The mammalian ovary is a structurally complex, heterogenous, and dynamic organ with follicles at different stages of development, remnants of ovulation (corpora lutea), and a heterogeneous stroma (Kinnear et al., 2020; Figure 1A). Very little is known about the long-term homeostasis and relative turnover of the ovarian tissue components during aging. To address this, we visualized the lifespan of ovarian macromolecules in mammals using a combination of stable isotope labeling and multi-isotope imaging mass spectrometry (MIMS) (Figure 1—figure supplement 1A). First, using a two-generational metabolic labeling of animals with ¹⁵N, we generated a cohort of fully ¹⁵N-labeled pups. After birth, labeled females were kept with the labeled dam until weaning, at which time their food source was switched to the ¹⁴N-chow chase. Using previously established methods (Duncan et al., 2017; Kimler et al., 2018; Perrone et al., 2023; Quan et al., 2020), we determined that chase periods of 6 and 10 months represented a biologically relevant reproductive aging continuum as mice within this age range begin to manifest ovarian aging phenotypes, including follicle loss (Figure 1—figure supplement 1B and D), decreased ovulation (Figure 1—figure supplement 1E and F), and increased fibrotic foci in the ovarian stroma (Figure 1—figure supplement 1G–J). Importantly, sufficient numbers of oocytes can still be collected at these timepoints for meaningful further downstream analyses of this rare cell type. We first performed MIMS on ovarian sections to visualize and quantify the abundance of ¹⁴N, representing molecules which have been replaced during the chase period (blue-teal), and ¹⁵N, which represents ¹⁵N-containing molecules that must have persisted through the chase period and therefore are long-lived (orange-pink) (Figure 1B, Figure 1—figure supplement 2A and B; Steinhauser and Lechene, 2013; Guillermier et al., 2014; Nuñez et al., 2018). Within the ovarian follicles, MIMS revealed a strikingly higher abundance of ¹⁵N containing molecules in primordial and primary stages relative to later stage follicles, suggesting that primordial follicles can persist for months with limited macromolecular turnover (Figure 1C). As follicles progress through the primary, secondary, and antral stages the ¹⁵N/¹⁴N ratio decreases due to signal dilution associated with the increase of cell number and follicle growth (Figure 1C, Figure 1—figure supplement 2C). This change in ¹⁵N abundance was most apparent in the granulosa cells where long-lived molecules were significantly higher in those within primordial and primary follicles compared to later follicle stages (Figure 1C and F). In addition to granulosa cells, long-lived molecules also localized to the basement membrane of some early growing follicles (Figure 1—figure supplement 1D). Beyond the follicle, other somatic compartments of the ovary were additionally found to have higher ¹⁵N:¹⁴N ratio suggesting enrichment in long-lived components, including steroidogenic cells (theca layer and corpora lutea), stromal cells, and cells within the ovarian surface epithelium (OSE) (Figure 1B and D). Our quantitative analysis revealed that the OSE had significantly higher ¹⁵N/¹⁴N ratio among the mentioned cell types (Figure 1G). Lastly, we analyzed the ¹⁵N/¹⁴N ratios in the nucleus relative to the cytoplasm, which revealed significant enrichment of long-lived, ¹⁵N- positive molecules in the nuclei of granulosa cells within primary follicles, and cells within the corpora lutea, the stroma, and the OSE (Figure 1E and H). As both proteins and nucleic acids contain nitrogen, the ¹⁵N-positive nuclear signal could correspond to known long-lived nuclear proteins, such as histones, nuclear pore proteins, and lamins, which were previously identified in neuronal, post-mitotic cells (Savas et al., 2012; Toyama et al., 2013). Alternatively, as these nuclear ¹⁵N-hotspots coincided with the ³¹P signal, which is enriched in DNA and correlates with DNA labeling (Guillermier et al., 2017a), this data may suggest that the DNA itself is long-lived (Figure 1E). These results reveal that distinct macromolecular components within select ovarian cells and tissue regions persist throughout the healthy reproductive stage, with limited renewal, and those long-lived molecules persist through the stage where ovaries manifest marked reproductive aging phenotypes.

Although MIMS analysis provides important spatial information on long-lived structures in the ovary, it does not provide the identity of ¹⁵N-containing macromolecules that comprise them. To address this, we performed liquid chromatography mass spectrometry (LC-MS/MS)-based proteomic analysis of ovarian tissues isolated from metabolically labeled mice. After 6 months of ¹⁴N-chase, we identified 36,222±6768 ¹⁴N-peptides mapping to 4106 proteins, and 13±5 ¹⁵ N-peptides, which collectively mapped to 33 LLPs across all the biological replicates (Figure 2A and B, Supplementary file 1). To gain a deeper insight into the persistence of LLPs beyond 6 months, we also analyzed ovaries isolated from females that remained on ¹⁴N chase for 10 months. Although the total numbers of both ¹⁴N and ¹⁵N-peptides and proteins were similar between the two timepoints (39,637±12005 ¹⁴N-peptides mapping to 4464 proteins and 13±6 ¹⁵ N-peptides mapping to 15 LLPs), the majority of LLPs identified at 6 months were no longer identified as long-lived at the 10-month chase timepoint. Only tubulins and select histones persisted and were identified as LLPs at this aged timepoint. Gene ontology (GO) enrichment analysis of LLPs identified at the 6-month chase timepoint revealed significant overrepresentation for terms related to chromatin, nucleosome, tubulin complex, and mitochondria (Figure 2C).

Next, we determined the quantity of each LLP remaining after the ¹⁴N-chase period by calculating the fractional abundance (FA; ¹⁵N-remaining, ¹⁵N/[¹⁴N + ¹⁵N]) for each LLP in the ovary using reconstructed MS1 chromatograms from LC-MS/MS analysis (Park et al., 2008; Figure 2D–F). We found that LLPs had significant differences in FA between the 6- and 10-month timepoints, with 1.13 ± 0.08% and 1.37 ± 0.12% ¹⁵N-remaining, respectively (Figure 2D). The discrete pool of proteins that persisted throughout both timepoints allowed for a unique direct comparison of ¹⁵N to ¹⁴N-peptide peak intensities. This analysis showed reduced abundance of ¹⁵N at the 10-month chase timepoint, consistent with continual, albeit slow, turnover of the protein pool (Figure 2E). Interestingly, at the 6-month chase timepoint, we identified ¹⁵N-peptides mapping to an oocyte-specific protein, zona pellucida-3 protein (ZP3), indicating that a pool persists without turnover for at least 6 months, but less than 10 months, as long-lived ZP3 was no longer identified at the 10-month timepoint (Figure 2F). While histones and tubulin have been previously identified as LLPs in the mammalian brain and heart tissues, the 6-month long persistence of ZP3 in mouse ovaries is unexpected and of potential importance to reproductive biology.

MIMS analysis of ovarian sections also captured oocytes at various stages of development, which in addition to the enrichment of ¹⁵N-signal within the oocyte nucleus revealed multiple smaller cytoplasmic ¹⁵N-hotspots (Figure 3A). Thus, to determine the identity of these ¹⁵N-enriched molecules we isolated fully grown oocytes from ovaries of labeled mice at 6- and 10-month chase timepoints followed by LC-MS/MS analysis (Figure 3B and Figure 3—figure supplement 1A). In oocytes, we identified a total of 2919 proteins at 6 months and 3234 proteins at 10 months (Figure 3B, Supplementary file 2). Although after 6 months of chase, we identified 146 LLPs in oocytes, only 11 LLPs were identified after 10 months, indicating that by this timepoint a vast majority of LLPs have been degraded and renewed. Interestingly, the GO analysis of LLPs in oocytes identified at 6-month timepoint revealed a significant enrichment of terms related to nucleosomes, myosin complex, and several additional terms related to mitochondria including OxPhos complexes, mitochondrial nucleoid, TCA cycle complexes, and mitochondrial permeability transition pore complex (Figure 3B).

Next, we quantified the fraction of each protein pool that persisted for 6 or 10 months by calculating FA values, where the higher the value the longer-lived the corresponding proteins are (Figure 3—figure supplement 1B). Overall, there was no significant difference in FA between the two chase timepoints, with 41±2.9 and 50.8±12.2 ¹⁵N-remaining at 6 and 10 months, respectively (Figure 3—figure supplement 1C). The higher average FA average at 10-month chase is likely due to turnover of proteins with lower FA values at 6 months, which by 10 months would leave the oocyte with the most persistent pool of proteins. In agreement, the FA values for four LLPs that were identified at both timepoints (Hba, Atp5a, Atp5B, and Hist1h4a) sharply decline between 6 and 10 months (Figure 3—figure supplement 1D), demonstrating protein degradation and replenishment. Considering the high abundance of mitochondrial proteins in our dataset, and recent reports showing that expression of mitochondrial proteins is suppressed in aging oocytes (Rodríguez-Nuevo et al., 2022), we compared the number of mitochondrial proteins and spectral counts identified (Figure 3—figure supplement 1E and F), as well as FA (Figure 3—figure supplement 1G) for mitochondrial proteins across the two timepoints. Our results indicate that while there are no significant differences in total mitochondrial protein abundance at 6- and 10-month post-chase, the pool of long-lived mitochondrial proteins decreased significantly, with the majority of proteins being turned over at the later timepoint.

Hierarchical clustering of the LLPs identified at the 6-month timepoint revealed mitochondrial proteins and myosins as the two protein groups with the highest FA in the oocyte (Figure 3C). In particular, mitochondria exhibited a wide range of FA values ranging from 1.10% to 98.9%, with an average of 55.9±35.1 of ¹⁵N-remaining at 6 months (Supplementary file 2). FA values for myosins were markedly higher than both actin and tubulin with an average FA value of 80.97 ± 19.8% for myosins, 31.4 ± 15.6% for actins, and only 8.3 ± 1.4% for tubulins. This indicates that while all three cytoskeletal components are long-lived, nearly 81% of the myosin protein pool persists throughout the 6-month timepoint, whereas only 31% of actin protein pool and 8.3% of tubulin protein pool persists throughout the same length of time. Histones were also identified as LLPs in oocytes with average FA values of 7.7±9.8. Mitochondrial proteins, histones, myosins, and tubulins have been previously identified as LLPs in brain and heart tissues, which are known to contain long-lived terminally differentiated cells (i.e. neurons in the brain and cardiac myocytes in the heart) (Bomba-Warczak et al., 2021). Importantly, however, this is the first time that a subset of the same proteins has been identified as long-lived in the germ cell (Figure 3D). Interestingly, the FA of mitochondrial LLPs in the brain (10.2 ± 6.6%) (Bomba-Warczak et al., 2021) and myosins in the heart (4.6 ± 8.1%) (Bomba-Warczak et al., 2021) is much lower than in the same LLPs quantified in the oocyte (55.9±35.1) (Figure 3D). In contrast, histones were less long-lived in the oocyte compared to the brain, and there was no significant difference observed for tubulins (Figure 3D). Thus, although the identity of LLPs may be conserved across tissues with long-lived cells, differences in FA may reflect tissue-specific functions and requirements.

Mice of the FVB strain were obtained from Jackson Laboratory (Bar Harbor, ME, USA). For the whole animal isotope pulse-chase labeling strategy, female (5 weeks of age) and male (10 weeks of age) mice were obtained. To validate the biological relevance of our chase period, reproductive aging parameters were evaluated in unlabeled female mice at 6 weeks, 6 months, and 10 months of age. Retired female FVB mouse breeders were obtained at 15 weeks of age and aged out to 6 and 10 months. Mice were acclimated in the vivarium upon arrival for at least 2 weeks prior to experimental use. All mice were housed at Northwestern University’s Center for Comparative Medicine under constant temperature, humidity, and light (14 hr light/10 hr dark). Mice were fed and provided with water ad libitum. All animal care and experimental protocols in this study were conducted under the guidelines set by the NIH Guide for the Care and Use of Laboratory Animals handbook, and the animal protocol was approved by the Animal Care and Use Committee of Northwestern University.

Mice were metabolically labeled using a two-generation metabolic pulse-chase labeling strategy as previously described (Bomba-Warczak et al., 2021). Briefly, three female FVB mice were fed a spirulina ¹⁵N-containg chow for the duration of the study (initial labeling, breeding, pregnancy, and weaning) (Cambridge Isotope Laboratories, Inc, Tewksbury, MA, USA). After 13 weeks of being on a ¹⁵N diet, these labeled females were co-housed with unlabeled males for 5 days to allow a complete estrous cycle for breeding. Pregnant females were housed separately to allow for accurate dating of litter births. All females were allowed to breed a total of three to five times to obtain sufficient pups for the chase period. The pulse period was defined as the timespan between gestation and weaning of pups. Thus, pups conceived from labeled females had nitrogen-containing molecules and proteins labeled with ¹⁵N. At 22 days of age, pups were weaned and fed a ¹⁴N diet ad libitum. Pups were maintained on a ¹⁴N diet (chase) until they reached 7 and 11 months of age at which point tissues were harvested for downstream analyses. This animal labeling strategy generated 19 female pups. For ovary studies, four mice were used at 6 months’ post-chase and four mice at 10 months’ post-chase. One ovary per mouse was used for MIMS, and the contralateral ovary was used for LC-MS/MS. For oocyte studies, four mice were hyperstimulated at 6 months’ post-chase and seven mice at 10 months’ post-chase. A total of 104 and 45 oocytes were collected at 6 and 10 months’ post-chase, respectively, and these samples were used for LC-MS/MS.

Ovaries were harvested and placed in a dish containing pre-warmed Leibovitz’s medium (L15) (Life Technologies Corporation, Grand Island, NY, USA) supplemented with 3 mg/mL polyvinylpyrrolidone (PVP) (Sigma-Aldrich, St. Louis, MO, USA), 0.5% penicillin-streptomycin (Life Technologies) (L15-PVP). Ovaries were either processed as described below for downstream analyses or used to isolate oocytes. Ovaries used to assess reproductive aging parameters were fixed for histological analysis in Modified Davidson’s (Electron Microscopy Sciences, Hatfield, PA, USA) at room temperature (RT) for 2–4 hr with agitation and overnight at 4°C. Ovaries used for MIMS were cut in half and fixed in 2% glutaraldehyde (Electron Microscopy Sciences, Hatfield, PA, USA) and embedded in LR white (EMS). Ovaries used for LC-MS/MS were snap-frozen and stored at –80°C until use.

To maximize the yield of a synchronized population of fully grown oocytes for LC-MS/MS, mice were hyperstimulated with an intraperitoneal injection of 5 IU pregnant mare serum gonadotropin (PMSG) (Prospec Bio, East Brunswick, N). Ovaries were harvested 44–48 hr post-PMSG injection and placed in L15-PVP supplemented with 0.025% milrinone to maintain oocyte meiotic arrest (Sigma-Aldrich, St. Louis, MO, USA).

Cumulus-oocyte-complexes (COCs) were released by puncturing antral follicles with insulin syringes. Oocytes were mechanically denuded from cumulus cells using a 75 μm stripper tip and washed thoroughly in L15/PVP/PS before being snap-frozen. For analysis of ovulated eggs, mice were hyperstimulated as described above with PMSG followed by superovulation induction 44–46 hr later with an intraperitoneal injection of 5 IU human chorionic gonadotropin (hCG) (Sigma, St. Louis, MO, USA). COCs were then collected from the oviducts of each mice 14–16 hr post-hCG injection, and the number of ovulated eggs was counted.

The decline in the number of ovarian follicles is a hallmark of reproductive aging (Broekmans et al., 2009). Therefore, we evaluated follicle counts in histological sections of ovaries as done previously (Duncan et al., 2017). In brief, fixed ovaries were washed three times with 70% ethanol and processed, dehydrated, and paraffin-embedded using an automated tissue processor (Leica Biosystems, Buffalo Grove, IL, USA). After embedding, ovaries were serial sectioned with every fifth tissue section stained with hematoxylin and eosin (H&E). All H&E-stained tissue sections were digitally scanned at the University of Washington’s Histology and Imaging Core using Hamamatsu-HT imaging system (Hamamatsu Photonics, Hamamatsu City, Japan) at ×20 magnification. Scanned images were uploaded and visualized using the NDP.view2 software (Hamamatsu Photonics, Hamamatsu City, Japan). Follicles were classified and counted in every fifth tissue section. Follicles were classified by stage (primordial, primary, secondary, and antral) according to established criteria (Duncan et al., 2017). Primordial follicles were comprised of an incomplete layer of squamous granulosa cells, while primary follicles contained a complete layer of cuboidal granulosa cells. Secondary follicles contained two or more layers of cuboidal granulosa cells. Antral follicles had more than eight layers of granulosa cells with the presence of an antrum, or fluid-filled cavity. All primordial and primary follicles were counted regardless of whether the oocyte’s nucleus was visible. Secondary and antral follicles were counted only if the nucleus was visible to avoid double counting. Only healthy follicles were included in the final counts. Atretic follicles containing abnormally shaped oocytes with dark, pyknotic granulosa cells were excluded. The average number of follicles per area of ovarian section was calculated and used to compare counts between each age cohort.

Ovarian fibrosis, characterized by excess collagen, is another hallmark of reproductive aging (Amargant et al., 2020). We evaluated fibrosis in ovaries by Picrosirius Red (PSR), a histological stain which detects collagen I and III (Amargant et al., 2020). Ovarian tissue sections were deparaffinized in Citrisolv (Fisher Scientific, Pittsburgh, PA, USA) and rehydrated in 100%, 70%, and 30% ethanol baths. Slides were submerged in PSR staining solution composed of Sirius Red F3BA (Direct Red 80, C.I. 357.82, Sigma-Aldrich, St. Louis, MO, USA) and picric acid (Sigma-Aldrich, St. Louis, MO, USA) at 0.1% wt/vol for 40 min at RT. The slides were then incubated in acidified water made of 0.05 M hydrochloric acid (Fisher Scientific) for 90 s. Tissue sections were dehydrated in 100% ethanol baths, three times for 30 s incubations. After dehydration, slides were immersed in Citrisolv for 5 min and mounted with Cytoseal XYL (Fisher Scientific). PSR-stained sections were then imaged with an EVOS FL Auto Imaging system (Thermo Fisher, Waltham, MA, USA) using a ×20 objective. Scans of whole ovarian tissue sections were performed to quantify the area of positive PSR staining using a threshold feature on ImageJ as previously described (Briley et al., 2016). PSR-positive staining was analyzed on two different ovarian tissue sections for three mice within each age cohort: 6 weeks, 6 months, and 10 months. Results were averaged to obtain average percent area of collagen.

Fixed LR white-embedded ovaries were sectioned to 0.5 µm and mounted on silicon wafers. At the Brigham and Women’s Hospital Center for NanoImaging, a NanoSims 50L (CAMECA Instruments Inc, Madison, WI, USA) instrument was tuned to simultaneously measure ¹²C¹⁴N^-, ¹²C¹⁵N^-, and ³¹P^- secondary ions as described previously for imaging of a wide range of mouse and human tissues (Steinhauser et al., 2012). Quantitative ¹²C¹⁴N images were used for histological representation of stereotypical ovarian structures and associated cell types. ³¹P images provided additional histological detail and were used for identification of nuclei due to the high phosphorus content of chromatin (Guillermier et al., 2017a). Quantitative mass images for ¹²C¹⁴N and ¹²C¹⁵N were used to generate quantitative ¹⁵N:¹⁴N ratio images. For imaging of swathes of the tissue section, images were acquired in chain analysis mode of sequential adjacent fields (dimensions of 45 μm × 45 μm or 50 μm × 50 μm). Sequential tiles were stitched together to make mosaic images for visualizing large sections of the ovary (Figure 1B). Some features were then imaged at higher resolution with smaller field sizes. All images were processed and further analyzed using the most recent version of the OpenMIMS 2.0 plugin (https://github.com/BWHCNI/OpenMIMS; BWHCNI, 2018) (RRID:SCR_001416, version 2.0.) to ImageJ (Guillermier et al., 2017b). ¹⁵N-labeling was visualized by a hue saturation intensity (HSI) transformation of the ¹²C¹⁵N/¹²C¹⁴N ratio. The color scale of HSI images is set such that the lower blue bound of the scale is at the ¹⁵N natural abundance of 0.37% (expressed as 0% above background). The upper magenta bound of the scale is set to reveal labeling differences. The quantitative isotope ratio measurements that form the basis for the images and that are used for statistical analyses are not affected by scaling changes that modify the visual appearance of the images (Figure 1—figure supplement 2A and B). A combination of ¹⁴N and ³¹P images were used to manually select regions of interest within each ovarian section (e.g. individual cells or subcellular structures). Structures and cell types were identified based on morphology and their anatomic location. Cells that were not identifiable or were not visualized due to low ion counts were excluded from the analysis. Typical reasons for difficult identifying cells include sectioning artifacts (cracks, wrinkles) or certain cells that were located at the juncture between adjacent imaging fields, where there is often lower yield of secondary ions. This edge effect is seen to variable degrees in the mosaic images as dark regions at the periphery of an imaging field. All quantitative data for ¹⁵N-labeling was presented as the ¹⁵N/¹⁴N ratio (percentage above natural abundance).

Isolated ovaries were homogenized directly in 6 M guanidine hydrochloride solution using bead-based Precellys 24 homogenizer, followed by processing with ProteaseMAX according to the manufacturer’s protocol. Samples were reduced with 5 mM Tris(2-carboxyethyl)phosphine (TCEP; vortex 1 hr at RT) alkylated in the dark with 10 mM iodoacetamide (IAA; 20 min), diluted with 50 mM ABC and quenched with 25 mM TCEP. Samples were digested with sequencing grade modified trypsin overnight at 37°C with shaking, spun down (15,000×g for 15 min at RT), placed in a new tube, and acidified with TFA to a final concentration of 0.1%. A total of 100 μg of digested and acidified sample was fractionated using High pH Reversed-Phase Peptide Fractionation Kit (Pierce, Cat# 84868). Fractions were step eluted in 300 μL buffer of increasing acetonitrile (ACN) concentrations with decreasing concentration of triethylamine (0.1%) as per the manufacturer’s instructions. Samples were dried down with vacuum centrifugation for future MS analysis.

Isolated oocytes were lysed directly in RIPA buffer, mixed with 6× SDS sample buffer, boiled for 5 min, and separated by SDS-PAGE using 10% Tris-glycine gels (Thermo Scientific, Cat# XV00100PK20). Gels were stained using Oriole fluorescent gel stain solution, scanned using Bio-Rad Chemidoc XRS system, and cut into sections, chopped into 1 mm × 1 mm cubes, and processed for in-gel digestion. The gel separating oocytes collected at t=6 months was cut into 36 individual pieces, whereas the gel separating oocytes collected at t=10 months was cut into 24 pieces. Gel pieces were incubated in 10 mM TCEP (in 50 mM ABC; 1 hr at 37°C). Liquid was replaced by 50 mM IAA (in 50 mM ABC; 45 min at RT in dark), followed by 50 mM TCEP (in 50 mM ABC; 30 min at RT). Gel pieces were washed with 50 mM ABC (3×) and digested with sequencing grade modified trypsin (1 μg in 50 mM ABC, overnight at 37°C, with shaking). The following day, supernatant was collected into new tube and the gel piece were subjected to three rounds of incubations with 50% ACN and 5% FA solution (30 min at RT, with shaking). Supernatant was collected after each incubation, combined, and dried down with vacuum centrifugation. Samples were re-suspended in 0.5% TFA, desalted with Pierce C18 spin columns (Thermo Scientific, Cat# 89873) per the manufacturer’s instructions, and dried down with vacuum centrifugation for future MS analysis.

Dried samples were re-suspended in 20 μL Buffer A (94.875% H₂O with 5% ACN and 0.125% FA) and 3 µg, as determined by microBCA assay (Thermo Scientific, Cat# 23235) of each fraction or sample were loaded via auto-sampler with either Thermo EASY nLC 100 UPLC or UltiMate 3000 HPLC pump, onto a vented Pepmap 100, 75 µm × 2 cm, nanoViper trap column coupled to a nanoViper analytical column (Thermo Scientific) with stainless steel emitter tip assembled on the Nanospray Flex Ion Source with a spray voltage of 2000 V. A coupled Orbitrap Fusion was used to generate MS data. Buffer A contained 94.785% H₂O with 5% ACN and 0.125% FA, and Buffer B contained 99.875 ACN with 0.125% FA. MS parameters were as follows: Ion transfer tube temp = 300°C, Easy-IC internal mass calibration, default charge state = 2 and cycle time = 3 s. Detector type set to Orbitrap, with 60K resolution, with wide quad isolation, mass range = normal, scan range = 300–1500 m/z, max injection time = 50 ms, AGC target = 200,000, microscans = 1, S-lens RF level = 60, without source fragmentation, and datatype = positive and centroid. MIPS was set as on, included charge states = 2–6 (reject unassigned). Dynamic exclusion enabled with n=1 for 30 s and 45 s exclusion duration at 10 ppm for high and low. Precursor selection decision = most intense, top 20, isolation window = 1.6, scan range = auto normal, first mass = 110, collision energy 30%, CID, detector type = ion trap, OT resolution = 30K, IT scan rate = rapid, max injection time = 75 ms, AGC target = 10,000, Q=0.25, inject ions for all available parallelizable time. For ovary samples the chromatographic run was 4.5 hr per fraction, with the following profile of Buffer B: 2% for 7 min, 2–7% for 1 min, 7–10% for 5 min, 10–25% for 160 min, 25–33% for 40 min, 33–50% for 7 min, 50–95% for 5 min, 95% for 15 min, then back to 2% for the remaining 30 min. For the GeLC/MS oocyte samples, the chromatographic run was 75 min per gel section with the following profile of Buffer B: 2–8% for 6 min, 8–24% for 10 min, 24–36% for 20 min, 36–55% for 10 min, 55–95% for 10 min, 95% for 10 min, then back to 2% for remaining 9 min.

Protein identification/quantification and analysis were performed with Integrated Proteomics Pipeline - IP2 (Integrated Proteomics Applications, Inc, San Diego, CA, USA) using ProLuCID (Eng et al., 1994; Xu et al., 2015), DTASelect2 (Cociorva et al., 2006; Tabb et al., 2002), Census, and QuantCompare. Spectral raw files were extracted into MS1, MS2 files using RawConverter 1.0.0.0 (http://fields.scripps.edu/downloads.php). The tandem mass spectra were searched against mouse database (downloaded on March 25, 2014). Searched spectra were matched to sequences using the ProLuCID/SEQUEST algorithm (ProLuCID version 3.1) with 50 ppm peptide mass tolerance for precursor ions and 600 ppm for fragment ions. ProLuCID searches included all fully and half-tryptic peptide candidates that fell within the mass tolerance window and had with unlimited mis-cleavages. Carbamidomethylation (+57.02146 Da) of cysteine was considered as a static modification. Peptide/spectrum matches were assessed in DTASelect2 using the cross-correlation score (XCorr), and normalized difference in cross-correlation scores (DeltaCN). Each protein identified was required to have a minimum of one peptide (-p1) of minimal length of six amino acid residues. False discovery rate (FDR) was set to 1% at the protein level, for all experiments. Peptide probabilities and FDR were calculated based on a target/decoy database containing the reversed sequences of all the proteins appended to the target database (Peng et al., 2003). Each dataset was searched twice, once against light (¹⁴N) and then against heavy (¹⁵N) protein databases, as described previously (Savas et al., 2012). In the light searches, all of the amino acid residues were considered to contain only ¹⁴N nitrogen, while in the heavy searches, all the amino acid residues were considered to contain only ¹⁵N nitrogen. After the results from ProLuCID were filtered using DTASelect2, the assembled search result file was used to obtain quantitative ratios between ¹⁴N and ¹⁵N using the software Census (Park et al., 2008; MacCoss et al., 2003).

LLPs were identified as previously described, with modifications (Savas et al., 2012). Briefly, in order for a protein to be considered as long-lived, the protein had to be identified in our heavy/light search by at least one long-lived peptide (¹⁵N-peptide). Peptide ratio measurements were filtered in Census based on a correlation threshold, and only peptides with correlation coefficient above 0.5 were used for further analysis. For singleton analysis, we required the ¹⁴N/¹⁵N ratio to be greater than 5.0 and the threshold score to be greater than 0.5. Identified peptides were further filtered based on their average peptide enrichment, which we set to 0.9, and peptide profile score, which was set to 0.8. Proteins were only identified as long-lived if they had more than three long-lived peptides that passed the above filtering (except for GeLC/MS experiments where one peptide was required). FA were calculated according to the following formula: ¹⁴N values: FA = 100 – (100∗(1/ (1+AR))), where FA = fractional abundance and AR = area ratio.

GO analysis was performed using the Pantherdb (Mi et al., 2019). The ‘query’ is defined as proteins identified as long-lived in the analyzed tissue (based on ¹⁴N-peptide identification), and the reference is defined as all proteins identified in the same tissue analyzed (¹⁴N- and ¹⁵N-peptide identification).

Statistical analyses were conducted using GraphPad Prism, version 9 (GraphPad Software, Inc). A Student’s t-test was performed for comparisons between two groups. For comparisons of more than two groups, a one-way ANOVA was used. A one-sample t-test was used to compare ¹⁵N/¹⁴N ratios of cytoplasmic and nuclear regions to a hypothetical value of 1. Values>1 signified a cell had greater ¹⁵N-labeling abundance in the nucleus compared to the cytoplasm. Values<1 represented cells with a lower ¹⁵N-labeling abundance in the nucleus compared to the cytoplasm. Values equal to 1 represented a cell with a ¹⁵N-labeling abundance that were equivalent in both nuclear and cytoplasmic regions. Data were considered significant with a p-value<0.05 (*p-value<0.05, **p-value<0.01, ***p-value<0.001). Variability of groups was denoted as standard error of mean (SEM).