Lysyl Oxidases Are Necessary for Myometrial Contractility and On-time Parturition in Mice

Abstract

The extracellular matrix (ECM) plays a pivotal role in the maintenance of tissue mechanical homeostasis. Collagens and elastic fibers are the most predominant fibrous ECM proteins providing tissue mechanical function through covalent cross-linking, which is mediated by the lysyl oxidase family of enzymes. In this study, the function of lysyl oxidases in maintaining the integrity of the extracellular matrix in the myometrium and its impact on parturition-timing was investigated. Gene and protein expression analyses demonstrate that a subset of the lysyl oxidase family of enzymes are highly induced in pregnant myometrium. Inhibition of the activity of the lysyl oxidase family of enzymes through β-aminopropionitrile (BAPN) delays parturition in mice, in part because of myometrial dysfunction. In BAPN-treated mice, the expression of genes encoding contraction-associated proteins such as connexin 43, oxytocin receptor, and prostaglandin synthase 2, is significantly reduced in the myometrium compared to the untreated control mice. Proteomic analysis revealed that the composition of the ECM is altered in response to BAPN treatment, which demonstrates that the inhibition of the activity of lysyl oxidases disrupted the integrity of the myometrial ECM. Our findings demonstrate that the lysyl oxidases-mediated ECM function is necessary for the myometrium to transition from a quiescent to a contractile phenotype at term for on-time parturition.

Preterm birth (PTB) is the primary cause of perinatal infant mortality. The rate of PTB ranges from 4% to 16% globally, with an estimated 13.4 million preterm babies born in 2020 [1]. In the United States, the rate of PTB in 2022 was 10.4% which is disturbingly high for a developed nation [2, 3]. Although the etiology of PTB is complex and multifactorial, premature myometrial contraction is a commonly observed clinical sign preceding preterm labor [4, 5]. Thus, understanding the mechanisms that regulate myometrial quiescence, and contractility will help us identify the risk factors and potential therapeutic targets to prevent premature myometrial contraction and its contribution to PTB. Our understanding of the mechanisms by which myometrium transitions from quiescence to contractility has improved over the decades including but not limited to fetal, hormonal, and inflammatory signaling pathways [6-9]. However, an understanding of the role of biomechanical signals in this process is still evolving.

Biomechanical signals are predominantly elicited by alterations to the structure and composition of extracellular matrix (ECM). In addition to its traditional role of providing tissue architectural and mechanical properties, the ECM is involved in the regulation of multiple cellular processes such as cell proliferation, differentiation, contractility, migration, and survival [10, 11]. The myometrium experiences progressive mechanical loading while supporting pregnancy [12]. To endure these biomechanical forces, the myometrial smooth muscle cells and fibroblasts presumably synthesize and remodel ECM. Among all known ECM components, collagen and elastic fibers are the predominant fibrous proteins known to provide tissue strength and resilience, respectively [13, 14]. The levels of collagen and elastin are increased in myometrium during pregnancy in mice, rats, and humans [15-18]. We have already demonstrated that the collagen and elastic fibers undergo structural reorganization in mouse myometrium from early to late pregnancy, indicating changes to synthesis, processing, and assembly [15].

The ultimate strength of collagen and elastic fibrous proteins is determined by the covalent crosslinking between lysine residues mediated by the lysyl oxidase family of enzymes [19-21]. This enzyme family consists of 5 genes: lysyl oxidase (LOX) and LOX like 1-4 (LOXL1-4). They encode enzymes with identical functions but exhibit tissue specific spatiotemporal expression patterns and functions. The primary structural difference among various LOX family members lie in their N-terminal sequences. LOX and LOXL1 have relatively simple N-termini, whereas LOXL2, LOXL3, and LOXL4 have distinct, extended N-terminal regions containing multiple scavenger receptor cysteine-rich domains. These domains contribute to the unique functions and tissue distributions of these LOX proteins. Notably, all LOX proteins share a highly conserved catalytic domain with a copper-binding site and a crucial lysyl tyrosylquinone cofactor, which is essential for their enzymatic activity [22-24]. Their role in ECM homeostasis is revealed by distinct phenotypes in knockout mouse models and through their aberrant expression and activity in disease conditions including fibrosis, tumor metastasis, pelvic organ prolapse, and endometriosis [19, 25-27]. Deletion of LOX causes aortic aneurysms and perinatal death in mice [28]. Mice lacking LOXL1 are viable but exhibit pelvic organ prolapse in adult life [26]. Although deletion of LOXL2 or LOXL3 leads to perinatal lethality, the deletion of both LOXL2/LOXL3 results in embryonic lethality [29-31]. LOXL4-deleted mice are viable [32]. On the other hand, increased activity of lysyl oxidase enzymes leads to collagen crosslinking and subsequent stiffening of the ECM, which are major drivers of fibrosis, tumor progression, and metastasis [19, 27, 33]. Inhibition of LOX activity through β-aminopropionitrile (BAPN)—a natural and irreversible inhibitor of activity of all lysyl oxidases—blocks collagen crosslinking and reduces tumor progression [19, 33].

Lysyl oxidases are widely expressed in rodent and human reproductive tissues including the uterus, placenta, fetal membranes, cervix, and vagina [25, 26, 34-38]. LOX is transcriptionally regulated by steroid hormones. Estrogen induces LOX in the cervix and vagina [36, 38]. The estrogen-induced LOX is downregulated by progesterone in the vagina [38]. In this study, we have analyzed the expression profile and activity of lysyl oxidases in the mouse myometrium during pregnancy. Consistent with elevated expression and activity of lysyl oxidase enzymes in mouse myometrium, inhibition of their activity delayed parturition in mice by altering ECM composition.

Materials and Methods

Animals

CD1 timed-pregnant mice used in the study were purchased from Charles River Laboratories. Mice strain C57B6/129sv mixed strain mice were maintained in a barrier facility at the University of Vermont under a 12-hour light/dark cycle. Timed mating was performed to collect tissues at different days of pregnancy. Breeding pairs were set up in the morning and checked for the presence of a vaginal plug in the afternoon. The presence of a vaginal plug was considered day 0 of pregnancy. For the BAPN experiment, the C57B6/129sv mixed strain mice were treated with a mixture of 0.8% BAPN and 5% sucrose in drinking water on the morning of gestation day 12 of pregnancy and maintained the BAPN water supply until the end of experiments. All animal studies were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals humane animal care standards. The Institutional Animal Care and Use Committee at the University of Vermont approved all animal experimental protocols used in this study.

Ovariectomized Mouse Model for Hormonal Administration

Ovariectomies were performed on female CD1 mice that were 7 to 8 weeks of age. These mice were given 2 weeks of rest following the surgery. After 2 weeks, the ovariectomized mice were administered a vehicle (corn oil) or estrogen (100 ng, subcutaneously); tissues were collected after 6 hours, 12 hours, 18 hours, or 24 hours. In a second experiment, the ovariectomized mice were treated with progesterone as follows: vehicle, P (1 mg), P (2 mg), P (1 mg) + E (10 ng), or E (100 ng) then after 2 days P (1 mg). Tissues were collected after 24 hours.

Tissue Collection

Myometrial tissues were collected from non-pregnant, pregnant (at various gestational time points), and 1 day postpartum mice. Nonpregnant tissue was collected during the metestrus stage of the estrous cycle determined by analysis of a vaginal smear. Gestational time points were determined using a timed mating procedure as described previously. All mice were euthanized using isoflurane overdose followed by cervical dislocation. Myometrial tissues were collected under a dissecting microscope, using a scalpel to scrape the endometrium off the uterus. Tissues were flash frozen in liquid nitrogen and stored at −80 °C until further processing. The uterine segments along with fetal sac units were immersed in OCT (Tissue-Tek) and solidified with liquid nitrogen. The OCT blocks were then cut using a cryostat to create 5-µm tissue sections that were mounted on glass slides and stored at −80 °C until further use.

RNA Isolation and Quantitative Polymerase Chain Reaction

Total RNA from frozen myometrial tissues was extracted using RNA STAT-60 (Tel-Test B) according to manufacturer's protocol. The total RNA was then treated with DNase I (Ambion) to remove genomic DNA. The iScript reverse transcription supermix (Bio-Rad Laboratories) was used to prepare complementary DNA. Quantitative polymerase chain reaction was performed using SYBR Green and respective primers designed for our genes of interest. Gene expression was calculated according to the 2^−ΔΔCt method. The target gene expression was normalized to the expression of a housekeeping gene, Rplp0. The primers used in this study are listed in Supplementary Table S1 [39].

Preparation of Tissue Lysates From Myometrial Tissues

RIPA extracts

Frozen myometrial tissues of specific gestational time points were pulverized, suspended in RIPA containing 1% protease inhibitors and EDTA (Thermo Fisher Scientific), and homogenized. The samples were centrifuged at 4 °C at 13 000 rpm for 15 minutes to collect the supernatant protein samples. Protein concentration of the samples was estimated using a BCA protein assay (Thermo Fisher Scientific).

PBS and urea extracts

Frozen myometrial tissues of specific gestational time points were pulverized, suspended in PBS containing 1% protease inhibitors and EDTA (Thermo Fisher Scientific), and homogenized. The samples were centrifuged at 4 °C at 13 000 rpm for 15 minutes. The supernatant was collected and labeled as PBS soluble protein fraction. The remaining pellet was resuspended in 6 M urea buffer containing 1% protease inhibitors and EDTA (Thermo Fisher Scientific) and homogenized. The samples were then left on a gentle rotation at 4 °C for 24 hours. The next day the samples were centrifuged at 4 °C at 13 000 rpm for 10 minutes. The supernatant was collected and labeled as the urea-soluble fraction. Both PBS and urea-extracted sample fractions were exchanged with equivalent volume of 0.5 M Tris/0.5% SDS, pH 8.0 buffer, and concentrated using Vivaspin ultrafiltration spin columns (Vivaspin 2, 5 kDa MWCO, # 28932359, Cytiva Life Sciences). Protein concentration of the samples was estimated using a BCA protein assay (Thermo Fisher Scientific).

Western Blot

Twenty micrograms of each protein sample in a 20:1 dilution of 4X Laemmli Sample Buffer (Bio-Rad): β-mercaptoethanol (Sigma) was boiled at 95 °C for 10 minutes. The samples were loaded, alongside a protein standard (Precision Plus Protein Kaleidoscope, Bio-Rad), into 10% Tris-HCl gel. Gel electrophoresis was run at 50 V, until the samples passed through the stacking gel, where the voltage was then increased to 100 V and run for another hour. Proteins were transferred onto a nitrocellulose membrane (Bio-Rad) at 100 V at 4 °C for 1 hour. Membranes were blocked in 3% nonfat dry milk in TBST (Blotting-grade blocker nonfat dry milk; Bio-Rad) at room temperature for 1 hour. The membranes were incubated with a primary antibody in blocking solution at 4 °C overnight. The following day, the membranes were washed with TBST and incubated with their respective horseradish peroxidase (HRP)-labelled secondary antibody at room temperature for 1 hour. The membranes were imaged with equal volume of Amersham ECL Western Blotting Detection Reagents (Cytiva Life Sciences) using ImageQuant 800 Western blot imaging system (Cytiva Life Sciences). The primary antibodies used were: anti-mouse collagen type I (Millipore Cat# AB765P, RRID:AB_92259), collagen type III (Proteintech Cat# 22734-1-AP, RRID:AB_2879158), tropoelastin (Elastin Products Company Cat# PR385, RRID:AB_3099530), LOX (Abcam Cat# ab174316, RRID:AB_2630343), LOXL1 (Thermo Fisher Scientific Cat# PA5-87701, RRID:AB_2804357), LOXL2 (Novus Cat# NBP1-32954, RRID:AB_2135089), Connexin 43 (Cell Signaling Technology Cat# 3512 [also used By NYUIHC-970], RRID:AB_2294590), and β-Actin (Cell Signaling Technology Cat# 3700 [also 3700P, 3700S], RRID:AB_2242334). The secondary antibodies used were anti-mouse IgG, HRP-linked antibody (Cell Signaling Technology Cat# 7076 [also 7076S, 7076V, 7076P2], RRID:AB_330924), anti-rabbit IgG, HRP-linked antibody (Cell Signaling, 7074S;), goat anti-rabbit IgG (H/L):HRP (Bio-Rad Cat# STAR208P, RRID:AB_3099534), or goat anti-mouse IgG (H/L):HRP (Bio-Rad Cat# 0300-0108P, RRID:AB_808614).

Confocal Microscopy Imaging

Frozen uterine tissue slides were fixed in cold acetone for 10 minutes, air dried for 20 minutes, and washed with PBS. The sections were blocked with 10% normal goat serum (Life Technologies) at room temperature in a moist chamber for 30 minutes. Primary antibodies in normal goat serum was added to each section and incubated at 4 °C overnight. The next day, the sections were washed with PBS and then incubated with Alexa Fluor conjugated secondary antibodies (Thermo Fisher Scientific) in normal goat serum at room temperature for 30 minutes. Slides were washed with PBS and coverslips were mounted using ProLong Gold Antifade Mountant with DNA Stain DAPI (Thermo Fisher Scientific). Images were captured using a Nikon A1R confocal microscope galvanometer scanner. The settings of the galvanometer scanner used were as follows: single illumination point scan as fast as 8 frames per second for a 512 × 512-pixel field, and up to 4096 × 4096 pixels image capture. The primary antibodies and subsequent secondary antibodies used were α-actin (1A4) (Santa Cruz Biotechnology Cat# sc-32251, RRID:AB_262054), LOX (Abcam Cat# ab174316, RRID:AB_2630343), LOXL1 (Thermo Fisher Scientific Cat# PA5-87701, RRID:AB_2804357), LOXL2 (Abcam Cat# ab96233, RRID:AB_10677617), Alexa-Fluor 488 goat anti-mouse IgG(H + L) (Thermo Fisher Scientific Cat# A-11029, RRID:AB_2534088) and goat anti-rabbit IgG(H + L) Alexa-Fluor 555 (Thermo Fisher Scientific Cat# A32732, RRID:AB_2633281).

In Situ Activity Assay for Lysyl Oxidases

The activity of lysyl oxidases was examined using an in situ activity assay as described [40]. Briefly, the frozen tissue sections prepared from mouse uterus were fixed with neutral buffered formalin for 10 minutes. The tissue sections were blocked with normal goat serum for 30 minutes in a moist chamber at room temperature. They were incubated with 200 μM EZ-Link Hydrazide-Biotin (BHZ) (Thermo Fisher Scientific, Cat#: 21339) overnight at 37 °C. After removing excess BHZ by washing with PBS, the sections were incubated with Streptavidin-Alexa Fluor 555 conjugate for 1 hour in a dark moist chamber. The sections were mounted with ProLong Gold Antifade Mountant with DNA Stain DAPI and imaged using Nikon A1R-ER Confocal Microscope with galvanometer scanner at 40×. The image settings were set for gestation day 15 sample, and the same settings were used for all other sections to compare staining intensity.

Proteomics

Sample preparation and trypsin digestion

Urea extracted protein samples were used in this experiment. Total protein extract (50 µg) from each sample was loaded onto the SDS-PAGE and run slightly (∼5 mm) into the separating gel that was stained with Bio-Safe Coomassie G-250. SDS-PAGE allowed the assessment of protein loading and removal of any incompatible detergents for subsequent in-gel trypsin digestion. Single-wide bands containing unseparated proteins from individual samples were excised, minced to approximately 1 mm³ cubes, and followed by destaining with 50% acetonitrile (CH₃CN)/50 mM triethyl ammonium bicarbonate. After dehydrating with CH₃CN, reduction and alkylation of disulfides were conducted with 10 mM dithiothreitol and 55 mM iodoacetamide in 100 mM triethyl ammonium bicarbonate, respectively. After repeated washing and dehydrating with triethyl ammonium bicarbonate and CH₃CN, the gel pieces were swelled in ∼12 ng/µL trypsin solution (Promega) and incubated overnight at 37 °C. Peptides were extracted successively with formic acid (FA) and CH₃CN and dried under vacuum.

Peptide labeling by tandem mass tag (TMT) and high pH fractionation

The labeling procedures were performed according to the manufacturer's protocols (Thermo Fisher Scientific). Briefly, 50 µg of the dried peptides from each sample were resuspended in 100 µL of 100 mM triethyl ammonium bicarbonate, and 0.8 mg of TMT reagents dissolved in 41 µL of anhydrous CH₃CN was added, followed by briefly vortexing and incubating for 1 hour at room temperature. The reactions were then quenched by adding 8 µL of 5% hydroxylamine. Short mass spectrometry runs and database searches confirmed that the labeling efficiency on average across all samples was >97%, verifying the number of peptides labeled in individual samples before combining. The combined labeled peptides (∼33 µg) were fractionated using the high-pH reversed-phase spin column (Cat. No.: 84868; Thermo Scientific) into 16 fractions (10%, 13%-25% with 1% increments, 35% and 50%). All fractions were dried under vacuum and kept at −80 °C until mass spectrometry analysis.

Liquid chromatography-tandem mass spectrometry

The fractionated TMT-labeled peptides were resuspended in 10 µL 2.5% CH₃CN and 2.5% FA in water. Half of the labeled peptides in each fraction were analyzed. Mass spectrometry was performed using an Orbitrap Eclipse mass spectrometer coupled to an EASY-nLC 1200 system (Thermo Scientific). Samples were loaded onto a 100 μm × 350 mm capillary column packed with UChrom C18 material (1.8-µm 120 Å Uchrom C18, nanoLC-MS Solutions) at a flow rate of 300 nL/min⁻¹. The column end was laser pulled to a ∼3-μm orifice and packed with minimal amount of 5-µm Magic C18AQ before packing with the 1.8-μm particles. Peptides were separated using a gradient of 4% to 20% CH₃CN/0.1% FA over 110 minutes, 20% to 32% CH₃CN/0.1% FA in 25 minutes, 32% to 76% CH₃CN/0.1% FA in 25 minutes then 76% CH₃CN/0.1% FA for 10 minutes, followed by an immediate return to 4% CH₃CN/0.1% FA and a hold at 4% CH₃CN/0.1% FA. Peptides were introduced into the mass spectrometer via a Nanospray Flex Ion Source with a spray voltage of 1.9 kV, an Ion Transfer Tube Temperature of 275 °C, and the RF set at 30%. “Cycle time in 3 seconds” acquisition mode, TMT-specific Real-Time Search and Synchronous Precursor Selection (SPS) workflows were used to acquire mass spectrometry data. Survey scans from m/z 400 to 1400 at 60 000 resolution (AGC target: standard; max IT: auto; profile mode) were acquired with data-dependent collision-induced dissociation tandem mass spectrometry scans on the most abundant ions (AGC target: standard; max IT: 35 ms; ion trap scan rate: turbo) in the ion trap with a normalized collision energy at 35% and an isolation width of 1 m/z. Real-Time Search was used for triggering MS3 acquisitions with TMT6plex, carbamidomethylation, and oxidation as modifications and the Mus musculus UP000000589 FASTA as database (downloaded from Uniprot on March 29, 2021). XCorr values of 1.4 and 2 were used as score thresholds for doubly and triply charged peptides, respectively. Protein “close-out” was not enabled. MS3 was performed using SPS with isolation widths of 1 and 2 Da for MS1 and MS2 isolation windows, respectively, and a normalized collision energy of 65% (AGC target: custom: 200%; max IT: 200). The product ions from MS3 were scanned from m/z 100 to 500 in the Orbitrap with a resolution of 50 000. Dynamic exclusion was enabled (exclude isotopes: on; exclusion duration: 45 seconds; charge states = 2-6). Precursor Fit window was set as 1 and the Fit Threshold as 30%. The samples were injected twice as technical replicates.

Database searches

The 32-mass spectrometry .RAW files (16 high-pH reversed phase separation fractions analyzed in 2 technical replicates) generated from each experiment were imported into the Proteome Discoverer 2.5 (Thermo Fisher Scientific) as “fractions.” Product ion spectra were searched using the SEQUEST in the Processing workflow against the Mus musculus UP000000589 FASTA database from Uniprot; downloaded on March 29, 2021). Search parameters were as follows: (1) full trypsin enzymatic activity; (2) maximum missed cleavages = 2; (3) minimum peptide length = 6; (4) mass tolerance at 10 ppm for precursor ions and 0.6 Da for fragment ions; (5) dynamic modification on methionine (+15.9949 Da: oxidation), dynamic modification on N-terminus (+42.01 Da: Acetyl), static TMT6plex modification (the TMT6plex and TMT10plex have the same isobaric mass) on N-termini and lysine residues (229.163 Da); (6) 4 maximum dynamic modifications allowed per peptide; and (7) static carbamidomethylation modification on cysteines (+57.021 Da). Percolator node was included in the workflow to limit the false discovery rate to less than 1% in the data set.

Quantification and statistical analysis

The biological replicates and technical replicates were assigned accordingly in the “Study Factors” tab (CNTRL-1, CNTRL-3, CNTRL-4, CNTRL-5, CNTRL-6, BAPN-1, BAPN-3, BAPN-4, BAPN-5, BAPN-6) and “Samples” tab, respectively, in the Proteome Discoverer. The abundances of TMT-labeled peptides were quantified with the Reporter Ions Quantifier node in the Consensus workflow and parameters were set as follows: (1) both unique and razor peptides were used for quantification; (2) Reject Quan Results with Missing Channels: False; (3) Apply Quan Value Corrections: True (values set according to the product spreadsheet (Lot # XD341063)); (4) Co-Isolation Threshold: 75; (5) Average Reporter S/N Threshold = 10; SPS Mass Matches [%] = 65; (6) Normalization mode: “Total Protein Amount”; and (7) Scaling Mode was set “On All Average.” Protein ratio calculation was “Protein Abundance Based.” Two-tailed t-test was used for hypothesis testing and associated P values were calculated for protein identities quantified across all samples. All the protein identification and quantification information (<1% false discovery rate) was exported from the Proteome Discoverer result files to Excel spreadsheets. Fold change (log₂) and P values (-log₁₀) information and scaled abundances (the abundances were scaled so that fold changes of proteins of varying abundances could be represented on the same color intensity scale of the heat map) were imported into Graph Pad Prism 10 (GraphPad Software Inc., CA) for constructing volcano plots and heat maps.

Enrichment analysis

The list of gene names of the differentially expressed proteins (fold change [BAPN/control] > 2, P < .05) was imported into pathfindR (https://github.com/egeulgen/pathfindR) [41] with STRING as the database for the protein interaction network. The output (enriched_terms.html) is included in Supplementary Table S2 [39].

Statistical Analysis

Statistical analysis was completed using GraphPad Prism software. The Student test was used to compare 2 groups. A 1-way ANOVA followed by a Tukey multiple comparisons test was used to compare multiple groups. The values were expressed as the mean ± the standard error of the mean. P < .05 was considered as significant between groups.

Results

LOX Is Significantly Induced in Mouse Myometrium During Pregnancy

To analyze the expression profile of Lox in mouse myometrium over the course of pregnancy, we utilized myometrial tissues collected from nonpregnant (NP) mice, at various gestational time-points starting from day 6 to day 18, and postpartum mice. The mRNA expression of Lox was significantly induced from gestation day 10 to term compared to NP myometrium (Fig. 1A). LOX is synthesized within the cell as a pro-enzyme, which is cleaved into an active enzyme after being exported into the ECM [19, 20]. To examine the levels of pro-LOX and active LOX in mouse myometrium, we prepared tissue lysates from the myometrial tissues by extracting with PBS followed by urea to release LOX that was integrated into the ECM. The pro-LOX was easily extractable and was enriched in PBS soluble fractions. However, the active LOX was highly enriched in urea-soluble fractions (Fig. 1B and 1C). Consistent with gene expression results, the protein levels of LOX were highly elevated in the pregnant myometrium. Our subsequent immunolocalization studies revealed that LOX was predominantly localized in myometrial smooth muscle cells (Fig. 1D). These results indicate that LOX is highly induced in the mouse myometrium during pregnancy and has a potential role in myometrial tissue function during pregnancy and parturition.

Figure 1.

Expression and localization of LOX in mouse myometrium. (A) Expression of Lox in mouse myometrium. Total RNA isolated from NP (nonpregnant), gestation day 6, 8, 10, 12, 15, 17, 18, IL (in-labor), and PP (postpartum) mice was used. The mRNA expression was normalized to Rplp0 and compared with NP samples (n = 5-6/group, *P < .05). (B) Western blot analysis of LOX protein levels in PBS and urea-soluble fractions of NP, gestation day 6, 8, 10, 12, 15, 18, IL, and PP mouse myometrial tissues. β-actin was used as a loading control. These are representative images from 3 independent replicates. (C) Quantification of protein levels of proLOX and mature LOX in urea fractions prepared from myometrial tissues of NP, gestation day 6, 8, 10, 12, 15, 18, IL, and PP mice normalized to β-actin levels and compared with NP samples (n = 3/group, *P < .05). (D) Confocal imaging analysis of LOX in gestation day 8, 12, 15, and 18 myometrium. Representative images from 3 independent replicates. Brightness and contrast for each image were optimized individually for optimal visualization of morphology. Scale bar: 50 μm. LOX, lysyl oxidase; SMA, smooth muscle actin. Note: LOX is predominantly localized in smooth muscle tissue of the myometrium.

LOXL1 and LOXL2 Are Also Significantly Induced in Mouse Myometrium During Pregnancy

We next analyzed the gene expression of Loxl1-4 in mouse myometrium. Loxl1mRNA was significantly induced on gestation days 10 through 17. The LOXL1 protein expression was significantly elevated from gestation day 15 through postpartum. Loxl2 mRNA was significantly induced on gestation day 15 (Figs. 2A and 3A). The gene expression of Loxl3 and Loxl4 could not be detected using quantitative PCR analysis. We next analyzed the protein levels in the whole tissue lysates prepared from myometrium. The protein levels of LOXL1 were significantly elevated in the pregnant myometrium consistent with its gene expression (Fig. 2B and 2C). Although LOXL2 protein levels appeared to be modestly elevated during pregnancy, there were high variations between replicates and thus were not statistically significant in the pregnant myometrium (Fig. 3B and 3C). Similar to LOX, both LOXL1 and 2 were localized in the smooth muscle compartment of the mouse myometrium (Figs. 2D and 3D). These results demonstrate that not only LOX, but also LOXL1-2 enzymes were significantly induced and could potentially contribute to the function of myometrium.

Figure 2.

Expression and localization of LOXL1 in mouse myometrium. (A) Expression of Loxl1 in mouse myometrium. Total RNA isolated from NP (nonpregnant), gestation day 6, 8, 10, 12, 15, 17, 18, IL (in-labor), and PP (postpartum) mice was used. The mRNA expression was normalized to Rplp0 and compared with NP samples (n = 4-6/group, *P < .05). (B) Western blot analysis of LOXL1 protein levels in myometrial whole tissue lysates of NP, gestation day 6, 8, 10, 12, 15, 18, IL, and PP mouse myometrial tissues. β-actin was used as a loading control. These are representative images from 4 independent replicates. (C) Quantification of protein levels of LOXL1 isolated from myometrial tissues of NP, gestation day 6, 8, 10, 12, 15, 18, IL, and PP mice normalized to β-actin levels and compared with NP samples (n = 3-4/group, *P < .05). (D) Confocal imaging analysis of LOXL1 in gestation day 8, 12, 15, and 18 myometrium. Representative images from 3 independent replicates. Brightness and contrast for each image were optimized individually for optimal visualization of morphology. Scale bar: 50 μm. LOXL1, lysyl oxidase like-1; SMA, smooth muscle actin. Note: LOXL1 is predominantly localized in smooth muscle tissue of the myometrium.

Figure 3.

Expression and localization of LOXL2 in mouse myometrium. (A) Expression of Loxl2 in mouse myometrium. Total RNA isolated from NP (nonpregnant), gestation day 6, 8, 10, 12, 15, 17, 18, IL (in-labor), and PP (postpartum) mice was used. The mRNA expression was normalized to Rplp0 and compared with NP samples (n = 3-6/group, *P < .05). (B) Western blot analysis of LOXL2 protein levels in myometrial whole-tissue lysates of NP, gestation day 6, 8, 10, 12, 15, 18, IL, and PP mouse myometrial tissues. β-actin was used as a loading control. These are representative images from 3 independent replicates. (C) Quantification of protein levels of LOXL2 isolated from myometrial tissues of NP, gestation day 6, 8, 10, 12, 15, 18, IL, and PP mice normalized to β-actin levels and compared with NP samples (n = 3/group, *P < .05). (D) Confocal imaging analysis of LOXL2 in gestation day 12, 15, and 18 myometrium. Representative images from 3 independent replicates. Brightness and contrast for each image were optimized individually for optimal visualization of morphology. Scale bar: 50 μm. LOXL2, lysyl oxidase like-2; SMA, smooth muscle actin. Note: LOXL2 is predominantly localized in smooth muscle tissue of the myometrium.

Enzyme Activity of Lysyl Oxidases Increased in Pregnant Myometrium

Our results so far demonstrated that the lysyl oxidases are significantly induced in the mouse myometrium during pregnancy. The catalytic functions of lysyl oxidases to crosslink collagen and elastin, and thus to modify the structure and function of the ECM, depends on their enzymatic activity. Therefore, we next examined the enzymatic activity of lysyl oxidases in the myometrium using an in situ activity assay. We measured the total activity of all lysyl oxidase enzymes because it is not possible to analyze the activity individually. The lysyl oxidases were active in all gestational time points as evidenced by signal elicited through the incorporation of BHZ in the myometrial tissues (Fig. 4). Consistent with the elevated levels of gene and protein levels, the total lysyl oxidase activity was much higher on gestation day 15 compared to all other samples (Fig. 4). These results indicate that the lysyl oxidases are more active in myometrium during late gestation.

Figure 4.

Activity of lysyl oxidases in mouse myometrium. (A) Uterine tissue sections prepared from gestation day 6, 12, 15, and 18 mice were incubated with 200 μM biotin-hydrazide (BHZ) overnight and then stained with Alexa Fluor-streptavidin. The representative images from confocal imaging analysis are shown. The staining intensity represents BHZ incorporation, which corresponds to activity of the lysyl oxidases. The image settings were set for day 15 sample and used for all other images for comparison. Scale bar: 50 μm. (B) The quantification of mean fluorescence intensity was analyzed by ImageJ. Data are expressed as means ± standard error of the mean of 3 independent experiments (*P < .05).

Steroid Hormones Differentially Induce Lox and Loxl1-2 in Mouse Myometrium

The lysyl oxidases were highly induced and active in the pregnant myometrium, indicating possible regulation by steroid hormones. We utilized ovariectomized mice exogenously administered with estrogen, progesterone, or in combination to induce the gene expression of lysyl oxidases in mouse myometrium. In response to 17β-estradiol treatment, Lox expression was significantly suppressed, whereas Loxl1 and Loxl2 were significantly induced (Fig. 5A). The robust induction of Loxl2 in response to 17β-estradiol treatment indicates that it could be a potential target of this hormone. Similar to 17β-estradiol, progesterone significantly suppressed the expression of Lox. In contrast to 17β-estradiol alone, progesterone or a combination of 17β-estradiol and progesterone did not exert any influence on the expression of Loxl1 and Loxl2 (Fig. 5B). It was also possible that 17β-estradiol-induced expression of Loxl1 and Loxl2 might have been suppressed by the progesterone. These results indicate that steroid hormones, 17β-estradiol and progesterone, differentially regulate the expression of lysyl oxidases in the myometrium.

Figure 5.

Gene expression of Lox, Loxl1 and Loxl2 in response to 17β-estradiol and progesterone in ovariectomized mouse myometrium. (A) Expression of Lox, Loxl1 and Loxl2 in response to 17β-estradiol in ovariectomized mouse myometrium. Ovariectomized mice were treated with vehicle (V), or 17β-estradiol (100 ng/mouse, subcutaneous) for 6, 12, 18, or 24 hours (n = 5-6/group, *P < .05). (B) Expression of Lox, Loxl1, and Loxl2 in response to progesterone (P) or progesterone and 17β-estradiol (E) in ovariectomized mouse myometrium. Ovariectomized mice were treated with P (1 mg/mouse, s/c), P (2 mg/mouse, s/c), P (1 mg) + E (10 ng), E (100 ng)—2 days’ rest—P (1 mg). The tissues were collected after 24 hours (n = 6/group, *P < .05).

Inhibition of Activity of Lysyl Oxidases Delays Parturition in Mice Due to Impaired Myometrial Contractility

To examine the impact of the activity of lysyl oxidases on myometrium and parturition, we treated mice with 0.8% BAPN (A3134, Sigma-Aldrich) through drinking water from gestation day 12 (day of vaginal plug = day 0) and monitored the timing of parturition (Fig. 6A). Lysyl oxidases were induced very early during pregnancy in the myometrium (Figs. 1-3) and were also expressed in multiple other reproductive tissues such placenta, fetal membranes, and the cervix. Using this information, we treated mice from gestation day 12 to avoid the effect of BAPN on placental, fetal membrane, and fetal development. In response to BAPN treatment, the parturition process was significantly delayed compared to an untreated control group of mice (Fig. 6B). These results demonstrate that the inhibition of activity of lysyl oxidases delayed parturition in mice. To examine the activity of lysyl oxidases on myometrial contractility, we treated pregnant mice with BAPN from gestation day 12 and collected myometrial tissues on day 18 (Fig. 6C). The expression levels of genes encoding contraction associated proteins—connexin 43 (Gja1), oxytocin receptor (Oxtr) and prostaglandin endoperoxide synthase 2 (Ptgs2)—were significantly reduced in response to BAPN (Fig. 6D). Consistent with mRNA expression, the protein level of connexin 43 was reduced in BAPN-treated tissues (Fig. 6E). BAPN treatment did not alter Col1a1, Eln, and Lox transcription (Fig 6D). Also, the genes encoding proteins related to myometrial quiescence were not altered in response to BAPN (Fig. 6F). These results demonstrate that activity of lysyl oxidases is necessary for myometrial contractility at term to facilitate on-time parturition in mice.

Figure 6.

Impact of inhibition of activity of lysyl oxidases in mouse myometrium on parturition-timing and myometrial contractility. (A) Schematic representation of BAPN treatment to monitor parturition timing in mice. Pregnant mice were treated with BAPN through drinking water from gestation day 12 until parturition. (B) The parturition-timings between control (n = 8) and BAPN treated (n = 10) mice were monitored, recorded, and analyzed. In the BAPN-treated group, the parturition was significantly delayed compared to the control group (*P < .05). (C) Schematic representation of BAPN treatment for tissue collection. Pregnant mice were treated with BAPN through drinking water from gestation day 12 until gestation day 18 for the collection of myometrial tissues to analyze gene and protein expression. Untreated mice served as control. (D) Expression of connexin 43 (Gja1), oxytocin receptor (Oxtr), prostaglandin endoperoxide synthase 2 (Ptgs2), lysyl oxidase (Lox), collagen type I alpha 1 chain (Col1a1), and elastin (Eln) in the myometrial tissues collected on gestation day 18 from control and BAPN-treated mice (n = 4-5/group). Expression of genes encoding contraction associated proteins—Gja1, Oxtr, and Ptgs2—are significantly reduced in response to BAPN treatment (*P < .05). The expression levels of Lox, Col1a1, and Eln are not altered in response to BAPN. (E) Western blot analysis of protein levels of connexin 43 (39-44 kDa) in control and BAPN-treated myometrial tissues collected on gestation day 18 (n = 3/group). β-actin was used as a loading control. The levels of connexin 43 are reduced in response to BAPN indicating myometrial contractile dysfunction. (F) Expression of Kcnn3, Zeb1 and 2, Igfbp5 and 6, and Trvp4—genes encoding factors known to be involved in the maintenance of uterine quiescence—are not altered in response to BAPN treatment. The total RNA isolated from myometrial tissues collected on gestation day 18 from control and BAPN-treated mice were used (n = 3-5/group).

Lysyl Oxidases Maintain Integrity of the ECM in Myometrium During Pregnancy and Parturition

To examine the activity of lysyl oxidases on the integrity of myometrial ECM, we compared the proteomic profiles of myometrial tissues isolated from control and BAPN-treated mice using isobaric tandem mass tags, a stable isotope-based quantitative proteomic approach (Fig 7A). A total of 5176 proteins were quantified and represented in a volcano plot (Fig. 7B). A total of 356 proteins were identified as differently expressed in BAPN vs control (2-tailed t-test; P < .05) (Fig. 7B). Twenty-two differentially expressed proteins had a 2-fold increase (20) or decrease (2) in abundance in the BAPN-treated samples are represented in a heat map (Fig. 7C) and also listed in Table 1. Term enrichment analysis on the 22 differentially expressed proteins indicated that collagens are involved in several pathways (Fig. 7D and Supplementary Table S2 [39]). The top 3 terms are protein digestion and absorption, ECM-receptor interaction, and focal adhesion (Fig. 7E). Other differentially expressed proteins include peroxidasin homolog (Pxdn), hemoglobin subunit epsilon-Y2 (Hbb-y), leucine-rich HEV glycoprotein (Lrg1), hemoglobin subunit zeta (Hbz) serum amyloid P-component (Apcs), alpha-1-acid glycoprotein 1 (Orm1), microfibril-associated glycoprotein 1 (Mfap2), sushi-repeat-containing protein (Srpx), aldo-keto reductase family 1 member C18 (Akr1c18), aquaporin-5 (Aqp5), OCIA domain-containing protein 2 (Ociad2), transforming growth factor beta (Tgfb2), alpha-fetoprotein (Afp), and microfibril-associated glycoprotein 4 (Mfap4). Consistent with gene expression, the protein level of Gap junction alpha-1 protein (Gja1) was decreased in BAPN treated samples. Of the 5176 quantified proteins, 131 were labeled as ECM-related according to Gene Ontology (Cellular Component), and 96 of which were classified in the matrisomeDB (https://matrisomedb.org/) [42] into categories under 2 divisions (Core matrisome and Matrisome-associated) (Fig. 8A and Table 2). ECM proteins that were quantified to have differential abundances (P < .05) in the BAPN samples are represented in heatmaps (Fig. 8B) and are also listed in Table 3. The majority of the differentially expressed (P < .05) ECM proteins were found to be upregulated. Elastin and LOXL1 were significantly increased in the BAPN samples with a 1.6-fold upregulation. These results revealed that proteomic composition of myometrial ECM is altered upon inhibition of the activity of lysyl oxidase enzymes.

Figure 7.

Proteomic analysis reveals alterations in the composition of myometrial proteome between control and BAPN-treated mice. (A) Experimental design. Urea-extracted protein samples from myometrial tissues of control and BAPN-treated mice (n = 6/group) were loaded onto SDS-PAGE slightly into the separating gel. Five samples per group were subsequently used. After in-gel digestion, peptide labeling with tandem mass tags (TMT), and high-pH reversed-phase (RP) fractionation, peptides were identified in tandem mass spectrometry (MS2) and their reporter ions quantified in MS3 scans after Real Time Search (RTS) and Synchronous Precursor Selection (SPS). (B) Volcano plot. A total of 5176 proteins were quantified and represented in a volcano plot; 356 proteins were identified as differently expressed in BAPN vs control (2-tailed t-test; P < .05) with 20 and 2 proteins having a 2-fold increase or decrease in abundance (BAPN/control: log2 < −1 or >1), respectively. Differentially expressed proteins are highlighted (upregulated in red; downregulated in blue). Collagens and Gja1 are labeled with their gene names. Cut-offs of fold change at 2 (log2 = −1 or 1) and P value at .05 (−log₁₀ 0.05 = 1.301) are indicated by dotted lines on the x and y-axis, respectively. (C) Heat maps of the differential proteomes. Scaled abundances of the 22 differentially expressed proteins are represented in a heat map, listed (top to bottom) according to their average log2 fold change from high to low. The scale of the color intensity is arbitrary (double gradient). (D) Enrichment analysis. Top 8 enrichment terms of the differentially expressed proteins (BAPN/Control > 2, P < .05). (E) Interactions between the top 3 enrichment terms. Upregulated proteins are highlighted in green.

Table 1.

Differential proteome of the myometrium in BAPN-treated mice

Accession	Gene symbol	Description	Fold change (BAPN/CNTL)	P (<.05)
P28481	Col2a1	Collagen alpha-1(II) chain	10.89	1.94E-04
Q3UQ28	Pxdn	Peroxidasin homolog	4.23	7.90E-04
P02104	Hbb-y	Hemoglobin subunit epsilon-Y2	3.79	1.90E-03
P11087	Col1a1	Collagen alpha-1(I) chain	3.32	2.49E-03
Q91XL1	Lrg1	Leucine-rich HEV glycoprotein	3.00	2.76E-03
P06467	Hbz	Hemoglobin subunit zeta	2.79	4.13E-03
B7ZNH7	Col14a1	Collagen alpha-1(XIV) chain	2.77	2.01E-02
Q63ZW6	Col4a5	Col4a5 protein	2.73	4.42E-02
Q01149	Col1a2	Collagen alpha-2(I) chain	2.70	2.90E-03
Q61245	Col11a1	Collagen alpha-1(XI) chain	2.59	3.74E-03
P12246	Apcs	Serum amyloid P-component	2.56	1.09E-02
Q60590	Orm1	Alpha-1-acid glycoprotein 1	2.48	3.16E-02
Q99PM0	Mfap2	Microfibril-associated glycoprotein 1	2.29	9.10E-05
Q9R0M3	Srpx	Sushi-repeat-containing protein SRPX	2.23	7.39E-05
Q8K023	Akr1c18	Aldo-keto reductase family 1 member C18	2.20	4.79E-02
Q9WTY4	Aqp5	Aquaporin-5	2.14	4.70E-02
Q9D8W7	Ociad2	OCIA domain-containing protein 2	2.10	1.94E-02
A0A0A6YXU9	Tgfb2	Transforming growth factor beta	2.11	2.70E-02
P02772	Afp	Alpha-fetoprotein	2.09	3.36E-02
Q9JLI2	Col5a3	Collagen type V alpha 3 chain	2.02	5.42E-03
Q9D1H9	Mfap4	Microfibril-associated glycoprotein 4	0.49	3.87E-03
P23242	Gja1	Gap junction alpha-1 protein	0.47	3.84E-04

Figure 8.

Myometrial extracellular matrix (ECM) proteome in BAPN-treated mice. (A) Volcano plot. ECM proteins were colored according to the categories they belong to and those that were quantified with a fold change (BAPN/control) > 2 and P < .05 were labeled. Cut-offs of fold change at 2 (log 2 = −1 or 1) and P value at .05 (−log₁₀ 0.05 = 1.301) are indicated by gray lines on the x and y-axis, respectively. (B) Heat map of the differentially expressed ECM proteins (P < .05) of various ECM categories. Scaled abundances of the differentially expressed proteins are represented and listed (top to bottom) according to their average log₂ fold change from large to small. The scale of the color intensity is arbitrary (double gradient).

Table 2.

Quantified extracellular matrix (ECM) proteome of the myometrium in BAPN-treated mice

Matrisome-DB	Gene symbol (total: 96)	#	Up	Down
Collagens	Col11a1, Col15a1, Col1a1, Col1a2, Col2a1, Col3a1, Col4a1, Col4a2, Col4a3, Col4a5, Col4a6, Col5a1, Col5a2, Col5a3, Col6a1, Col6a2, Col6a3, Col6a3, Col6a4	19	9	0
Glycoproteins	Agrn, Angptl2, Bmper, Crispld2, Efemp1, Efemp2, Eln, Emilin1, Fbln1, Fbln5, Fbln7, Fbn1, Fbn2, Fgl2, Lama1, Lama2, Lama3, Lama4, Lama5, Lamb1, Lamb2, Lamc1, Lamc1, Lamc2, Ltbp1, Ltbp3, Matn2, Mfap1b, Mfap2, Mfap4, Mfap5, Mmrn2, Nid1, Nid2, Pxdn, Sparc, Sparcl1, Spon1, Svep1, Tgfbi, Thbs1, Thsd4, Tnc, Tnxb, Vtn, Vwa1, Vwf	47	18	1
Proteoglycans	Aspn, Bgn, Dcn, Fmod, Hspg2, Lum, Ogn	7	1	1
ECM regulators	Adamts1, Adamts2, Lox, Loxl1, Loxl2, Mmp14, Mmp17, Mmp19, Plod3, Mmp2, Mmp23, Serpinf1, Serpine1, Tgm2	14	2	1
ECM-affiliated	Anxa2, Frem2, Elfn2, Gpc1, Lgals3, Mbl2	6	0	0
Secreted factors	Tgfb2, Tgfb1, Wnt5a	2	1	1

Abbreviation: BAPN, β-aminopropionitrile.

Table 3.

Differential extracellular matrix (ECM) proteome of the myometrium in BAPN-treated mice

Accession	Gene symbol	Description	Fold Change (BAPN/CNTL)	P (<.05)
		Core matrisome/collagens
P28481	Col2a1	Collagen alpha-1(II) chain	10.89	1.94E-04
P11087	Col1a1	Collagen alpha-1(I) chain	3.32	2.49E-03
Q63ZW6	Col4a5	Col4a5 protein	2.73	4.42E-02
Q01149	Col1a2	Collagen alpha-2(I) chain	2.70	2.90E-03
Q61245	Col11a1	Collagen alpha-1(XI) chain	2.59	3.74E-03
Q9JLI2	Col5a3	Collagen type V alpha 3 chain	2.02	5.42E-03
Q3U962	Col5a2	Collagen alpha-2(V) chain	1.41	4.64E-02
P02463	Col4a1	Collagen alpha-1(IV) chain	1.20	2.52E-02
P08122	Col4a2	Collagen alpha-2(IV) chain	1.14	4.52E-02
		Core matrisome/ECM Glycoproteins
Q3UQ28	Pxdn	Peroxidasin homolog	4.23	7.90E-04
Q99PM0	Mfap2	Microfibril-associated glycoprotein 1	2.29	9.10E-05
A0A3Q4EGK3	Ltbp1	Latent-transforming growth factor beta-binding protein 1	1.97	3.44E-02
Q99K41	Emilin1	EMILIN-1	1.94	5.91E-04
Q60675	Lama2	Laminin subunit alpha-2	1.77	2.44E-02
P19137	Lama1	Laminin subunit alpha-1	1.72	2.50E-02
P54320	Eln	Elastin	1.57	6.35E-03
A0A1Y7VJW9	Fbln5	Fibulin-5	1.55	3.77E-03
Q8R2Z5	Vwa1	von Willebrand factor A domain-containing protein 1	1.44	1.47E-03
E9QN70	Lamb1	Laminin subunit beta-1	1.40	1.13E-02
Q8BPB5	Efemp1	EGF-containing fibulin-like extracellular matrix protein 1	1.39	5.27E-04
G5E874	Lamc2	Laminin subunit gamma-2	1.39	4.33E-02
Q3UTY6	Thsd4	Thrombospondin type-1 domain-containing protein 4	1.36	3.00E-03
F8VQJ3	Lamc1	Laminin subunit gamma-1	1.33	4.60E-03
Q61001	Lama5	Laminin subunit alpha-5	1.30	1.41E-04
Q61292	Lamb2	Laminin subunit beta-2	1.27	7.11E-04
P10493	Nid1	Nidogen-1	1.23	5.14E-03
P97927	Lama4	Laminin subunit alpha-4	1.17	1.58E-02
Q9D1H9	Mfap4	Microfibril-associated glycoprotein 4	0.49	3.87E-03
		Core matrisome/proteoglycans
E9PZ16	Hspg2	Heparan sulfate proteoglycan core protein	1.18	2.12E-02
P28653	Bgn	Biglycan	0.79	3.20E-03
		Matrisome-associated/ECM regulators
A0A0R4J0Q4	Loxl1	Lysyl oxidase homolog	1.56	8.88E-04
P21981	Tgm2	Protein-glutamine gamma-glutamyltransferase 2	1.38	3.10E-02
P53690	Mmp14	Matrix metalloproteinase-14 OS = Mus musculus	0.82	4.25E-02
		Matrisome associated/secreted factors
A0A0A6YXU9	Tgfb2	Transforming growth factor beta OS = Mus musculus	2.11	2.70E-02
P04202	Tgfb1	Transforming growth factor beta-1 proprotein	0.83	1.39E-02

Abbreviations: BAPN, β-aminopropionitrile; CNTL, control; ECM, extracellular matrix.

To further validate our results obtained from the proteomic profiling, we performed western blot analysis. The myometrial tissues isolated on gestation day 18 from control and BAPN-treated mice were subjected to PBS and followed by urea extraction to analyze the protein levels of collagen I, III, and elastin. The protein levels of collagen I, III, and elastin were higher in the BAPN-treated group compared to control (Fig. 9). This result demonstrates that inhibition of LOX activity presumably leads to less crosslinked and weaker collagen and elastin, which makes them easily extractable. Collectively, these results demonstrate that the lysyl oxidases maintain integrity of the ECM in myometrium during pregnancy.

Figure 9.

Impact of inhibition of activity of lysyl oxidases on the integrity of myometrial extracellular matrix. Western blot analysis of collagen I (140-210 kDa), collagen III (140-180 kDa), and tropoelastin (monomer: 72 kDa) in urea fractions of myometrial tissues collected from gestation day 18 mice (left). Pregnant mice were treated with BAPN through drinking water from gestation day 12 until gestation day 18. Untreated mice served as control (n = 3/group). Quantification of levels of collagen I, III, and elastin in BAPN-treated samples compared with control samples (right). The protein levels were normalized to β-actin levels (n = 3/group, *P < .05). Note: The protein levels of collagen I, III, and elastin were higher in the BAPN-treated group compared to the untreated control group because these extracellular matrix proteins were easily extractable in BAPN-treated tissues.

Discussion

This study expands our understanding of myometrial tissue function beyond smooth muscle cells to include ECM to better outline the myometrial transition from quiescence to contractility at parturition. The predominant ECM components such as fibrillar collagen and elastic fibers undergo structural remodeling in the mouse myometrium during pregnancy [15]. Among all factors, lysyl oxidase enzymes are emerging as critical targets because of their ability to crosslink collagen and elastic fibers and thus mediate the strength and stability of the ECM. The results of this study demonstrate that a subset of these lysyl oxidase enzymes is highly induced in the pregnant mouse myometrium. The inhibition of activity of these enzymes results in delayed parturition because of the failure of myometrium to mount a contractile response at an appropriate time to initiate parturition. In this study, we have uncovered a novel role of lysyl oxidases mediating myometrial contractility for parturition timing in mice.

ECM remodeling with respect to fibrillar collagen and elastic fiber reorganization in the cervix and myometrium has unique differences to accomplish their function during pregnancy and parturition [15, 37, 43, 44]. In myometrium, thin and scattered collagen fibers are reorganized into thick and straight bundles as pregnancy progresses, whereas in the cervix these collagen fibers are reorganized from thin and linear fibers to thick and wavy bundles [15, 37]. These structural alterations support their respective functions in the myometrium and cervix to accommodate fetal growth and to undergo softening to prepare for birth, respectively. Notably, the crosslinking lysyl oxidase enzymes that determine strength and stability of collagen and elastic fibers are differentially expressed in myometrium and cervix during pregnancy. In mouse cervix, the level of LOX transcript and activity are high during NP, decline early in pregnancy, and remain low until gestation day 18 [37]. This temporal regulation has direct functional implications in the cervix. In the NP mouse cervix, the high level of LOX increases the crosslinks of collagen and elastic fibers to maintain mechanical strength. In early pregnancy, LOX activity declines in the cervix, leading to a reduction in the level of crosslinks, change in the ECM structure, and a decline in mechanical stiffness that subsequently leads to progressive softening [36, 37, 45]. In mouse myometrium, Lox transcripts are significantly induced in the quiescent phase of myometrial remodeling, similar to the temporal expression of collagen and elastin. These elevated activities presumably increase collagen and elastin crosslinks, promote change in the ECM structure and increase mechanical stiffness of the myometrium during pregnancy to support mechanical loading exerted by the growing fetus and also to mount forceful contractile response at term for the expulsion of the fetus. Our previous and current studies indicate that many genes encoding ECM proteins are differentially regulated by steroid hormones in mouse myometrium. Comparison of the expression pattern of LOX in the cervix and myometrium reveals that LOX is differentially regulated in the same steroid hormonal milieu in distinct reproductive tissues to achieve different functions.

This study has a few limitations. The lysyl oxidases are highly induced very early in pregnancy. However, BAPN treatment was administered on gestation day 12 to avoid its impact on the fetus, fetal membrane, and placental development. Therefore, complete inhibition of lysyl oxidases in myometrium might not be possible through this approach. As previously mentioned, lysyl oxidases are widely expressed in almost all reproductive and fetal tissues [25, 26, 34-38]. Therefore, the results of this study might have been confounded by widespread effect of BAPN on all of these tissues. In addition, there are 5 lysyl oxidases and 3 of them are highly induced in pregnant mouse myometrium. Because of their functional redundancy, identifying their unique function might be very challenging. To circumvent all these confounding factors, the lysyl oxidases should be specifically deleted in the myometrium to understand their distinct function in the myometrium during pregnancy and parturition. Despite these limitations, this study has provided important information on the function of lysyl oxidases in regulation of the structure and function of myometrial ECM and their impact in the processes of pregnancy and parturition.

Abbreviations

BAPN

β-aminopropionitrile

BHZ

Hydrazide-Biotin

ECM

extracellular matrix

FA

formic acid

HRP

horseradish peroxidase

LOX

lysyl oxidase

NP

nonpregnant

PTB

preterm birth

SPS

Synchronous Precursor Selection

TMT

tandem mass tag

Funding

This work was supported by funding from the National Institutes of Health (Grant No. R00HD090301) to S.N. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Confocal microscopy was performed on a Nikon A1R-HD point scanning confocal microscope supported by the National Institutes of Health (Grant No. 1S10OD025030-01). The Vermont Biomedical Research Network Proteomics Facility (RRID:SCR_018667) is supported through the National Institutes of Health Grant P20GM103449 from the INBRE Program of the National Institute of General Medical Sciences.

Author Contributions

S.N. conceptualized the study, designed the experiments, and wrote the manuscript. S.N., A.O., C.D., S.G., and Y.L. conducted experiments and analyzed the data. S.N., A.O., C.D., S.G., M.M., and Y.L. reviewed, edited, and approved the manuscript.

Disclosures

The authors have declared that no conflict of interest exists.

Data Availability

Original data generated and analyzed during this study are included in this published article or in supplementary materials. A link to the supplementary materials (Supplementary Tables S1 and S2) is included in reference [39]. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD055537.