42:87–97.

52. Kokenyesi R, Woessner JF Jr.. Effects of hormonal perturbations on thesmall dermatan sulfate proteoglycan and mechanical properties of theuterine cervix of late pregnant rats. Connect Tissue Res 1991; 26:199–205.

53. Leppert PC, Kokenyesi R, Klemenich CA, Fisher J. Further evidence of adecorin-collagen interaction in the disruption of cervical collagen fibersduring rat gestation. Am J Obstet Gynecol 2000; 182:805–811; discussion811–812.

54. Iams JD, Goldenberg RL, Mercer BM, Moawad AH, Meis PJ, Das AF,Caritis SN, Miodovnik M, Menard MK, Thurnau GR, Dombrowski MP,Roberts JH. The preterm prediction study: can low-risk women destinedfor spontaneous preterm birth be identified? Am J Obstet Gynecol 2001;184:652–655.

55. Owen J, Yost N, Berghella V, Thom E, Swain M, Dildy GA 3rd,Miodovnik M, Langer O, Sibai B, McNellis D. Mid-trimester endovaginalsonography in women at high risk for spontaneous preterm birth. JAMA2001; 286:1340–1348.

1062 AKINS ET AL.

Dow

nloaded from w

ww

.biolreprod.org.