Fused in sarcoma (FUS) inhibits milk production efficiency in mammals

Shao, Haili; Huang, Jipeng; Wang, Hui; Wang, Guolei; Yang, Xu; Cheng, Mei; Sun, Changjie; Zou, Li; Yang, Qin; Zhang, Dandan; Liu, Zhen; Jiang, Xuelong; Shi, Lei; Shi, Peng; Han, Baowei; Jiao, Baowei

doi:10.1038/s41467-024-48428-5

Download PDF

Article
Open access
Published: 10 May 2024

Fused in sarcoma (FUS) inhibits milk production efficiency in mammals

Nature Communications volume 15, Article number: 3953 (2024) Cite this article

578 Accesses
Metrics details

Subjects

Abstract

Efficient milk production in mammals confers evolutionary advantages by facilitating the transmission of energy from mother to offspring. However, the regulatory mechanism responsible for the gradual establishment of milk production efficiency in mammals, from marsupials to eutherians, remains elusive. Here, we find that mammary gland of the marsupial sugar glider contained milk components during adolescence, and that mammary gland development is less dynamically cyclic compared to that in placental mammals. Furthermore, fused in sarcoma (FUS) is found to be partially responsible for this establishment of low efficiency. In mouse model, FUS inhibit mammary epithelial cell differentiation through the cyclin-dependent kinase inhibitor p57Kip2, leading to lactation failure and pup starvation. Clinically, FUS levels are negatively correlated with milk production in lactating women. Overall, our results shed light on FUS as a negative regulator of milk production, providing a potential mechanism for the establishment of milk production from marsupial to eutherian mammals.

Transient loss of Polycomb components induces an epigenetic cancer fate

Article Open access 24 April 2024

Evolution of a novel adrenal cell type that promotes parental care

Article 15 May 2024

Multiplex profiling of developmental cis-regulatory elements with quantitative single-cell expression reporters

Article Open access 09 May 2024

Introduction

The mammary gland, a defining feature of mammals, plays a crucial role in facilitating the efficient delivery of nutrition from mother to offspring through lactation¹. Throughout mammalian evolution, there has been a gradual increase in delivery efficiency from marsupials to eutherians^2,3. Although both marsupials (e.g., tammar wallabies and sugar gliders) and eutherians (e.g., mice) produce milk, they exhibit distinct differences in lactation strategies. Marsupials, characterized by a short gestation period with limited energy resources available for the developing fetus, rely on a prolonged suckling period in the mother’s pouch for nourishment and growth, lasting ~300 days in the tammar wallaby (Macropus eugenii)⁴ and 70–90 days in the sugar glider (Petaurus breviceps). In contrast, rats of comparable weight to sugar gliders exhibit a considerably shorter lactation period of only 21 days⁵. The efficient energy transfer gained during mammary gland evolution in rats allows their young to complete development within a relatively short period, and then survive independently.

In humans, breastfeeding reduces the risk of many diseases in mothers, including breast cancer, ovarian cancer, postpartum depression, and cardiometabolic disorders^6,7. In addition, breastfeeding protects infants from gastrointestinal infections, adaptive dermatitis, respiratory diseases, and obesity⁸. However, less than 25% of infants are exclusively breast-fed⁹, with insufficient maternal milk secretion being an important reason¹⁰. Therefore, elucidating the precise regulatory mechanism underlying lactation during evolution is critical for understanding the etiology of postpartum hypogalactia.

Mammary epithelial cells (MECs) increase exponentially during pregnancy and differentiate into alveolar cells in late pregnancy, when MEC proliferation rapidly declines as the glands devote themselves to differentiation^11,12,13. Through ductal branching and extension, the development and remodeling of the mammary duct, crucial for lactation, rely on coordinated MEC proliferation and the regulation and maintenance of cell differentiation¹⁴. However, the specific factors driving MEC differentiation and mammary gland development remain unclear. Previous studies have suggested that post-transcriptional regulation may play a critical role in adaptive divergence and evolution^15,16,17. RNA-binding proteins (RBPs) are key players in the post-transcriptional regulation of gene expression^18,19,20.

In this work, we discover distinct differences regarding mammary gland development in marsupial sugar gliders and eutherian mice. Notably, unlike the dynamic cycle observed in mice, sugar gliders exhibit consistent morphological characteristics throughout the different stages of development. Furthermore, we identify Fused in sarcoma (FUS) as a key regulatory protein for the above lactation efficiency. Overall, our study reveals the pivotal role of FUS as a regulator of lactation, providing valuable insights into the underlying mechanisms governing milk production processes.

Results

Sugar gliders exhibit lower milk production control ability

Our current understanding of mammary gland development in marsupials is limited. To investigate the differences in mammary glands between marsupials and eutherian mammals, we explored mammary gland development in sugar gliders compared to that in mice. We first conducted a detailed morphological analysis of the mammary glands of female sugar gliders at four stages (virgin, pregnancy, lactation, and involution) (Fig. 1a). In contrast to mice, which demonstrated a dynamic presence of milk-producing alveolar units, the sugar gliders featured evident mammary gland alveolar structures not only during the lactation period but also during the virgin and pregnancy stages (Fig.1b). Furthermore, the number of mammary gland ducts in the sugar gliders, both during the virgin and lactation stages, was significantly lower compared to that in mice (Fig. 1c, d), as demonstrated by immunofluorescence staining of the ductal marker CD133^21,22,23 (Supplementary Fig. 1a, b). Interestingly, milk-like secretions were observed in the mammary gland of sugar gliders during the virgin stage, whereas mice exhibited milk secretion exclusively during the lactation period (Fig. 1e). To ascertain whether the secretions observed in virgin mammary glands of sugar gliders were indeed milk, we conducted a comprehensive composition analysis. Milk is known to contain essential nutritional components such as proteins, fats, and lactose²⁴. Immunofluorescence staining revealed that mice exclusively produced milk protein in their mammary glands during lactation, while, in sugar gliders, milk protein was also detected in the virgin stage of mammary gland development (Fig. 1f). Oil Red staining further demonstrated the presence of mammary gland milk fat in both the virgin and lactation stages in sugar gliders, while this was only present during the lactation period in mice (Fig. 1g). Furthermore, we used Raman spectroscopy^25,26 to detect the Raman spectra of milk from both mice and sugar gliders, as well as the mammary gland contents during the virgin and lactation. Only the constituents from the lactating mouse mammary glands corresponded precisely with the milk spectra, there were significantly different peak shape and intensity variations in the Raman spectra of mammary contents from virgin compared to mice milk spectra. Conversely, for sugar gliders, the Raman spectra of mammary gland contents during lactation matched perfectly with sugar glider milk spectra, and similarly, the spectra from virgin mammary gland contents also aligned completely with milk spectra (Fig. 1h). These findings confirmed that sugar gliders were capable of secreting milk components during non-lactating periods. Unlike in mice, the lack of precise regulation of milk secretion in sugar gliders, which occurs even when offspring do not require milk, suggests an inefficient utilization of energy by the parent. Additionally, sugar gliders contained significantly fewer mammary ducts than mice. Overall, these results indicate that sugar gliders exhibit a lower regulatory capacity for milk production compared to mice.

**Fig. 1: Sugar glider mammary gland produces milk during adolescence.**

FUS is a potential regulator of lactation efficiency in mammals

The regulation of lactation in eutherian mammals follows a dynamic cyclic pattern, with milk components being synthesized during gestation and secreted during lactation²⁷. In contrast, our results showed that lactation regulation in sugar gliders remained unchanged, with milk components present at various stages of mammary gland development. Based on these disparities in lactation strategies between eutherians and marsupials, we hypothesized that certain key factors involved in the regulation of lactation may remain unchanged throughout marsupial mammary gland development. To explore these key factors and their evolutionary significance, we aimed to identify genes that were differentially expressed in eutherian mammals but did not show significant expression changes during marsupial mammary gland development. We first collected mammary gland samples from sugar gliders and mice during the virgin and lactation stages for RNA-seq analysis. By comparing the gene expression profiles, we identified genes that displayed no significant changes (P > 0.05; |fold-change|<2) between the two stages in sugar glider mammary glands, while exhibiting significant changes (P < 0.05; |fold-change|> 2) between the two stages in mouse mammary glands, yielding a total of 2214 genes (Supplementary Data 1). To further refine this gene list, we incorporated differentially expressed genes (DEGs) between lactating and non-lactating mammary glands in well-recognized dynamic species, including cattle (GSE116079), pigs (GSE30704), and mice (GSE12247)²⁸. Among the DEGs across the three species, a total of 26 genes exhibited significant fluctuations (P < 0.05) in expression levels between lactation and virginity (Supplementary Table 1). In combination with the above two gene lists, five genes demonstrated significant changes in mice but not in sugar gliders during both stages (Fig. 2a). Notably, among these genes, Fus displayed the smallest fold-change (−1.02) and the least significant level of change (P = 0.87) in the sugar glider mammary gland (Supplementary Data 2).

**Fig. 2: FUS is a potential inhibitor of lactation in mammals.**

FUS expression in the sugar glider mammary gland remained largely stable over time. This was consistent at the protein level, as shown by immunohistochemical staining (Fig. 2b) and western blotting (Supplementary Fig. 1c), and at the mRNA level, as indicated by quantitative real-time polymerase chain reaction (RT-qPCR) results (Fig. 2c). In contrast, eutherian mice exhibited a significant reduction in FUS protein levels during late pregnancy and lactation compared to levels observed during the virgin and involution stages (Fig. 2b). These findings were confirmed at the protein level by western blot analysis (Supplementary Fig. 1d, e) and immunofluorescence staining (Supplementary Fig. 1f) and at the mRNA level by RT-qPCR (Supplementary Fig. 1g). Furthermore, low FUS expression during lactation was also evident in eutherians beyond mice, including horseshoe bats (Rhinolophus sinicus) (Fig. 2b), suggesting a common pattern across eutherian species. Taken together, these findings suggest that the identified FUS protein potentially serves as a regulator of lactation efficiency in mammals.

FUS expression levels are crucial for lactation efficiency in mammals

We next explored the reasons behind the distinct expression patterns of FUS between marsupials and eutherians. As genetic diversity is a key parameter for understanding the evolutionary process²⁹, we compared the amino acid sequences of FUS between eutherians and marsupials. Our findings revealed that a glycine and serine enrichment sequence (GS-sequence, PGS: GGGGGGGGSG for sugar glider, Petaurus breviceps, MGS: GGGGSGSGGG for tammar wallaby, Macropus eugenii, and SGS: GSSSGG for tasmanian devil, Sarcophilus harrisii) was absent in the glycine-rich domain (GR) of the FUS gene coding region from species with a strong lactation capability (Fig. 2d and Supplementary Fig. 2). These results imply that the loss of the GS-sequence in the GR decreases the FUS protein level during the lactation stage in those species with strong lactation ability. Conversely, in the sugar glider, this sequence was retained, and the protein level of FUS remained stable over time in the mammary gland stages. To determine the impact of the GS-sequence on FUS protein levels, the GS-sequence from marsupials was inserted into the GR region of human (hFUS) and mouse (mFUS) FUS sequences, named as mFUS-PGS, mFUS-MGS, mFUS-SGS, hFUS-PGS, hFUS-MGS, and hFUS-SGS. These constructs were then expressed in various cell lines (Fig. 2e). Of note, the presence of inserts significantly increased both the mRNA and protein levels of FUS (Fig. 2f, g and Supplementary Fig. 1h, i). The longer inserts (PGS and MGS) induced much more stable FUS levels compared to the shorter insert (SGS), further indicating that GS-enriched sequences are required for maintaining FUS expression. Furthermore, we investigated whether the GS-sequence had an effect on the mRNA stability of Fus through mRNA stability experiments. The results showed that the insertion of GS-sequence promoted the mRNA stability of Fus (Fig. 2h and Supplementary Fig. 1j) and FUS protein levels (Supplementary Fig. 1k). Thus, these results suggest that FUS expression is regulated by GS-sequence across different species.

High FUS expression in MECs results in lactation failure

To validate the role of FUS in lactation efficiency, we recovered FUS expression during the late pregnancy and lactation stages of the mammary gland using a transgenic mouse model overexpressing FUS. FUS expression was enhanced in Fus floxed mice (Fus^fl/fl) crossed with transgenic mice expressing Cre recombinase (Cre) under the control of the whey acidic protein (WAP) promoter, specifically targeting luminal epithelial cells from the mid-pregnancy to lactation stages³⁰ (Supplementary Fig. 3a). The efficiency of the Fus overexpression system in mice was validated through various assays (Fig. 3a, b and Supplementary Fig. 3b, c). In Fus^fl/wt/WAP-Cre (Fus-OE) mice, FUS predominantly localized in the nucleus, as revealed by immunofluorescence staining (Supplementary Fig. 3d). We conducted further investigations to verify the specificity of Cre recombinase expression driven by the WAP promoter. This involved isolating and analyzing different brain regions and the spinal cord for the presence of FUS and Cre recombinase using western blotting. The analysis confirmed that Cre recombinase was not present in these samples, indicating no unintended Cre activity in the cerebral cortex, limbic system, or spinal cord (Supplementary Fig. 3e, f). Additionally, in the Fus-OE group, the FUS levels were comparable to those in the WAP-Cre group within the brain and spinal cord, suggesting no overexpression of FUS in these areas (Supplementary Fig. 3e, f). GFP expression, exclusive to Fus-OE mice, was also detected (Supplementary Fig. 3c), with no GFP presence in the samples from the WAP-Cre and Fus^fl/fl groups. These findings confirm the successful establishment of a Fus gene overexpression mouse model.

**Fig. 3: FUS overexpression induces failure of both pup viability and growth.**

To assess the effects of FUS on offspring survival, we initially analyzed the survival rates of pups from mixed litter sizes. Notably, we observed significantly lower survival rates among first-litter pups born to Fus-OE mice compared to pups born to WAP-Cre mice. Most pups in the Fus-OE group did not survive beyond lactation day 2 (L2) (Supplementary Fig. 3g), with the decreased survival rates stabilizing after the fourth day (Fig. 3c). Next, to eliminate any variations in nutrition caused by different litter sizes, we calculated the pup survival rates for the same-sized litters (6, 7, or 8 pups per litter), which showed a similar pattern (Fig. 3d and Supplementary Fig. 3g–i). Furthermore, despite improvements in pup survival in the overexpression group during the second lactation, a similar trend of declining pup survival was observed (Fig. 3e and Supplementary Fig. 3j).

To gain a better understanding of the role of FUS, we closely monitored the weight of surviving pups by adjusting the litter size to six at the end of L2. When comparing the average weight and weight gain of surviving pups born to Fus-OE mothers during their first (Fig. 3f, black curves) and second lactations (Fig. 3g and Supplementary Fig. 3k), the significant decreases in weight and weight gain were observed in comparison to pups born to WAP-Cre mothers. To exclude possible defects in pups, we conducted cross-fostering experiments, whereby pups born to Fus-OE and WAP-Cre mothers were switched and nursed by WAP-Cre and Fus-OE mothers, respectively, starting at either L2 or L10. The results demonstrated that pups born to Fus-OE mothers but fostered by WAP-Cre mice from L2 onwards exhibited normal body weights and survival rates (Fig. 3f, red curves). In addition, the lower average weight of surviving pups born to Fus-OE mothers could be rescued when fostered by WAP-Cre mothers starting at L10 (Fig. 3f, blue curves). Thus, these findings suggest that lower FUS expression is required for offspring survival in mice.

High FUS expression induces insufficient milk production

To identify factors contributing to the mortality of pups nursed by Fus-OE mice, we next focused on milk, the primary source of energy for offspring survival. The stomachs of Fus-OE contained minimal milk (Fig. 4a). To exclude the possibility of muscle weakness affecting suckling ability, we conducted cross-fostering experiments. For the pups given birth by Fus-OE mothers, 1/3 of the pups were fed by the foster mothers of WAP-Cre genotype or Fus^fl/fl genotype on day 1, respectively. The breastfeeding status was observed after 8 h of cross-fostering. Results demonstrated that the stomachs of pups born to Fus-OE mothers but fostered by WAP-Cre mice or Fus^fl/fl mice contained sufficient milk, implying that pups born to Fus-OE mothers had normal suckling abilities (Supplementary Fig. 4a). Conversely, control, pups born to WAP-Cre mothers but fostered by Fus-OE mice contained minimal milk in their stomachs (Supplementary Fig. 4b). These results suggest that the pups of both Fus-OE and WAP-Cre genotypes could suckle milk normally, thereby excluding impaired suckling as a contributing factor to their mortality. We collected milk from lactating mice using established protocols for volume analysis, as described in our previous study³¹. Notably, compared with WAP-Cre mice, Fus overexpression mice showed a significant decrease in milk volume at both L2 and L10 (Fig. 4b and Supplementary Fig. 4c). Immunofluorescent and immunohistochemical analyses further confirmed that Fus overexpression inhibited milk production (Fig. 4c and Supplementary Fig. 4d, e). Furthermore, the expression levels of the α-casein, β-casein, and Wap genes, which encode milk protein, as well as the Prlr gene, a key regulator of lactation^32,33,34, were significantly downregulated in the MECs of the Fus-OE group (Fig. 4d).

High FUS expression results in sparse alveolar structures

We next examined whether alveolar dysplasia occurred in Fus-OE mice using whole-mount and hematoxylin-eosin (H&E) staining. Whole-mount analysis revealed notable atrophy and sparsity of secretory lobuloalveolar structures in Fus-OE mammary glands, even immediately after delivery, with the sparse alveolar pattern persisting until L20 (Fig. 4e). H&E staining further confirmed the sparse distribution of alveolar structures in the Fus-OE group (Fig. 4f). In addition to the WAP-Cre system, we verified these phenotypes in another Fus overexpression system driven by K14 under tamoxifen induction (Supplementary Fig. 3a). Whole-mount and H&E staining again demonstrated the presence of sparse alveolar structures in the Fus-OE group (Supplementary Fig. 4f, g). Upon closer examination at higher magnification, all observed alveolar structures appeared intact without signs of collapse or premature degeneration (Fig. 4f and Supplementary Fig. 4g), suggesting that the sparse alveolar phenotype in the Fus-OE group was not attributed to involution.

Collectively, these data demonstrate that increased FUS expression in the mammary gland can lead to a sparse distribution of alveolar structures.

FUS inhibits MEC differentiation

Upon pregnancy, MECs undergo differentiation into alveolar cells, which subsequently form lactation lobules responsible for milk secretion¹². Thus, we next tested whether the sparse alveolar structures were the result of reduced MEC differentiation. Stat5, Elf5, and Nfib, markers of MEC differentiation, play crucial roles in regulating the expression of milk proteins and other differentiation-related proteins involved in secretion during alveolar differentiation^35,36. Our results showed that the expression levels of the three markers were significantly decreased in the Fus-OE mammary gland at L0 (Fig. 5a). Immunostaining results confirmed the absence of these marker proteins in the Fus-OE group (Fig. 5b and Supplementary Fig. 5a), indicating that FUS inhibits MEC differentiation.

**Fig. 5: FUS overexpression inhibits MEC differentiation.**

To further investigate the impact of FUS on MEC differentiation, we utilized the well-established MEC differentiation model in vitro. HC11 is a normal immortalized PRL-positive MECs derived from the mammary tissue of pregnant BALB/C mice³⁷. HC11 cells can be induced to differentiate into alveolar-like cells in the presence of prolactin and other factors, closely mimicking the process of milk production in vivo (Fig. 5c). Alveolar-like cells can form milk-like dome-shaped structures, which can serve as indicators of MEC differentiation³⁸. Therefore, we performed Fus knockdown in HC11 cells using short hairpin RNA (shRNA) (Supplementary Fig. 5b), then evaluated dome formation. Results showed that Fus knockdown significantly promoted dome formation (Fig. 5d, e), while overexpressing Fus in HC11 cells resulted in reduced dome formation (Supplementary Fig. 5c, d). Dome formation assays were also carried out in primary mammary cells, revealing that Fus overexpression led to a reduction in dome formation (Fig. 5f, g), whereas Fus knockdown led to a significant increase in dome formation (Supplementary Fig. 5e, f). We analyzed the expression of milk-related genes, including Wap, α-Casein, β-Casein, and Prlr. These genes showed significant up-regulation after treatment with prolactin in the Fus knockdown group, as indicated by RT-qPCR analysis (Supplementary Fig. 5g). These findings are consistent with previous research showing that HC11 cells can synthesize milk proteins when stimulated by lactogenic hormones, differentiate into alveolar-like cells, and form dome-shaped structures^38,39. Furthermore, the expression of Fus was significantly reduced upon HC11 differentiation (Fig. 5h, i). These findings provide additional evidence supporting the inhibitory role of Fus in MEC differentiation.

FUS regulates coordination between cell cycle exit and MEC proliferation

To further explore the mechanism underlying the impact of Fus on milk secretion disorders and MEC differentiation, we extracted RNA from the breast tissue of Fus-OE and WAP-Cre mice at L1 and performed transcriptome sequencing. Based on the Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis of the transcriptome data, we found that DEGs between the Fus-OE and WAP-Cre groups were primarily associated with cell cycle and DNA replication (Fig. 6a). Cellular division is essential for generating appropriate numbers of cells, but subsequent exit from the cell cycle is necessary to permit differentiation^40,41. The duration of the G1 phase is pivotal in establishing a window of opportunity for DNA replication, pluripotency exit, and differentiation commencement⁴². To determine whether FUS regulates MEC differentiation by controlling the cell cycle, we detected the cell cycle in HC11 cells following Fus knockdown. Flow cytometry revealed that cell cycle progression was arrested in the G0/G1 phase upon Fus knockdown (Fig. 6b). Furthermore, the expression of negative cell-cycle regulators (P21) was up-regulated, and the expression of positive cell-cycle regulators (p-RB, Cyclin D1 and Cyclin E1) was downregulated after Fus knockout, resulting in cell cycle arrest in the G1 phase (Fig. 6c). These results suggest that proper cell-cycle exit and MEC differentiation rely on the appropriate expression level of FUS. In addition, the clone formation assay demonstrated that Fus knockdown in HC11 and primary cells impaired clone formation ability (Fig. 6d and Supplementary Fig. 5h, i).

As proliferation and differentiation exhibit an inverse relationship during development⁴², we further detected the effects of FUS on proliferation. Immunofluorescence analysis using the proliferation marker Ki67 revealed a significantly higher number of proliferating MECs in Fus-OE mice compared to WAP-Cre mice at the lactating stage (Supplementary Fig. 5j), as confirmed by immunohistochemical staining and analysis (Fig. 6e, f and Supplementary Fig. 5k, l). To substantiate the regulatory effects of FUS in vitro, we assessed cell proliferation using the 3-(4,5-dimethylthiazol-2-yl)−5-(3-carboxymethoxyphenyl)−2-(4-sulfophenyl)−2H-tetrazolium (MTS) assay in HC11 and primary mammary gland cells. Results showed that cell proliferation was significantly reduced after Fus knockdown in both cell types (Fig. 6g and Supplementary Fig. 5m). In addition, proliferating cell nuclear antigen (PCNA), a cell proliferation marker, was also downregulated during differentiation (Fig. 5i).

Taken together, these findings indicate that high FUS expression inhibits MEC differentiation by inhibiting cell cycle exit.

FUS binds to p57Kip2 mRNA and decreases its stability

To further elucidate the mechanism by which FUS regulates the cell cycle, we investigated its RNA-binding properties. As a known RNA/DNA-binding protein, FUS binds to and interacts with RNA in a variety of ways, thereby modulating downstream signaling⁴³. Using online tools (http://www.tartaglialab.com/), we predicted the potential binding between FUS and mRNAs related to the cyclin-dependent kinase (CDK) inhibitor (CKI) (Supplementary Data 3), consistent with our observations that Fus inhibited cell differentiation and promoted cell proliferation. We selected p57Kip2, which had a high ranking in terms of combination ability, for the follow-up study. Notably, FUS binds to the 3’UTR of p57Kip2, as confirmed by RNA pull-down assay (Fig. 6h and Supplementary Fig. 6a) and RNA immunoprecipitation (RIP) assay (Fig. 6i and Supplementary Fig. 6b). The p57Kip2 protein is a Cip/Kip family member of CKIs, encoded by the Cdkn1c gene, and plays an important role in the regulation of cell proliferation and differentiation during mammalian development⁴⁴.

RNA-binding proteins can regulate mRNA stability through interaction with the 3’UTR of mRNA⁴⁵. To further investigate whether FUS affects the stability of p57Kip2 by binding to its 3’UTR, we examined the decay of p57Kip2 mRNA following treatment in transfected HC11 cells with the RNA Pol II inhibitor actinomycin D to stop transcription. Results revealed a significant increase in p57Kip2 mRNA levels in the absence of Fus (Fig. 6j), suggesting that the loss of FUS promoted p57Kip2 mRNA stability. To further determine whether FUS regulates p57Kip2 mRNA stability through the 3’UTR, the 3’UTR of p57Kip2 mRNA was inserted into downstream of the green fluorescent protein (GFP) gene in a mammalian expressive vector (Fig. 6k). Changes in GFP expression were then assessed upon Fus knockdown, serving as an indicator of FUS-mediated regulation of p57Kip2 via its binding to the 3’UTR. As expected, Fus knockdown using shRNA resulted in significantly higher GFP expression in the GFP-p57Kip2-3’ UTR group compared with the scramble control (Fig. 6l). Conversely, Fus overexpression reduced GFP expression (Fig. 6m). Furthermore, to confirm the regulatory effects of FUS on p57Kip2 in vitro, knockdown of Fus in HC11 cells led to a substantial increase in p57Kip2 protein expression (Supplementary Fig. 6c). To further verify the effects of p57Kip2 on lactation, we detected the expression levels of p57Kip2 during in vitro differentiation of HC11 cells. Results showed that p57Kip2 expression was significantly increased at both the mRNA (Supplementary Fig. 6d) and protein levels (Fig. 5i), opposite to the FUS expression pattern observed during differentiation (Fig. 5h, i).

These findings suggest that FUS binds to the 3’UTR of p57Kip2 mRNA, reducing mRNA stability and inhibiting p57Kip2 expression, thereby regulating the cell cycle of MEC. In addition to p57Kip2, FUS also regulated the expression of other cell cycle proteins, including p21, p27, Cyclin B1, Cyclin D1, and Cyclin E1 (Supplementary Fig. 6c), confirming the regulatory role of FUS in the cell cycle. RIP-qPCR assay confirmed that FUS could bind to the mRNA of cell cycle proteins (Supplementary Data 3 and Supplementary Fig. 6e), which may directly regulate their protein levels by regulating their mRNAs translation by interacting with sequence elements in the 5′ and 3′ untranslated regions^46,47.

High FUS expression is correlated with human lactation deficiency

To further clarify the potential implications of FUS expression in human lactation, we collected fresh milk samples from 55 healthy lactating women within 5 days after giving birth to a full-term infant. mRNA expression in milk fat globules (MFGs) can serve as an indicator of gene expression in milk-secreting epithelial cells^1,48. To ensure the exclusion of RNA contamination from immune cells, we assessed MFG mRNA levels of specific markers (Supplementary Fig. 6f). Among the diverse expression patterns of various RBPs, we observed a significant reduction in FUS levels in women with sufficient milk compared to those with insufficient milk (Fig. 6n). Thus, these findings indicate that Fus levels are negatively correlated with milk production in lactating women.

Mechanically, we discovered that the downregulation of FUS is necessary for MEC differentiation during the late pregnancy and lactation stages by stabilizing p57Kip2 mRNA. Conversely, FUS overexpression during the lactation stage results in a reduction in p57Kip2 mRNA stability and inhibition of MEC differentiation, leading to sparse alveolar distribution and negative regulation of lactation (Fig. 7).

**Fig. 7: Working model of roles of FUS in regulating milk production in mammals.**

Discussion

Throughout mammalian evolution, there has been a progressive increase in the efficiency of milk production, serving as a vital carrier of intergenerational energy transfer, from marsupials to eutherians³. The composition of milk is complex, and its synthesis and secretion in eutherians begin during gestation in a tightly regulated process^49,50. In the current study, we noted the existence of milk components in sugar gliders during the pubertal stage (Fig. 1e–h), suggesting that lactation regulation in sugar gliders may exhibit inefficiencies that emerged early in the evolution of mammary glands. This evolutionary shift prompted us to investigate the key factors driving the evolution of lactation.

Notably, we observed the absence of a GS-sequence in FUS from mammals harboring strong lactation ability (Fig. 2d and Supplementary Fig. 2). In contrast, this sequence was retained in marsupials known to exhibit lower milk production efficiency³. Based on serial findings (Fig. 2 and Supplementary Fig. 1), we concluded that the absence of the GS-sequence in FUS during mammary gland evolution resulted in low expression of FUS during lactation, thereby promoting high milk production efficiency. For the mechanism by which GS-sequence regulates the expression levels of FUS, various studies have emphasized the importance of strict regulation of mRNA stability in controlling gene expression^51,52. Through mRNA stability experiments, we discovered that the insertion of GS-sequence enhanced the mRNA stability of Fus, suggesting that this may directly influence FUS expression levels (Fig. 2 and Supplementary Fig. 1). Furthermore, why does the insertion of GS-sequence affect mRNA stability? The stability of mRNA is primarily determined by its nucleotide sequence, which not only shapes its secondary and tertiary structure but also determines its accessibility by various RBPs^53,54. In addition, RNA modification can affect the activity, localization, and stability of mRNA^55,56,57. Mutations in the nucleotide sequence of Fus during mammary gland evolution may have influenced its RNA modification, thereby altering mRNA stability. However, further experiments are required to verify this hypothesis. In addition to the potential changes in RNA modification, GS-sequence may also impact FUS expression by affecting the multidimensional structure of FUS and regulating its stability and degradation^58,59.

When overexpressed FUS in MECs during the lactation stage, the mice exhibited a distinct phenotype characterized by sparse alveolar structures, insufficient milk production, and subsequent lactation failure, leading to poor newborn survival (Figs. 3, 4 and Supplementary Fig. 3, 4). Notably, the negative impact of Fus overexpression was less pronounced in the second lactation (Fig. 3e and Supplementary Fig. 3j) compared to the first (Fig. 3c, d and Supplementary Fig. 3h, i). A potential explanation for this observation is the typically higher volume of milk produced in the second pregnancy (including colostrum and mature milk) compared to the first, as documented in mice⁶⁰, human⁶¹, rabbits⁶², sows⁶³, and cows⁶⁴. Successive pregnancies significantly re-organize both the rate and pattern of mammary gland alveolar development⁶⁵ and the overall size of the mammary gland in mice⁶⁰, potentially contributing to the increase in milk production between the first and second lactations. In addition, pregnancy may alter the receptivity of the gland to pregnancy-related hormones, possibly through long-term epigenetic modifications. Genes affected by pregnancy-related epigenome changes are likely to be reactivated more rapidly in subsequent pregnancies²².

The mammary epithelium is organized into a basal layer of myoepithelial cells and a luminal layer of secretory cells¹¹. FUS expression was noted in both basal/myoepithelial cells and luminal cells (Supplementary Fig. 1f). Overexpression of FUS in luminal cells (WAP-Cre system) resulted in impaired lactation, linked to changes in MEC proliferation and differentiation (Fig. 4e, f and Supplementary Fig. 3a). Similarly, overexpression of FUS in basal/myoepithelial cells (K14 system) led to a sparse alveolar phenotype (Supplementary Figs. 3a, 4f, g), but the detailed mechanism affecting lactation needs to be further explored. In addition to their role in contraction, basal/myoepithelial cells have the potential to regulate alveolar cell development via paracrine pathways¹¹. For instance, the absence of p63 in basal/myoepithelial cells can lead to complete lactation failure due to obstructed cell proliferation and differentiation in the lumen⁶⁶. The K5 promoter has been used to target the expression of stable N-terminal truncated β-catenin in the basal/myoepithelial cell layer of the mammary gland⁶⁷. Activation of β-catenin signaling is accompanied by precocious lobular alveolar development during pregnancy, continuous proliferation, and accelerated involution of luminal cells during lactation⁶⁷. In addition, the deletion of kindin-2⁶⁸ or the overexpression of JAG1⁶⁹ in basal/myoepithelial cells can regulate the differentiation and lactation of alveolar cells by affecting Notch signaling pathway activation.

The mammary gland undergoes dynamic changes during different stages of development. In parallel, FUS also shows a dynamic pattern of expression. Our study revealed that the dynamic expression pattern of FUS plays a crucial role in coordinating the balance between MEC proliferation and differentiation (Figs. 5, 6 and Supplementary Figs. 5, 6). This coordination ensures appropriate cell cycle exit and differentiation of MEC by regulating the RNA stability of p57Kip2 (Fig. 6 and Supplementary Fig. 6). Furthermore, to understand the regulatory mechanisms underlying the dynamic expression pattern of FUS, we explored its hormonal regulation, considering the close relationship between mammary gland development and the endocrine system. Notably, previous studies in mice have shown that prolactin, progesterone, and estrogen regulate alveolar formation and secretion differentiation²⁷. By re-analyzing previously published RNA-seq data⁷⁰, we found that exogenous progesterone injections had no significant effect on FUS expression in ovariectomized mice. Similarly, in vitro experiments treating cells with progesterone or oxytocin did not significantly impact FUS expression (Supplementary Fig. 7). These results suggest that low FUS expression during pregnancy is not hormonally regulated. Thus, further investigations are needed to unravel the underlying factors driving the dynamic expression pattern of FUS and its intricate relationship with mammary gland development.

In conclusion, we observed the gradual establishment of milk production from the marsupial to the eutherian development of the mammary gland. During this process, FUS was identified as a key negative regulator in mammary gland evolution. Clinically, the expression of Fus in human milk is inversely correlated with lactation ability (Fig. 6n). These findings enhance our understanding of mammary gland development and provide a new avenue for exploring the relationship between mammary gland evolution and the factors underlying inadequate breast milk production.

Methods

Animals

The wild-type (WT) mouse FUS coding sequence (CDS) (NM_139149.2) was inserted into the CAG-IRES-EGFP vector. The resulting construct was purified and used for microinjection to generate Fus overexpression transgenic mice. The CAG-IRES-EGFP vector contains a CAG promoter followed by a loxP-flanked sequence and transcriptional STOP sequence, driving the expression of enhanced green fluorescent protein (EGFP)-coding regions linked by an internal ribosomal entry site (IRES) (Supplementary Fig. 3a). The inclusion of an IRES following the FUS open reading frame (ORF) enables independent translation of the EGFP ORF, thereby preventing the translation of the FUS-GFP fusion protein. To induce mammary gland-specific overexpression of FUS, the Fus overexpression transgenic mice were bred with WAP-Cre mice (stock #007905; Jackson Laboratory) and K14-Cre^ERT mice to yield Fus-OE mice. Tamoxifen (80 mg/kg/d) dissolved in ethanol and then in corn oil was administered to mice through intraperitoneal injection every other day for 3 days. All mice were bred and maintained under specific pathogen-free conditions. Genotyping was determined via PCR using the primers: Fus-LoxP: forward 5ʹ-AGTCGCTCTGAGTTGTTATCAG-3ʹ, reverse 1 5ʹ-TGAGCATGTCTTTAATCTACCTCGATG-3ʹ, reverse 2 5ʹ-AGTCCCTATTGGCGTTACTATGG-3ʹ; Cre: forward 5ʹ-AATGCTTCTGTCCGTTTGCCGG-3ʹ, reverse 5ʹ-CCAGGCTAAGTGCCTTCTCTACA-3ʹ. The mice were raised in a specific pathogen-free (SPF) environment with an ambient temperature of 18–22 °C, a humidity of 50–60%, and a 12 h light-dark cycle. The sugar gliders were obtained from the pet trade and thereafter were maintained in breeding colonies at the Kunming Institute of Zoology, Chinese Academy of Sciences. Horseshoe bats were captured in Yunnan, China, between 2022 and 2023. Mice, sugar gliders, and horseshoe bats were euthanized by asphyxiation using a chamber with carbon dioxide and cervical dislocation as a secondary method. All experimental procedures and animal care and handling were performed per the protocols approved by the Ethics Committee of the Kunming Institute of Zoology, Chinese Academy of Sciences (IACUC-PA-2023-03-049).

RNA sequencing and data analysis

Raw RNA sequencing data from mouse and sugar glider mammary glands in the virgin and lactation periods were processed through standard Illumina pipelines for base-calling and fastq file generation. Reads were then mapped against the genome assembly using HISAT2 v2.1.0⁷¹. The alignment results were converted to BAM format using SAMtools v1.9⁷². FeatureCounts v2.0.3⁷³ was used to assign sequence reads to annotation. Genes possessing less than 10 raw counts were removed. Differential expression analysis was performed with the Bioconductor DESeq2 package v1.34.0⁷⁴. Significant genes in mice were determined by an adjusted P value <0.05 and fold-change >2 or <−2. Non-significant genes in sugar gliders were determined by an adjusted P value >0.05 and fold-change <−2 and <2. The reference genome and annotation versions used for analysis were GRCm39.108 for mice, Large_White_v1 for pigs, and ARS-UCD1.2 for cattle, with the sugar glider reference genome and annotation versions sourced from the FigShare repository: https://figshare.com/s/d6c585fbae0c1f22e8df⁷⁵.

Genomic data

The FUS coding gene sequences (CDSs) of the ten species used in this study, including Homo sapiens, Mus musculus, Capra hircus, Camelus bactrianus, Sus scrofa, Rhinolophus sinicus, Bos taurus, Petaurus breviceps, Macropus eugenii, and Sarcophilus harrisii, were downloaded from the National Center for Biotechnology Information database (www.ncbi.nlm.nih.gov/nuccore).

Whole-mount staining

The inguinal mammary glands were placed on a slide and fixed in 25% glacial acetic acid and 75% ethanol for 1–2 d. The tissue was then stained in carmine alum solution overnight at 4 °C, dehydrated with a graded series of ethanol solutions, cleared in xylene, and coverslipped with Neutral Balsam (Solarbio, #G8590).

Histological analysis and immunostaining

Mammary glands were fixed in a 10% neutral-buffered formalin solution and embedded in paraffin. The paraffin-embedded tissue sections (5-μm thickness) were then dried at 65 °C for 1 h and stored at −20°C. The paraffin-embedded tissue sections were dehydrated with xylene and graded alcohol and boiled in 10 mM sodium citrate for 20 min for antigen retrieval. The sections were then used for immunofluorescence and immunohistochemical staining. For immunofluorescence, sections were blocked with 10% goat serum for 1 h, then incubated with primary antibodies overnight at 4 °C. The slides were then washed three times in phosphate-buffered saline (PBS) and incubated with secondary antibodies for 1 h at room temperature. The slides were counterstained with 4’,6-diamidino-2-phenylindole dihydrochloride. For immunohistochemical analysis, sections were inactivated with endogenous peroxidases with 3% H₂O₂ for 10 min, then incubated with primary antibodies overnight at 4 °C and secondary antibodies for 1 h at room temperature after blocking for 1 h. Primary antibodies used for immunostaining were FUS (1:100, Abcam, #ab124923), FUS (1:800, Bethyl, #A300-294), CD133 (1:100, ABclonal, #A0219), EpCAM (1:100, Abcam, #ab71916), K14 (1:50, Abcam, #ab118685), K18 (1:50, Abcam, #ab133263), milk (1:500, Nordic Immunology, #5941), Ki67 (1:500, Abcam, #ab15580), ELF5 (1:500, ABclonal, #A7181), and E-cadherin (1:100, Abcam, #ab40772). Secondary antibodies used in immunostaining included Alexa Fluor 555 goat anti-rabbit IgG (1:500, Invitrogen, #A21428), Alexa Fluor 488 goat anti-mouse IgG (1:500, Abcam, #ab150113), and anti-rabbit IgG (1:200, Abcam, #ab6721).

Raman spectroscopy analysis

Fresh breast tissues from mice and sugar gliders during virgin and lactation were frozen with the embedding agent (SAKURA, #4583) and then sliced. Milk secretion from mice and sugar gliders was evenly applied to the slides as a positive control for Raman spectroscopy. The slides are submitted to the company for Raman spectroscopic analysis. Then, the difference in Raman characteristic peaks between milk secretion and mammary gland contents inside the ductal cavity was compared.

Oil Red staining

Fresh mammary tissue was placed in 4% paraformaldehyde and fixed at 4 °C. The fixed tissue was then dehydrated by soaking in a 30% sucrose solution for 24 h. After embedding the tissue in a tissue embedding medium (Surgipath, #3801480), 20-μm thick slices were prepared and stored at −80 °C. For staining, the slices were immersed in 60% isopropanol for 30 s, followed by incubation in prepared Oil Red working solution (Servicebio, #G1015) in the dark for 10 min. The slices were then taken out and immersed in 60% isopropanol for 5 s, repeating the process twice. After a brief double-distilled water rinse for 10 s, the slices were stained with hematoxylin. Finally, the slices were rinsed with running water and mounted with glycerol-alum (Servicebio, #G1402).

Cell lines and culture

HC11 cells (obtained from the Bernd Groner Lab, Ludwig Institute for Cancer Research) were grown in RPMI-1640 medium (Gibco, #C11875500BT) with 10% fetal bovine serum (FBS), 5 μg/ml Insulin (Sigma, #I5500), 5 μg/ml gentamycin sulfate, and 10 ng/ml epidermal growth factor (EGF, Gibco, #PGH0315). The HEK293T cells were purchased and authenticated from the American Type Culture Collection (ATCC) and cultured in Dulbecco’s Modified Eagle Medium (DMEM) (Gibco, #C11995500BT) supplemented with 10% FBS and 1% penicillin-streptomycin solution (PS). The MCF10A cells were purchased and authenticated from the ATCC and cultured in DMEM/F12 (Gibco, #C11330500BT) supplemented with 10% Horse Serum (HS), 20 ng/ml EGF, 100 ng/ml Cholera toxin, 0.008 mg/ml Insulin, 50 ng/ml Hydrocortisone and 1% PS.

RNA extraction and RT-qPCR

Total RNA was prepared using TRIzol reagent (Life Technologies), then converted to cDNA using a HiScript®III RT SuperMix for qPCR Kit (Vazyme, #R323-01). RT-qPCR analysis was performed using a SYBR qPCR Master Mix (Vazyme, #Q711-02) on a QuantStudio 3 instrument and normalized against GAPDH. Primers used are listed in Supplementary Table 2.

Protein extraction and western blot analysis

Cell and breast tissue homogenates were subjected to lysis in RIPA buffer containing protease inhibitors. The samples were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to a polyvinylidene difluoride (PVDF) membrane. The membranes were blocked with 5% nonfat dry milk for 1 h and incubated with primary antibodies at 4 °C overnight, then incubated with horseradish peroxidase (HRP)-linked secondary antibodies (Sigma) for 1 h at room temperature. A chemiluminescent HRP substrate (Millipore) was used to detect protein expression. The following primary antibodies were used for western blot analysis: α-Tubulin (1:5000, Sigma, #T5168), GAPDH (1:2000, Santa Cruz, #sc-25778), Flag (1:2000, CST, #14793), Cre (1:1000, Abcam, #ab190177), GFP (1:1000, Abcam, #ab290), FUS (1:1000, Abcam, #ab124923), p57Kip2 (1:1000, Abcam, #ab75974), PCNA (1:1000, Abcam, #ab29), p-RB (1: 50,000, ABclonal, #AP0484), Cyclin B1 (1:1000, Abcam, #ab72), Cyclin D1 (1:1000, CST, #2978 S), Cyclin E1 (1:1000, CST, #4129 S), P21 (1:1000, Abcam, #ab109199), and P27 (1:1000, BD biosciences, #610241). The secondary antibodies used were anti-mouse IgG (1: 5000, Abcam, #ab6728) and anti-mouse IgG (1:5000, Abcam, #ab6721).

Plasmid construction, overexpression, and knockdown

For overexpression, full-length (FL) and mutant (GS-sequence inserted) human (NM_004960.4) or mouse (NM_139149.2) FUS CDS with Flag or GFP tags was cloned into pCDH-CMV-MCS-EF1-Puro lentiviral expression vector. For knockdown, shRNAs were cloned into the pLKO.1 vector. The shRNA sequences are listed in Supplementary Table 2. All plasmids were sequenced to rule out mutations. The vectors were co-transfected with psPAX2 and pMD2.G (4:3:1) into 293T cells to produce lentiviral particles, which were then transfected into HC11 or MCF10A cells. After 48–72 h, the cells were collected for further analysis or experiments.

Primary MEC preparation

Breast tissue was cleaved and digested in lysate with DMEM/F12 containing 300 U/ml collagenase I (Sigma, #C0130), 100 U/ml hyaluronidase (Sigma, #H3506), 5% FBS, and 1% penicillin-streptomycin-glutamine for 1–3 h at 37 °C and 100 rpm. The cells were then treated with pancreatin for 5 min for dispersal, and DNA was removed with 5 mg/ml dispase (Sigma, #D4693) containing 0.1 mg/ml DNase I (Roche, #11248932001) for 5 min at 37 °C with gentle pipetting. Finally, red blood cells were removed in 0.8% NH₄Cl, followed by filtration through a 40-mm filter. Primary MECs were grown in DMEM/F12 medium with 1 mM Glutamine, 5 µg/ml Insulin, 500 ng/ml Hydrocortisone, 10 ng/ml EGF, and 20 ng/ml Cholera toxin.

mRNA stability analysis and differentiation induction

Cells were treated with 5 μg/ml actinomycin D (Sigma, #SBR00013) and collected at different times for RNA analysis to examine mRNA decay. For differentiation induction, a small number of cells were plated in six-well plates and allowed to reach confluence after 1–2 weeks. Once confluence was reached, cells were grown for 24 h in a medium without epidermal growth factor (EGF), followed by growth in a medium containing 5 μg/ml prolactin (ProSpec, #cyt-240), 1 μM Dexamethasone, and 5 μg/ml Insulin.

Mouse milk collection

Milk collection was performed according to previously reported methods⁷⁶. Female mice were separated from their pups for 3–5 h at L2 or L10 and injected intraperitoneally with 0.2 U oxytocin (Sigma) after anesthesia. After 20 min, as much milk as possible was collected manually from each mammary gland nipple. Subsequently, milk-specific antibodies were utilized for staining to confirm that all milk had been completely collected from the mice. Samples in which milking was deemed incomplete, indicated by a substantial amount of milk remaining within the mammary gland, were excluded from further analysis.

Cell cycle

HC11 and primary cells were cultured in a complete medium for 24 h at 37 °C after growth in a blank medium for 12 h at 37 °C. The cells were immobilized with 75% ethanol overnight at 4 °C, followed by incubation with propidium iodide (PI) mixture (0.1 mg/ml PI, 1 mg/ml RNase A, and 0.6% NP40) at 37 °C for 30 min away from light. The cells were then moved into a flow tube for cell cycle analysis by flow cytometry.

Clone and MTS assays

Cells were grown in a medium containing Matrigel (Corning, #354234), and clone formation was observed and photographed after 10 days. For the MTS assay, primary and HC11 cells were incubated with an MTS mixture (Promega, #G5421) (MTS: blank medium = 1:5) at 37 °C for 1 h away from light after growth on a 96-well plate for 2 days. Absorbance at OD₄₉₀ was measured and analyzed.

RNA immunoprecipitation

HC11 cells were collected and incubated with RIP lysis buffer overnight at −80 °C using the RIP Kit (Millipore, #17-701). The cell lysate was then immunoprecipitated with FUS (5 μg) (Bethyl, #A300-294A) and IgG (5 μg, Millipore, #CS200621) beads for 6–8 h at 4 °C, followed by washing with RIP buffer. RNA was then purified for RT-qPCR analysis and RIP-seq.

RNA pull-down

The 3’UTR of p57Kip2 mRNA was synthesized in vitro using a HiScribe T7 High Yield RNA Synthesis Kit (NEB, #E2040S). Single biotinylated nucleotide was then attached to the RNA using a Pierce RNA 3’ End Desthiobiotinylation Kit (Thermo, #20163). The protein-RNA complexes were subsequently washed three times with wash buffer after the HC11 cell extract was incubated with biotin-labeled RNAs for 1 h at 4 °C using an RNA Pull-Down Kit (Thermo, #20164). Interaction between the FUS protein and 3’UTR of p57Kip2 mRNA were detected by protein elution and western blot analysis.

Human milk sample collection and analysis

Written informed consent was obtained from all participants. Fresh milk (1–5 ml) was collected from 55 healthy females who gave birth to full-term babies 3–5 days after delivery and immediately placed on ice. The subsequent MFG separation steps were completed within 3 h before the sample was transferred to a −80 °C refrigerator. The upper-layer phase containing MFGs was carefully transferred to another RNase-free tube after centrifugation of the milk for 10 min at 13,000 rpm and 4 °C, followed by washing the MFGs with cold PBS three times. TRIzol lysis reagent was added to the MFGs for RNA extraction. Mothers were followed up by phone at 7–8 weeks postpartum to enquire about breastfeeding and formula feeding. According to the follow-up results, the samples were divided into three groups: insufficient milk (predominantly formula feeding), sufficient milk (meeting the baby’s needs), and spare milk (more than the baby’s needs). This project was approved by the Ethics Committee of Weifang People’s Hospital or Luoyang Maternal and Child Health Hospital.

Statistics and reproducibility

All animals were randomly assigned to the experimental groups, except when the purpose of the experiment was to compare the difference between controls and Fus-OE mice. In histological staining, more than four visual fields in each of the five directions of the upper, lower, left, right, and middle were randomly selected for statistics. No data were excluded from the experiments. The investigators were not blinded to outcome assessments. All experiments were performed to have at least three biological replicates to ensure power for statistical analysis using a two-sided student t-test. Sample groups for all experiments were not blinded. P values equal to or <0.05 were considered to be significant. All immunoblots are performed at least three biological replicates unless specified in figure legends. Graphs and error bars reflect means ± SD. All statistical analyses were carried out using GraphPad Prism 8.0.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

All data used in this study are available within the Article and Supplementary information, or available from the corresponding authors on request. Source data are provided as Source Data file. Source data are provided with this paper. The raw transcriptomics data produced and analyzed in this study have been deposited in the Genome Sequence Archive in the National Genomics Data Center, China National Center for Bioinformation / Beijing Institute of Genomics, Chinese Academy of Sciences that are publicly accessible at https://bigd.big.ac.cn/gsa/browse/CRA014528. The accession code is GSA: CRA014528. Source data are provided with this paper.

References

Goldman, A. S. Evolution of immune functions of the mammary gland and protection of the infant. Breastfeed. Med. 7, 132–142 (2012).
Article PubMed Google Scholar
Lefèvre, C. M., Sharp, J. A. & Nicholas, K. R. Evolution of lactation: ancient origin and extreme adaptations of the lactation system. Annu. Rev. Genom. Hum. Genet. 11, 219–238 (2010).
Article Google Scholar
McClellan, H. L., Miller, S. J., & Hartmann, P. E. Evolution of lactation: nutrition v. protection with special reference to five mammalian species. Nutr. Res. Rev. 21, 97–116 (2008).
Sharp, J. A. et al. The tammar wallaby: a marsupial model to examine the timed delivery and role of bioactives in milk. Gen. Comp. Endocrinol. 244, 164–177 (2017).
Article CAS PubMed PubMed Central Google Scholar
Crispel, Y. et al. Effects of breastfeeding on body composition and maturational tempo in the rat. BMC Med. 11, 114 (2013).
Article PubMed PubMed Central Google Scholar
Horta, B. L., Loret de Mola, C. & Victora, C. G. Breastfeeding and intelligence: a systematic review and meta-analysis. Acta Paediatr. 104, 14–19 (2015).
Article PubMed Google Scholar
Ip, S. et al. Breastfeeding and maternal and infant health outcomes in developed countries. Evid. Rep. Technol. Assess. 153, 1–186 (2007).
Molès, J.-P. et al. Breastmilk cell trafficking induces microchimerism-mediated immune system maturation in the infant. Pediatr. Allergy Immunol. 29, 133–143 (2018).
Article PubMed Google Scholar
Dietz, W. H. The response of the US Centers for Disease Control and Prevention to the obesity epidemic. Annu. Rev. Public Health 36, 575–596 (2015).
Article PubMed Google Scholar
Maningat, P. D. et al. Gene expression in the human mammary epithelium during lactation: the milk fat globule transcriptome. Physiol. Genomics 37, 12–22 (2009).
Article CAS PubMed Google Scholar
Fu, N. Y. et al. Stem cells and the differentiation hierarchy in mammary gland development. Physiol. Rev. 100, 489–523 (2020).
Article CAS PubMed Google Scholar
Macias, H. & Hinck, L. Mammary gland development. Wiley Interdiscip. Rev. Dev. Biol. 1, 533–557 (2012).
Article CAS PubMed PubMed Central Google Scholar
Joshi, K. et al. Cellular proliferation in the rat mammary gland during pregnancy and lactation. Lab. Investig. 54, 52–61 (1986).
CAS PubMed Google Scholar
Hadsell, D. L. et al. Changes in secretory cell turnover, and mitochondrial oxidative damage in the mouse mammary gland during a single prolonged lactation cycle suggest the possibility of accelerated cellular aging. Exp. Gerontol. 41, 271–281 (2006).
Article CAS PubMed Google Scholar
Jiang, H., Guan, W. & Gu, Z. Tinkering evolution of post-transcriptional RNA regulons: puf3p in fungi as an example. PLoS Genet. 6, e1001030 (2010).
Article PubMed PubMed Central Google Scholar
Patraquim, P., Warnefors, M. & Alonso, C. R. Evolution of Hox post-transcriptional regulation by alternative polyadenylation and microRNA modulation within 12 Drosophila genomes. Mol. Biol. Evol. 28, 2453–2460 (2011).
Article CAS PubMed Google Scholar
Verta, J.-P. & Jacobs, A. The role of alternative splicing in adaptation and evolution. Trends Ecol. Evol. 37, 299–308 (2022).
Article CAS PubMed Google Scholar
Corbett, A. H. Post-transcriptional regulation of gene expression and human disease. Curr. Opin. Cell Biol. 52, 96–104 (2018).
Harvey, R. et al. Post-transcriptional control of gene expression following stress: the role of RNA-binding proteins. Biochem. Soc. Trans. 45, 1007–1014 (2017).
Article CAS PubMed PubMed Central Google Scholar
Ray, D. et al. A compendium of RNA-binding motifs for decoding gene regulation. Nature 499, 172–177 (2013).
Article ADS CAS PubMed PubMed Central Google Scholar
dos Santos, C. O. et al. Molecular hierarchy of mammary differentiation yields refined markers of mammary stem cells. Proc. Natl Acad. Sci. USA 110, 7123–7130 (2013).
Article ADS PubMed PubMed Central Google Scholar
Dos Santos, C. O. et al. An epigenetic memory of pregnancy in the mouse mammary gland. Cell Rep. 11, 1102–1109 (2015).
Article PubMed PubMed Central Google Scholar
Li, C. M.-C. et al. Aging-associated alterations in mammary epithelia and stroma revealed by single-cell RNA sequencing. Cell Rep. 33, 108566 (2020).
Article CAS PubMed PubMed Central Google Scholar
Eriksen, K. G. et al. Human milk composition and infant growth. Curr. Opin. Clin. Nutr. Metab. Care 21, 200–206 (2018).
Article CAS PubMed Google Scholar
Butler, H. J. et al. Using Raman spectroscopy to characterize biological materials. Nat. Protoc. 11, 664–687 (2016).
Article CAS PubMed Google Scholar
Dodo, K., Fujita, K. & Sodeoka, M. Raman spectroscopy for chemical biology research. J. Am. Chem. Soc. 144, 19651–19667 (2022).
Article CAS PubMed PubMed Central Google Scholar
Hannan, F. M. et al. Hormonal regulation of mammary gland development and lactation. Nat. Rev. Endocrinol. 19, 46–61 (2023).
Article PubMed Google Scholar
Stein, T. et al. Involution of the mouse mammary gland is associated with an immune cascade and an acute-phase response, involving LBP, CD14 and STAT3. Breast Cancer Res. 6, R75–R91 (2004).
Article CAS PubMed Google Scholar
Osada, N. Genetic diversity in humans and non-human primates and its evolutionary consequences. Genes Genet. Syst. 90, 133–145 (2015).
Article CAS PubMed Google Scholar
Wagner, K. U. et al. Cre-mediated gene deletion in the mammary gland. Nucleic Acids Res. 25, 4323–4330 (1997).
Article CAS PubMed PubMed Central Google Scholar
Zhao, L. et al. TDP-43 facilitates milk lipid secretion by post-transcriptional regulation of Btn1a1 and Xdh. Nat. Commun. 11, 341 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
Kim, U. et al. Mammary-enriched transcription factors synergize to activate the Wap super-enhancer for mammary gland development. Int. J. Mol. Sci. 23, 11680 (2022).
Lemay, D. G. et al. RNA sequencing of the human milk fat layer transcriptome reveals distinct gene expression profiles at three stages of lactation. PLoS ONE 8, e67531 (2013).
Article ADS CAS PubMed PubMed Central Google Scholar
Robinson, G. W. et al. Coregulation of genetic programs by the transcription factors NFIB and STAT5. Mol. Endocrinol. 28, 758–767 (2014).
Oakes, S. R., Hilton, H. N. & Ormandy, C. J. The alveolar switch: coordinating the proliferative cues and cell fate decisions that drive the formation of lobuloalveoli from ductal epithelium. Breast Cancer Res. 8, 207 (2006).
Article PubMed PubMed Central Google Scholar
Oakes, S. R. et al. The Ets transcription factor Elf5 specifies mammary alveolar cell fate. Genes Dev. 22, 581–586 (2008).
Article CAS PubMed PubMed Central Google Scholar
Sornapudi, T. R. et al. RNA sequencing of murine mammary epithelial stem-like cells (HC11) undergoing lactogenic differentiation and its comparison with embryonic stem cells. BMC Res. Notes 11, 241 (2018).
Article PubMed PubMed Central Google Scholar
Morrison, B. & Cutler, M. L. Mouse mammary epithelial cells form mammospheres during lactogenic differentiation. J. Vis. Exp. 32, 1265 (2009).
Wirl, G. et al. Mammary epithelial cell differentiation in vitro is regulated by an interplay of EGF action and tenascin-C downregulation. J. Cell Sci. 108, 2445–2456 (1995).
Article CAS PubMed Google Scholar
Soufi, A. & Dalton, S. Cycling through developmental decisions: how cell cycle dynamics control pluripotency, differentiation and reprogramming. Development 143, 4301–4311 (2016).
Article CAS PubMed PubMed Central Google Scholar
Sainz de la Maza, D. et al. Cell-cycle exit and stem cell differentiation are coupled through regulation of mitochondrial activity in the Drosophila testis. Cell Rep. 39, 110774 (2022).
Article CAS PubMed PubMed Central Google Scholar
Ruijtenberg, S. & van den Heuvel, S. Coordinating cell proliferation and differentiation: antagonism between cell cycle regulators and cell type-specific gene expression. Cell Cycle 15, 196–212 (2016).
Article CAS PubMed PubMed Central Google Scholar
Portz, B., Lee, B. L. & Shorter, J. FUS and TDP-43 phases in health and disease. Trends Biochem. Sci. 46, 550–563 (2021).
Article CAS PubMed PubMed Central Google Scholar
Pateras, I. S. et al. p57KIP2: “Kip“ing the cell under control. Mol. Cancer Res. 7, 1902–1919 (2009).
Article CAS PubMed Google Scholar
Zhou, H. et al. RNA-binding proteins in neurological diseases. Sci. China Life Sci. 57, 432–444 (2014).
Article ADS CAS PubMed Google Scholar
Thoreen, C. C. et al. A unifying model for mTORC1-mediated regulation of mRNA translation. Nature 485, 109–113 (2012).
Article ADS CAS PubMed PubMed Central Google Scholar
Harvey, R. F. et al. Trans-acting translational regulatory RNA binding proteins. Wiley Interdiscip. Rev. RNA 9, e1465 (2018).
Article PubMed PubMed Central Google Scholar
Heid, H. W. & Keenan, T. W. Intracellular origin and secretion of milk fat globules. Eur. J. Cell Biol. 84, 245–258 (2005).
Article CAS PubMed Google Scholar
McManaman, J. L. & Neville, M. C. Mammary physiology and milk secretion. Adv. Drug Deliv. Rev. 55, 629–641 (2003).
Article CAS PubMed Google Scholar
Shennan, D. B. & Peaker, M. Transport of milk constituents by the mammary gland. Physiol. Rev. 80, 925–951 (2000).
Article CAS PubMed Google Scholar
Liu, H., Luo, M. & Wen, J.-K. mRNA stability in the nucleus. J. Zhejiang Univ. Sci. B 15, 444–454 (2014).
Article CAS PubMed PubMed Central Google Scholar
Bicknell, A. A. & Ricci, E. P. When mRNA translation meets decay. Biochem. Soc. Trans. 45, 339–351 (2017).
Article CAS PubMed Google Scholar
Bae, H. & Coller, J. Codon optimality-mediated mRNA degradation: linking translational elongation to mRNA stability. Mol. Cell 82, 1467–1476 (2022).
Article CAS PubMed PubMed Central Google Scholar
Durmaz, Y. T., Shatadal, A. & Friend, K. Geneticin reduces mRNA stability. PLoS ONE 17, e0272058 (2022).
Article CAS PubMed PubMed Central Google Scholar
Boo, S. H. & Kim, Y. K. The emerging role of RNA modifications in the regulation of mRNA stability. Exp. Mol. Med. 52, 400–408 (2020).
Article CAS PubMed PubMed Central Google Scholar
Lee, M., Kim, B. & Kim, V. N. Emerging roles of RNA modification: m(6)A and U-tail. Cell 158, 980–987 (2014).
Article CAS PubMed Google Scholar
Li, X., Peng, J. & Yi, C. Acetylation enhances mRNA stability and translation. Biochemistry 58, 1553–1554 (2019).
Article CAS PubMed Google Scholar
Timms, R. T. et al. A glycine-specific N-degron pathway mediates the quality control of protein N-myristoylation. Science 365, eaaw4912 (2019).
Pesiridis, G. S., Lee, V. M. Y. & Trojanowski, J. Q. Mutations in TDP-43 link glycine-rich domain functions to amyotrophic lateral sclerosis. Hum. Mol. Genet. 18, R156–R162 (2009).
Article CAS PubMed PubMed Central Google Scholar
Knight, C. H., Maltz, E. & Docherty, A. H. Milk yield and composition in mice: effects of litter size and lactation number. Comp. Biochem. Physiol. A Comp. Physiol. 84, 127–133 (1986).
Article CAS PubMed Google Scholar
Ingram, J., Woolridge, M. & Greenwood, R. Breastfeeding: it is worth trying with the second baby. Lancet 358, 986–987 (2001).
Article CAS PubMed Google Scholar
McNitt, J. I. & Lukefahr, S. D. Effects of breed, parity, day of lactation and number of kits on milk production of rabbits. J. Anim. Sci. 68, 1505–1512 (1990).
Article CAS PubMed Google Scholar
Speer, V. C. & Cox, D. F. Estimating milk yield of sows. J. Anim. Sci. 59, 1281–1285 (1984).
Article CAS PubMed Google Scholar
Gonçalves, J. L. et al. Milk losses associated with somatic cell counts by parity and stage of lactation. J. Dairy Sci. 101, 4357–4366 (2018).
Article PubMed Google Scholar
Lang, S. L. C., Iverson, S. J. & Bowen, W. D. Primiparous and multiparous females differ in mammary gland alveolar development: implications for milk production. J. Exp. Biol. 215, 2904–2911 (2012).
Article PubMed Google Scholar
Forster, N. et al. Basal cell signaling by p63 controls luminal progenitor function and lactation via NRG1. Dev. Cell 28, 147–160 (2014).
Article CAS PubMed PubMed Central Google Scholar
Teulière, J. et al. Targeted activation of beta-catenin signaling in basal mammary epithelial cells affects mammary development and leads to hyperplasia. Development 132, 267–277 (2005).
Article PubMed Google Scholar
Wang, Z. et al. Kindlin-2 in myoepithelium controls luminal progenitor commitment to alveoli in mouse mammary gland. Cell Death Dis. 14, 675 (2023).
Article CAS PubMed PubMed Central Google Scholar
Gomi, K. et al. JAG1-mediated Notch signaling regulates secretory cell differentiation of the human airway epithelium. Stem Cell Rev. Rep. 12, 454–463 (2016).
Article CAS PubMed Google Scholar
Fernandez-Valdivia, R. et al. Transcriptional response of the murine mammary gland to acute progesterone exposure. Endocrinology 149, 6236–6250 (2008).
Article CAS PubMed PubMed Central Google Scholar
Kim, D., Langmead, B. & Salzberg, S. L. HISAT: a fast spliced aligner with low memory requirements. Nat. Methods 12, 357–360 (2015).
Article CAS PubMed PubMed Central Google Scholar
Li, H. et al. The sequence alignment/map format and SAMtools. Bioinformatics 25, 2078–2079 (2009).
Article PubMed PubMed Central Google Scholar
Liao, Y., Smyth, G. K. & Shi, W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics 30, 923–930 (2014).
Article CAS PubMed Google Scholar
Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014).
Article PubMed PubMed Central Google Scholar
Feigin, C. Y. et al. Convergent deployment of ancestral functions during the evolution of mammalian flight membranes. Sci. Adv. 9, eade7511 (2023).
Article CAS PubMed PubMed Central Google Scholar
Willingham, K. et al. Milk collection methods for mice and Reeves’ muntjac deer. J. Vis. Exp. 89, 51007 (2014).

Download references

Acknowledgements

This work was supported by the National Key Research and Development Program of China (2023YFA1800500, B.J.), National Natural Science Foundation of China (U23A20452, B.J., 31701306, H.W., and 82272727, H.W.), Yunnan Applied Basic Research Key Projects (202201AS070039, B.J., and 202401AS070079, H.W.), Yunnan Applied Basic Research Projects (202301AT070280, H.W.), Spring City Project from Kunming Science and Technology Bureau (2022SCP007, B.J.), CAS “Light of West China” Program (H.W.) and Yunnan Revitalization Talent Support Program Young Talent Project (H.W.). We would like to thank Nengyin Sheng and Ceshi Chen for constructive suggestions and Christine Watts for English editing. We thank the Institutional Center for Shared Technologies and Facilities of Kunming Institute of Zoology (KIZ), Chinese Academy of Sciences (CAS) for providing us with Confocal Microscopy image acquisition and flow cytometric analysis from Cong Li, Guolan Ma, and Xiuzhen Yang.

Author information

These authors contributed equally: Haili Shao, Jipeng Huang, Hui Wang, Guolei Wang.

Authors and Affiliations

National Key Laboratory of Genetic Evolution & Animal Models, Kunming Institute of Zoology, Chinese Academy of Sciences, Kunming, Yunnan, 650201, China
Haili Shao, Jipeng Huang, Hui Wang, Xu Yang, Mei Cheng, Changjie Sun, Li Zou, Qin Yang, Zhen Liu, Xuelong Jiang, Lei Shi, Peng Shi & Baowei Jiao
Kunming College of Life Science, University of the Chinese Academy of Sciences, Kunming, Yunnan, 650201, China
Haili Shao, Jipeng Huang, Xu Yang, Mei Cheng, Changjie Sun & Li Zou
Department of Obstetrics, Weifang People’s Hospital, Weifang, Shandong, 261042, China
Guolei Wang
Luoyang Maternal and Child Health Hospital, Luoyang, Henan, 471000, China
Dandan Zhang & Baowei Han
KIZ-CUHK Joint Laboratory of Bioresources and Molecular Research in Common Diseases, Kunming Institute of Zoology, Chinese Academy of Sciences, Kunming, Yunnan, 650203, China
Peng Shi & Baowei Jiao

Authors

Haili Shao
View author publications
You can also search for this author in PubMed Google Scholar
Jipeng Huang
View author publications
You can also search for this author in PubMed Google Scholar
Hui Wang
View author publications
You can also search for this author in PubMed Google Scholar
Guolei Wang
View author publications
You can also search for this author in PubMed Google Scholar
Xu Yang
View author publications
You can also search for this author in PubMed Google Scholar
Mei Cheng
View author publications
You can also search for this author in PubMed Google Scholar
Changjie Sun
View author publications
You can also search for this author in PubMed Google Scholar
Li Zou
View author publications
You can also search for this author in PubMed Google Scholar
Qin Yang
View author publications
You can also search for this author in PubMed Google Scholar
Dandan Zhang
View author publications
You can also search for this author in PubMed Google Scholar
Zhen Liu
View author publications
You can also search for this author in PubMed Google Scholar
Xuelong Jiang
View author publications
You can also search for this author in PubMed Google Scholar
Lei Shi
View author publications
You can also search for this author in PubMed Google Scholar
Peng Shi
View author publications
You can also search for this author in PubMed Google Scholar
Baowei Han
View author publications
You can also search for this author in PubMed Google Scholar
Baowei Jiao
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

B.J., H.S., J.H., and H.W. conceived the project, designed the experiments, and wrote the manuscript. H.S., H.W., J.H., and M.C. conducted all experiments. H.S., H.W., J.H., M.C., X.Y., P. S., B.H., and L.S. discussed or analyzed the data. H.W., G.W., D.Z., and B.H. collected human milk and conducted postpartum follow-up. C.S., Z.L., and X.J. helped obtain horseshoe bat samples. L.Z. and Q.Y. purchased the reagents.

Corresponding authors

Correspondence to Baowei Han or Baowei Jiao.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks the anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Peer Review File

Description of Additional Supplementary Files

Supplementary Data 1

Supplementary Data 2

Supplementary Data 3

Reporting Summary

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Shao, H., Huang, J., Wang, H. et al. Fused in sarcoma (FUS) inhibits milk production efficiency in mammals. Nat Commun 15, 3953 (2024). https://doi.org/10.1038/s41467-024-48428-5

Download citation

Received: 25 August 2023
Accepted: 25 April 2024
Published: 10 May 2024
DOI: https://doi.org/10.1038/s41467-024-48428-5

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.