Failure of pelvic organ support in mice deficient in fibulin-3

Abstract

Fibulin-5 is crucial for normal elastic fiber synthesis in the vaginal wall; more than 90% of fibulin-5-knockout mice develop pelvic organ prolapse by 20 weeks of age. In contrast, fibulin-1 and -2 deficiencies do not result in similar pathologies, and fibulin-4-knockout mice die shortly after birth. EFEMP1 encodes fibulin-3, an extracellular matrix protein important in the maintenance of abdominal fascia. Herein, we evaluated the role of fibulin-3 in pelvic organ support. Pelvic organ support was impaired significantly in female Efemp1 knockout mice (Fbln3(-[supi]/-)), and overt vaginal, perineal, and rectal prolapse occurred in 26.9% of animals. Prolapse severity increased with age but not parity. Fibulin-5 was up-regulated in vaginal tissues from Fbln3(-[supi]/-) mice regardless of prolapse. Despite increased expression of fibulin-5 in the vaginal wall, pelvic organ support failure occurred in Fbln3(-[supi]/-) animals, suggesting that factors related to aging led to prolapse. Elastic fiber abnormalities in vaginal tissues from young Fbln3(-[supi]/-) mice progressed to severe elastic fiber disruption with age, and vaginal matrix metalloprotease activity was increased significantly in Fbln3(-[supi]/-) animals with prolapse compared with Fbln3(-[supi]/-) mice without prolapse. Overall, these results indicate that both fibulin-3 and -5 are important in maintaining pelvic organ support in mice. We suggest that increased vaginal protease activity and abnormal elastic fibers in the vaginal wall are important components in the pathogenesis of pelvic organ prolapse.

Figure 1

Pelvic anatomy of Fbln3^−/− mice with pelvic organ prolapse. Perineal bulge (stage 3; A, arrow) and rectal prolapse (B) are observed in Fbln3^−/− females. C: Uterine horns (Ut) descend beneath the pubic symphysis (PS). The bulge (arrow) is comprised of enlarged bladder and vagina. D: Dissected female reproductive tract of Fbln3^−/− mice with prolapse. Note patulous, enlarged vaginal wall (Vag). Ut, uterine horns; Bl, bladder.

Figure 2

Pelvic organ support in Fbln3^−/− with or without pelvic organ prolapse. Magnitude of perineal bulge, perineal body length, and vaginal diameter were measured in WT and Fbln3^−/− mice without (KO without prolapse) and with (KO with prolapse) prolapse as part of the MOPQ scoring system. Prolapse was defined as the presence of ≥stage 1 perineal bulge (visually evident bulge). *P ≤ 0.05 compared with WT.

Figure 3

Rate and severity of pelvic organ prolapse in Fbln3^−/− mice. A: Percentage of Fbln3^−/− mice with ≥stage 1 pelvic organ prolapse as a function of age. Magnitude of perineal bulge (B) and perineal body length (C) are increased significantly in Fbln3^−/− mice with age. Measurements were derived from 197 animals >11 weeks of age. *P < 0.05 compared with 12 weeks.

Figure 4

Elastic fiber morphology in Fbln3^−/− and Fbln5^−/− mice. Transverse sections of mid-vagina were stained with Hart’s stain in WT mice (A, E), Fbln3^−/− mice without prolapse (B), Fbln3^−/− mice with prolapse (C, F), and Fbln5^−/− mice with prolapse (D). Arrowheads indicate elastic fibers. A–D represent epithelium (Epi), lamina propria (LP), and muscularis (M). E and F represent deep muscularis and adventitial layers of the vaginal wall adjacent to the pubocaudalis. Scale bars = 20 μm.

Figure 5

Ultrastructural morphology of elastic fibers in vaginal tissues from WT, Fbln3^−/−, and Fbln5^−/− mice at 6 to 8 months of age. A and B: Elastic fibers in vaginal wall from WT animals are comprised of a central elongated core of elastin (e) surrounded by numerous fine microfibrils (arrows, mf), and collagen fibers (coll) in the extracellular matrix. C and D: Elastic fibers in vaginal tissues from Fbln3^−/− mice without prolapse. Elastic fibers are disrupted resulting in evident microfibrils within the elastin core. Longitudinal section reveals significant disruption of the elastic fiber, which is porous and frayed. E and F: Elastic fibers in vaginal tissues from Fbln3^−/− mice with prolapse. Disrupted elastin core permits visualization of microfibrils throughout the elastin core. Severely disrupted and frayed elastic fibers are noted on longitudinal section. G and H: Elastic fibers in vaginal tissues from Fbln5^−/− mice. Large amorphous elastin aggregates and displaced clusters of microfibrils. Scale bars: 0.2 μm (A, C, E, G); 0.5 μm (B, D, F, H).

Figure 6

Ultrastructural morphology of elastic fibers in vaginal tissues from young WT and Fbln3^−/− mice. Elastic fibers from WT (A, B) and Fbln3^−/− (C, D) mice at 5 weeks of age. The elastin core (e) is more fragmented in vaginal tissues from young Fbln3^−/− females compared with WT. Microfibrils (mf) are noted with arrows. coll, collagen. Scale bars: 0.2 μm (A, C); 0.5 μm (B, D).

Figure 7

Localization of fibulin-3 in the vaginal wall. Transverse sections of mid vagina from WT (left) and Fbln3^−/− mice (right) at 6 months of age were immunostained with a polypeptide polyclonal antibody to fibulin-3. Immunostaining was visualized in subepithelial cells and basement membrane (arrows), subepithelial elastic fibers (arrowheads), and stromal cells of the vaginal muscularis (small arrowheads). Epi, epithelium; musc, muscularis. Scale bars = 20 μm.

Figure 8

MMP2 and MMP9 activity in vaginal tissues from Fbln3^−/− mice. A: Gelatin zymogram indicating pro- and active MMP9 and MMP2 in vaginal tissue extracts from WT mice, Fbln3^−/− mice without prolapse, and Fbln3^−/− mice with prolapse. Immunoblot for α-tubulin is placed beneath the zymogram. B: Quantification of MMP2 and MMP9 in vaginal tissues from WT (n = 4) and Fbln3^−/− (KO) mice without (n = 4) and with prolapse (n = 5). Enzyme activity is expressed in relative units and normalized to α-tubulin. *P ≤ 0.05 compared with WT.

Figure 9

Expression of TIMPs in the vaginal wall. Relative levels of Timp1 (A), Timp2 (B), and Timp3 (C) mRNA were quantified in vaginal tissues from WT (n = 5) and Fbln3^−/− mice without (n = 5) and with prolapse (n = 6) and expressed relative to that of β2-microglobulin (B2M). *P ≤ 0.05 compared with WT.

Figure 10

Expression of fibulin-5 (FBLN-5) and tropoelastin (TE) in vaginal tissues from Fbln3^−/− mice. A: Immunoblot of FBLN-5 was conducted in urea extract proteins from vaginal tissues of WT and Fbln3^−/− mice without and with prolapse. Coomassie staining was conducted in side-by-side gels to ensure even protein loading. B: Immunoblot of tropoelastin (TE) in urea extract proteins from vaginal tissues of WT and Fbln3^−/− mice without and with prolapse. Coomassie staining was conducted in side-by-side gels to ensure even protein loading. C: Quantification of FBLN-5 and TE in urea extract proteins from vaginal tissues of WT and Fbln3^−/− mice without and with prolapse. Data represent mean ± SEM of five to six tissues in each group. *P ≤ 0.05 compared with WT.

Figure 11

Desmosine content in the vaginal wall. Desmosine levels were determined in vaginal tissues from age-, cycle-, and strain-matched WT (n = 5) and Fbln3^−/− (KO) mice without (n = 5) and with prolapse (n = 7). Desmosine levels were expressed as the total amount of desmosine per vagina. *P ≤ 0.05 compared with WT.