Biomechanical function of the murine cervix with elastic fiber deficiency

Abstract

Preterm birth (PTB) is the leading cause of infant death globally. Improper softening and dilation of the cervix due to cervical insufficiency is a major contributor to PTB. However, the processes describing why or how the cervix prematurely remodels are not well understood. Elastic fibers and smooth muscle cells contribute to contractile and mechanical stability in soft tissues, but their relationship in the cervix is not fully characterized. Towards this end, this study uses the fibulin-5 global knockout mouse model to investigate the impact of elastic fiber deficiency on the contractile and biomechanical behavior of the cervix. Cervixes from nulliparous fibulin-5 wildtype (Fbln5^+/+), haploinsufficient (Fbln5^+/−), and deficient (Fbln5^−/−) mice were explanted and subjected to biaxial contractile and biomechanical extension-inflation testing. Immunofluorescence was used to quantify the localization and expression of elastin and alpha-smooth muscle actin throughout the cervix. Elastin expression and the number of elastic fibers decreased with Fbln5 deficiency. Additionally, contractility and modulus were significantly reduced in the Fbln5^−/− cervix as a consequence of elastic fiber deficiency. Overall, this study showed that intact elastic fibers significantly contribute to biomechanical function of the cervix and may be an essential target for future PTB investigations.

1. Introduction

Preterm birth (PTB) – defined as delivery prior to 37 weeks gestation – affects 1 in 10 babies globally and is the leading cause of infant death (Perin et al., 2022; Ohuma et al., 2023). Premature delivery can also result in adverse, prolonged physical, psychological, and financial effects for both the infant and parents (Pace et al., 2020; Pravia and Benny, 2020; Lakshmanan et al., 2022). Cervical insufficiency occurs when the cervix prematurely dilates and softens asymptomatically leading to PTB (Vink and Feltovich, 2016). While difficult to clinically diagnose, the result is the mechanical dysfunction of the cervix which may be due, in part, to improper remodeling of the cervical smooth muscle cells (SMC) and extracellular matrix (ECM) (Vink and Mourad, 2017).

To facilitate labor and delivery, the ECM and cells within the cervix undergo a drastic remodeling process that takes an originally stiff tissue into a compliant structure (Leppert, 1995; Word et al., 2007). One of the principal cellular components of the cervix are SMCs which are critical for contractile behavior and maintaining proper cervical shape (Gravina et al., 2014; Vink et al., 2016). In the nonpregnant state, the cervix remains contracted and closed to protect the uterus from the external environment. During pregnancy, the cervix must also remain closed to protect the uterus and withstand the mechanical forces imposed by the growing fetus (Myers et al., 2015); however, at the onset of labor, the cervix rapidly softens and dilates for safe passage and delivery of the fetus (Word et al., 2007; Nott et al., 2016). Functional elastic fibers – defined as an elastin core surrounded by a mantle of elastin-associated microfibrils – promote cervical dilation as they permit the cervix to both stretch during delivery and recoil back to its original shape and size postpartum (Danforth, 1983; Leppert et al., 1986). Additionally, proper elastic fiber formation and composition may also contribute to contractile homeostasis in multiple soft tissues (Karnik et al., 2003; Murtada et al., 2016; Clark-Patterson et al., 2021; White et al., 2024). Pelvic floor disorders, such as pelvic organ prolapse, are associated with disrupted elastic fibers and decreased elastic fiber components (Karam et al., 2007; Li et al., 2009; Jameson et al., 2020). Elastic fibers are also decreased in the human cervix affected by cervical insufficiency (Leppert et al., 1987).

Higher rates of preterm birth exist for those with connective tissue disorders, but the cause of cervical insufficiency in those without established risk factors remains unclear (Anum et al., 2009; Curry et al., 2014). Further, it is unknown if tissue contractility changes following loss of elastic fiber functionality in the cervix. Prior reports have evaluated uniaxial contractile and biomechanical properties of the cervix from rodents (Darios et al., 2012; Ferland et al., 2015; Yoshida et al., 2016) and humans (Petersen et al., 1991; Ekerhovd et al., 2000; Vink et al., 2016). While these methods provided crucial information about cervical behavior, it should be noted that the cervix is loaded multi-axially in the body and demonstrates anisotropic behavior which cannot be fully captured by uniaxial testing alone (Aspden, 1988; Myers et al., 2010; Wu et al., 2024). Thus, there is a need to evaluate contractile and biomechanical properties of the cervix in multiple directions. Biaxial extension-inflation testing enables systematic evaluation of circumferential and longitudinal behavior via pressurization and extension while keeping the geometry of the murine cervix intact (Conway et al., 2021). Furthermore, this loading environment is better suited to recapitulate physiologically relevant deformations and thus obtain meaningful data on cervical function. However, to better understand the role of cervical elastic fibers, there is also a need to quantify these direction-dependent properties in the presence of elastic fiber disruption in the nulliparous (never pregnant) cervix. Establishing the role of elastic fibers in promoting cervical behavior in the nonpregnant state is the first step toward understanding critical relationships of elastic fibers in normal cervical function with implications for future cervical insufficiency research.

While not a complete representation of human tissue, mouse models are ubiquitous tools for better understanding the biological, cellular, and microstructural processes underlying tissue remodeling. Mice are widely accepted models in reproductive and preterm birth research due to a well-established estrous cycle and similarities in tissue composition and function to humans (Dudley et al., 1996; Elovitz et al., 2003; Yoshida et al., 2016; O'Brien et al., 2017; Conway et al., 2021; Suarez et al., 2024, 2026; Domingo et al., 2025). Further, knockout mouse models enable selective evaluation of the contribution of native ECM proteins – such as fibulin-5 (gene: Fbln5), a microfibril glycoprotein critical for proper elastic fiber formation and maturation (Nakamura et al., 2002; Yanagisawa et al., 2002) – on tissue structure and function. Fbln5-deficient mice are a model of severe elastinopathy in both pelvic floor and vasculature research as they readily develop pelvic organ prolapse and hallmarks of accelerated cardiovascular aging including tortuous arteries (Yanagisawa et al., 2002; Drewes et al., 2007; Wan et al., 2010; Ferruzzi et al., 2015; Jennings et al., 2024). Consistent with its critical role in elastic fiber formation, Fbln5 deficient mice exhibit disorganized elastic fibers in common carotid arteries, compared to the wildtype (Wan et al., 2010). Similarly, in the Fbln5 deficient vagina, elastic fibers present as large elastin globules rather than the mature elastic fibers with an intact elastin core observed in the wildtype (Rahn et al., 2009). Elastic fiber integrity correlates with changes in other ECM components such as collagen organization and microstructure, the component responsible for imparting tensile strength to soft biological tissues. Indeed, alterations in elastic fiber integrity and subsequent collagen microstructure coincide with an increase in tissue modulus and reduced contractility (Murtada et al., 2016; Clark et al., 2019; White et al., 2024), suggesting well-formed elastic fibers are crucial to maintaining the ECM environment necessary for functional contractility in soft tissues. For example, collagen fiber undulation decreased alongside malformed elastic fibers, resulting in arterial stiffening (Wan et al., 2010; Ferruzzi et al., 2011). This is consistent with observations of increased passive modulus of the murine vagina following elastic fiber disruption via porcine pancreatic elastase (Akintunde et al., 2019; White et al., 2022). Fbln5 deficient mice exhibit decreased contractility in the vagina (Clark-Patterson et al., 2021) alongside molecular disorganization and morphological disruption of cervical collagen (Jennings et al., 2024); however, the impact of improper elastic fiber formation on cervical contractility remains unknown. Thus, the study herein seeks to leverage the Fbln5-deficient mouse to determine the impact of compromised elastic fibers on nulliparous cervical contractility and biomechanical properties.

The objectives of this study are to evaluate the effect of Fbln5 deficiency on the contractile and passive biomechanical properties and microstructural composition of the nulliparous murine cervix. Biaxial extension-inflation testing will be used to quantify contractility and modulus in the circumferential and longitudinal directions, and changes in cervical microstructure will be evaluated by immunofluorescent staining for elastin and alpha smooth muscle actin (αSMA). Improper elastic fiber formation coincides with decreased aorta (Murtada et al., 2016) and vaginal (Clark-Patterson et al., 2021) contractility. Furthermore, enzymatic disruption of elastic fibers in the vagina resulted in decreased contractility and increased passive modulus (Akintunde et al., 2019; Clark et al., 2019; White et al., 2022, 2024). As such, we hypothesize that a similar structure-function relationship exists in the cervix wherein cervical contractility decreases and passive modulus increases with elastic fiber deficiency. Improving our understanding of the relationship between biomechanical behavior and ECM components in the nulliparous cervix may help build fundamental knowledge related to cervical function. This may, in turn, provide insight into potential biological targets for intervention to reduce the risk of cervical insufficiency and PTB.

2. Methods

2.1. Animal use

This study used a total of 43 nulliparous female mice aged 3-6 months. The Institute of Animal Care and Use Committee (IACUC) at Tulane University approved all animal care and experimental protocols in this study. Mice used in this study were generated from mating male and female Fbln5 haploinsufficient breeding pairs on mixed genetic backgrounds (C57BL/6 x 129SvEv) (Yanagisawa et al., 2002). Animals were housed in microisolators under a 12-h light-dark cycle, with access to standard chow and water ad libitum. At weaning, tail snips were sent to a commercial vendor (Transnetyx, Cordova, TN, USA) for genotyping via real-time PCR. Mice were divided into three age-matched experimental groups based on Fbln5 genotype: wildtype (Fbln5^+/+), haploinsufficient (Fbln5^+/−), and deficient (Fbln5^−/−). All experiments in this study were performed on cervical tissue at estrus. Visual inspection of the vagina for estrous cycle staging is a widely accepted method in rodents as it is noninvasive, does not mechanically manipulate the tissue, prevents pseudopregnancy, and is a relatively quick process (Champlin et al., 1973; Byers et al., 2012). To determine estrus, the tail of the mouse was lifted and the external vaginal introitus was inspected for dilation, lack of pink coloration, and pronounced striations (Byers et al., 2012).

2.2. Pressure catheterization

To inform biomechanical testing parameters, in vivo transcervical pressure measurements were collected in Fbln5^+/+ (n = 5), Fbln5^+/− (n = 5), and Fbln5^−/− (n = 3) mice at estrus as previously described (Conway et al., 2021). Briefly, mice were anesthetized with 1.5% isoflurane mixed with 100% O₂ under a heat lamp. A 2F Millar Mikro-tip^® Catheter (ADInstruments, Colorado, USA) was inserted into the cervix, approximately 5-6 mm from the vaginal introitus. Proper catheter placement was determined by an increase in pressure and the presence of phasic contractions in the cervix, which occur during the estrus phase of the estrous cycle (Gravina et al., 2014; Robuck et al., 2018). In vivo pressure was determined as the average pressure between subsequent phasic contractions during a 5-min recording period (see Fig. 1A) (Robuck et al., 2018; Conway et al., 2021).

Fig. 1.

(A) Representative in vivo cervical pressure recording. Boxed section indicates baseline pressures used for biomechanical characterization; values were averaged between contractions to report baseline in vivo pressure values. Circled point indicate maximum pressure; peaks of each contraction were averaged to report maximum in vivo pressure. (B) Average baseline pressures for the Fbln5^+/+ (10.0 ± 2.0 mmHg, black; n = 5), Fbln5^+/− (11.0 ± 2.0 mmHg, grey; n = 5), and Fbln5^−/− (7.0 ± 3.0 mmHg, light grey; n = 3) cervix. (C) Average maximum in vivo pressures for the Fbln5^+/+ (39.0 ± 11.0 mmHg), Fbln5^+/− (47.0 ± 21.0 mmHg), and Fbln5^−/− (44.0 ± 19.0 mmHg) cervix.

2.3. Contractility Testing

To ensure consistent cervical contractility measurements, mice were sacrificed via guillotine (n = 5/group) at estrus as described in Section 2.1 (Conway et al., 2021). The entire reproductive tract was immediately dissected and placed in 4°C Hanks Balanced Saline Solution (HBSS). The cervix was isolated and cannulated onto an extension-inflation device (Danish MyoTechnologies, Aarhus, Denmark) as previously described (Suppl. Fig. 1) (Conway et al., 2021). HBSS was replaced with 37°C Krebs Ringer Buffer (KRB) and aerated with 95% O₂ and 5% CO₂ to maintain a physiologic pH of 7.4. To establish a reference configuration for biomechanical analysis, unloaded length and diameter were recorded at the longitudinal position where the organ appeared slightly buckled and was not collapsed (Robison et al., 2017; White et al., 2019).

Following protocols established in the vasculature, multi-step preconditioning is performed to minimize hysteresis and ensure a consistent, repeatable, and mathematically interpretable response (Fung, 1993; Humphrey, 2010; Ferruzzi et al., 2013; Robison et al., 2017; Conway et al., 2021). Cervixes were first preconditioned circumferentially at the unloaded length by continuous pressurization from 0 to a maximum pressure of 28.0 mmHg at 1.5 mmHg/s for five cycles. Maximum pressure (28 mmHg) was calculated as one standard deviation below the mean maximum pressure from Fbln5^+/+ pressure catheter experiments (Fig. 1C). No statistically significant differences in maximum pressure were observed between genotypes; thus, all samples were pressurized to 28.0 mmHg. Consistent pressurization to 28.0 mmHg at a rate of 1.5 mmHg/s maintained tissue viability during physiologically relevant load application (Conway et al., 2021). The pressurization rate ensured any time-dependent effects were minimized (Zhou et al., 2015; Robison et al., 2017; Conway et al., 2021).

Explanting the cervix and removing neighboring organs results in tissue retraction (Ferruzzi et al., 2013; Conway et al., 2021). As such, the tissue was manually extended using a micrometer on a linear stage to identify and restore the estimated physiological length, assumed to be the length where the force remained constant over a range of increasing pressures (Van Loon, 1977; Ferruzzi et al., 2013; Conway et al., 2019) and measure properties under physiologically relevant conditions. Five cycles of circumferential and longitudinal preconditioning were performed to minimize hysteresis and ensure reproducible measurements in both directions (Ferruzzi et al., 2013; Robison et al., 2017; White et al., 2019). Circumferential preconditioning (cyclic pressurization from 0 to 28.0 mmHg) was repeated at the estimated physiologic length, and longitudinal preconditioning (cyclic extension between −1% and +1% of the estimated physiologic length) was performed at 9.0 mmHg, approximately 1/3 of the maximum pressure in the mechanical testing protocol (Conway et al., 2021).

Repeatable contractile measurements and tissue viability were confirmed with cervical contractility preconditioning (Zhou et al., 2015; Murtada et al., 2016; Clark et al., 2019; Conway et al., 2021) that included induction of maximum sustained contractions for 5 min with an optimal dose of 20 mM of potassium chloride (KCl) (Conway et al., 2021). After confirmation, the bath was replaced with fresh KRB, and the unloaded geometry was re-established, followed by a 10 min equilibration period.

Isometric–isobaric dose-response experiments were conducted using KCl to evaluate biaxial cervical contractility (Tankó et al., 1998; Clark et al., 2019). Cervixes from each Fbln5 genotype were loaded to an internal pressure equal to the baseline value determined from the pressure catheter experiments (Fbln5^+/+ = 10.0 mmHg, Fbln5^+/− = 11.0 mmHg, Fbln5^−/− = 7.0 mmHg; see Fig. 1B), and the tissue was set to the physiological length. 10 mM of KCl was added to the bath to allow the tissue to spontaneously contract while changes in outer diameter and longitudinal force were continuously recorded. After 5 min, the bath was replaced with fresh KRB, and the tissue was allowed to equilibrate until the outer diameter returned to its pre-contraction value. This process was repeated for increasing KCl doses of 15, 20, 30, 40, 50, 60, 80, and 100 mM to generate biaxial dose-response curves (Conway et al., 2021).

2.4. Passive biomechanical testing

A separate cohort of mice were sacrificed via guillotine at estrus, and the cervix was isolated and cannulated (n = 5/group) for passive biomechanical characterization. Cervixes were first incubated in calcium-free KRB with 2 mM of egtazic acid for 30 min to deactivate contractile components, such as SMC, and their contractile contributions to the passive biomechanical response (Guan et al., 1988). After removing contractile activity, the cervix was subjected to biaxial preconditioning and estimation of the unloaded and physiological geometry (see 2.3 Contractility Testing). Passive biomechanical characterization consisted of seven mechanical testing protocols. Briefly, cervixes underwent pressure-diameter tests at three fixed longitudinal lengths consisting of three pressurization cycles. Intraluminal pressure was applied in a stepwise manner from 0 to 28 mmHg in 1 mmHg increments at a rate of 1.5 mmHg/s at the physiologic length and ±1% of the physiologic length. This quasistatic loading approach ensures data is collected at each pressure in the experimental testing range (Humphrey, 2010; White et al., 2019). Next, force–length tests at four fixed internal pressures consisted of three cycles of longitudinal extension from 1% below to 1% above the physiologic length at a constant pressure of 3 (tare), 9, 18, and 28 mmHg.

Data from the final loading cycle from each testing protocol was used in the biomechanical analysis. After testing, the cervix was returned to the unloaded state, and cross-sectional ultrasound B-mode images (Vevo2100, FUJIFILM VisualSonics, Inc., Toronto, Canada) were taken with a 40 MHz transducer (LZ550, resolution: 30 μm) to measure the cervical wall thickness. Images were acquired at the center of the cervix near the location where the diameter was optically tracked during biomechanical testing. Using ImageJ (NIH, Bethesda, MD), the lengths of 25 transmural lines around the circumference of the tissue were manually traced (Suppl. Fig. 2) and averaged to provide an estimate of the unloaded wall thickness. This process was performed in triplicate and averaged to determine the thickness of the tissue.

2.5. Contractility and biomechanical testing data analysis

Under fixed intraluminal pressure (isobaric), circumferential contractility was calculated as the change in outer diameter between the maximally contracted state (OD_contract) and the pre-contraction baseline state (OD_baseline) (Eq (1)). Because tissue length was fixed longitudinally (isometric), longitudinal contractility was defined as the change in longitudinal force between the contracted (F_contract) and baseline state (F_baseline), as measured by the inline force transducer (Eq (2)) (Tankó et al., 1998; Ferruzzi et al., 2016; Murtada et al., 2016; Conway et al., 2021; White et al., 2024).

$$CircumferentialContractiliy\to \Delta OuterDiameter=O{D}_{contract}-O{D}_{baseline}$$ (1)

$$LongitudinalContractility\to \Delta LongitudinalForce=F_{contract}-F_{baseline}$$ (2)

Cervical geometry was assumed to be a cylinder with uniform diameter and wall thickness. Unloaded volume ($\overline{V}$, Eq (3)) was determined from the experimentally measured unloaded geometry where $R_o$ is the unloaded outer radius, $H$ is the unloaded thickness, and $L$ is the unloaded length.

$$\overline{V}=\pi (R_o^2-{(R_o-H)}^2)L$$ (3)

Following the incompressibility assumption that volume remains constant across all quasistatic loading conditions during mechanical test, the deformed inner radius ($r_i$) and wall thickness ($h$) were thus calculated from the deformed outer radius ($r_0$) and longitudinal length ($l$).

$$r_i=\sqrt{r_o^2-\frac{\overline{V}}{\pi l}}$$ (4)

$$h=r_o-r_i$$ (5)

Circumferential stretch (${\lambda }_{\theta }$) was defined based on the mid-wall radius and was calculated as the deformed to unloaded mid-wall radius ratio (Eq (6)). Longitudinal stretch (${\lambda }_z$) was similarly defined as the ratio of the deformed to unloaded longitudinal length (Eq (7)).

$${\lambda }_{\theta }=\frac{r_i+h∕2}{R_i+H∕2}$$ (6)

$${\lambda }_z={}^l∕_L$$ (7)

Further, wall-averaged circumferential (${\sigma }_{\theta }$; Eq (8)) and longitudinal (${\sigma }_z$; Eq (9)) Cauchy stresses were determined from the intraluminal pressure ($P$), transducer-measured longitudinal force ($F_t$) and the corresponding deformed outer radius ($r_o$) and inner radius ($r_i$) (Humphrey, 2010; Ferruzzi et al., 2013).

$${\sigma }_{\theta }=\frac{P{r}_i}{(r_o-r_i)}$$ (8)

$${\sigma }_z=\frac{F_t+\pi {Pr}_i^2}{\pi (r_o^2-r_i^2)}$$ (9)

Passive modulus was calculated as the local tangent, or slope, of the nonlinear stress-stretch curve in the circumferential (Eq (10)) or longitudinal (Eq (11)) directions within a physiological pressure range as determined by in vivo pressure catheterization (see Section 2.2 Pressure Catheterization) (Stoka et al., 2018; Conway et al., 2021). The upper pressure boundary (UPB) was defined as one standard deviation above, and the lower pressure boundary (LPB) was one standard deviation below, the average baseline pressure for Fbln5^+/+: 10.0 ± 2.0 mmHg, Fbln5^+/−: 11.0 ± 2.0 mmHg, and Fbln5^−/−: 7.0 ± 3.0 mmHg (Fig. 1B). The slope within this range of pressures on the stress-stretch curves quantified modulus (Suppl. Fig. 3) (White et al., 2019, 2024; Conway et al., 2021).

$$CircumferentialModulus=\frac{{\sigma }_{\theta }^{UPB}-{\sigma }_{\theta }^{LPB}}{{\lambda }_{\theta }^{UPB}-{\lambda }_{\theta }^{LPB}}$$ (10)

$$LongitudinalModulus=\frac{{\sigma }_z^{UPB}-{\sigma }_z^{LPB}}{{\lambda }_z^{UPB}-{\lambda }_z^{LPB}}$$ (11)

2.6. Immunofluorescence

Cervix tissue was embedded in OCT and cryosectioned at 8 μm in the cross-sectional plane. The tissue was then fixed/permeabilized in a solution of 10% neutral buffered formalin (Millipore Sigma, HT501128) and 0.1% Triton X-100 (Sigma-Aldrich, x100) prior to blocking with 5% bovine serum albumin (BSA; Sigma-Aldrich, A3059) in phosphate buffered saline (PBS). Staining with a primary antibody against murine tropoelastin exons 6-17 (1:300) was done for 4 h at room temperature, followed by a 1-h incubation with a Cy5 secondary antibody (Abcam, ab6564) and conjugated αSMA-Cy3 monoclonal antibody (Millipore Sigma, C6198) (both at 1:300) (Kozel et al., 2003; J DeMarsilis, 2014). Slides were mounted with Prolong gold anti-fade reagent with DAPI to label cell nuclei. Images were acquired at 20× magnification with an Olympus BX53 microscope and Olympus DP80 dual-sensor CCD camera. Images in each fluorescent channel were acquired at consistent exposure settings to allow for quantitative image analysis.

Four regions of interest spanning the stromal area from the inner cervical lumen to the outer vaginal lumen, were identified and imaged within each sample. Elastin and smooth muscle area fractions were calculated from positive immunofluorescent staining within each region of interest using FIJI (NIH) (Schindelin et al., 2012). In addition to analysis of total elastin area, elastin staining was partitioned into smaller elastin particles (structures with area ranging from 5 to 15 pixels) and larger elastin bundles (structures with area >15 pixels). Elastin particles represent shortened elastin fibers, or small clusters of soluble tropoelastin monomers, while elastin bundles are more representative of traditional mature elastic fibers. Individual elastic fibers were identified from the elastin immunofluorescence images via CT-FIRE, an open-source MATLAB-based image analysis tool (Bredfeldt et al., 2014).

2.7. Statistical analysis

All statistical analyses were performed in RStudio (4.3.1). Levene and Shapiro-Wilk tests assessed the homogeneity of variances and the normality of each experimental measure, respectively. Mixed ANOVAs designed with a within-subject factor of KCl concentration and a between-subject factor of genotype, followed by multiple comparison testing with Bonferroni correction, evaluated the differences in circumferential and longitudinal contractility. A one-way ANOVA evaluated differences in geometry, biomechanical properties, and microstructural properties across genotypes. If applicable, multiple comparison testing with Bonferroni corrections were performed. If the normality or equal variance was not guaranteed across groups, the nonparametric Kruskal–Wallis test followed by post hoc Dunn's test for multiple comparisons – a nonparametric alternative to the one-way ANOVA – was performed.

3. Results

3.1. The effect of Fbln5 genotype on cervix contractility

To investigate changes in cervical contractile function due to Fbln5 genotype, the change in outer diameter and longitudinal force were measured in response to varied KCl concentrations. In the circumferential direction, genotype had a significant effect on cervical contractility (p < 0.0001), but KCl concentration did not (p = 0.22) (Fig. 2A). Circumferential contractility was significantly decreased in the Fbln5^−/− cervix when compared to the Fbln5^+/+ (p < 0.0001) and Fbln5^+/− cervix (p < 0.0001). No significant differences were identified between genotype or KCl concentration in the longitudinal direction (Fig. 2B). Regardless, the Fbln5-deficient cervix had the lowest contractile response, and the Fbln5^+/− cervix had the highest contractile response in both directions for all concentrations studied. In each genotype, SMCs were likely the primary cell type responsible for the observed changes in cervical contractility (Bryman et al., 1985; Vink et al., 2016; Conway et al., 2021). All samples exhibited phasic contractions once mounted onto the extension-inflation device in standard KRB containing 4.7 mM KCl. Maximum contractility was identified in all genotypes at 20 mM KCl, consistent with the nulliparous CD-1 cervix (Conway et al., 2021). From KCl doses 4.7 mM to 20 mM, samples exhibited a phasic response (Suppl. Fig. 4A); for KCl doses at and above 30 mM, all samples exhibited a tonic response (Suppl. Fig. 4B).

Fig. 2.

Biaxial contractility results showing KCl dose responses for the Fbln5^+/+ (black filled circle), Fbln5^+/− (grey-filled circle), and Fbln5^−/− (black open circle) cervix (n = 5/group). (A) Mixed ANOVA showed a significant decrease in the change in outer diameter following KCl treatment (circumferential contractility) in the Fbln5^−/− cervix compared to Fbln5^+/+ (p < 0.0001) and Fbln5^+/− (p < 0.0001). No significant differences were found between the Fbln5^+/+ and Fbln5^+/− cervix. (B) The change in longitudinal force in response to KCl treatment (longitudinal contractility), as measured via the force transducer, showed no significant differences between genotypes. Data is reported as mean ± SEM, ****p < 0.0001, denotes statistical significance, n = 5/group.

3.2. The effect of Fbln5 genotype on geometry and passive biomechanics

Outer diameter at the physiologic in vivo pressure was decreased in the Fbln5^−/− cervix compared to Fbln5^+/− (p = 0.03) and Fbln5^+/+ (p = 0.07) (Fig. 3A). While not significant (p = 0.07), physiologic length increased with Fbln5 deficiency (Fig. 3B). Outer diameter as a function of pressure at the estimated in vivo length demonstrated a decrease in outer diameter in the Fbln5^−/− cervix, as represented by the downward shift of the curve (Fig. 3C). While pressure and genotype had a significant interaction, a simple main effect identified that genotype had a significant effect on outer diameter at all pressures (p < 0.05).

Fig. 3.

Cervical geometry measured during passive mechanical testing. Outer diameter at the measured physiologic pressure determined via pressure catheter (A) and the estimated physiological length (B) of the Fbln5^+/+ (black), Fbln5^+/− (grey), and Fbln5^−/− (light grey) cervix. Outer diameter decreased in Fbln5^−/− tissue compared to Fbln5^+/− (p = 0.03) and Fbln5^+/+ (p = 0.07). While not significant, length tended to increase with Fbln5 deficiency. (C) Outer diameter as a function of pressure from the Fbln5^+/+ (black filled circle), Fbln5^+/− (grey filled circle), and Fbln5^−/− (black open circle) cervix at the in vivo length. Genotype significantly affected outer diameter at all pressures (p < 0.05). Fbln5^−/− decreased cervical outer diameter as shown by the downward shift of the curve. Data are reported as mean ± SEM, *p < 0.05 denotes statistical significance, n = 5/group.

The Fbln5^−/− cervix was more distensible than that from Fbln5^+/+ and Fbln5^+/− mice in both the circumferential (Fig. 4A) and longitudinal (Fig. 4B) directions, as represented by the rightward shift of the stress-stretch curves. When calculated at the physiological length and mean in vivo pressure, circumferential stress (Fig. 4C) was significantly decreased in the Fbln5^−/− cervix compared to Fbln5^+/+ (p = 0.006) and Fbln5^+/− (p = 0.04). Fbln5 deficiency also significantly affected longitudinal stress at the physiologic length and mean in vivo pressure. Namely, Fbln5^−/− cervix exhibited significantly lower longitudinal stress than Fbln5^+/+ (p = 0.004) and Fbln5^+/− (p = 0.03) cervix (Fig. 4D).

Fig. 4.

Circumferential (circles) (A) and longitudinal (squares) (B) stress versus circumferential stretch curves from the Fbln5^+/+ (black filled), Fbln5^+/− (grey filled), and Fbln5^−/− (black open) cervix. Fbln5 deficiency increased distensibility in both the circumferential and longitudinal directions. Data are reported as mean ± SEM, n = 5/group. Stress analysis in the circumferential (C) and longitudinal (D) direction at physiologic length and pressure for the Fbln5^+/+ (black), Fbln5^+/− (grey), and Fbln5^−/− (light grey) cervix. Both circumferential and longitudinal stress were significantly decreased in the Fbln5^−/− cervix compared to the Fbln5^+/+ (p = 0.006 and p = 0.004, respectively) and Fbln5^+/− (p = 0.04 and p = 0.03, respectively) cervix. **p < 0.01, *p < 0.05 denotes statistical significance, n = 5/group.

Modulus was quantified as the slope of nonlinear stress-stretch curves within a physiologic pressure range of loading (Fbln5^+/+ = 10.0 ± 2.0 mmHg, Fbln5^+/− = 11.0 ± 2.0 mmHg, Fbln5^−/− = 7.0 ± 3.0 mmHg, see Fig. 1) (Conway et al., 2021). Similar to stress, modulus was significantly decreased in the Fbln5^−/− cervix compared to Fbln5^+/+ in both the circumferential (Fig. 5A) and longitudinal (Fig. 5B) direction (p = 0.007 and p = 0.007, respectively). While not statistically significant, biaxial modulus was also reduced in the Fbln5^+/− cervix compared to Fbln5^+/+. Altogether, this data suggests that proper elastic fiber formation is critical for passive mechanical behavior of the cervix.

Fig. 5.

Circumferential (circles) (A) and longitudinal (squares) (B) modulus of the Fbln5^+/+ (black), Fbln5^+/− (grey) and Fbln5^−/− (light grey) cervix. Modulus was significantly decreased in the Fbln5^−/− cervix relative to Fbln5^+/+ in both the circumferential (p = 0.007) and longitudinal (p = 0.007) direction. **p < 0.01 denotes statistical significance, n = 5/group.

3.3. Changes in cervical microstructure with Fbln5 deficiency

Immunofluorescent labeling of cervical regions of interest (ROIs) revealed alterations in tissue microstructure (Fig. 6A) and elastin composition (Fig. 6B) across Fbln5 genotypes. First, the overall tissue ROI area examined between cervical and vaginal lumens was statistically similar across groups (Fig. 6C). Further, the area percentage of αSMA-positive staining did not significantly differ across genotypes (Fig. 6D). However, the area percentage of elastin-positive staining was significantly impacted by Fbln5 deficiency (p = 0.02) (Fig. 6E). Post hoc tests showed that elastin percentage was reduced Fbln5^+/− cervix and significantly decreased in the Fbln5^−/− cervix, compared to that of Fbln5^+/+ (p = 0.02). Elastin bundles (longer mature elastin fibers, structures with area >15 pixels) and particles (shorter fibers or tropoelastin monomers, structures with area ranging from 5 to 15 pixels) areas were also found to be significantly impacted by Fbln5 deficiency (p = 0.001, p = 0.001) (Fig. 6F). The percentage of elastin classified as bundles was significantly decreased compared to Fbln5^+/+ (p = 0.003) and Fbln5^+/− (p = 0.002) cervix while the normalized area of elastin particles significantly increased in the Fbln5^−/− cervix compared to Fbln5^+/+ (p = 0.003) and Fbln5^+/− (p = 0.002) (Fig. 6G).

Fig. 6.

(A) Immunofluorescence staining of αSMA (yellow), elastin (red), and cell nuclei (DAPI, blue) in the Fbln5^+/+, Fbln5^+/−, and Fbln5^−/− cervix. Images were taken spanning a ROI between the cervical lumen (CL) and vaginal lumen (VL). Scalebar:100 μm. (B) Elastin images were further classified into the fraction of either elastin bundles (>15 pix) or elastin particles (5-15 pix). Scalebar: 50 μm. (C) Tissue area, and (D) αSMA-positive area did not significantly differ across groups. (E) Elastin-positive area in the cervical wall significantly decreased in the Fbln5^−/− compared to Fbln5^+/+ (p = 0.02). Further, the Fbln5^−/− cervix demonstrated a decrease in the (F) fraction of elastin bundles and an increase in the fraction of (G) elastin particles compared to Fbln5^+/+ (p = 0.003 and p = 0.003, respectively) and Fbln5^+/− (p = 0.002 and p = 0.002, respectively). At least 4 independent ROIs were images per sample. **p < 0.01, *p < 0.05 denotes statistical significance, n = 5/group.

The open-source image processing software CT-FIRE was used to identify elastin fibers and measure the width and length of each individual fiber within the cervical stroma of each genotype (Fig. 7A). Histograms of fiber length and width showed minimal changes in both parameters across the three genotypes. However, downward shifts in the Fbln5^−/− curves for both elastin length and width suggested that Fbln5 deficiency resulted in a significant decrease in the number of elastin fibers identified (Fig. 7B-C). The total number of fibers significantly decreased in the Fbln5^−/− cervix compared to the Fbln5^+/+ (p = 0.01) and Fbln5^+/− (p = 0.04) cervix (Fig. 7D).

Fig. 7.

(A) Immunofluorescent labeling of elastin (red) and CT-FIRE fiber overlay (green) in the cervical stroma, as defined by the luminal boundaries (CL = cervical lumen, VL = vaginal lumen). Scalebar: 100 μm. Histograms demonstrate the distribution of elastin fiber (B) length and (C) width within the Fbln5^+/+ (black), Fbln5^+/− (grey), and Fbln5^−/− (light grey) cervix. Mean length and width are also presented. (D) The number of elastin fibers significantly decreased in the Fbln5^−/− cervix compared to the Fbln5^+/+ (p = 0.01) and Fbln5^+/− genotypes (p = 0.04). At least 4 independent ROIs were imaged per sample. **p < 0.01, *p < 0.05 denotes statistical significance, n = 5/group.

The midstroma region of the cervix was defined as the area where αSMA-positive staining indicated the presence of contractile cells (Fig. 8A) (Vink et al., 2016; Conway et al., 2021).Using this definition, no statistically significant differences were identified in the midstroma area between genotypes (p = 0.09) (Fig. 8B). However, the elastin-positive percentage (Fig. 8C) and elastin bundle area (Fig. 8D) within the midstroma decreased with Fbln5 deficiency compared to Fbln5^+/+ (p = 0.005 and p = 0.009, respectively). The subepithelial stroma demonstrated similar trends in elastin area percentage as observed in the midstroma region, but no statistically significant differences were identified (Fig. 8E). However, subepithelial stromal elastin bundle area (Fig. 8F) was significantly decreased in the Fbln5^−/− cervix compared to Fbln5^+/+ (p = 0.04).

Fig. 8.

(A) The midstroma region (M) of a representative Fbln5^+/+ cervix was identified by the presence of αSMA-positive staining (yellow), while the remainder of the cervical wall was defined as the subepithelial-stroma region (S). Scalebar: 100 μm. (B) Midstroma area did not vary with Fbln5 genotype. (C) Elastin area percentage and (D) elastin bundle area significantly decreased in the Fbln5^−/− cervix compared to Fbln5^+/+ (p = 0.005 and p = 0.009, respectively). (E) Elastin expression in the subepithelial stroma region did not vary with Fbln5 disruption, whereas (F) elastin bundle area was significantly decreased in the Fbln5^−/− cervix compared to Fbln5^+/+ (p = 0.04). At least 4 independent ROIs were imaged per sample. **p < 0.01, *p < 0.05 denotes statistical significance, n = 5/group.

4. Discussion

This study investigated the contractile properties, passive mechanical behavior, and microstructural composition of the cervix in nulliparous Fbln5^+/+, Fbln5^+/−, and Fbln5^−/− mice. Using biaxial extension-inflation approaches, we showed that contractility, stress, and modulus at physiologic levels of load were decreased with Fbln5 deficiency compared to wildtype controls. We also determined that cervical elastin content decreased with Fbln5 deficiency. Together, these findings suggest that elastic fibers may contribute to maintaining tissue stability in the nonpregnant cervix with further implications for better understanding the underlying biological processes that may contribute to cervical insufficiency and PTB.

4.1. Contractility was dependent on Fbln5 genotype and ECM properties

Contractile potential decreased in the Fbln5^−/− cervix compared to Fbln5^+/+ and Fbln5^+/− tissue (Fig. 2). However, αSMA expression did not change with genotype (Fig. 6D). Thus, the reduction in contractile behavior may be due to a combination of either a decrease in passive modulus of the tissue (Fig. 5) or compromised SMC function associated with Fbln5 deficiency. In vasculature, Fbln5 deficiency also decreases contractility compared to wildtype controls, but no changes in smooth muscle area fraction were identified (Ferruzzi et al., 2015; Murtada et al., 2016). However, modulus of the vasculature significantly increases with Fbln5 deficiency (Wan et al., 2010; Ferruzzi et al., 2015). Further, culturing isolated human cervical SMCs on hard pillars increases contractility compared to culture on soft pillars (Vink et al., 2021), suggesting matrix stiffness may influence the contractile potential of the cervix. Additionally, specific ECM components may be responsible for contractility, as cervical SMC cultured on soft collagen-coated pillars exhibit increased contractile force relative to those cultured on soft fibronectin pillars (Vink et al., 2021). No differences in contractility were identified between SMCs isolated from patients with premature cervical failure compared to those without (Vink et al., 2021). Rather than an inherent SMC defect, these results suggest ECM biomechanical properties and composition of specific proteins may be responsible for proper SMC behavior in the cervix.

In the current study, a significant decrease in elastin expression (Fig. 6E) and an overall reduction in the number of elastic fibers (Fig. 7D) were observed in the Fbln5-deficient cervix. Combined with the altered contractility in Fbln5 tissues, these results may suggest that elastin contributes, at least in part, to normal contractile behavior in the cervix. For example, contractility decreased in the Fbln5-deficient vagina in conjunction with a decrease in elastin area fraction (Clark-Patterson et al., 2021); whereas in the vasculature, elastin regulates contractility through phenotypic modulation and proliferation of SMCs (Karnik et al., 2003). Interestingly, the Fbln5^+/− group had the highest contractile potential at all doses of KCl in both directions (Fig. 2) despite having similar levels of elastin to the wildtype cervix (Figs. 6E and 7D). This may suggest a compensatory mechanism whereby the loss of one copy of the Fbln5 gene preserves contractile behavior. While results from this study offer insights into critical structure-function relationships in the nonpregnant cervix with elastic fiber deficiency, future studies are needed to investigate the role of pregnancy on cervical contractility with elastic fiber deficiency to gain a more comprehensive understanding of the relationship between these components and to ultimately better inform PTB research.

4.2. Cervical modulus decreased with Fbln5 deficiency

In this study, biaxial modulus was significantly decreased in the Fbln5-deficient cervix compared to the wildtype group (Fig. 5). Further, distensibility increased with Fbln5-deficiency compared to the Fbln5^+/+ and Fbln5^+/− groups (Fig. 4). We anticipated that the loss of functional elastic fibers would decrease distensibility thereby increasing the modulus of the cervix. However, our opposing results suggest that further exploration between elastic fiber assembly and the organization of other structural proteins in the cervix, such as collagen, is necessary. In support of these findings, collagen anisotropy and collagen fiber alignment have been shown to decrease in the Fbln5^−/− cervix compared to Fbln5^+/−, suggesting a higher degree of disordered collagen in the Fbln5^−/− cervical ECM (Jennings et al., 2024). Additionally, improper collagen fibril assembly in the cervix decreases both modulus and elastic fiber organization (Nallasamy et al., 2017) and collagen fiber undulation decreases after elastic fiber disruption in the common carotid artery (Ferruzzi et al., 2011). In the vagina, distensibility changes throughout prolapse progression with Fbln5 deficiency; with grade 2-3 prolapse, distensibility increases and collagen fiber alignment ratio decreases compared to wildtype control (Clark-Patterson et al., 2021). The decrease in collagen alignment and matrix organization may explain why modulus decreased in the Fbln5^−/− tissue, further emphasizing an important relationship between collagen and elastic fibers in the cervix.

When comparing our findings to preterm delivery in humans, the absence of Fbln5, decreased modulus (Fig. 5), and reduced elastin expression (Fig. 6E) aligns with prior observations of cervical softening prior to delivery (Nallasamy et al., 2017; Jayyosi et al., 2018). Women with a history of premature cervical failure often have an absence of intact elastic fibers (Leppert et al., 1987) as well as a decrease in collagen concentration and crosslink maturity (Vink et al., 2016; Wu et al., 2024). Collagen concentration and crosslink maturity ratios are directly correlated to softer cervical tissue mechanical properties (Read et al., 2007; Yoshida et al., 2014). Altogether, these results suggest softer cervical tissue may be due to microstructural alterations in elastic fibers and collagen content may be an indicator of elevated potential for cervical insufficiency. Future studies should consider collagen and elastic fiber content and organization in the cervix as potential targets when investigating the etiology of cervical insufficiency.

4.3. Elastin content was significantly altered with Fbln5 deficiency

In this study, Fbln5 deficiency significantly decreased elastin area percentage (Fig. 6E), elastin bundle area (large clusters; Fig. 6F) and the number of elastin fibers in the cervix (Fig. 7D) compared to Fbln5^+/+ and Fbln5^+/− groups. Elastin particle area (small clusters; Fig. 6G) was however significantly increased with Fbln5 deficiency, suggesting an accumulation of malformed elastic fibers in the cervical tissue. Furthermore, the decrease in elastin area was evident throughout the cervix as shown by consistent reductions in the midstroma and subepithelial stroma expression (Fig. 8C-F). Historically, the cervix was considered to be a primarily collagenous structure with minimal cellular and elastic fiber contributions (Danforth, 1947). Despite only making up 1-2% of the cervix (Leppert et al., 1982), the current study demonstrates that elastin is a critical component in retaining cervical contractility (Fig. 2) and biomechanical function (Figs. 4 and 5). Prior work in other pelvic organs demonstrates that elastin has a significant role in promoting tissue stability. In the vagina, reduced elastin expression and elastic fiber width was detected in patients with pelvic organ prolapse (Karam et al., 2007). Additionally, enzymatic disruption of elastic fibers with elastase treatment, which results in fragmented elastic fibers, reduces both contractility and distensibility and increases modulus in the murine vagina (Akintunde et al., 2019; Clark et al., 2019; White et al., 2022, 2024). Altogether, this may suggest reduced elastin expression influences elastic fiber architecture and functionality in soft tissues. As such, the role of elastic fibers may be valuable in future research investigating the mechanical failure of the cervix in PTB. Indeed, attempts to recapitulate the cervical microenvironment in microphysiological systems should consider the importance of elastic fibers in retaining the modulus of the cervix at homeostatic levels.

4.4. Contractility was significantly affected in the circumferential direction

Interestingly, circumferential contractility but not longitudinal contractility was significantly decreased with Fbln5 deficiency (Fig. 2A). However, immunofluorescence staining of cervical cross-sections did not detect changes in circumferential αSMA-positive expression between Fbln5 genotypes (Fig. 6C). In the murine cervix, a larger amount of SMCs are organized in the longitudinal direction which may explain why contractility persists in that direction despite the Fbln5 deficiency (Conway et al., 2021). Immunofluorescence of longitudinal tissue sections was not performed in this study, but future studies should determine direction dependent differences in SMC and elastic fiber content in the cervix. This study emphasizes direction-dependent behavior under normal cervical function but further highlights the importance of considering directional differences when designing PTB prevention and intervention strategies.

4.5. Limitations

The study presented herein is not without limitations. Notably, the cervix is innervated in vivo, and SMCs are receptive to neurotransmitters and hormones (Gravina et al., 2014). Although contractility in response to nerve-mediated or hormonal stimuli was not evaluated, KCl stimulation of SMCs at estrus enabled consistent assessment of contractility across all genotypes. The Fbln5-deficient vagina also consistently and repeatably contracts in response to KCl (Clark-Patterson et al., 2021). Investigating baseline contractile behavior with a general agonist provided insights into the relationship between elastic fibers and contractility independent of nerve or hormone mediated responses.

In this study, SMC could not be directly identified as αSMA expression could be attributed to any cell with a contractile apparatus, including myofibroblasts. In the absence of changes in the amount of SMCs, changes in cervical contractility may be attributed to SMC phenotypic modulation – the reversible change between contractile and synthetic SMC phenotypes due to hormonal, biomechanical, and environmental effects (Owens et al., 2004). This phenomenon is established in the vasculature, gastrointestinal, and uterine SMCs (Rensen et al., 2007; Shynlova et al., 2009; Nair et al., 2011) but has yet to be fully evaluated in the cervix. Future work is needed to characterize the specific SMC phenotypes throughout the cervix and determine whether modulation occurs in response to dynamic changes in the cervical ECM. Failure to retain a proper equilibrium between the SMCs and ECM may lead to reduced contractile behavior. In turn, this could contribute to a biomechanical failure of the cervix to remain stiff and closed during pregnancy thus leading to PTB.

Furthermore, due to the setup of the extension-inflation device, circumferential and longitudinal contractility were reported as different metrics and could not be directly compared to examine anisotropy. Specifically, in the circumferential direction, the cervix was maintained in isobaric conditions with the outer diameter allowed to move freely. During contraction, the cervical SMCs constrict to effectively reduce the tissue outer diameter. Therefore, the circumferential contractile response was quantified as the change in outer diameter (Ferruzzi et al., 2016; Clark et al., 2019; Conway et al., 2021). For the longitudinal direction, the cervix was maintained in isometric conditions and longitudinally fixed at each end of the tissue. Therefore, geometric changes during contraction could not be tracked. Readings from an in-line force transducer, however, enable measurement of the force exerted by the cervix on the cannula during contraction. As such, longitudinal contractility was quantified as the change in force (Ferruzzi et al., 2016; Clark et al., 2019; Conway et al., 2021). As the cervix is an anisotropic tissue (Myers et al., 2010; Wu et al., 2024), direct comparison of circumferential and longitudinal contractility will be useful in future studies to delineate if one has a larger contribution to cervical function. The study presented herein establishes a first step in identifying changes in cervical contractility in response to improper elastic fiber formation associated with Fbln5 deficiency.

Results from this study cannot directly calculate SMC contributions to the contractile response or pressure-diameter mechanical response. Basal tone mechanical testing permits evaluation of pseudoelastic mechanical properties while the SMC and/or other contractile cells are still in an “active” or contractile state (Clark et al., 2019). By subtracting the passive mechanical contribution from the basal mechanics, contribution from contractile cells could be inferred however, basal tone testing was not performed. c. Rather, cervical contractile function was evaluated by inducing maximum contraction via KCl dose-response as a complement to passive mechanics. While we infer SMC are the primary driver of contractile function in the cervix (Bryman et al., 1985; Vink et al., 2016; Conway et al., 2021), we could not directly determine how much of the maximum contraction was due to the SMC, other cells with a contractile apparatus (such as myofibroblasts), or the ECM itself. The pressure-diameter mechanical responses reported in this study were only evaluated in the “passive” state, wherein contractile responses were inhibited (Guan et al., 1988; Murtada et al., 2016). While the SMC themselves were still present in the tissue and may, in part, contribute to passive resistance to changes in shape (Gravina et al., 2014; Vink et al., 2016), we cannot directly delineate how much of the mechanical response was due contributions from SMC or other contractile cells versus the passive contribution from the ECM. Future studies characterizing the specific SMC contribution to cervical mechanical responses will be useful to further elucidate any potential relationship between SMC and biomechanical failure of the cervix in the case of spontaneous PTB.

Although elastic fibers impart the ability to recover after deformation in soft tissues, including the cervix (Mecham and Heuser, 1991; Jayyosi et al., 2018; Trębacz and Barzycka, 2023), the present study was designed as an initial characterization of pseudoelastic mechanics of the cervical ECM to understand how functional elastic fibers contribute to baseline biomechanical cervical function (Humphrey, 2010). Future studies characterizing viscoelastic behavior may be useful to understand elastic fiber recovery and elucidate any potential relationship between contractility and deleterious remodeling of the cervix in the case of spontaneous PTB and premature cervical remodeling.

Finally, the mouse model used in this study was a Fbln5 global knockout; therefore, all soft tissues were affected, including vasculature, skin, and other pelvic organs (Yanagisawa et al., 2002; Ferruzzi et al., 2015). These mice spontaneously develop pelvic organ prolapse with age such that severe prolapse often appears by 6 months of age. Excessive tissue remodeling is likely driven by aberrant ECM turnover and a heightened inflammatory response in animals that develop prolapse (Nakasaki et al., 2015; Bersi et al., 2017). All Fbln5^−/− mice included in this study were evaluated for prolapse via the mouse pelvic organ prolapse quantification system (Wieslander et al., 2009). Only Fbln5^−/− mice at grade 1 prolapse were used herein. Through investigation of Fbln5-deficient tissue prior to prolapse formation, this animal model provided critical information on the role of elastic fibers in the contractile and mechanical function of the non-pregnant cervix. Although hormones heavily influence the cervix and investigation of cervical contractility throughout gestation is of interest, we submit that initial studies in the nulliparous state must first be completed to establish a baseline understanding of cervical structure and function in health and disease.

5. Conclusions

In summary, proper elastic fiber formation is critical to the contractile and mechanical function of the murine cervix. The elastic fiber-deficient cervix of the Fbln5 mouse model resulted in a decreased contractile response and modulus. Additionally, the elastin area fraction and overall number of elastic fibers in the cervix decreased. Taken together, this study established a potential role for elastic fibers in maintaining proper ECM microstructure which may, in turn, contribute to cervical contractile and biomechanical function. This may provide insights into potential matrix-driven mechanisms of contractile dysfunction as seen in the case of PTB due to cervical insufficiency.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.jmbbm.2026.107562.

Data availability

The experimental data that support the findings of this publication are available in the Texas Data Repository (https://doi.org/10.18738/T8/JRXLPE).