Kindlin-2 in myoepithelium controls luminal progenitor commitment to alveoli in mouse mammary gland

Wang, Zhenbin; Zhang, Lei; Li, Bing; Song, Jiagui; Yu, Miao; Zhang, Jing; Chen, Ceshi; Zhan, Jun; Zhang, Hongquan

doi:10.1038/s41419-023-06184-2

Download PDF

Article
Open access
Published: 13 October 2023

Kindlin-2 in myoepithelium controls luminal progenitor commitment to alveoli in mouse mammary gland

Cell Death & Disease volume 14, Article number: 675 (2023) Cite this article

1376 Accesses
1 Altmetric
Metrics details

Subjects

Abstract

Myoepithelium plays an important role in mammary gland development, but less is known about the molecular mechanism underlying how myoepithelium controls acinus differentiation during gestation. Herein, we found that loss of Kindlin-2 in myoepithelial cells impaired mammary morphogenesis, alveologenesis, and lactation. Using five genetically modified mouse lines combined with single-cell RNA sequencing, we found a Kindlin-2–Stat3–Dll1 signaling cascade in myoepithelial cells that inactivates Notch signaling in luminal cells and consequently drives luminal progenitor commitment to alveolar cells identity. Single-cell profiling revealed that Kindlin-2 loss significantly reduces the proportion of matured alveolar cells. Mechanistically, Kindlin-2 depletion in myoepithelial cells promotes Stat3 activation and upregulates Dll1, which activates the Notch pathway in luminal cells and inhibits luminal progenitor differentiation and maturation during gestation. Inhibition of Notch1 with tangeretin allowed luminal progenitors to regain commitment ability in the pregnant mice with Kindlin-2 depletion in myoepithelium. Taken together, we demonstrated that Kindlin-2 is essential to myoepithelium-controlled luminal progenitors to alveoli transition during gestation.

VANGL2 regulates luminal epithelial organization and cell turnover in the mammary gland

Article Open access 08 May 2019

Transcription factor FoxO1 regulates myoepithelial cell diversity and growth

Article Open access 11 January 2024

Role of Snai2 and Notch signaling in salivary gland myoepithelial cell fate

Article 09 July 2022

Introduction

Kindlin-2, encoded by the gene FERMT2, is an important integrin-binding protein that is highly expressed in mesoderm-derived organs, including smooth muscle, heart, blood vessel, and bone, but has no or weak expression in endoderm- and ectoderm-derived organs [1]. Global ablation of Kindlin-2 results in peri-implantation lethality at embryonic development day 7.5 (E7.5) in mice [2, 3]. Specific loss of Kindlin-2 impairs smooth muscle formation during embryonic development through the induction of apoptosis. Loss of Kindlin-2 in smooth muscle impairs the contraction of adult smooth muscle by blocking Ca²⁺ influx, which ultimately leads to intestinal obstruction [4]. Kindlin-2 also acts as a critical regulator of myocyte elongation [5] and vascular maintenance and branching [6]. Overexpression of Kindlin-2 in mammary luminal epithelium can promote the growth of gland ductal trees and induce tumor formation [7]. However, the specific role of Kindlin-2 in mammary glands has not been fully elucidated.

Breastfeeding, the normative feeding process for infants, is a public health issue, as it improves the survival of infants and the health of mothers [8, 9]. However, breastfeeding rates are lower than recommended and 5–15% of women suffer from lactation failure worldwide [10]. Development of the mammary gland occurs mainly after birth, including ductal morphogenesis during puberty and acinar morphogenesis during pregnancy [11]. During puberty, morphogenesis is characterized by ductal elongation and branching to form an elaborate ductal tree comprising an inner layer of luminal cells and an outer layer of myoepithelial cells adjacent to the basement membrane. During pregnancy, the mammary epithelium undergoes massive expansion, with the formation of alveolar structures that terminally differentiate into secretory units in late pregnancy to enable lactation. Following weaning, these alveolar structures regress through programmed cell death, returning the gland to a virgin-like state [12].

The Notch signaling pathway is widely reported to participate in mammary stem cell maintenance, cell fate specification, and differentiation during mammary gland development and maturation [13]. A clonal tracing study revealed that Notch1 activation in mammary stem cells is essential for promoting luminal cell specification and driving a progressive transition into unipotent estrogen receptor-negative luminal progenitors [14]. During the final trimester of pregnancy and lactation, Notch pathway activation appears to be reduced, including both the distributions of receptors and ligands in mammary epithelial cells and expression of target genes [15, 16]. These findings demonstrate the dynamic regulation of the Notch pathway in luminal cells, but the mechanisms through which these ligands of the Notch pathway are regulated require further investigation.

In this study, using single-cell profiling in combination with multiple genetically modified mouse models, we demonstrate that Kindlin-2 is an essential regulatory element in the myoepithelium that maintains a suitable maturation niche for the luminal epithelium of the mammary gland. We describe a novel mechanism through which Kindlin-2 plays an indispensable role in the fate determination of luminal and alveolar progenitors in myoepithelial cells via regulation of the Jak2/Stat3/Dll1 axis and Notch pathway in alveolar cells.

Results

Mammary epithelial-specific knockout of Kindlin-2 inhibits lactation in female mice

We first acquired a report of the single-cell landscape of the mammary gland during puberty, gestation, lactation, and the involution stage to determine the physiological distribution of Kindlin-2 in mammary glands [17]. Interestingly, Kindlin-2 was mainly expressed in basal and myoepithelial cells, consistent with the important function of Kindlin-2 in smooth muscle tissues and basal-like breast cancer (Fig. 1A).

**Fig. 1: Mammary epithelial-specific knockout of Kindlin-2 inhibits lactation of female mice.**

To fully explore the function of Kindlin-2 in the mammary gland, we generated myoepithelial cell-specific (K14-Cre) and mammary luminal cell-specific (MMTV-Cre) Kindlin-2 gene knockout mice using the Cre-LoxP system (Supplementary Fig. 1A). The K14-Cre + ; Kindlin-2^flox/flox genotype was distinguished through a polymerase chain reaction (PCR) assay (Supplementary Fig. 1B). Western blotting showed that the level of Kindlin-2 was reduced in breast tissues but not kidney or liver tissues (Supplementary Fig. 1C). Immunohistochemistry demonstrated a reduction of Kindlin-2 in the basal layer of mammary ducts in the K14-Cre + ; Kindlin-2^flox/flox group (Supplementary Fig. 1D). Due to the difficulty of distinguishing myoepithelial cells in the immunohistochemistry assay, we applied multi-color immunofluorescence staining to explore the localization and expression of Kindlin-2 in myoepithelial cells of wild-type and K14-Cre + ; Kindlin-2^flox/flox adult mice (Fig. 1B). Consistent with the previously reported single-cell landscape, Kindlin-2 showed higher expression levels in myoepithelial cells, which were marked with cytokeratin 5 (CK5), of the Kindlin-2^flox/flox group and was apparently knocked out in the K14-Cre + ; Kindlin-2^flox/flox group (Fig. 1C). The genotype of MMTV-Cre could be distinguished based on mouse coat color, with a yellow coat representing the MMTV-Cre + ; Kindlin-2^flox/flox genotype and a black coat representing the Kindlin-2^flox/flox genotype (Supplementary Fig. 1E). In addition, the genotype was characterized using PCR (Supplementary Fig. 1F). To confirm the specificity and efficiency of MMTV promoter-based Cre expression, we detected the levels of Kindlin-2 in the lung, breast, and kidney. Western blotting showed that the level of Kindlin-2 was significantly down-regulated in breast tissue but had no change in the other tissues tested (Supplementary Fig. 1G). Immunohistochemistry results supported the down-regulation of Kindlin-2 in mammary luminal cells of the MMTV-Cre + ; Kindlin-2^flox/flox group (Supplementary Fig. 1H). These results demonstrated that Kindlin-2 knockout mouse models for mammary myoepithelial or luminal cells were successfully generated.

Furthermore, we monitored the morphology of the mammary gland in K14-Cre + ; Kindlin-2^flox/flox, MMTV-Cre + ; Kindlin-2^flox/flox, and Kindlin-2^flox/flox female virgin mice for 4 months, and whole-mount staining with carmine showed that the number of tertiary branches was significantly reduced in both the K14-Cre + ; Kindlin-2^flox/flox group and MMTV-Cre + ; Kindlin-2^flox/flox group compared with the Kindlin-2^flox/flox group (Supplementary Fig. 1I, J). These results suggest that depletion of Kindlin-2 impairs morphogenesis of the mammary gland.

Analysis of mammary glands during pregnancy in either K14-Cre + ; Kindlin-2^flox/flox or MMTV-Cre + ; Kindlin-2^flox/flox females indicated an important role of Kindlin-2 in alveolar expansion. The formation of alveolar units was significantly compromised in Kindlin-2-deficient females at 18.5 days of pregnancy in the K14-Cre + ; Kindlin-2^flox/flox group, as reflected by the presence of small, sparsely distributed alveoli (Fig. 1D, E). The number of mammary alveoli and the number of cells that make up mammary alveoli were significantly reduced in K14-Cre + ; Kindlin-2^flox/flox mice (Fig. 1F). Similar to the K14-Cre + ; Kindlin-2^flox/flox group, the formation of alveolar units was significantly compromised in Kindlin-2-deficient females at 18.5 days of pregnancy in the MMTV-Cre + ; Kindlin-2^flox/flox group (Fig. 1G–I). These results indicate that loss of Kindlin-2 affects alveologenesis in the mouse mammary gland during gestation.

The ability to feed pups is one of the most important indicators for evaluating mammary function. We found that all pups produced by K14-Cre + ; Kindlin-2^flox/flox female mice died within 9 days, whereas pups produced by Kindlin-2^flox/flox and MMTV-Cre + ; Kindlin-2^flox/flox survived (Fig. 1J). These findings indicated that K14-Cre + ; Kindlin-2^flox/flox dams were completely incapable of nursing their pups, whereas MMTV-Cre + ; Kindlin-2^flox/flox and Kindlin-2^flox/flox dams could feed their pups until ablactation.

Examination of K14-Cre + ; Kindlin-2^flox/flox and Kindlin-2^flox/flox dams revealed that pups nursed by K14-Cre + ; Kindlin-2^flox/flox dams were significantly stunted, with little milk present in their stomachs (Fig. 1K). Severely affected mice began dying on the day of birth and attempting to supplement the dams with nutrition did not improve the condition of their pups (Fig. 1K). We found that the mice in the experimental group of the same age were significantly smaller than mice in the control group at 8 days after birth (Fig. 1L). Average weight was significantly reduced in pups nursed by K14-Cre + ; Kindlin-2^flox/flox dams compared to Kindlin-2^flox/flox dams at the early stage of lactation (Fig. 1M). These results suggest that loss of Kindlin-2 in myoepithelial cells catastrophically impairs lactation.

Examination of MMTV-Cre + ; Kindlin-2^flox/flox and Kindlin-2^flox/flox dams revealed that pups nursed by MMTV-Cre + ; Kindlin-2^flox/flox dams were relatively stunted, with a little milk present in their stomachs (Fig. 1N). Average weight was significantly lower in pups nursed by MMTV-Cre + ; Kindlin-2^flox/flox dams compared to Kindlin-2^flox/flox dams, especially after 10 days of nursing (Fig. 1O). To exclude effects of pup genotype, pups of MMTV-Cre + ; Kindlin-2^flox/flox were nursed by Kindlin-2^flox/flox dams beginning at 14 days after parturition, and the results showed that pups nursed by Kindlin-2^flox/flox dams regained normal weight (Supplementary Fig. 1K). In contrast with pups fed by K14-Cre + ; Kindlin-2^flox/flox dams, none of the pups fed by MMTV-Cre + ; Kindlin-2^flox/flox dams starved to death, with most instead showing malnutrition relative to the Kindlin-2^flox/flox group.

Based on the distribution of Kindlin-2 and the difference in nursing ability among epithelium-specific knockout mouse models, we proposed that Kindlin-2 in the myoepithelium plays an essential role in controlling the development and lactation function of the mammary gland.

Single-cell RNA-seq reveals the impact of Kindlin-2 depletion on the mammary gland microenvironment during the final trimester of pregnancy

To reveal the mechanism through which knockout of Kindlin-2 in myoepithelial cells induced disorders during gestation and lactation, we isolated a total of six mammary glands from K14-Cre + ; Kindlin-2^flox/flox, MMTV-Cre + ; Kindlin-2^flox/flox, and Kindlin-2^flox/flox mice at 18.5 days of gestation for single-cell RNA-seq analysis (Fig. 2A). After low-quality cells were filtered out through Seurat analysis, 15,719 high-quality single-cell transcriptomes were retained and profiled, and an average of 2120 genes and 11,385 counts per sample were detected (Supplementary Fig. 2A) [18]. We reduced dimensionality through principal component analysis and visualized the data using t-distributed stochastic neighbor embedding. Cells with similar expression patterns were clustered using a graph-based clustering method (Supplementary Fig. 2B). Using marker genes for major cell lineages, cells were annotated as epithelial cells, immune cells, and stromal cells (Fig. 2B and Supplementary Fig. 2C).

**Fig. 2: Single cell RNAs-seq reveals the impactions of loss of Kindlin-2 in mammary gland microenvironment during late trimester of pregnancy.**

Based on differential gene expression analysis and known cell type references, cells were divided into four immune compartments (T/NK cells, dendritic cells, macrophages, and monocytes), a stromal compartment (fibroblasts), and six epithelial compartments (alveolar differentiated cells, alveolar progenitors, luminal progenitors, hormone-sensing luminal cells, myoepithelial cells, and endothelial cells). We found that the mammary glands of different groups showed distinct distributions of cell lineages (Fig. 2C). The most highly expressed genes are shown in Supplementary Fig. 2D, E. Aside from myoepithelial cells, mammary epithelial cells displayed a similar expression pattern of a gradual shift from immature luminal progenitors to alveolar differentiated cells fully differentiated for lactation (Fig. 2D). Based on the distribution of cell lineages, we determined that K14-Cre + ; Kindlin-2^flox/flox mammary glands contained few mature alveolar cells and fewer progenitor cells compared to other groups. This disproportionate cell type distribution was improved in the MMTV-Cre + ; Kindlin-2^flox/flox mammary gland. This change in the cell distribution is in accordance with our histological observations and estimates of feeding capacity among the three groups.

To explore the function of Kindlin-2 in mammary epithelial maturation and differentiation in detail, we collected mammary luminal epithelial cells and took a census of cell lineages (Fig. 2E). We also inspected the expression levels of Kindlin-2 in myoepithelial cells and luminal cells. Kindlin-2 was accurately knocked out in the K14-Cre + ; Kindlin-2^flox/flox myoepithelial cells, while the MMTV-Cre + ; Kindlin-2^flox/flox group showed no significant difference in luminal epithelium compared with other groups due to the relatively low Kindlin-2 expression levels typical of luminal cells (Supplementary Fig. 3A).

Monocle 2 was used to construct a pseudotime map [19,20,21], which revealed the trajectory of the luminal cell maturation process (Fig. 2F). Mammary luminal epithelial cells were divided into five states with two branch points (Supplementary Fig. 3B). Cells from luminal progenitors were mainly distributed in state 4, one of the initial branches along the pseudotime trajectory, while hormone-sensing luminal cells settled in state 3, the other early branch. Alveolar differentiated cells were located in state 1, at the terminal of the pseudotime trajectory, and alveolar progenitors were continuously situated along the branch from luminal progenitors to alveolar differentiated cells (Fig. 2G and Supplementary Fig. 3C). The trajectory we constructed was consistent with the reported physiological differentiation trajectory. We also observed that the expression levels of milk-related genes (Cd36, Csn2, Fabp3, Fasn, Lpl, and Olah) increased gradually with maturation and terminal differentiation (Fig. 2H) [22]. Mapping group information to the trajectory, we observed that luminal cells in the K14-Cre + ; Kindlin-2^flox/flox group were mainly located along the two early branches, while cells from MMTV-Cre + ; Kindlin-2^flox/flox and Kindlin-2^flox/flox dominated the terminal development trajectory (Fig. 2I and Supplementary Fig. 3D).

To verify whether our luminal cell differentiation trajectory was in accordance with previous research, we integrated our dataset with reports of mammary epithelial cells, covering mammary glands at the virgin, pregnancy, lactation, and involution stages (Fig. 2J) [17]. The integrated results covered the processes of luminal cell differentiation, maturation, and involution throughout the reproductive cycle (Fig. 2K). Our dataset complements the terminal differentiation of luminal cells during late gestation (Supplementary Fig. 3E). Combined with the data presented above, the findings indicate that loss of Kindlin-2 in myoepithelial cells impedes the terminal differentiation and maturation of luminal progenitors (Fig. 2L).

Myoepithelium-specific knockout of Kindlin-2 activates the Notch pathway in the luminal epithelium

Mammary gland development and lactation require complex and dynamic stimulation from intercellular signals, intracellular regulatory processes, and the extracellular matrix. Intercellular communication is important for maintaining the balance of the mammary gland microenvironment and disequilibrium of intercellular communication might induce dysfunction. We used CellChat to investigate intercellular communication disparities between K14-Cre + ; Kindlin-2^flox/flox and Kindlin-2^flox/flox mouse mammary epithelial cells (Supplementary Fig. 4A, B) [23]. Some signaling pathways, including Notch, Fgf, and Ptn, were generally enriched in the K14-Cre + ; Kindlin-2^flox/flox group, while other pathways such as Lifr and Tgf-beta were mainly enriched in the Kindlin-2^flox/flox group (Supplementary Fig. 4C).

Knockout of Kindlin-2 in myoepithelial cells strikingly increased activity of the Notch pathway, a canonical pathway mediating the proliferation and stemness of mammary luminal cells, as a result of the elevation of Dll1, a classical Notch pathway ligand (Fig. 3A, B and Supplementary Fig. 4D-F) [13, 24, 25]. The major senders of the Notch signal were myoepithelial cells and the receivers included all types of immature luminal cells (Fig. 3B). We checked other pairs of ligands and receptors associated with the Notch pathway to verify that Dll1 and Notch1 were the only ligand and receptor regulated in our experiment (Supplementary Fig. 4D). We also estimated the expression of Dll1 and Notch1 in the mammary gland throughout the reproductive cycle and found that Dll1 was mainly expressed in the basal and myoepithelial cells, while Notch1 was primarily expressed in immature progenitors, which exhibited lower levels of canonical lactation-related markers (Fig. 3C), echoing previous research [16].

**Fig. 3: Myoepithelial cells specific knockout of Kindlin-2 activates Notch signaling pathway in P18.5 mammary gland by upregulating Dll1 in myoepithelium and Notch1 in luminal epithelium.**

To verify the alterations in the Notch pathway induced by loss of Kindlin-2 in myoepithelial cells, we employed multi-color immunofluorescence staining to compare the localization and expression of Notch1 and Dll1 between myoepithelial cells, marked with CK5, and luminal cells, marked with CK8, in 18.5-day pregnant Kindlin-2^flox/flox and K14-Cre + ; Kindlin-2^flox/flox mice (Fig. 3D). Protein levels of Dll1 were significantly upregulated in the myoepithelium of the K14-Cre + ; Kindlin-2^flox/flox group, and Notch1 was more highly expressed in luminal cells compared with the Kindlin-2^flox/flox group (Fig. 3E–G). Similar results were obtained in mammary alveoli (Supplementary Fig. 5A, B). Using immunohistochemistry assays, we also found that the target gene of the Notch pathway, Hes1, was elevated in luminal cells of K14-Cre + ; Kindlin-2^flox/flox mice (Supplementary Fig. 4G).

Consistent with the single-cell RNA-seq results, we confirmed that the mammary gland of the K14-Cre + ; Kindlin-2^flox/flox group exhibited strong Notch pathway communication and activation during the final trimester of pregnancy, with high expression levels of the ligand in myoepithelial cells and luminal cells remaining in a poorly differentiated state with high levels of Notch1 expression.

Myoepithelium-specific Kindlin-2 knock-in inhibits Notch pathway activation in the luminal epithelium

Based on our initial results, we observed that depletion of Kindlin-2 in myoepithelial cells significantly impacts mammary gland development and maturation. To clarify whether overexpression of Kindlin-2 in myoepithelial cells boosts mammary gland development, we constructed K14 Cre + ; Kindlin-2^LSL/LSL mice, which expressed higher levels of Kindlin-2 in myoepithelial cells than Kindlin-2^LSL/LSL (Supplementary Fig. 6A–C). We detected mammary gland morphology in K14 Cre + ; Kindlin-2^LSL/LSL and Kindlin-2^LSL/LSL female virgin mice at 5 months of age, and whole-mount carmine staining showed that the number of tertiary branches was greater in the K14 Cre + ; Kindlin-2^LSL/LSL group than the Kindlin-2^LSL/LSL group (Supplementary Fig. 6D, E). We calculated the average intensity of Dll1, Notch1 and Hes1 expression in certain unit cells and found that the expression levels of Dll1 in myoepithelial cells and of Notch1 and Hes1 in luminal epithelial cells were apparently reduced in mammary glands of the K14 Cre + ; Kindlin-2^LSL/LSL group (Fig. 4A, B). Meanwhile, the relative wall thickness of conduits was greater in the K14 Cre + ; Kindlin-2^LSL/LSL group (Fig. 4C). We obtained insights into the development of mammary glands during pregnancy, and demonstrated that the number of mammary alveoli and the number of cells that make up mammary alveoli were similar in the K14 Cre + ; Kindlin-2^LSL/LSL group and the Kindlin-2^LSL/LSL group; this result revealed that overloading of Kindlin-2 in the myoepithelium did not significantly promote alveologenesis. These findings indicate that overexpression of Kindlin-2 inhibits Dll1/Notch1 cross-talk between myoepithelial cells and luminal cells, thereby enhancing mammary duct development. However, in maturation, the intrinsic mechanism of Kindlin-2 is to safeguard the development of the mammary gland during gestation. The formation of alveoli showed no distinct difference between the K14 Cre + ; Kindlin-2^LSL/LSL group and the K14 Cre-; Kindlin-2^LSL/LSL group (Fig. 4D–F).

**Fig. 4: Myoepithelium specific Kindlin-2 knock-in inhibits the activation of the Notch signaling pathway in mammary glandular ducts.**

Kindlin-2 interacts with and inhibits the activation of Stat3

To identify the key molecular mechanisms underlying the function of Kindlin-2 in the mammary gland, we compared transformation of myoepithelial cells between the K14-Cre + ; Kindlin-2^flox/flox and Kindlin-2^flox/flox groups. We analyzed differentially expressed genes between these two groups (Fig. 5A) but did not directly identify factors explaining the regulatory process involving Kindlin-2 and Dll1. Kindlin-2 plays a role as an intracellular scaffold in multiple cellular processes [26, 27]; therefore, we examined whether Kindlin-2 could directly bind to and impact transcription factors to influence Dll1 expression. We used DoRothEA software to estimate transcription factor activities (Fig. 5B). In total, 96 transcription factors had activities that were higher in the myoepithelial cells of K14-Cre + ; Kindlin-2^flox/flox mice compared with MMTV-Cre + ; Kindlin-2^flox/flox and Kindlin-2^flox/flox mice [28]. Combining these findings with the results described above, two of those 96 transcription factors, namely Stat3 and Rela, may possess the ability to bind Kindlin-2 (Fig. 5C) [29].

**Fig. 5: Kindlin-2 interacts with and inactivates STAT3 by competitively blocking JAK2/STAT3 interaction.**

Stat3 is one of a few reported transcription factors that directly regulate the expression of Dll1 [30]. Previous research indicated that Jak2–Stat3 activation was essential for Dll1 expression in neural precursor cells; in this process, Stat3 directly binds to the promoter of Dll1, enhancing its transcription. As the same FRMD domain is present in Kindlin-2 and the Jak family [31, 32], we speculated that Kindlin-2 might regulate the phosphorylation and activation of Stat3 by disrupting the interaction and catalytic reaction of Jak and Stat3. We checked the expression levels of Stat3 canonical target genes, such as Irf1, Junb, Socs3, Fos, and Hspb1, and found that they were significantly higher in myoepithelium of K14-Cre + ; Kindlin-2^flox/flox mammary gland, which is consisted with previous analysis of transcription factor activity (Fig. 5D). We confirmed that endogenous Kindlin-2 could interact with Stat3 in MDA-MB-231 and HEK293T cells (Fig. 5E, F). However, knockdown of Kindlin-2 promoted Stat3 phosphorylation at the Y705 and S727 sites, both of which are important for the activation of Stat3 in MDA-MB-231 cells (Fig. 5G), but not in HEK293T cells (Supplementary Fig. 7A, B). We assume that the regulatory interaction between Kindlin-2 and Stat3 may occur mainly in myoepithelial and basal cells, consistent with previous analysis that knockout of Kindlin-2 in MMTV-Cre + ; Kindlin-2^flox/flox mice did not interrupt Stat3 activation. Deletion of Kindlin-2 enhanced the amount of Stat3 co-immunoprecipitated with anti-JAK2 antibody in MDA-MB-231 cells (Fig. 5H).

To determine whether depletion of Kindlin-2 promotes Stat3 phosphorylation and expression of Dll1 through a common mechanism in smooth muscle tissue, we obtained liver, kidney, colorectal, and skeletal muscle tissue from Kindlin-2 knockout-first (KO-First), and WT mice. In the colorectal tissue, which contains abundant smooth muscle in the muscular layer, the levels of Dll1 expression and Stat3 phosphorylation were significantly elevated in Kindlin-2 knockout-first (KO-First) group (Fig. 5I). Furthermore, we detected Stat3 Y705 phosphorylation and Dll1 in Kindlin-2^fl/fl; MYH-cre+ mice, whose Kindlin-2 was specifically knocked out in smooth muscle upon tamoxifen injection. Both Stat3 phosphorylation sites and Dll1 expression showed higher levels in colorectal tissue of Kindlin-2^fl/fl; MYH-cre+ mice (Fig. 5J).

Loss of Kindlin-2 activates Stat3 and upregulates Dll1 in the myoepithelium

To confirm the regulatory interaction between Kindlin-2 and Stat3 in the mammary gland microenvironment in situ, we used multi-color immunofluorescence staining to qualify the phosphorylation levels of Stat3 in myoepithelial cells of the K14-Cre + ; Kindlin-2^flox/flox and Kindlin-2^flox/flox groups. When marked with CK5, myoepithelial cells exhibit a relatively high expression level of Stat3, with significantly elevated phosphorylation levels at the Y705 and S727 sites (Fig. 6A–D and Supplementary Fig. 5C, D). Interestingly, the phosphorylation levels of Y705 and S727 showed a consistent correlation with Dll1 expression, especially in myoepithelial cells from the K14-Cre + ; Kindlin-2^flox/flox group (Fig. 6E). These results demonstrate that loss of Kindlin-2 in myoepithelial cells induced the phosphorylation of Stat3, and the phosphorylated Stat3 functioned as an activated transcription factor to induce the expression of Dll1 and thus regulate Notch pathway activation in luminal cells in situ.

**Fig. 6: Loss of Kindlin-2 activates STAT3 and upregulates Dll1 in myoepithelium.**

From these results, we determined that Kindlin-2 is an essential regulator in myoepithelial cells that controls the maturation of luminal cells through the Jak2–Stat3–Dll1–Notch1 axis.

Tangeretin treatment improves the lactation function of K14-Cre + ; Kindlin-2^flox/flox female mice

Absence of Kindlin-2 in the myoepithelium activates Stat3 and sharply increases communication along the Notch pathway between myoepithelial cells and luminal progenitors. Compared with the extensive distribution of phosphorylated Stat3, the relatively precise expression of Notch1 in luminal and alveolar progenitor cells provides a better potential drug target. Herein, we assumed that targeting Notch activation with an effective Notch1 inhibitor, tangeretin [33, 34], would rescue the dysfunction and empower differentiation of the naïve luminal epithelium. We divided the pregnant female mice of Kindlin-2^flox/flox and K14-Cre + ; Kindlin-2^flox/flox into four groups: (1) K14-Cre + ; Kindlin-2^flox/flox pregnant mice treated with solvent beginning at E13.5; (2) K14-Cre + ; Kindlin-2^flox/flox pregnant mice treated with tangeretin at 50 mg/kg/day beginning at E13.5; (3) K14-Cre-; Kindlin-2^flox/flox pregnant mice treated with solvent beginning at E13.5; and (4) K14-Cre-; Kindlin-2^flox/flox pregnant mice treated with tangeretin at 50 mg/kg/day beginning at E13.5 (Fig. 7A). Consistent with our hypothesis, treatment with tangeretin partially restored the nursing ability of K14-Cre + ; Kindlin-2^flox/flox dams (Fig. 7B). The histology of the mammary gland from those four groups showed that alveologenesis was partly rescued with tangeretin treatment in the K14-Cre + ; Kindlin-2^flox/flox group (Fig. 7C–E). To verify that tangeretin inhibits luminal epithelium Notch pathway, we detected the expression levels of Notch1, Hes1, and Hey1 in four group (Fig. 7F–G). The Immunohistochemical staining showed that Notch pathway was significantly inhibited in the K14-Cre + ; Kindlin-2^flox/flox group upon treatment with tangeretin.

**Fig. 7: Inhibitor of Notch signaling pathway tangeretin treatment can improve the lactation function of K14Cre + ; Kindlin-2^flox/flox female mice.**

These results suggested that tangeretin treatment blocked abnormal activation of the Notch pathway in luminal cells induced by loss of Kindlin-2 in myoepithelial cells, supporting luminal epithelium function in lactation.

Discussion

Researchers have made great efforts to reconstruct the trajectory of mammary gland development and its disequilibrium associated with cancer [35,36,37,38]. Most research has focused on the physiological or pathological state of the luminal epithelium, which is considered the most important factor affecting mammary gland function. With deepening research into the mammary gland, more attention is being paid to its microenvironment, including cross-talk among immune cells, stromal cells, epithelial cells, and the extracellular matrix, rather than focusing solely on luminal epithelial cells, by combining a fairly comprehensive model with high-resolution methods [39]. In this study, we demonstrate that Kindlin-2 is a critical regulator located in the myoepithelium that prevents the phosphorylation of Stat3 and expression of Dll1 to preserve the order of luminal progenitor maturation during gestation (Fig. 8A). Our findings also indicate that the regulatory axis of Kindlin-2–Stat3–Dll1 is common to smooth muscle-enriched tissues. Through targeting Notch1 with tangeretin, we partially reversed the retardation of mammary luminal cells caused by the loss of Kindlin-2 in myoepithelial cells.

Myoepithelium is a specialized muscle-like differentiated epithelium [40, 41]. Many difficulties arise in investigations of the cross-talk between the myoepithelium and luminal epithelium. One such difficulty is that highly terminal differentiated myoepithelial cells are present in low proportions and have no stable cell lines. Most research has been performed with genetically modified mouse models to capture in vivo mechanisms [42, 43]. However, many myoepithelium-specific mechanisms can only be revealed at single-cell resolution due to the distinct histological pattern of the myoepithelium, a layer of cells that is very thin and clings to luminal epithelial cells. Therefore, previous research using bulk sequencing or total lysis might have been contaminated with other cells, concealing key mechanisms. Using single-cell RNA-seq, we successfully determined the mechanism through which loss of Kindlin-2 induces overactivation of Stat3, a transcription factor that is reported to regulate Dll1 expression, and demonstrated that Kindlin-2 could impact the activation of Stat3 through binding with Jak2 and Stat3 in the basal-like cell line MDA-MB-231. We also used multiple-color staining to demonstrate the impact of Stat3 and Dll1 after Kindlin-2 depletion in myoepithelial cells in situ and confirmed this mechanism in smooth muscle tissues of Kindlin-2 knockout mice. The important role of Kindlin-2 in Jak2–Stat3 signaling is specific to smooth muscle and muscle-like cell lineages and is not present in the luminal cells, epithelial cells (HEK293T), or other tissues with a low abundance of smooth muscle.

Numerous questions remain to be resolved, such as how the dynamic regulation of both Notch pathway ligands and receptors occurs throughout the development and pregnancy cycle, as signaling activation does not affect only the recipient cells. Deficiency of Dll1 in basal cells leads to a more complete block of development of mammary ducts, and the expression of Dll1 in mammary gland stem cells (MaSCs) can activate the Notch pathway in macrophages, which promote the stemness of MaSCs through Wnt signaling [44]. However, our data did not show this type of regulation between myoepithelial cells and macrophages, which might be due to the specific focal periods of this study. Aside from Dll1, the Notch ligand Jag1 in basal cells can also receive Robo1 and β-catenin signaling regulation, mediating luminal progenitor maturation through the Notch pathway [45]. Our findings suggest that at 14.5 days of pregnancy, the luminal epithelium produces alveolar progenitors with Notch1 expression, and thus we applied tangeretin in dams since the later trimester of pregnancy.

During mammary gland development and maturation, the main process known to regulate mammary duct development and alveologenesis is the variation of several hormones, including estrogen, progestin, and prolactin [46]. Previous research reported that a high level of estrogen could protect Dll1 from the ubiquitin proteasomal degradation pathway in breast cancer [47]. The expression levels of receptors and ligands of the Notch signaling pathway changed dynamically and were consistent with the levels of progesterone and estradiol in the mouse oviduct during the estrous cycle [48]. Another study reported that estrogen could improve the proliferation and differentiation of bone marrow stem cells through the Notch signaling pathway [49]. Whether a similar regulatory network exists between hormones and the Notch pathway during mammary gland development and the detailed regulation mechanisms of Notch signaling ligands and receptors during the hormonal cycle remain to be resolved.

Activation of Stat3 in the luminal epithelium is associated with cell death and modulates the inflammatory microenvironment during involution [50]. As a common inflammatory disease during lactation, mastitis reduces milk supply. Previous research noted that the activation Il6–Jak2–Stat3 axis is associated with plasma cell mastitis [51]. Our study further elucidates the function of Stat3 in Notch pathway regulation. Whether the activation of Stat3 in myoepithelium, triggered by Il6 and interferon, in mothers with mastitis, is the etiology of lactation failure via cross-cell signal transduction requires further investigation.

Based on the efficacy of tangeretin treatment in our research, we successfully constructed a mouse model of lactation disorder driven by abnormal activation of the Notch signaling pathway, which can be broadly applied in further molecular mechanism and drug discovery research. The application of tangeretin might benefit mothers who are unable to feed their infants due to abnormality of the Notch pathway and Jak2–Stat3 signaling.

Fully demonstrating the regulation of intercellular signaling and intracellular transduction in mammary gland development and cancer is a worthwhile topic for future research. Further investigations are required to determine whether the regulatory interactions between Kindlin-2 and the Jak2–Stat3–Dll1 axis are common across cell lineages with high levels of Kindlin-2, such as myoepithelial cells, basal cells, and fibroblast cells. Whether the complex regulatory processes among Kindlin-2, the Jak–Stat signaling pathway (related to inflammation), and the Notch signaling pathway (associated with cell stemness and fate differentiation) are involved in breast-related diseases such as mastitis and breast cancer is another important topic of further investigation.

Methods

Mice and genotyping

All animal experiments were approved by the Peking University Animal Care and Use Committee. We bought Kindlin-2 knockout-first mice (ID: 38007, C57BL/6 genetic backgroud) from International Knockout Mouse Consortium (IKMC). Our lab developed the Kindlin-2 floxed mice. LoxP sites flanked the fifth and sixth exons of FERMT2. We bought Kindlin-2^LSL/LSL transgenic mice from Shanghai Model Organisms Center, Inc. MMTV-Cre+ transgenic mice were bought from Nanjing Biomedical Research Institute of Nanjing University. K14-Cre+ mice were provided by Zhengquan Yu lab in China Agricultural University. Kindlin-2 floxed (Kindlin-2^flox/flox) mice were crossed with K14-Cre+ and MMTV-Cre+ mice to generate Kindlin-2 heterozygous mice (K14-Cre + ; Kindlin-2^flox/wt and MMTV-Cre + ; Kindlin-2^flox/wt), which were further backcrossed with Kindlin-2^flox/flox mice to generate mammary myoepithelial-specific and luminal epithelium-specific Kindlin-2 knockout mice. Kindlin-2 overexpressed (Kindlin-2^LSL/LSL) mice were crossed with K14-Cre+ to generate Kindlin-2 heterozygous mice (K14-Cre + ; Kindlin-2^LSL/wt), which were further backcrossed with Kindlin-2^LSL/LSL mice to generate mammary myoepithelial-specific Kindlin-2 knock-in mice. Kindlin-2^fl/fl; MYH-cre+ mice were described previously [4]. Genotyping was determined via PCR using primers: Kindlin-2 knockout-first: P1 5ʹ- TACAGGTGGCTGACAAGATCC -3ʹ, P2 5ʹ- GTGAGGCTCACCTTTCAGAGG -3ʹ, P3 5ʹ- CAACGGGTTCTTCTGTTAGTCC -3ʹ; Kindlin-2: forward 5ʹ- TACAGGTGGCTGACAAGATCC -3ʹ, reverse 5ʹ- GTGAGGCTCACCTTTCAGAGG -3ʹ; Cre: forward 5ʹ- ATTTGCCTGCATTACCGGTCG -3ʹ, reverse 5ʹ- CAGCATTGCTGTCACTTGGTC -3ʹ; Kindlin-2^LSL/LSL: P1 5ʹ- TCAGATTCTTTTATAGGG-GACACA -3ʹ, P2 5ʹ- TAAAGGCCACTCAATGCTCACTAA -3ʹ, P3 5ʹ- GGTGTTGTCGGGG-AAATCATCGTC -3ʹ, P4 5ʹ- AGGAGCCTGCCAAGTAAC -3ʹ; MYH-cre: SMWT1 5ʹ- TGACCCCATCTCTTCACTCC -3ʹ, SMWT2 5ʹ- AACTCCACGACCACCTCATC -3ʹ, phCREAS1 5ʹ- AGTCCCTCACATCCTCAGGTT -3ʹ.

Dosing experiment

We divided the pregnant female mice from K14-Cre-; Kindlin-2^flox/flox and K14-Cre + ; Kindlin-2^flox/flox into 4 groups: (1) K14-Cre + ; Kindlin-2^flox/flox pregnant mice were treated with solvent by oral gavage from E13.5 and equivalent solvent were used every day. (2) K14-Cre + ; Kindlin-2^flox/flox pregnant mice were treated with Tangeretin (Bi De Yi Yao, BD33755), a Notch1 inhibitor, 50 mg/kg/day by oral gavage from E13.5 and equivalent drugs were used every day. (3) K14-Cre-; Kindlin-2^flox/flox pregnant mice were treated with solvent by oral gavage from E13.5 and equivalent solvent were used every day. (4) K14-Cre-; Kindlin-2^flox/flox pregnant mice were treated with Tangeretin 50 mg/kg/day by oral gavage from E13.5 and equivalent drugs were used every day.

Cell culture

Human breast cancer (MDA-MB-231) and human embryonic kidney (HEK293T) cell lines were cultured in DMEM medium containing 10% FBS (Invitrogen, Carlsbad, CA, USA), supplemented with 100 units/ml penicillin and 100 mg/ml streptomycin. These cell lines were purchased from ATCC. All cells were cultured at 37 °C in 5% CO₂. According to the manufacturer’s instructions (Invitrogen), 50%–70% of confluent cells were transfected with indicated siRNA using Lipofectamine RNAiMAX for transient transfection. Specific siRNA targeting human Kindlin-2 (Kindlin-2 siRNA) was designed by Guangzhou Rui Bo (China). Kindlin-2si_001: AAGCUGGUGGAGAAACUCG; Kindlin-2si_002: CAGCGAGAAUCUUGGAGGC.

Western blot analysis

Cell lysates or tissues were treated using RIPA lysis buffer supplemented with 1× inhibitor cocktail (Roche) and 1× Roche PhosSTOP phosphatase inhibitor cocktail (Roche). Aliquots of 30 μg protein were separated by SDS-PAGE and transferred to the polyvinylidene fluoride (PVDF) membrane. Membranes were blocked in 5% non-fat powdered milk at room temperature for 1 h and incubated with specific antibodies overnight at 4 °C. Antibodies used for western blot analysis in this study were: rabbit anti-Kindlin-2 (Sigma, K3269, 1:1000 dilution); mouse anti-GAPDH (Zhong Shan Jin Qiao, TA-08, 1:1000 dilution); mouse anti-tubulin (Sigma, clone B-5-1-2, 1:1000 dilution); rabbit anti-DDDDK tag (Abcam, ab205606, 1:1000 dilution); rabbit anti-STAT3 (Abcam, ab68153, 1:1000 dilution); rabbit anti-pSTAT3-S727 (Abcam, ab32143, 1:1000 dilution); rabbit anti-pSTAT3-Y705 (Abcam, ab76315, 1:1000 dilution); rabbit anti-Dll1 (Abcam, ab10554, 1:1000 dilution); rabbit anti-JAK2 (Abcam, ab108596, 1:1000 dilution). After washing three times in TBST buffer, the membranes were incubated with appropriate HRP-conjugated secondary antibodies at room temperature for 1 h. The Super Signal Chemiluminescence kit (Thermo Fisher Scientific) was used to detect immobilized antibodies.

Immunoprecipitation

Immunoprecipitation methods were described previously [52]. MDA-MB-231 and HEK293T cell lysates were treated using RIPA lysis buffer supplemented with 1× inhibitor cocktail (Roche) and 1× Roche PhosSTOP phosphatase inhibitor cocktail (Roche). After centrifugation at 12,000 rpm for 15 min at 4 °C, cell debris was removed. Precleared lysates were incubated with indicated antibodies and normal IgG (as controls) overnight at 4 °C. 40 μl of protein A or protein G agarose (Santa Cruz Biotechnology, Inc.) were incubated with Lysates with rotation at 4 °C for 6 h. The complex was washed with RIPA buffer five times, resuspended with 60 μl 2× loading buffer, and cooked at 100 °C for 8 min, which was detected by western blot analysis.

Histology and immunohistochemistry

Mouse mammary glands were fixed with 4% buffered formalin (Fisher), embedded in paraffin, and sectioned at 5 μm before staining. The prepared sections were stained with immunohistochemistry to detect different proteins or with H&E to examine the structure. Briefly, sections were deparaffinized with xylene and rehydrated with ethanol. 3% hydrogen peroxide was used to block the sections for 15 min at room temperature. After washing in phosphate buffer saline (PBS, pH 7.6) three times, antigen-retrieval was performed by heating slides in 0.01 M citrate buffer (pH 6.0) for 20 min in a 100 °C water bath. When slides cooled at room temperature, using 10% normal goat serum blocked the non-specific binding of antibodies for 20 min. The sections were probed with anti-Kindlin-2(Sigma, K3269, 1:100 dilution), anti-Dll1 (Abcam, ab10554, 1:200 dilution), anti-Hes1 (Abcam, Ab71559, 1:200 dilution), anti-Notch1 (EPITOMICS, Cat#1935-1, 1:200 dilution), anti-STAT3 (Abcam, ab68153, 1:200 dilution); rabbit anti-pSTAT3-Y705 (Abcam, ab76315, 1:100 dilution), anti-CK5 (Abcam, ab52635, 1:200 dilution), or anti-CK8 (Abcam, ab53280, 1:400 dilution) overnight. After washing in PBS three times, the sections were incubated with biotinylated IgG (Zhongshanjinqiao, PV6001) for 60 min at room temperature. After washing in PBS three times, Diaminobenzidine (DAB) was used as a substrate, and hematoxylin was used as a counterstain. A BX51 microscope (Olympus) was used to examine and photograph for immunostainings.

Whole-mounting of mammary glands

Mammary glands were fixed in Carnoy’s solution containing 10% glacial acetic acid, 30% chloroform, and 60% ethanol for at least 24 h at room temperature. The sections were rehydrated with ethanol and then stained with carmine solution (HARVEYBIO, C21940-1G) overnight at room temperature.

Quantitative analysis of tissue section image

Performed multi-color staining according to the kit instructions (TissueGnostics, TGFP550). Using StrataQuest v7.0.158 software (TissueGnostics, Vienna, Austria) performed Immunofluorescence image analysis. Total cells were identified based on the DAPI staining.

Statistical analysis

Using GraphPad Prism 9.0 software performed statistical analysis. All the results are expressed as the mean values ± SEM or mean values ± SD. Using unpaired one- or two-tailed Student’s t tests performed quantitative PCR analysis. Statistical significance was defined as p < 0.05. All data statistics were tested for homogeneity of variance, and then we chose whether to do a nonparametric test or a parametric test.

Tissue dissociation and preparation

The fresh mammary gland tissues were washed with Hanks Balanced Salt Solution (HBSS) for 3 times and minced into 1–2 mm pieces. Then the tissue pieces were digested with 2 ml GEXSCOPE® Tissue Dissociation Solution (Singleron) at 37 °C for 15 min in 15 ml centrifuge tube with sustained agitation. After digestion, using 40-micron sterile strainers to filter the samples and centrifuging the samples at 1000 rpm for 5 min. Then the supernatant was discarded, and the sediment was resuspended in 1 ml PBS (HyClone). To remove the red blood cells, 2 mL GEXSCOPE® red blood cell lysis buffer (Singleron) was added at 25 °C for 10 min. The solution was then centrifuged at 500 × g for 5 min and suspended in PBS. Cells were incubated with 7-AAD (1:1000, Invitrogen) on ice for 30 min, and sorted by flow cytometry to filter dead cells.

Single cell RNA sequencing

Single-cell suspensions with 1 × 10⁵ cells/mL in concentration in PBS (HyClone) were prepared. Single-cell suspensions were then loaded onto microfluidic devices and scRNA-seq libraries were constructed according to Singleron CLindex® protocol by GEXSCOPE® Single-Cell RNA Library Kit (Singleron Biotechnologies). Mixture libraries were diluted to 4 nM and pooled for sequencing. Pools were sequenced on Illumina HiSeq X with 150 bp paired end reads.

scRNA-seq quantifications and statistical analysis

Raw reads were processed and aligned to the genome mm10, to generate gene expression profiles using an internal pipeline. The Seurat program (http://satijalab.org/seurat/, R package, v.4.1.0) was applied for analysis of RNA-Sequencing data.

A total of 16,702 cells were sequenced (5230 from Kindlin-2^flox/flox, 6,052 from MMTV-Cre + ; Kindlin-2^flox/flox and 5,420 from K14-Cre + ; Kindlin-2^flox/flox). First, we removed low-quality cells by modeling mitochondrial to nuclear gene content to <10% and gene numbers higher than 1000 and lower than 4000, as a result 16,261 high quality cells (5,155 from Kindlin-2^flox/flox, 5876 from MMTV-Cre + ; Kindlin-2^flox/flox and 5,230 from K14-Cre + ; Kindlin-2^flox/flox) for downstream analysis. After normalization and scale the expression matrix, we selected 2000 high variable genes for PCA analysis. Cells were separated into 24 clusters, using the top 30 principal components and resolution parameter at 0.8. Further, we integrated our dataset with GSE106273 using Fast integration using reciprocal PCA function in Seurat.

Differentially expressed genes (DEGs) analysis

Genes expressed in more than 50% of the cells in a cluster and with average log (Fold Change) of greater than 1 were selected as DEGs based on Wilcox likelihood-ratio test with default parameters.

Cell type annotation

The main lineage identification of each cluster was determined with the expression of canonical markers of Epithelial cells (Epcam), immune cells (Ptprc) and stroma cells (Col1a1). The cell lineages were determined by SingleR (https://github.com/LTLA/SingleR. v.1.8.1) with MouseRNAseq Data as reference cell type from celldex (http://github.com/LTLA/celldex. v.1.4.0) and detailed cell markers from CellMarker (https://bio-bigdata.hrbmu.edu.cn/CellMarker/).

Trajectory analysis

To map differentiation of cell subtypes during lactation, pseudotime trajectory analysis was performed with Monocle2. For constructing the trajectory, top 400 highly variable genes among 4 types of luminal epithelial cells, and dimension-reduction was performed by DDRTree.

Cell-cell interaction

CellChat was used to perform cell-cell communication calculation and analysis for the mammary epithelial clusters. The CellChat results were used to reveal both incoming communication patterns of target cells and outgoing communication patterns of the secreting cells. To reveal the strength of specific pathways among mammary epithelial cells clusters, we selected the most relevant Notch signaling pathway for further visualization.

Transcription factor regulatory analysis

Activity of the most variable transcription factors in the myoepithelial was determined using the DoRothEA (v1.8.0) algorithm for single-cell RNA-sequencing. Mouse regulons (grades A–C) were obtained from the DoRothEA R package. We extracted the 12 most variable transcription factors over groups. The predicted transcription factors were compared to their corresponding gene expression, and those with low real expression (normalized expression < 0.5) were removed from analysis.

Data availability

The raw RNA-seq data used in this study are available in the Sequence Read Archive. The raw data and processed data of single-cell transcriptomic data generated in this study have been deposited in the Gene Expression Omnibus (GEO) database under accession code GSE222028. The publicly available single cell dataset used in this study are available from the Gene Expression Omnibus (accession numbers GSE106273). All the other data supporting the findings of this study are available within the article and its Supplementary Information files.

Code availability

The code applied in this study is available upon request.

References

Zhan J, Yang M, Chi X, Zhang J, Pei X, Ren C, et al. Kindlin-2 expression in adult tissues correlates with their embryonic origins. Sci China Life Sci. 2014;57:690–7.
Article CAS PubMed Google Scholar
Montanez E, Ussar S, Schifferer M, Bösl M, Zent R, Moser M, et al. Kindlin-2 controls bidirectional signaling of integrins. Genes Dev. 2008;22:1325–30.
Article CAS PubMed PubMed Central Google Scholar
Dowling JJ, Gibbs E, Russell M, Goldman D, Minarcik J, Golden JA, et al. Kindlin-2 is an essential component of intercalated discs and is required for vertebrate cardiac structure and function. Circ Res. 2008;102:423–31.
Article CAS PubMed Google Scholar
He X, Song J, Cai Z, Chi X, Wang Z, Yang D, et al. Kindlin-2 deficiency induces fatal intestinal obstruction in mice. Theranostics. 2020;10:6182–200.
Article CAS PubMed PubMed Central Google Scholar
Dowling JJ, Vreede AP, Kim S, Golden J, Feldman EL. Kindlin-2 is required for myocyte elongation and is essential for myogenesis. BMC Cell Biol. 2008;9:36.
Article PubMed PubMed Central Google Scholar
Rozario T, Mead PE, DeSimone DW. Diverse functions of kindlin/fermitin proteins during embryonic development in Xenopus laevis. Mech Dev. 2014;133:203–17.
Article CAS PubMed PubMed Central Google Scholar
Li B, Chi X, Song J, Tang Y, Du J, He X, et al. Integrin-interacting protein Kindlin-2 induces mammary tumors in transgenic mice. Sci China Life Sci. 2019;62:225–34.
Article CAS PubMed Google Scholar
Mosca F, Gianni ML. Human milk: composition and health benefits. Pediatr Med Chir. 2017;39:155.
Article PubMed Google Scholar
Victora CG, Bahl R, Barros AJ, Franca GV, Horton S, Krasevec J, et al. Breastfeeding in the 21st century: epidemiology, mechanisms, and lifelong effect. Lancet. 2016;387:475–90.
Article PubMed Google Scholar
Marasco LA. Unsolved mysteries of the human mammary gland: defining and redefining the critical questions from the lactation consultant’s perspective. J Mammary Gland Biol Neoplasia. 2014;19:271–88.
Article PubMed Google Scholar
Oakes SR, Gallego-Ortega D, Ormandy CJ. The mammary cellular hierarchy and breast cancer. Cell Mol Life Sci. 2014;71:4301–24.
Article CAS PubMed PubMed Central Google Scholar
Watson CJ, Khaled WT. Mammary development in the embryo and adult: new insights into the journey of morphogenesis and commitment. Development. 2020;147:dev169862.
Article CAS PubMed Google Scholar
Edwards A, Brennan K. Notch signalling in breast development and cancer. Front Cell Dev Biol. 2021;9:692173.
Article PubMed PubMed Central Google Scholar
Lilja AM, Rodilla V, Huyghe M, Hannezo E, Landragin C, Renaud O, et al. Clonal analysis of Notch1-expressing cells reveals the existence of unipotent stem cells that retain long-term plasticity in the embryonic mammary gland. Nat Cell Biol. 2018;20:677–87.
Article CAS PubMed PubMed Central Google Scholar
Bouras T, Pal B, Vaillant F, Harburg G, Asselin-Labat ML, Oakes SR, et al. Notch signaling regulates mammary stem cell function and luminal cell-fate commitment. Cell Stem Cell. 2008;3:429–41.
Article CAS PubMed Google Scholar
Raafat A, Goldhar AS, Klauzinska M, Xu K, Amirjazil I, McCurdy D, et al. Expression of Notch receptors, ligands, and target genes during development of the mouse mammary gland. J Cell Physiol. 2011;226:1940–52.
Article CAS PubMed PubMed Central Google Scholar
Bach K, Pensa S, Grzelak M, Hadfield J, Adams DJ, Marioni JC, et al. Differentiation dynamics of mammary epithelial cells revealed by single-cell RNA sequencing. Nat Commun. 2017;8:2128.
Article PubMed PubMed Central Google Scholar
Hao Y, Hao S, Andersen-Nissen E, Mauck WM 3rd, Zheng S, Butler A, et al. Integrated analysis of multimodal single-cell data. Cell. 2021;184:3573–87.e29.
Article CAS PubMed PubMed Central Google Scholar
Trapnell C, Cacchiarelli D, Grimsby J, Pokharel P, Li S, Morse M, et al. The dynamics and regulators of cell fate decisions are revealed by pseudotemporal ordering of single cells. Nat Biotechnol. 2014;32:381–6.
Article CAS PubMed PubMed Central Google Scholar
Qiu X, Hill A, Packer J, Lin D, Ma YA, Trapnell C. Single-cell mRNA quantification and differential analysis with Census. Nat Methods. 2017;14:309–15.
Article CAS PubMed PubMed Central Google Scholar
Qiu X, Mao Q, Tang Y, Wang L, Chawla R, Pliner HA, et al. Reversed graph embedding resolves complex single-cell trajectories. Nat Methods. 2017;14:979–82.
Article CAS PubMed PubMed Central Google Scholar
Pal B, Chen Y, Milevskiy MJG, Vaillant F, Prokopuk L, Dawson CA, et al. Single cell transcriptome atlas of mouse mammary epithelial cells across development. Breast Cancer Res. 2021;23:69.
Article CAS PubMed PubMed Central Google Scholar
Jin S, Guerrero-Juarez CF, Zhang L, Chang I, Ramos R, Kuan CH, et al. Inference and analysis of cell-cell communication using CellChat. Nat Commun. 2021;12:1088.
Article CAS PubMed PubMed Central Google Scholar
Phoon YP, Chivukula IV, Tsoi YL, Kanatani S, Uhlen P, Kuiper R, et al. Notch activation in the mouse mammary luminal lineage leads to ductal hyperplasia and altered partitioning of luminal cell subtypes. Exp Cell Res. 2020;395:112156.
Article CAS PubMed Google Scholar
Chen W, Wei W, Yu L, Ye Z, Huang F, Zhang L, et al. Mammary development and breast cancer: a notch perspective. J Mammary Gland Biol Neoplasia. 2021;26:309–20.
Article CAS PubMed PubMed Central Google Scholar
Wang W, Kansakar U, Markovic V, Sossey-Alaoui K. Role of Kindlin-2 in cancer progression and metastasis. Ann Transl Med. 2020;8:901.
Article PubMed PubMed Central Google Scholar
Zhan J, Zhang H. Kindlins: Roles in development and cancer progression. Int J Biochem Cell Biol. 2018;98:93–103.
Article CAS PubMed Google Scholar
Garcia-Alonso L, Holland CH, Ibrahim MM, Turei D, Saez-Rodriguez J. Benchmark and integration of resources for the estimation of human transcription factor activities. Genome Res. 2019;29:1363–75.
Article CAS PubMed PubMed Central Google Scholar
Song J, Wang T, Chi X, Wei X, Xu S, Yu M, et al. Kindlin-2 inhibits the hippo signaling pathway by promoting degradation of MOB1. Cell Rep. 2019;29:3664–77.e5.
Article CAS PubMed Google Scholar
Yoshimatsu T, Kawaguchi D, Oishi K, Takeda K, Akira S, Masuyama N, et al. Non-cell-autonomous action of STAT3 in maintenance of neural precursor cells in the mouse neocortex. Development. 2006;133:2553–63.
Article CAS PubMed Google Scholar
Chishti AH, Kim AC, Marfatia SM, Lutchman M, Hanspal M, Jindal H, et al. The FERM domain: a unique module involved in the linkage of cytoplasmic proteins to the membrane. Trends Biochem Sci. 1998;23:281–2.
Article CAS PubMed Google Scholar
Ferrao R, Lupardus PJ. The Janus Kinase (JAK) FERM and SH2 domains: bringing specificity to JAK-receptor interactions. Front Endocrinol (Lausanne). 2017;8:71.
Article PubMed Google Scholar
Xu S, Kong YG, Jiao WE, Yang R, Qiao YL, Xu Y, et al. Tangeretin promotes regulatory T cell differentiation by inhibiting Notch1/Jagged1 signaling in allergic rhinitis. Int Immunopharmacol. 2019;72:402–12.
Article CAS PubMed Google Scholar
Liu LL, Li FH, Zhang Y, Zhang XF, Yang J. Tangeretin has anti-asthmatic effects via regulating PI3K and Notch signaling and modulating Th1/Th2/Th17 cytokine balance in neonatal asthmatic mice. Braz J Med Biol Res. 2017;50:e5991.
Article PubMed PubMed Central Google Scholar
Fu NY, Nolan E, Lindeman GJ, Visvader JE. Stem cells and the differentiation hierarchy in mammary gland development. Physiol Rev. 2020;100:489–523.
Article CAS PubMed Google Scholar
Ogra PL. Immunology of human milk and lactation: historical overview. Nestle Nutr Inst Workshop Ser. 2020;94:11–26.
Article PubMed Google Scholar
Lee E, Piranlioglu R, Wicha MS, Korkaya H. Plasticity and potency of mammary stem cell subsets during mammary gland development. Int J Mol Sci. 2019;20:2357.
Article CAS PubMed PubMed Central Google Scholar
Sakakura T, Suzuki Y, Shiurba R. Mammary stroma in development and carcinogenesis. J Mammary Gland Biol Neoplasia. 2013;18:189–97.
Article PubMed Google Scholar
Dawson CA, Visvader JE. The cellular organization of the mammary gland: insights from microscopy. J Mammary Gland Biol Neoplasia. 2021;26:71–85.
Article PubMed Google Scholar
Moumen M, Chiche A, Cagnet S, Petit V, Raymond K, Faraldo MM, et al. The mammary myoepithelial cell. Int J Dev Biol. 2011;55:763–71.
Article PubMed Google Scholar
Gieniec KA, Davis FM. Mammary basal cells: stars of the show. Biochim Biophys Acta Mol Cell Res. 2022;1869:119159.
Article CAS PubMed Google Scholar
Zhang Y, Li F, Song Y, Sheng X, Ren F, Xiong K, et al. Numb and Numbl act to determine mammary myoepithelial cell fate, maintain epithelial identity, and support lactogenesis. Faseb j. 2016;30:3474–88.
Article CAS PubMed Google Scholar
Raymond K, Cagnet S, Kreft M, Janssen H, Sonnenberg A, Glukhova MA. Control of mammary myoepithelial cell contractile function by α3β1 integrin signalling. Embo j. 2011;30:1896–906.
Article CAS PubMed PubMed Central Google Scholar
Chakrabarti R, Celia-Terrassa T, Kumar S, Hang X, Wei Y, Choudhury A, et al. Notch ligand Dll1 mediates cross-talk between mammary stem cells and the macrophageal niche. Science. 2018;360:eaan4153.
Article PubMed PubMed Central Google Scholar
Cazares O, Chatterjee S, Lee P, Strietzel C, Bubolz JW, Harburg G, et al. Alveolar progenitor differentiation and lactation depends on paracrine inhibition of notch via ROBO1/CTNNB1/JAG1. Development. 2021;148:dev199940.
Article CAS PubMed PubMed Central Google Scholar
Brisken C, Ataca D. Endocrine hormones and local signals during the development of the mouse mammary gland. Wiley Interdiscip Rev Dev Biol. 2015;4:181–95.
Article CAS PubMed Google Scholar
Kumar S, Srivastav RK, Wilkes DW, Ross T, Kim S, Kowalski J, et al. Estrogen-dependent DLL1-mediated Notch signaling promotes luminal breast cancer. Oncogene. 2019;38:2092–107.
Article CAS PubMed Google Scholar
Murta D, Batista M, Trindade A, Silva E, Mateus L, Duarte A, et al. Dynamics of Notch signalling in the mouse oviduct and uterus during the oestrous cycle. Reprod Fertil Dev. 2015;28:1663–78.
Article Google Scholar
Fan JZ, Yang L, Meng GL, Lin YS, Wei BY, Fan J, et al. Estrogen improves the proliferation and differentiation of hBMSCs derived from postmenopausal osteoporosis through notch signaling pathway. Mol Cell Biochem. 2014;392:85–93.
Article CAS PubMed PubMed Central Google Scholar
Hughes K, Watson CJ. The multifaceted role of STAT3 in mammary gland involution and breast cancer. Int J Mol Sci. 2018;19:1695.
Article PubMed PubMed Central Google Scholar
Liu Y, Zhang J, Zhou YH, Zhang HM, Wang K, Ren Y, et al. Activation of the IL-6/JAK2/STAT3 pathway induces plasma cell mastitis in mice. Cytokine. 2018;110:150–8.
Article CAS PubMed Google Scholar
Wan J, Xu W, Zhan J, Ma J, Li X, Xie Y, et al. PCAF-mediated acetylation of transcriptional factor HOXB9 suppresses lung adenocarcinoma progression by targeting oncogenic protein JMJD6. Nucleic Acids Res. 2016;44:10662–75.
Article CAS PubMed PubMed Central Google Scholar

Download references

Acknowledgements

This work was supported by grants from the National Key Research and Development Program of China (2022YFA1104003, 2021YFC2501003, 2020YFA0112300), the National Natural Science Foundation of China (no. 82230094, 81621063, 81730071, 81972616, 82172972, 81972609, 81301802, 81670626, and 82072903, 81830087, U2102203, 82002811, and 81872145), and the Beijing Natural Science Foundation (no. 7202084, 7171005, 7120002). This study was also supported by and the Yunnan Fundamental Research Projects (202101AS070050). Peking University grants BMU2020JC003, PKU2020LCXQ007 and PKU2021LCXQ023.The authors appreciate the academic support from the AME Breast Cancer Collaborative Group.

Author information

These authors contributed equally: Zhenbin Wang, Lei Zhang, Bing Li, Jiagui Song.

Authors and Affiliations

Program for Cancer and Cell Biology, Department of Human Anatomy, Histology and Embryology, School of Basic Medical Sciences; Peking University International Cancer Institute; MOE Key Laboratory of Carcinogenesis and Translational Research and State Key Laboratory of Natural and Biomimetic Drugs, Peking University Health Science Center, 100191, Beijing, China
Zhenbin Wang, Lei Zhang, Bing Li, Jiagui Song, Miao Yu, Jing Zhang, Jun Zhan & Hongquan Zhang
Department of Histology and Embryology, Hebei University of Chinese Medicine, Shijiazhuang, Hebei, 050200, China
Bing Li
Academy of Biomedical Engineering, Kunming Medical University, Kunming, 650500, China
Ceshi Chen
Key Laboratory of Animal Models and Human Disease Mechanisms of the Chinese Academy of Sciences and Yunnan Province, Kunming Institute of Zoology, Chinese Academy of Sciences, Kunming, Yunnan, 650223, China
Ceshi Chen

Authors

Zhenbin Wang
View author publications
You can also search for this author in PubMed Google Scholar
Lei Zhang
View author publications
You can also search for this author in PubMed Google Scholar
Bing Li
View author publications
You can also search for this author in PubMed Google Scholar
Jiagui Song
View author publications
You can also search for this author in PubMed Google Scholar
Miao Yu
View author publications
You can also search for this author in PubMed Google Scholar
Jing Zhang
View author publications
You can also search for this author in PubMed Google Scholar
Ceshi Chen
View author publications
You can also search for this author in PubMed Google Scholar
Jun Zhan
View author publications
You can also search for this author in PubMed Google Scholar
Hongquan Zhang
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

Study design: HZ, J Zhan, and CC. Study conduct and data collection and analysis: ZW, LZ, BL, and JS. Data interpretation: ZW, LZ, JS, J Zhan, MY, and J Zhang. Drafting the manuscript: LZ, ZW, HZ, J Zhan, and CC.

Corresponding authors

Correspondence to Ceshi Chen, Jun Zhan or Hongquan Zhang.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

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

Edited by Professor Anastasis Stephanou

Supplementary information

Supplementary files

Original western blots

Reproducibility checklist

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 license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license 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 license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Wang, Z., Zhang, L., Li, B. et al. Kindlin-2 in myoepithelium controls luminal progenitor commitment to alveoli in mouse mammary gland. Cell Death Dis 14, 675 (2023). https://doi.org/10.1038/s41419-023-06184-2

Download citation

Received: 01 July 2023
Revised: 11 September 2023
Accepted: 26 September 2023
Published: 13 October 2023
DOI: https://doi.org/10.1038/s41419-023-06184-2