Analysis of obstetric complications and uterine connective tissue in tenascin-X-deficient humans and mice
- First Online:
- Cite this article as:
- Egging, D.F., van Vlijmen-Willems, I., Choi, J. et al. Cell Tissue Res (2008) 332: 523. doi:10.1007/s00441-008-0591-y
- 410 Downloads
Tenascin-X (TNX) is a large, multi-domain, extracellular matrix glycoprotein. Complete deficiency of TNX in humans leads to a recessive form of Ehlers-Danlos syndrome (EDS), and TNX haploinsufficiency is a cause of hypermobility type EDS. EDS patients appear to have a higher risk of several complications during pregnancy, such as pelvic instability, premature rupture of membranes, and postpartum hemorrhage. Here, we present a study of genitourinary and obstetric complications in TNX-deficient women of reproductive age. We have found complications, such as uterus prolapses, that are in agreement with previous findings in other EDS types. In TNX knockout (KO) mice, we have observed mild pregnancy-related abnormalities. Morphological and immunohistological analysis of uterine tissues has not revealed obvious quantitative or spatial differences between TNX KO and wildtype mice with respect to collagen types I, III, V, and XII or elastic fibers. We conclude that TNX-deficient women are at risk of obstetric complications, but that TNX KO mice show only a mild phenotype. Furthermore, we show that TNX is involved in the stability of elastic fibers rather than in their initial deposition.
KeywordsEhlers-Danlos syndromeTenascin-XCollagenElastinPregnancyHumanMouse (TNX knockout; C57BL/6)
Tenascin-X (TNX) is a large, multi-domain, extracellular matrix (ECM) glycoprotein composed of epidermal growth factor (EGF) like-repeats, fibronectin type III repeats, and a C-terminal fibrinogen domain (Bristow et al. 1993; Lethias et al. 1996; Elefteriou et al. 1997; Ikuta et al. 1998; Tucker et al. 2006). Complete deficiency of TNX in humans leads to a rare recessive form of Ehlers-Danlos Syndrome (EDS), and TNX haploinsufficiency is a cause of hypermobility type EDS. Patients of both EDS types exhibit mild to severe joint hypermobility. The skin of TNX-deficient patients is markedly lax with poor recoil properties and shows easy bruising. The collagen density appears reduced in the dermis of these patients, and the elastic fibers are abnormal (Burch et al. 1997; Schalkwijk et al. 2001; Zweers et al. 2003; Peeters et al. 2004; Lindor and Bristow 2005). Most of the literature concerning pregnancy in EDS uses the old classification of nine subgroups. In our study, we use the revised classification of six subtypes of EDS (Beighton et al. 1998). Tissue fragility is a hallmark of EDS and is present in all the different subtypes. EDS patients appear to have a higher risk of several complications during pregnancy, such as pelvic instability, premature rupture of membranes, and postpartum hemorrhage. Pregnancy in classical and hypermobility type EDS has a relative favorable maternal and neonatal outcome. Pregnancy in vascular type EDS patients is associated with severe complications including maternal mortality (Pepin et al. 2000; Carley and Schaffer 2000; Roop and Brost 1999; Wegrowski et al. 1999; Ramos-e-Silva et al. 2006; Lind and Wallenburg 2002; Kuczkowski 2005; Parry and Strauss 1998). The effect of TNX deficiency on pregnancy has not been studied so far, and only one case report of pregnancy in a TNX-deficient woman has been reported in the literature (Lindor and Bristow 2005). Here, we present a study of all known pregnancies and genito-urinary (GU) abnormalities in TNX-deficient woman of reproductive age identified in our clinic (Schalkwijk et al. 2001) and in the literature (Lindor and Bristow 2005; Schalkwijk et al. 2001). Furthermore, we have extended our studies to a TNX knockout (KO) mouse model in which we have investigated uterine development during pregnancy. We have previously established that TNX is essential for the development of tissue strength in the skin and is able to bind to elastin, a major component of elastic fibers, and to collagens type I, III, V, and XII (Egging et al. 2006a, b; Veit et al. 2006). We have investigated the expression of these molecules at the protein level.
Materials and methods
We investigated pregnancies and GU abnormalities in all currently identified female tenascin-X deficient patients of reproductive age identified in our clinic and in the literature. The study protocol was approved by the local medical ethics committee, and written informed consent was obtained from the patients.
TNX knockout (KO) mice were obtained as described previously (Mao et al. 2002). For all studies, we used TNX KO mice that were crossed back with six generations of C57BL/6N mice. Wildtype C57BL/6 (WT) mice were used as a control. All mice investigated for pregnancy abnormalities and morphological changes in uterus tissue were between 2 and 6 months of age, reflecting a normal age distribution of a breeding mice population. Skin samples were taken from 2-month-old and 9-month-old mice, and aortic tissue was obtained from 9-month-old mice. The experimental design was approved by the Animal Use Committee of Radboud University, Nijmegen.
Breeding pairs were kept in a 5:1 female:male ratio. Females were inspected daily for vaginal plugs, and those with plugs were isolated and inspected each day for abnormalities and progression of pregnancy. Litter size was determined after birth. Neonatal survival is defined as the difference in litter size after partus and 2 weeks postpartum.
Processing of samples for (immuno)histochemistry
Mice were killed in a sealed compartment by exposure to a mixture of carbogen gas and increasing concentrations of CO2. Skin samples for frozen sections were embedded in Tissue-Tek O.C.T. compound (Sakura Finetek Europe, The Netherlands) and snap frozen in liquid nitrogen. Skin samples for paraffin sections were fixed in 4% buffered formalin for 4 h.
Affinity-purified TNX antibody production
TNX FNIII repeats 27–32 were amplified by polymerase chain reaction (PCR; forward primer: 5’-GGAATTCGAGCTACCTCCCCAC-3’, reverse primer: 5’-CAGGTCGACTCAGGTGAAAGAGGTGGA-3’) by using a previously described 2.7-kb human TNX cDNA as a template (Morel et al. 1989). The PCR product was ligated into the pCR2.1TOPO vector (Invitrogen, Breda, The Netherlands) according to the manufacturer’s instructions for easy digestion with restriction enzymes. The pCR2.1 TOPO vector with insert was digested with EcoRI and SalI. The region coding for FNIII27-32 was inserted into the EcoRI/SalI site of the pET28(a)+ plasmid (Brunschwig Chemie, Amsterdam, The Netherlands). The sequence of the TNX domains was verified by dideoxy sequencing with a 3730 DNA analyzer (Applied Biosystems, Nieuwekerk a/d IJssel, The Netherlands). TNX FNIII27-32 protein was expressed and purified according to the manufacturer’s instructions (Brunschwig Chemie). Purified TNX FNIII27-32 protein was used for immunization of a rabbit. Aliquots of 500 μg TNX FNIII27-32 in 500 μl phosphate-buffered saline were mixed with an equal volume of Freund’s complete adjuvant for the first injection and Freund’s incomplete adjuvant for boosters. Three injections were administered subcutaneously at 3-week intervals. Polyclonal antibodies against TNX FNIII27-32 were purified by affinity chromatography on a column with the antigen coupled to CNBr-activated sepharose 4B (GE Healthcare Life Sciences, Diegem, Belgium).
Histopathology, histochemistry, and immunohistochemistry
Staining of elastic fibers was performed by modified Hart’s stain on paraffin sections (7 μm; Starcher et al. 2005). Paraffin sections were also stained with hematoxylin and eosin (H&E) stain to assess uterus morphology and pathology, in collaboration with a professional pathologist. Frozen sections (7 μm) were stained with antibodies against collagen types I, III, and V (SouthernBiotech, USA) and elastin (Elastin Products, USA) as previously described by Egging et al. (2006b). Collagen XII was stained as previously described by Veit et al. (2006). TNX was stained with affinity-purified antibodies directed against FNIII27-32 of TNX. Detection was performed with antibodies labeled with fluorescein isothiocyanate (FITC, green; DakoCytomation, Denmark) and Alexa Fluor 594 (AF594, red; Molecular Probes, The Netherlands). Cell nuclei were made visible with 4,6-diamidino-2-phenylindole (DAPI, blue; Molecular Probes) counterstaining. Assesment of immunostaining was performed as previously described (Egging et al. 2007, 2006b). Images were captured and processed with Axiovision software (Carl Zeiss, The Netherlands).
For electron microscopy, WT and TNX KO samples from uterine wall at 3 weeks post-partum, skin from 2-month-old and 9-month-old mice, and aortas from 9-month-old mice were fixed in 3% glutaraldehyde in 0.1 M sodium cacodylate overnight and washed in 0.1 M sodium cacodylate buffer. The tissues were then sequentially treated with osmium tetroxide, tannic acid, and uranyl acetate and then dehydrated and embedded in Epon as previously described (Davis 1993). Thin sections (60 nm) were counterstained with 7% uranyl acetate in absolute methanol and lead citrate and examined in a Tecnai 12 transmission electron microscope, at 120 kV.
For statistical analyses, Fisher’s exact or Chi-square tests were used where appropriate for proportional data. Student’s t-test was used to compare mean litter size after partus and neonatal survival (difference in litter size after partus and 2 weeks postpartum). A P-value of <0.05 was considered statistically significant.
Pregnancy in TNX-deficient patients
Pregnancy in TNX-deficient woman (VUE vaginal uterine extirpation, IUD intra-uterine death, CAH congenital adrenal hyperplasia). Age represents the age of the patient at the anamnesis
Gravida number/Para number
2 (sister of patient 1)
VUE after uterine prolapse (at age 49)
Vaginal prolapse 2 weeks after VUE
Was born prematurely (week 35) with shoulder luxation, rectal prolapse at age 1
4 (sister of patient 3)
One incident of hemorrhage postpartum (>1000 ml blood loss)
IUD (24 weeks) with relatively large blood loss
Deceased at age 51, severe co-morbidity including cardiomyopathy, arrhythmia, arteriosclerosis, and amputation of left leg (Schalkwijk et al. 2001).
Undefined prolapse (at age 36)
Precipitious second stage at term for all births
Lindor and Bristow 2005
Uterine prolapse (at age 20), also recurrent rectal prolapses
Pregnancy in TNX KO mice
Pregnancy in TNX-deficient mice
TNX KO mice
Number of mice
Number of mice
Maternal death during pregnancy
Obstruction during partus
Excessive bleeding of vagina (as seen at vaginal plug check)
Cannibalism after partus
Total number of evaluated pregnancies
Structure of mouse uterus
TNX distribution in mouse uterus
Collagen distribution in mouse uterus
Elastic fibers in mouse tissues
GU complications occur frequently in patients with various types of EDS (Carley and Schaffer 2000; Ramos-e-Silva et al. 2006; Lind and Wallenburg 2002; Kuczkowski 2005; Parry and Strauss 1998; Pepin et al. 2000). In classical and hypermobility types of EDS, which are the most common, the outcome of pregnancy is generally favorable. Maternal complications however, such as postpartum hemorrhage and pelvic instability, are more common than in the general population (Ramos-e-Silva et al. 2006; Kuczkowski 2005; Parry and Strauss 1998). Furthermore, EDS patients appear to have an elevated risk of uterine prolapse (Carley and Schaffer 2000; Lind and Wallenburg 2002). Pregnancy in patients with vascular type EDS may even lead to maternal death through uterus or vessel rupture (Pepin et al. 2000). We have investigated pregnancies and GU abnormalities in TNX-deficient woman of reproductive age. Generally, pregnancy is without major complications in TNX-deficient patients, apart from one noted incident of postpartum hemorrhage. However, uterine and vaginal prolapse regularly occur in TNX-deficient women, even at a young age, suggesting laxity of GU tissues. Premature rupture of fetal membranes is a risk in pregnancy with EDS affected fetuses (Lind and Wallenburg 2002; Parry and Strauss 1998). No premature births have been observed in the offspring of the TNX-deficient patients; however, some TNX-deficient patients had been born prematurely. Obviously, some caution must be taken in making conclusions and extrapolating data from such a small group of TNX-deficient woman.
Complete TNX deficiency in humans is a rare condition, and so far, only a few patients have been identified. We have therefore investigated pregnancy and uterine tissue structure in our TNX KO mouse model. No gross significant differences have been found with regard to abnormalities during pregnancy or reproduction between the TNX KO and WT mice. A trend toward a reduction in the length of pregnancy in TNX KO mice. The only significant difference has been observed in the location of the vaginal plugs, which suggests laxity in the vaginal wall.
In a previous study, we have not found any differences in cutaneous collagen deposition between TNX KO and WT mice (Egging et al. 2006b), as has been observed in the skin of TNX-deficient patients (Schalkwijk et al. 2001; Zweers et al. 2004). Alterations in collagen deposition in the skin of TNX KO mice is, however, a matter of debate (Egging et al. 2006b; Mao et al. 2002; Minamitani et al. 2004). In the present study, we have found elastin aggregates and enlarged elastic fibers in 9-month-old skin of TNX KO mice at the ultrastructural level; this is in accordance with our previously published data concerning increased elastin staining in aging TNX KO mouse skin (Egging et al. 2006b). We have detected no differences in elastic fibers in the uterus or elastic laminae of the abdominal aorta between TNX KO and WT mice suggesting a specific role for TNX in the maturation or maintenance of elastic fibers in skin. However, abnormal elastic fibers might not be present in the uterus, since the ECM is constantly being remodeled during the menstrual cycle and pregnancy.
Although the TNX KO mice appear to have a much milder phenotype compared with TNX-deficient humans, the skin of TNX KO mice is significantly weaker and more lax compared with WT mice (Bristow et al. 2005; Burch et al. 1997; Egging et al. 2006b; Lindor and Bristow 2005; Mao et al. 2002; Matsumoto et al. 2001; Minamitani et al. 2004; Schalkwijk et al. 2001; Zweers et al. 2004). Our data suggest a more lax vaginal canal in TNX-deficient patients and to a certain degree in TNX KO mice. An attractive speculation, considering the observations from our studies in mice, is that TNX is essential for tissue integrity, independent of the quantity of collagen deposition or elastic fiber structure. In conclusion, we would advise obstetricians and gynecologists to monitor TNX-deficient patients closely, as they are likely to have weaker GU connective tissue and could therefore be at risk of GU complications.
We are grateful to Ilona van den Brink, Geert Poelen, Debby Smits, and Brenda van Vliet of the Central Animal laboratory of the Radboud University Nijmegen for their technical assistance. Willeke Blokx of the Department of Pathology of the UMC Nijmegen, is acknowledged for the histopathological evaluation of mouse uteri. Jim Bristow of the Lawrence Berkeley National Laboratory, USA, is acknowledged for providing the original TNX KO mice and discussion.
This article is distributed under the terms of the Creative Commons Attribution Noncommercial License which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.