Search

Aims: To determine the localisation and distribution of connective tissue growth factor (CCN2; CTGF) and transforming growth factor β type 1 (TGF-β1) in uterine tissues from cycling and early pregnant pigs. Methods: In situ hybridisation and immunohistochemistry were used to localise CCN2 (CTGF) or TGF-β1 in uteri obtained from gilts on days 0, 5, 10, 12, 15, and 18 of the oestrous cycle or days 10, 12, 14, 16, 17, and 21 of gestation. Results: In cycling animals, CCN2 (CTGF) mRNA and protein were abundant in luminal epithelial cells (LECs) and glandular epithelial cells (GECs), with lesser amounts in stromal fibroblasts and little or none in endothelial cells. A similar pattern of staining was seen up to day 10 of pregnancy, except that overall staining intensities for CCN2 (CTGF) mRNA or protein were higher and that stromal and endothelial cells were CCN2 (CTGF) positive. However, on days 12–17 there was a striking decrease in the amount of CCN2 (CTGF) in LECs at the utero–conceptus interface, which was associated with maternal stromal matrix reorganisation and the onset of subepithelial neovascularisation. This differential distribution of CCN2 (CTGF) was localised to those LECs that were in close proximity to or in apposition with trophoblast cells. This decrease in CCN2 (CTGF) staining was transient in nature and high amounts of CCN2 (CTGF) were again apparent in LECs on days 17–21, when endometrial neovascularisation and matrix remodelling were complete. The expression of uterine TGF-β1 was comparable to that of CCN2 (CTGF) at most stages of the oestrous cycle or early pregnancy. Pre-elongation blastocysts recovered on day 10 were positive for both CCN2 (CTGF) and TGF-β1 in the extra-embryonic trophectoderm, endoderm, and inner cell mass. On day 12, trophectoderm expressed low amounts of TGF-β1 mRNA and non-detectable amounts of TGF-β1 protein or CCN2 (CTGF) mRNA or protein. By days 17–21, the expression of both growth factors in the extra-embyronic/placental membranes increased and frequently exceeded that seen in LECs. Conclusions: The pattern of CCN2 (CTGF) production during the initial attachment phase supports a role for this factor in stromal remodelling and neovascularisation, although alternative functions at later stages such as epithelial–epithelial interactions are also possible. In most major cell types in the uterus or utero–placental unit, CCN2 (CTGF) expression was highly correlated with that of TGF-β1, indicating that CCN2 (CTGF) may mediate some of the functions of TGF-β in the reproductive tract during the oestrous cycle and pregnancy. The data further highlight epithelium as an important source of CCN2 (CTGF) in the regulation of uterine function.

Connective tissue growth factor (CTGF) stimulates cell proliferation, migration, adhesion and extracellular matrix production, and functions in processes such as development, differentiation, angiogenesis, implantation, wound healing and fibrosis. CTGF is a 38 kDa protein that comprises four discrete structural modules (modules 1-4) but is susceptible to limited proteolysis in utero yielding bioactive isoforms that comprise either modules 3 and 4 (16-20 kDa) or module 4 (10 kDa). Here we report the development of a stable cell line, termed DB1, that was generated by transfecting cDNA encoding full-length human CTGF into Chinese hamster ovary cells that were mutant for heparin sulphate and chondroitin sulphate. DB1 cells produced 38 kDa CTGF and low molecular mass CTGFs that had N-termini between modules 2 and 3 at Ala(181) (20 kDa), Leu(184) (18 kDa) or Ala(197) (16 kDa) or between modules 3 and 4 at Gly(253) (10 kDa). CTGF was exported from DB1 cells as early as 5 min after synthesis and all isoforms were readily purified from conditioned medium by sequential steps of heparin affinity, cation exchange, and reverse-phase chromatography. The 38 kDa CTGF was faithfully glycosylated and underwent limited proteolysis in the presence of thrombin, kallikrein or uterine fluids, the last of which was antagonized by anti-thrombin III. All CTGF isoforms promoted cell adhesion, mitosis and epithelial transdifferentiation in vitro as well as subcutaneous fibrosis in vivo. The establishment of this recombinant expression system allows for mass-scale production of all previously reported uterine CTGF isoforms, demonstrates that module 4 contains functional domains involved in a broad range of biological activities, and will facilitate studies of CTGF processing in vitro.

Aims —To determine mechanisms regulating the production of connective tissue growth factor (CCN2; CTGF) and transforming growth factor β1 (TGF-β1) in the mouse uterus. Methods —In situ hybridisation and immunohistochemistry were used to localise CCN2 (CTGF) and TGF-β1 in uteri from sexually mature female mice that had either been (1) mated with sterile males to induce pseudopregnancy or (2) ovariectomised (OVX) and administered estradiol-17β (E 2 ) or progesterone (P 4 ), either alone or in combination. Uteri collected on days 0.5, 1.5, 2.5, 3.5, 4.5, or 5.5 of pseudopregnancy or at one, three, six, 12, or 24 hours after steroid administration were fixed, sectioned, and incubated with specific riboprobes or antibodies to permit detection and localisation of mRNA or protein for CTGF and TGF-β1. Results —On days 0.5–2.5 of pseudopregnancy, CCN2 (CTGF) and TGF-β1 were principally colocalised to uterine epithelial cells, with much smaller amounts in the stroma. On days 3.5–4.5, there was a reduction of CCN2 (CTGF) and TGF-β1 in the epithelium but an increase in stromal and endothelial cells, corresponding to a period of extracellular matrix remodelling and neovascularisation within the endometrium. In OVX mice, epithelial cells were weakly positive for both CCN2 (CTGF) and TGF-β1 in the absence of steroid hormones. Epithelial CTGF mRNA production were strongly but transiently stimulated in OVX mice cells by E 2 . These effects were antagonised by P 4 , which itself transiently stimulated epithelial CCN2 (CTGF) production, although less robustly than E 2 . CTGF and TGF-β1 protein amounts were high in epithelial cells throughout steroid treatment and were increased in the stroma, where they were relatively long lived. Stromal CCN2 (CTGF) and TGF-β1 were lower after co-administration of E 2 and P 4 than in response to each hormone individually. Although ccn2 (ctgf) is a TGF-β1 inducible gene in other systems, and both growth factors were often co-localised in uterine tissues in these studies, several treatment regimens resulted in high amounts of TGF-β1 protein in stromal cells without the concomitant production of ccn2 (ctgf) mRNA. Conclusions— Maternal factors are principal cues for CCN2 (CTGF) and TGF-β1 production in the uterus because (1) their expression during pseudopregnancy is comparable to that seen in pregnancy and (2) they are regulated by ovarian steroids. TGF-β dependent and independent mechanisms of ccn2 (ctgf) gene transcription exist in the uterus that are variably regulated by steroid hormones. Collectively, the data support a role for CCN2 (CTGF) in mediating the effects of steroid hormones and TGF-β on endometrial function.