A successful pregnancy is orchestrated by the sequential and coordinated events happening in the FRT [1]. Each of these events is crucial to advance to the next step in pregnancy. For example, as a first step, the ovary undergoes ovulation to release a mature ovum into the oviduct where it gets fertilized by the sperm, and a zygote is formed in the ampullary region of the oviduct. The newly formed zygote undergoes several rounds of cleavage to form a morula while being transported through the isthmus region of the oviduct. Timely transport of the embryos from the oviduct into the uterus and synchronous development of the implantation-competent blastocyst and a receptive uterus is a prerequisite for implantation initiation [2]. Any perturbations in these early pregnancy events can lead to adverse ripple effects that can compromise pregnancy outcomes. Previous studies have reported that impaired oviductal transport of embryos can lead to pregnancy failure and cause infertility in mice [36]. However, the underlying mechanism is not yet completely understood.

Autophagy is a cellular process, evolutionarily conserved from yeast to mammals, that recycles long-lived proteins and organelles to maintain cell energy homeostasis. Autophagy is classically activated through the nutrient sensor or mammalian target of the rapamycin complex. Genetic screens for autophagy-defective mutants in yeast and other fungi have currently identified 41 autophagy-related (ATG) genes that play a primary role in autophagy [7, 8]. Approximately half of these genes have homologs in higher organisms, and Atg14 (also known as Barkor for Beclin 1 (Becn1)-associated autophagy-related key regulator) is one of them [9, 10]. ATG14 is part of a protein complex that is composed of Beclin 1, vacuolar sorting protein 15 (VPS15), and VPS34 (also named as Pik3c3, the catalytic subunit of the class III phosphatidylinositol 3-kinase), and this ATG14-containing complex plays an important role in the initiation process of autophagy.

With a better understanding of the fundamental process of autophagy, its pathophysiological functions have begun to be appreciated in the female reproductive tract. For example, mice deficient in key autophagy genes such as genetic knockout of Atg7 or Becn1 result in primary ovarian insufficiency and reduced progesterone production [11, 12]. Similarly, recent studies from our group established the roles of three different autophagy-specific genes: ATG16L, FIP200/RB1CC1, or BECN1 in endometrial physiological processes, including receptivity and decidualization [1315]. However, none of these proteins exhibited any discernible impact on oviduct function. Surprisingly, in this study, we revealed a critical role for Atg14 in maintaining proper oviduct function, specifically enabling the transport of embryos to the uterus—a function distinct from that of other autophagy-related proteins. Loss of Atg14 in the oviduct resulted in severe structural abnormalities, compromising its cellular plasticity and integrity, ultimately leading to embryo retention and infertility.

Materials and Methods

Animal care and use

All animal studies were approved by the Institutional Animal Care and Use Committee of Washington University School of Medicine, Saint Louis, MO, USA and Use Committee of Baylor College of Medicine, Houston, TX, USA. Atg14 f/f mice were provided by gift from Dr. Shizuo Akira at the Department of Host Defense, Research Institute for Microbial Diseases (RIMD), Osaka University, and previously described [16]. Wild type Pr-cre mice were provided by Dr. John Lydon at Baylor College of Medicine, Houston and previously described [17]. Atg14flox/flox mice, in which exon 4 was flanked by loxp sites, were bred to progesterone receptor cre (PRcre/+) mice to generate (Atg14flox/flox; PRcre/+ mice), hereafter referred to as Atg14 cKO mice. Both control and conditional knockout females were generated by crossing females carrying homozygous Atg14flox/flox alleles with Atg14 cKO males. Foxj1-cre mice were a generous gift from Dr. Michael J. Holtzman at Washington University St. Louis and previously described [18]. Foxj1/Atg14 cKO mice were generated by crossing females carrying homozygous Atg14 f/f/Foxj1f/f females with Atg14f/f/Foxj1-cre males. All mice were age-matched and on a C57BL/6 genetic background (The Jackson Laboratory, Bar Harbor, ME). Mice were genotyped by PCR analysis of genomic DNA isolated from tail clippings using the gene-specific primers listed in Supplementary Table 1.

Fertility analysis and timed mating

Female fertility was determined by mating cohorts of Atg14 cKO experimental (n=6) and control Atg14 f/f (n=4) females individually starting at 8 weeks of age with sexually mature males of proven fertility. Similarly, breeding trials were set up for Foxj1/Atg14-cre females (n=6) and Foxj1/Atg14 control 8-week-old f/f females (n=6). The numbers of litters and pups were tracked over 6 months for each female. Pups per litter for each genotype are reported as mean ± SEM. For timed mating, the morning on which the copulatory plug was first observed was considered 1 dpc. To visualize implantation sites, mice received a tail vein injection of 50 μL of 1% Chicago Sky Blue dye (Sigma-Aldrich, St. Louis, MO, USA) at 5 dpc just before sacrifice.

Steroid hormone treatments

The hormonal profile of pregnancy at the time of implantation was done using a previously described experimental scheme [14]. Briefly, Atg14 cKO and control females (8 weeks old) were bilaterally ovariectomized under ketamine anesthesia with buprenorphine-SR as an analgesic. Mice were allowed to rest for two weeks to dissipate all endogenous ovarian hormones. After the resting period, mice were injected with 100 ng of estrogen (E2; Sigma-Aldrich) dissolved in 100 µL of sesame oil on two consecutive days and then allowed to rest for two days. At this point, mice were randomly divided into three groups of five: Vehicle-treated (E2 priming) mice received four consecutive days of sesame oil injections; E2 group mice received three days of sesame oil injections followed by a single injection of 50 ng of E2 on the fourth day; The E2/P4 mice received 1 mg of progesterone (P4; Sigma-Aldrich) for three consecutive days followed by a single injection of 1 mg P4 plus 50 ng E2 on the fourth day. All hormones were delivered by subcutaneous injection in a 90:10 ratio of sesame oil: ethanol. Mice were euthanized 16 hours after the final hormone injection to collect the uteri. A small piece of tissue from one uterine horn was processed in 4% neutral buffered paraformaldehyde for histology, and the remaining tissue was snap-frozen and stored at -80 °C.

Hematoxylin and eosin staining

Tissues were fixed in 4% paraformaldehyde, embedded in paraffin, and then sectioned (5 μm) with a microtome (Leica Biosystem, Wetzlar, Germany). Tissue sections were deparaffinized, rehydrated, and stained with Hematoxylin and Eosin (H&E) as described previously [19]. All the histology was performed on three sections from each tissue of individual mice, and one representative section image is shown in the respective figures.

Histological analysis

For histological analysis, the collected tissues (oviduct or uteri) were fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned. Sections (5 µm) were immunostained (n=5 per group) as described previously [19]. Briefly, after deparaffinization, sections were rehydrated in an ethanol gradient and then boiled for 20 min. in citrate buffer (Vector Laboratories Inc., Newark, CA, USA) for antigen retrieval. Endogenous peroxidase activity was quenched with Bloxall (Vector Laboratories Inc.), and tissues were blocked with 2.5% goat serum in PBS for 1 hr (Vector Laboratories Inc.). After washing in PBS three times, tissue sections were incubated overnight at 4 °C in 2.5% goat serum containing the primary antibodies listed in Supplementary Table 2. Sections were incubated for 1 hr with biotinylated secondary antibody, washed, and incubated for 45 min with ABC reagent (Vector Laboratories Inc.). Color was developed with 3, 3’-diaminobenzidine (DAB) peroxidase substrate (Vector Laboratories Inc.), and sections were counter-stained with hematoxylin. Finally, sections were dehydrated and mounted in Permount histological mounting medium (Thermo Fisher Scientific, Waltham, MA, USA).

Transmission electron microscopy (TEM)

For ultrastructural analysis, oviducts were fixed in 2% paraformaldehyde/2.5% glutaraldehyde (Ted Pella Inc., Redding, CA) in 100 mM cacodylate buffer, pH 7.2 for 1 hr at room temperature and then overnight at 4°C. Samples were washed in cacodylate buffer and postfixed in 1% osmium tetroxide (Ted Pella Inc.) for 1 hr. Samples were then rinsed extensively in dH20 prior to en bloc staining with 1% aqueous uranyl acetate (Ted Pella Inc.) for 1 hr. Following several rinses in dH20, samples were dehydrated in a graded series of ethanol and embedded in Eponate 12 resin (Ted Pella Inc.). For initial evaluation semithin sections (0.5 μm) were cut with a Leica Ultracut UCT7 ultramicrotome (Leica Microsystems Inc., Bannockburn, IL) and stained with toluidine blue. Sections of 95 nm were then cut and stained with uranyl acetate and lead citrate and viewed on a JEOL 1200 EX II transmission electron microscope (JEOL USA Inc., Peabody, MA). Images at magnifications of 3,000X to 30,000X were taken with an AMT 8-megapixel digital camera (Advanced Microscopy Techniques, Woburn, MA).

Immunofluorescence analysis

Formalin-fixed and paraffin-embedded sections were deparaffinized in xylene, rehydrated in an ethanol gradient, and boiled in a citrate buffer (Vector Laboratories Inc.) for antigen retrieval. After blocking with 2.5% goat serum in PBS (Vector laboratories) for 1 hr at room temperature, sections were incubated overnight at 4°C with primary antibodies (Supplementary Table 2) diluted in 2.5% normal goat serum. After washing with PBS, sections were incubated with Alexa Fluor 488-conjugated secondary antibodies (Life Technologies, Carlsbad, CA, USA) for 1 hr at room temperature, washed, and mounted with ProLong Gold Antifade Mountant with DAPI (Thermo Fisher Scientific). All Immunofluorescence images were obtained using a Zeiss LSM 880 confocal microscope (10x and 40x objective lens).

Western blotting

Protein lysates (40 µg per lane) from uteri or oviducts were loaded on a 4-15% SDS-PAGE gel (Bio-Rad, Hercules, CA, USA), separated in 1X Tris-Glycine Buffer (Bio-Rad), and transferred to PVDF membranes via a wet electro-blotting system (Bio-Rad), all according to the manufacturer’s directions [20]. PVDF membranes were blocked for 1 hour in 5% non-fat milk in Tris-buffered saline containing 0.1% Tween-20 (TBS-T, Bio-Rad), then incubated overnight at 4 °C with antibodies listed in Supplementary Table 2 in 5% BSA in TBS-T. Blots were then probed with anti-Rabbit IgG conjugated with horseradish peroxidase (1:5000, Cell Signaling Technology, Danvers, MA, USA) in 5% BSA in TBS-T for 1 hr at room temperature. Signal was detected with the Pierce™ ECL Western Blotting Substrate (Millipore, Billerica, MA, USA), and blot images were collected with a Bio-Rad ChemiDoc imaging system.

RNA Isolation and Quantitative Real-Time RT-PCR Analysis

Tissues/cells were lysed in RNA lysis buffer, and total RNA was extracted with the Purelink RNA mini kit (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. RNA was quantified with a Nano-Drop 2000 (Thermo Fisher Scientific). Then, 1 µg of RNA was reverse transcribed with the High-Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific). The amplified cDNA was diluted to 10 ng/µL, and qRT-PCR was performed with primers listed in the Supplementary Table 1 and TaqMan 2X master mix (Applied Biosystems/Life Technologies, Grand Island, NY, USA) on a 7500 Fast Real-time PCR system (Applied Biosystems/Life Technologies). The delta-delta cycle threshold method was used to normalize expression to the reference gene 18S.

Treatment of mice with Polyphyllin VI

Polyphyllin VI (Selleck chemicals, Houston, TX, USA), a pharmacological agent that induces caspase-1-mediated pyroptosis, was used to study its effects on embryo transport in the oviduct [21]. Eight-week-old C57BL/6 mice were injected for three consecutive days starting from 1 dpc with Polyphyllin VI activator dissolved in 40% PEG-300 (15mg/kg body weight). Dimethyl sulfoxide with 40% PEG-300 was administered as a vehicle.

Transfection and human endometrial stromal cell (HESC) decidualization

HESCs were plated in six-well culture plates at a cell density of 1x105 per well in triplicates. For siRNA-mediated knockdown of Atg14, HESCs were treated with Lipofectamine RNAiMAX reagent (Thermo Fisher Scientific, Waltham, MA)) and 60 pmol of either non-targeting siRNA (D-001810-10-05) or siRNAs targeting Atg14 (L-020438-01-0005). Six hours after the transfection, the media was replaced with complete HESC media. 48 hrs later, HESCs were treated with Opti-MEM (Thermo Fisher Scientific) reduced serum media containing 2% fetal bovine serum (FBS), E2 (100 nM), 10 mM medroxyprogesterone acetate (MPA, Sigma-Aldrich) and 50 mM cAMP (Sigma-Aldrich), which constitutes the decidualization media. The first day that HESCs were cultured in decidualization media was assigned day 0. Decidualization media was renewed every two days. Cells were harvested at appropriate time points as per experimental conditions. Total RNA was isolated to assess transcript levels of the decidualization markers: prolactin (PRL) and insulin-like growth factor binding protein-1 (IGFBP-1).


A two-tailed paired student t-test was used to analyze data from experiments with two experimental groups and one-way ANOVA followed by Tukey’s post hoc multiple range test was used for multiple comparisons. All data are presented as mean ±SEM. GraphPad Prism 9 software was used for all statistical analyses. Statistical tests, including p values, are reported in the corresponding figure legends or, when possible, directly on the data image. To ensure the reproducibility of our findings, experiments were replicated in a minimum of three independent samples, to demonstrate biological significance, and at least three independent times to ensure technical and experimental rigor and reproducibility.


Conditional deletion of Atg14 in the FRT results in infertility despite the normal ovarian function

To explore the role of Atg14 in uterine function, we first determined its expression levels in uterus during early pregnancy (Day 1-7) in mice. We found distinct expression of ATG14 in all the uterine compartments (luminal epithelium, glands, and stroma) by day 1 of pregnancy, which disappeared by day 2 of pregnancy (Fig. 1A). However, the ATG14 expression in uterus reappeared by day 3 and persisted through the day 7 of pregnancy. The period from day 3 to day 7 is critical, as it is during this time when uterus begins to prepare for embryo implantation and undergoes decidualization process. Consistent with protein expression, similar Atg14 expression at mRNA levels was noted in uteri from early pregnant mice (Fig. 1B). This analysis suggests a potential role for ATG14 protein in the uterine physiologic adaptations during early pregnancy. Thus, to study the role of ATG14 in uterine function, we generated a conditional knockout (PRcre/+ /Atg14flox/flox) mouse model by crossing Atg14 flox/flox mice with mice expressing Cre recombinase under the control of progesterone receptor promoter (PRcre/+). Histological examination of the uterus from adult females showed no gross morphological differences between Atg14 cKO and control mice (Fig. S1A). Further, we did not find any overt defects in ovary as cKO mice had normal follicles and corpus luteum as like their corresponding controls (Fig. S1B). Analysis of transcript levels from the uterus, ovary, and liver showed that while Atg14 levels were efficiently depleted in uteri from cKO mice the levels were unaltered in the ovary and liver samples in control and cKO mice (Fig. 1C). Immunofluorescence analysis further confirmed the efficient deletion of ATG14 in all uterine compartments (epithelium and stroma) of cKO mice uteri compared to corresponding control groups (Fig. 1D).

Atg14 loss in the FRT results in infertility.

(A) Representative immunofluorescent images of uteri from pregnant mice at the indicated days of pregnancy stained with an ATG14-specific antibody (green). LE: luminal epithelium, G: glands, S: stroma, Scale bar: 200 µm. (B) Relative transcript levels of Atg14 mRNA in uteri from pregnant mice at indicated days of pregnancy. mRNA levels are normalized to levels of 18S m-RNA. Data are presented as mean ±SEM. (C) Relative mRNA levels of Atg14 in 8-week-old virgin control and cKO mice uteri, ovary, and liver (n=5). mRNA levels are normalized to levels of 18S mRNA. Data are presented as mean ±SEM; ***P<0.001, P>0.05, ns=non-significant. (D) Representative immunofluorescent images of ATG14 expression in different uterine compartments in control and Atg14 cKO mice. (E) (left panel) Relative number of pups/female and (right panel) and relative number of total pups/months of Atg14 control and cKO mice sacrificed after the breeding trial. Data are presented as mean ±SEM; P>0.05, ns=non-significant.

Considering the effective ablation, a 6-month breeding study was performed mating virile wild-type male mice with adult Atg14 cKO and control female mice. We found that Atg14 cKO females did not deliver any litter during 6-months trial period. However, the control female (Atg14 flox/flox) mice delivered an average of ∼7-8 pups per litter every month (Table 1) (Fig. 1E & F. These findings suggest an indispensable role of Atg14 in female fertility with intact ovarian function.

Six-month breeding trial of Atg14 control and cKO females with wild-type males

Six-month breeding trial of Foxj1/Atg14 control and cKO females with wild-type males

Loss of Atg14 results in impaired embryo implantation and uterine receptivity in mice

Based on the normal ovarian morphology, we posited that the infertility observed in females with Atg14 cKO status could be attributed to compromised uterine functioning. Analysis of the implanting embryos at Day 5 of pregnancy found no implantation sites in the uteri of Atg14 cKO females, whereas ∼8 to 9 implantation sites were seen in the uteri of control mice at 5 dpc (Fig. 2A, left panel). Whilst there was no embryo present in the uterine lumen of Atg14 cKO mice, a fully attached embryo encapsulated by the luminal uterine epithelium, were seen in control mice uteri were (Fig. 2A, middle panel). Additionally, MUC1, a receptive marker expression was persisted in the luminal epithelium of cKO mice uteri at 5 dpc as well as 4 dpc (Fig. 2A, right panel, Fig. S2A). The process of successful embryo attachment and implantation within the uterus necessitates a transition from a non-receptive to a receptive state, a transformation orchestrated under the regulated influence of steroid hormones. Thus, the effects of Atg14 loss on uterine responsiveness to steroid hormones E2 and P4 was characterised. We performed an established hormone-induced uterine receptivity experiment, which involved ordered and co-stimulatory actions of E2 and P4 leading to the initiation of a receptive phase [22]. In response to E2 treatment, uteri from both Atg14 control and cKO mice showed similar proliferation as evidenced from Ki-67 staining (Fig. 2B, middle panel). Following the co-stimulation with E2+P4 treatment, epithelial proliferation was inhibited with concomitant induction of stromal cell proliferation in control mice uteri indicating fully receptive uteri (Fig. 2C). Interestingly, cKO mice uteri failed to elicit sub-epithelial stromal cell proliferation and showed intact P4-driven inhibition of epithelial proliferation. Consistently, uteri from Atg14 cKO mice at 4 dpc showed reduced number of proliferating stromal cells (Fig. S2B). These results demonstrate that ATG14 is required for P4-mediated but not for E2-mediated actions during uterine receptivity.

Atg14 is critical for embryo implantation, uterine receptivity, and stromal cell decidualization.

(A) Gross images of 5.0 dpc uteri of control and Atg14 cKO mice injected with Chicago Sky Blue dye to visualize implantation sites (denoted by black arrows) (left panel). H&E-stained cross-sections (4X & 40X) of 5.0 dpc uteri of control and Atg14 cKO mice to visualize embryo implantation (Middle panel). Immunofluorescence analysis of uterine tissues from control and Atg14 cKO mice, stained with MUC1 and KRT8 (right panel). (B) Representative immunofluorescence images of uteri from control and Atg14 cKO mice stained for Ki-67 following Oil or E2 or E2+P4 treatment (n=5 mice/group); scale bar: 100 μm. (C) Representative images of uteri (left) and uterine histology after artificial decidualization (middle). Immunofluorescence analysis of PH3 in decidualized horn of Control and Atg14 cKO mice. (D) Relative transcript levels of Atg14, and decidualization markers (Bmp2 and Wnt4) after artificial decidualization. (n=4-6 per group). Data are presented as mean ±SEM. *P < 0.05; **P < 0.01; ***P < 0.001 compared with controls. 18S was used as an internal control. (F) Morphology of human endometrial stromal cells transfected with control or Atg14 siRNA followed by treatment with decidualization media and (G) qPCR analysis of ATG14 and the indicated decidualization markers. Data are presented as mean ±SEM. *P < 0.05; **P < 0.01; ***P < 0.001 compared with controls.

Given that stromal cell proliferation is a prerequisite for successful decidualization, we wondered whether cKO mice uteri can undergo the decidualization process. The uteri of Atg14 cKO mice failed to respond to oil injection and did not show any increase in size, reflecting their failure to undergo decidualization. (Fig. 2C, left panel). Further, a significant reduction in proliferative pH3-positive and multinucleated decidualized stromal cells were noted in cKO mice stimulated uterine horns (Fig. 2C). Also, the decidualization markers Wnt4 and Bmp2 was significantly reduced in stimulated horns of cKO mice compared to control mice (Fig. 2D). We determined next whether ATG14 is likewise required for HESCs decidualization. Control siRNA transfected cells when treated with EPC seemed to change their morphological transformation from fibroblastic to epithelioid (Fig. 2E) and had increased expression of the decidualization markers IGFBP1 and PRL by day three only (Fig. 2F). In contrast, HESCs with ATG14 knockdown failed to undergo morphological transformation with concomitant reduction in PRL and IGFBP1 marker expression (Fig. 2F). Together, these results underscore the critical role of ATG14 in embryo implantation, uterine receptivity, and the decidualization.

Atg14 is required for maintaining oviductal cell plasticity and embryo transport

Given that Atg14 cKO mice had impaired embryo implantation, we wondered whether embryos are reaching the uterus timely through oviduct. To determine this, we flushed embryos from both the control and cKO mice uteri on day 4 of pregnancy. Interestingly, in cKO mice, we could retrieve only 1-2% of embryos from their uteri, whereas in control mice, 100% of well-developed blastocysts were retrieved from their uteri (Fig. 3A, Upper and lower panel). To ensure the timely transport of all embryos from the oviducts to the uteri, we also flushed oviducts from both control and cKO mice. As expected, in the control mice oviduct flushing, we could not recover any embryos or blastocysts, indicating their precise and timely transport to the uterus. Unexpectedly, oviduct flushing from cKO mice resulted in the retrieval of approximately 80-90% of blastocysts, suggesting their potential entrapment within the oviducts, impeding their transit to the uterus (Fig. 3B). Similarly, when we super-ovulated the mice and looked for the embryos on the day 4 of pregnancy, we could recover ∼20-25 embryos from the cKO mice oviducts compared to only 4-5 embryos that were able to reach the uterus. In comparison, in control mice, 90% of embryos were able to complete their pre-destined and timely transport to the uterus (Fig. 3C), except 1-2% of unfertilized embryos, which remained in their oviducts.

Atg14 is critical for embryo transport in oviduct.

(A) Representative images (upper panel) and percentage of embryos (lower panel) collected at 4 dpc from the uteri or the oviducts of control or Atg14 cKO mice. (B) Histological analysis using H&E staining of the ampullary and isthmic region of the oviduct from control and Atg14 cKO female mice at 4 dpc (n=3 mice/genotype). (C) Embryos retrieved from the oviduct and uterus of super-ovulated Atg14 control or cKO mice at 4 dpc. (D) Immunofluorescence analysis of KRT8 (green), and α-SMA (Red), in oviduct of 4 dpc control and Atg14 cKO mice. (E) TEM of oviducts at 4 dpc from control and Atg14 cKO.

To understand the underlying cause for retained embryos in cKO mice oviducts, we performed histology analysis and determined the structural morphology of their oviducts. The cKO mice oviduct lining shows marked eosinophilic cytoplasmic change akin to decidualization in human oviducts. Some of the cells showed degenerative changes with cytoplasmic vacuolization and nuclear pyknosis, loss of nuclear polarity, and loss of distinct cell borders giving an appearance of fusion of cells (Fig. 3D). The marked cytologic enlargement appears to cause luminal obliteration or narrowing resulting in a completely unorganized, obstructed, and narrow lumen, thereby, hampering the path of embryos to reach the uterus as evident from an entrapped embryo shown in Fig. 3B. Further, epithelial (cytokeratin KRT8-positive) and myosalpinx (α-smooth muscle active [SMA]-positive,) marker analysis revealed completely distorted epithelial structures (in terms of loss of epithelial cell integrity or plasticity). However, there were no overt defects in the muscle organization in oviducts of cKO mice (Fig. 3D). To gain more insights into structural defects, we performed TEM analysis on oviducts collected from both control and cKO females on day 4 of pregnancy. Interestingly, oviducts from cKO females had numerous altered mitochondrial structures with abnormally enlarged mitochondria and a less dense matrix compared to control oviducts that had small and compact mitochondria with tight cristae and a dense matrix (Fig. 3E). Together, these results demonstrate that Atg14 is required for maintaining oviductal cell plasticity and embryo transport.

Selective loss of Atg14 in oviduct cilia is dispensable for female fertility

The oviduct epithelium primarily consists of two types of cells: ciliated and secretory (non-ciliated) cells. Ciliated cells play a role in embryo and oocyte transport by means of ciliary beat, and non-ciliated/secretory cells produce an oviductal fluid that is rich in amino acids and various molecules, thereby providing an optimal micro-environment for sperm capacitation, fertilization, embryonic survival, and development [23]. Therefore, we determined whether oviducts from cKO mice possess normal ciliated and secretory cell composition. To do so, we examined the expression of various markers, including acetylated-α-tubulin (cilia marker), FOXJ1 (ciliogenesis markers), and PAX8 (non-ciliated/ secretory cell marker). As shown in Fig. 4A, while we observed normal ciliary structures in the ampulla of both control and cKO oviducts, there was a remarkable reduction in the ciliary cells (indicated by fewer α-tubulin and FOXJ1-positive cells) as well as secretory cells (indicated by fewer PAX-8 positive cells) in the isthmus of cKO oviducts (Fig. 4A & B).

Atg14 loss in oviduct cilia is dispensable for embryo transport.

(A) Immunofluorescence analysis of acetylated a-tubulin (green), and DAPI (blue) in oviduct of 4 dpc control and Atg14 cKO mice. (B) Immunohistochemical analysis of FOXJ1 and PAX8 at 4 dpc. (C & D) Relative number of pups/femalsoyale and relative number of total pups/months of control and Foxj1/Atg14 cKO mice sacrificed after the breeding trial. Data are presented as mean ± SEM; P>0.05, ns=non-significant. (E) Gross images of 5.0 dpc uteri of control and Atg14 cKO mice injected with Chicago Sky Blue dye to visualize implantation sites (denoted by black arrows).

Given the importance of cilia in embryo transport, we wondered whether the loss of Atg14 in oviduct cilia has any impact on embryo transport. To address this, we generated a Foxj1-cre/+; Atg14 f/f mouse model, wherein Atg14 will be ablated only in ciliary epithelial cells. Interestingly, the 6-month breeding trial revealed that loss of Atg14 in oviductal cilia had no impact on fertility (Table 2, Fig. 4C & D). Consistently, D5 implantation study analysis showed ∼7-8 visible embryo implantation sites in Foxj1/Atg14 cKO mice like their corresponding controls suggesting that embryos were able to make their way to the uterus in a timely manner and undergo implantation despite the ablation of Atg14 in the oviduct ciliary epithelial cells (Fig. 4E). These findings suggest that ciliary expression of Atg14 is dispensable for embryo transport.

Atg14 prevents unscheduled pyroptosis activation in the oviduct to enable embryo transport

Having identified the cellular mechanisms of Atg14, we aimed to delineate the underlying molecular mechanisms associated with Atg14 function in oviduct. Since, we found mitochondrial defects with ATG14 loss and altered mitochondrial structures have been linked to the activation of pyroptosis, we determined if ATG14 regulates pyroptosis in oviduct [2426]. First, we assessed the key primary executors of the pyroptosis pathway, GSDMD and caspase-1. Immunofluorescence analysis revealed a remarkable upregulation in GSDMD and caspase-1 expression in the cKO oviducts compared to controls (Fig. 5A & B). Consistently, western blot analysis also confirmed elevated expression of caspase-1 and GSDMD in cKO oviducts in comparison to control oviducts as shown in Fig. 5C. The qPCR analysis further demonstrated elevated levels of inflammatory markers, such as Tnf-α and Cxcr3 in cKO oviducts compared to control ones (Fig. 5D). Based on these findings, we posit that Atg14 exerts a regulatory control over the pyroptotic pathway and activation of pyroptosis owing to loss of ATG14 within the oviduct adversely impacts embryo transport. However, to substantiate the aforementioned notion, we also evaluated the impact of unscheduled activation of pyroptosis on the embryo transport. To test this, we employed a pyroptosis inducer, Polyphyllin VI and chose the optimal dose based on the established studies [21]. Wild type females were mated with virile males and following the plug detection treated with Polyphyllin VI for three consecutive days from 1 to 4 dpc (Fig. 5E). We chose to treat to 1 to 4 dpc for treatment to activate unwarranted pyroptosis in oviduct prior to embryo transport. Following the treatments, the oviducts, and uteri from both vehicle-treated and Polyphyllin VI-treated were flushed. We found that pregnant females treated with Polyphyllin VI showed ∼50% embryo retention in the oviduct, whereas in the vehicle-treated group, no embryos were retained in the oviduct (Fig. 5F & G). Precisely activating the unscheduled pyroptosis during the critical period of embryo transport is technically challenging. Nonetheless, our findings provide evidence that unscheduled pyroptosis adversely affects embryo transport through the oviduct. Taken together, these results demonstrate that Atg14 safeguards pyroptosis activation in the oviduct and allows the smooth transport of embryos to the uterus (Fig. 5H).

Atg14 promotes embryo transport in oviduct by preventing pyroptosis activation.

Immunofluorescence analysis of GSDMD (red) + KRT8 (green), Caspase-1 (red) + KRT8 (green) in oviducts of adult control and Atg14 cKO mice. Tissues were counterstained with DAPI (blue) to visualize nuclei; scale bars, 100 µm. (B) Immunohistochemical analysis of GSDMD in adult oviduct tissues. (C) Western blotting to show protein levels of Caspase-1, and GSDMD in oviduct tissues. β-actin is used as a loading control. (D) Relative transcript levels of Tnf-α and Cxcr3 in oviduct tissues. Data are presented as mean ±SEM. *P<0.05; **P<0.01; ***P<0.001 compared with controls. 18S was used as an internal control. (E) Experimental strategy for pyroptosis activation in pregnant female mice. (F) Embryos flushed from vehicle or polyphyllin VI treated D-4 pregnant females. (G) Percentage of embryos recovered from oviducts or uteri. (H) Graphical illustration to show embryo transport and pyroptosis regulation in oviduct.


In this study, we delineated the essential role of ATG14 in maintaining the structural integrity of the oviduct by preventing pyroptosis, which enables smooth embryo transit during early pregnancy. Specifically, we dissected the tissue-specific functions of ATG14 in the FRT using Cre driver mice targeting female reproductive tract. Given that PR-Cre expresses postnatally in different tissues of FRT such as corpus luteum, oviducts, and different cellular compartments (epithelial, stromal, and myometrium) of uteri [17], we found that conditional ablation of ATG14-mediated functions in FRT resulted in infertility owing to hampered transport of embryos from the oviduct. Additionally, Atg14 loss in uteri causes impaired embryo implantation, receptivity, and decidualization. Its ablation in the oviduct causes activation of pyroptosis, an inflammatory-associated apoptosis pathway that causes the retention of embryos in the oviduct and prevents their timely transport to the uterus. However, the ovarian-specific functions were found to be unaffected despite the loss of Atg14 in the corpus luteum.

The human endometrium is a complex dynamic tissue that undergoes sequential phases of proliferation and differentiation to support embryo implantation during the conceptive. However, in absence of implanting embryo, endometrium sheds (menstruation) and initiates regeneration process [27]. Previous reports indicated that autophagy is modulated in the human endometrium during the menstrual cycle. For example, autophagy altered during the proliferative and secretory phases of the menstrual cycle with its highest activity occurring during the secretory phase when the stroma is decidualized [28]. Similarly, the level of autophagy was higher in postmenopausal human uterine epithelial cells compared to premenopausal uterine epithelial cells indicating the onset of autophagy upon estrogen deprivation [29]. Although all these studies reported that autophagy is hormonally regulated, however, the role of autophagy-specific proteins in the endometrium is not established till our group’s recent reports. Specifically, studies from our group reported the role of three different autophagy-associated proteins: FIP200 [13], ATG16L [15], and BECLIN-1 in uterine functions [14]. Conditional ablation of Fip200 and Atg16l in the uterus displayed fertility defects owing to impaired implantation, uterine receptivity, and decidualization defects. On contrary, loss of Beclin1 in the uterus caused progressive loss of endometrial progenitor stem cells resulting in severe uterine developmental defects and rendered the mice infertile [14]. Although the autophagy-related proteins we studied so far influenced uterine functions, [1315], we found a distinct role for ATG14 in both the uterine and oviduct-specific functions. It is intriguing to note that the absence of ATG14 did not affect the tissue integrity of the uterus. However, the severe structural abnormalities in the oviduct due to Atg14 ablation unearthed a unique function of Atg14 in maintaining oviduct homeostasis. This distinctive role of ATG14, unlike other autophagy proteins, might be due to its pivotal role in the assembly of PtdIns3K complexes which is not the case for either FIP200 or ATG16L [27, 30, 31]. Nevertheless, understanding the specific contributions of each core autophagy protein in reproductive tract functions is necessary which requires substantial efforts.

The retention of preimplantation embryos in the oviduct established as a significant contributor to implantation failures, presenting challenges to the overall reproductive health of women [46]. Embryo transport in the oviduct is known to be controlled primarily by two major physiological responses: ciliary activity and muscle contractility. In our study, a dramatic decrease in the number of Foxj1+ve ciliary epithelial cells and Pax-8+ve secretory cells in Atg14 cKO mice implied cell-type specific actions for ATG14 in oviduct. Interestingly, the specific ablation of Atg14 in Foxj1+ve ciliary epithelial cells of the oviduct does not appear to impact fertility. This suggests that the absence of ATG14 within ciliary cells does not impact the process of embryo transport. Although unexpected, this is consistent with other reports that have demonstrated the dispensability of cilia for embryo transport in the oviduct [4, 5].

Pyroptosis is a highly inflammatory form of programmed cell death, characterized by cell swelling, flattening of the cytoplasm and large bubbles like protrusions on the plasma membrane [3234]. Several recent studies found a link between autophagy and pyroptosis [35]. In vivo studies also found that mice lacking the autophagy genes Atg14, Fip200, Atg5, or Atg7 in myeloid cells had more pronounced lung inflammation [36]. In our study, cellular swelling, and fused membranous structures (a unique feature of activation of pyroptosis) observed in the oviductal epithelial folds from Atg14 cKO mice. This aberrant activation of the pyroptosis pathway due to loss of Atg14 appear to adversely affect the oviduct cellular plasticity. These findings have implications beyond the oviduct, as autophagy can modulate the inflammatory signalling pathway through different avenues. For example, first, autophagy prevents mitochondrial reactive oxygen species release that can activate the inflammasome by inhibiting IL1-β and IL18 production through the digestion of dysfunctional mitochondria [37]. Second, autophagy is capable of targeting inflammasome complexes for degradation, which prevents the cleavage of pro-IL-1β and pro-IL-18 into biologically active forms. Finally, autophagy machinery further regulates IL-1β levels by engulfing and degrading pro-IL-1β proteins [37]. In our study, accumulation of abnormally enlarged and dysfunctional mitochondria in absence of ATG14 in oviduct clearly suggests induction of severe inflammatory conditions hampering the embryo transit through the oviduct. These findings are further substantiated by the upregulation of pivotal mediators of the pyroptosis program, including GSDMD and caspase-1, in Atg14 cKO mice. Notably, the induction of pyroptosis in the absence of ATG14 is specifically localized to the oviducts, with no parallel activation detected in either the ovary or uterus. This distinct activation pattern suggests a unique interaction between oviduct-specific proteins and ATG14, which keeps pyroptosis at the bay in the oviduct. Nevertheless, a comprehensive understanding of the oviduct-cell-specific roles of ATG14 necessitates additional investigations, employing oviduct-specific cKO mouse models.

In this report, we revealed tissue-specific roles of ATG14 in governing oviductal transport, and uterine receptivity which are necessary for embryo implantation and pregnancy establishment. Of mechanistic insights, we delineated a novel function of ATG14 as a negative regulator of pyroptosis, which is vital for proper embryo transport in the oviduct (Fig. 5 H). Thus, this study not only sheds light on the involvement of autophagy in oviductal embryo transport but also underscores the importance of autophagy in preserving the oviduct tissue integrity. Such insights hold promise for advancing our understanding of gynaecological pathologies associated with the oviduct including tubal pregnancies.


We thank Neil Ghosh, and Goutham Davuluri from our group for the technical assistance. We thank Dr. Robert Lawrence (senior editor at Baylor College of Medicine) for assistance with manuscript editing. We thank WashU Molecular Microbiology Imaging Facility for assisting us with transmission electron microscopy. This work was funded, in part, by the National Institutes of Health/National Institute of Child Health and Human Development (grants R01HD102680, R01HD104813, and R01HD065435) to RK, R01 HD-042311 to JPL, and in part by National Institutes of Health/National Heart, Lung, and Blood Institute (grant R35HL145242) to MJH.

Authors’ roles

PP and RK designed experiments, conducted most of the studies, analysed the data, and wrote the manuscript. AKO, VKM and MNR conducted some of the experiments. MJH, JL, SA, YZ, and KHM, provided critical reagents for the study. RM helped us with oviduct pathology phenotype analysis. RK conceived the project, supervised the work, and wrote the manuscript. All authors critically reviewed the manuscript.

Conflict of Interest

The authors have declared that no conflict of interest exists.

Non-standard Abbreviations

  • WT: Wild Type

  • HESCs: Human Endometrial stromal cells

  • dpc: Days post coitum

  • E2: Estrogen

  • P4: Progesterone

  • cKO: conditional knockout