During aging, cellular plasticity and senescence play important roles in tissue regeneration and the pathogenesis of different diseases, including cancer. We have recently shown that senescent breast luminal cells can activate their adjacent stromal fibroblasts. In the present report, we present clear evidence that these senescence-related active fibroblasts can dedifferentiate proliferating primary human luminal cells to multipotent stem cells in an interleukin-8 (IL-8)-dependent manner. This was confirmed using recombinant IL-8, while the truncated protein was not active. This IL-8-related dedifferentiation of luminal cells was mediated through the STAT3-dependent downregulation of p16INK4A and the microRNA miR-141. Importantly, these in vitro-generated mammary stem cells exhibited high molecular and cellular similarities to human mammary stem cells. They have also shown a long-term mammary gland-reconstituting ability and the capacity to produce milk postdelivery. Thereby, these IL-8-generated mammary stem cells could be of great value for autologous cell therapy procedures and also for biomedical research as well as drug development.
The mammary gland is a complex organ composed of various types of cells organized in different functional tissues. The mammary epithelium contains two major types of cells, luminal and myoepithelial cells. In addition to these terminally differentiated cell lineages, cell-tracing experiments have shown the presence of unipotent as well as multipotent stem cells, which can give rise to both luminal and myoepithelial cells (1, 2). This hierarchical organization could be seriously affected in response to various stresses or diseases. The well-known one is the procarcinogenic aging process, during which there is an accumulation of senescent cells, known to have proinflammatory and -tumorigenic paracrine effects mediated through the secretion of various cytokines and growth factors (3–5). Indeed, several lines of evidence indicated the presence of age-dependent accumulation of multipotent progenitor cells with a decline in function and alteration in the luminal-to-basal cell ratio (6, 7). Moreover, senescence promoted cellular reprogramming in vivo through the secretion of high levels of interleukin-6 (IL-6) (8, 9). The plasticity of somatic cells is a very important and complex physiological process with implications for organismal rejuvenation, tissue regeneration, and disease (10, 11). This cellular reprogramming is genetically programmed and is also promoted through paracrine effects from the microenvironment comprising the surrounding cells.
We have recently shown that senescent breast luminal cells can activate their adjacent stromal fibroblasts in an IL-8-dependent manner (12). IL-8, a key senescence and inflammatory cytokine, is a very important cell-cell signaling transmitter with potent reprogramming capacities (13). Active fibroblasts are known to have potent procarcinogenic activities through paracrine signaling (14). Therefore, in the present study, we sought to investigate the effects of these senescence-related active fibroblasts on their breast luminal cells. We show that these active fibroblasts dedifferentiate luminal cells to multipotent mammary stem cells in an IL-8-dependent manner. This effect is mediated through STAT3 activation and the microRNA (miRNA) miR-141 downregulation. These IL-8-generated mammary stem cells were able to generate active mammary glands in mice.
Active breast stromal fibroblasts trigger stemness properties in luminal cells in an IL-8-dependent fashion.
We have recently shown that while senescent human breast luminal cells can transactivate stromal fibroblasts (senescent luminal cell-activated fibroblasts [SLAF]), young proliferative luminal cells had only minimal effects on these fibroblastic cells (young luminal cell-activated fibroblasts [YLAF]) (12). Since active fibroblasts are known to have strong procarcinogenic paracrine effects, we sought to investigate these effects on breast normal primary luminal cells in order to shed more light on cell-cell cross talk during the complex aging process. To this end, normal primary human breast luminal cells (NBL-10), purified as previously described (12), were incubated for 24 h in serum-free conditioned medium (SFCM) from SLAF (SLAF-SFCM) or their corresponding control cells (YLAF), while serum-free medium (SFM) was used as a negative control. Figure 1A shows clear increases in the proliferation rates as well as the migration/invasion abilities of NBL-10 cells exposed to SFCM from SLAF cells compared to SFCM from YLAF, which had an effect similar to that of SFM.
Next, we investigated the possible induction of mesenchymal features in luminal cells upon exposure to SFCM from SLAF cells. Indeed, the immunoblot analysis shows a strong decrease in the epithelial markers E-cadherin and epithelial cell adhesion molecule (EpCAM), while the mesenchymal markers N-cadherin, vimentin, Twist1, ZEB1, Snail, and Slug were upregulated in cells that were exposed to SLAF-SFCM compared to YLAF-SFCM and SFM (Fig. 1B). Similar results were obtained with SFCM from IL-8-activated stromal fibroblasts (ILAF) (Fig. 1B). Similar findings were also obtained when NBL-10 cells were exposed to SFCM from 3 different IL-8-activated breast stromal fibroblasts (NBF-20, NBF-21, and NBF-34) (Fig. 1C). This indicates that active breast stromal fibroblasts trigger epithelial-to-mesenchymal transition (EMT) in breast luminal cells in a paracrine manner.
Since the EMT is a transition phase toward stemness and because links exist between the EMT and certain properties of mammary stem cells (MaSC) (15–17), we investigated this possibility in response to SFCM from SLAF and ILAF. Figure 1B shows a clear decrease in CD24, while CD44 and aldehyde dehydrogenase 1/2 (ALDH1/2) were upregulated compared to their levels in luminal cells treated with YLAF-SFCM or SFM. Similarly, the levels of the stemness transcription factors BMI-1, OCT-4, and KLF-4 were also strongly increased in NBL-10 cells exposed to SLAF-SFCM and ILAF-SFCM compared to their respective controls (Fig. 1B). This suggests the possible induction of stemness in breast luminal cells exposed to SFCM from active breast stromal fibroblasts.
Active breast stromal fibroblasts and IL-8 dedifferentiate luminal cells to multipotent mammary stem cells.
To confirm the stemness of luminal cells that were exposed to SFCM from SLAF and ILAF, single NBL-10 cells were cultured in suspension in the presence of stem cell-specific culture medium. Figure 2A shows the formation of typical hollow mammospheres within 2 days of culture, with a self-renewal capacity that has been lost over time with passaging (Fig. 2B). Figure 2B also shows that 45 and 47 mammospheres were formed from 1,000 single cells that were pretreated with ILAF and SLAF, respectively. This indicates that the dedifferentiation efficiencies reached 4.5% and 4.7%, respectively. Next, we tested for the presence of some mammosphere-specific markers and found that the spheres were positive for cytokeratin 18 (CK18) and CK14 (Fig. 2C). This indicates the presence of biprogenitor features in these cells. To confirm this, we investigated the surface markers by flow cytometry and found that exposure of luminal cells to ILAF-SFCM and SLAF-SFCM increased the proportion of EpCAMlow/CD49fhigh (a typical marker of mammary stem cells) compared to control (SFM) cells (Fig. 2D). Interestingly, these cells also exhibited a high and exponential proliferative capacity as well as high telomerase activity when grown as a monolayer (Fig. 2E and F). To further confirm the stemlike nature of these cells, we assessed their differentiation capacity. To this end, freshly obtained mammospheres were plated on gelatin for 1 week, and they were then characterized by flow cytometry and immunostaining. Figure 3A shows the presence of cells with luminal (red arrows) and myoepithelial (green arrows) shapes. These cells stained positive for the myoepithelial marker (CK14), the luminal marker (CK19), or both, with a lower proportion of cells that were positive for both (Fig. 3B). This indicates the ability of these mammospheres to differentiate into specialized cells, namely, luminal and myoepithelial cells, in addition to some bipotent cells. This was confirmed by flow cytometry analysis, which also showed 2 different cell populations: one population exhibited myoepithelial features (EpCAMlow CD49fhigh CD10+ Muc1−), while the other one showed luminal characteristics (EpCAMhigh CD49flow CD10− Muc1+) (Fig. 3C). Next, we tested the branching capacities of the SLAF- and ILAF-treated luminal cells. To this end, single cells from mammospheres were first differentiated and then covered with Matrigel containing prolactin. Figure 3D shows three-dimensional (3D) branching structures of both SLAF- and ILAF-treated luminal cells. Importantly, the SLAF- and ILAF-related generation of mammospheres, but not SFM-treated luminal cells, reconstituted cleared mammary fat pads upon transplantation into female SCID mice (Fig. 3E). After resection of mammary outgrowth, whole-mount analysis showed the generation of mammary ductal trees with distal-end buds only in the branching epithelia formed with ILAF- and SLAF-related mammospheres but not in the control (SFM) (Fig. 3E). Similar branching structures were also found in mouse fat pads (Fig. 3E). Hematoxylin and eosin (H&E) histology showed clear mammary ducts surrounded by adipose tissue in ILAF- and SLAF-related fat pads but in not the control one (Fig. 3E). Next, the human nature of the formed ducts was confirmed by immunostaining using specific anti-human anti-EpCAM and anti-CK18 antibodies. Interestingly, while the SLAF-related ducts were positively stained, mouse ducts were negative for this staining (Fig. 3E and F), which confirmed that the formed ducts originated from the implanted cells.
We then set out to investigate the possible implication of IL-8 in the dedifferentiation of luminal cells into mammary stem cells. To this end, NBL-10 cells were incubated for 24 h in SFM (used as a negative control), SFM containing IL-8 (2.5 ng/ml), as determined in our previous study (12), or SFCM from ILAF containing either neutralizing anti-IL-8 antibody or IgG (used as a negative control). The immunoblot analysis shows a clear IL-8-dependent decrease in the level of E-cadherin and an increase in the N-cadherin level (Fig. 4A). This effect was partially inhibited by IL-8-specific neutralization (Fig. 4A). Similarly, recombinant IL-8 upregulated the stem cell markers CD44, ALDH, KLF-4, and OCT-4, while CD24 was downregulated (Fig. 4A). On the other hand, neutralizing IL-8 antibody inhibited these effects (Fig. 4A). Flow cytometry analysis showed similar IL-8-related changes in the CD44high/CD24low subpopulation (Fig. 4B). This suggests that IL-8 is responsible, at least partially, for the ILAF-related induction of stemness in breast luminal cells. When these cells were cultured in low-attachment plates in the presence of stem cell medium, mammospheres were formed (Fig. 4C) and were sustained for several generations, indicating their self-renewal capacity, with a dedifferentiation efficiency similar to those obtained with SLAF and ILAF (Fig. 2B). Interestingly, the spheres costained positive for both cytokeratins 14 and 19 (Fig. 5A). To show the biprogenitor nature of these cells, we assessed their differentiation capacity. Figure 4C shows the presence of cells with luminal and myoepithelial shapes. These cells stained positive for the myoepithelial marker (CK14), the luminal marker (CK19), or both, with a lower proportion of cells that were positive for both (Fig. 5B). This indicates the ability of these mammospheres to be differentiated into luminal and myoepithelial cells, in addition to some bipotent cells. Moreover, we observed the differentiation of these IL-8-generated mammospheres into branching structures in Matrigel containing prolactin (Fig. 5C). Importantly, like ILAF and SLAF treatments, IL-8-treated luminal cells reconstituted cleared mammary fat pads upon transplantation into female SCID mice. After resection, whole-mount analysis showed the generation of mammary ductal trees with distal-end buds (Fig. 5D). To confirm the multipotent nature of these mammary stem cells, mammary outgrowths were collected and retransplanted for secondary mammary fat pad reconstitution. Interestingly, 9 out of 9 of such transplantations reconstituted secondary glands and showed typical mammary ductal trees (Fig. 5E). This shows that the primary transplants retained their regenerative potentials through secondary transplants. This indicates that the generated MaSC exhibit a long-term reconstituting ability.
Similar experiments were performed on different breast primary luminal cells, which also generated MaSC. Figure 6A shows that IL-8-dependent treatment of NBL-34 cells also generated mammospheres that stained positive for both cytokeratins 14 and 18 and were differentiated into luminal (cytokeratin 18-positive) and myoepithelial (cytokeratin 14-positive) cells.
Interestingly, IL-8 treatment induced MaSC features in luminal cells even when these cells were grown in 3D structures (Fig. 6E). Indeed, SLAF-SFCM, ILAF-SFCM, as well as IL-8 enhanced the proliferative as well as the invasive capacities of luminal cells relative to controls (Fig. 6E, H&E). Furthermore, these treatments enriched for cells positive for both CK14 and CK19 and enhanced the expression of CD44 and ALDH1, while CD24 was downregulated (Fig. 6E).
To confirm this IL-8 dependence in dedifferentiating human luminal cells into MaSC, NBL-10 cells were treated with IL-8 or a truncated form of the protein (IL-8t) (2.5 ng/ml each), while SFM was used as a negative control. Figure 6F shows that while IL-8 promoted EMT and stemness features (upregulation of N-cadherin, CD44, and ALDH and downregulation of E-cadherin and CD24), IL-8t had no effect on these markers. Similarly, while IL-8 treatment generated mammospheres under stem cell growth conditions, no spheres were obtained when NBL-10 was exposed to IL-8t (Fig. 6G). Similar to IL-8t, two different concentrations of IL-6 (1 ng/ml and 3.5 ng/ml) were not able to generate mammospheres under stem cell growth conditions (Fig. 6H, right), although both concentrations increased the level of the mesenchymal and stemness markers (Fig. 6H, left). This shows the importance of active IL-8 in the dedifferentiation process. Together, these results indicate that IL-8 can dedifferentiate human luminal cells into multipotent mammary stem cells.
It is noteworthy that luminal cells were exposed to IL-8 as attached cells under conditions where mammary stem cells do not attach, which, in principle, should remove all potential contaminant MaSC and further support the IL-8-dependent dedifferentiation process.
IL-8-related generation of MaSC is mediated through STAT3 activation and p16/miR-141 downregulation.
In order to delineate the molecular pathway implicated in the IL-8-dependent dedifferentiation of luminal cells into MaSC, we sought to evaluate the role of the STAT3 signaling effect via knocking down the gene using specific small interfering RNA (siRNA). Figure 7A shows that IL-8 treatment of luminal cells did not induce EMT and stemness in STAT3-deficient cells relative to controls. Indeed, STAT3 siRNA-1 prevented the upregulation of mesenchymal markers as well as CD44 and ALDH (Fig. 7A). Similar results were obtained when a different STAT3 siRNA was utilized (STAT3 siRNA-2) (Fig. 7B). Specific STAT3 knockdown also suppressed the IL-8-related activation of the proliferation and migration/invasion of luminal cells (Fig. 7C). Consequently, while IL-8-treated control cells formed mammospheres when cultured in ultra-low-attachment flasks, STAT3-deficient cells did not (Fig. 7D). This indicates that IL-8-dependent dedifferentiation of luminal cells into MaSC is mediated through STAT3 signaling.
To further elucidate the molecular mechanism underlying the IL-8-dependent dedifferentiation of luminal cells, we sought to delineate the role of the STAT3 downstream effector miR-141. Therefore, we first tested the effect of IL-8 treatment on miR-141 expression. Figure 8A shows a potent reduction in the level of miR-141. To test the implication of STAT3 in this process, the gene was knocked down in NBL-10 cells by specific siRNA, while a scrambled sequence was used as a control, and the cells were then exposed or not to IL-8. Real-time reverse transcription-PCR (RT-PCR) confirmed the IL-8-dependent downregulation of miR-141, while STAT3 knockdown enhanced the expression of this microRNA relative to controls (Fig. 8B). Interestingly, exposure of STAT3-deficient cells to IL-8 had only a slight effect on the expression of miR-141 (Fig. 8B). This shows that IL-8-dependent miR-141 downregulation is STAT3 related. Similarly, IL-8 treatment upregulated AUF1 (a direct target of STAT3 [18–20]), CD44, and ZEB1 and reduced the level of p16 in control cells (Fig. 8C). However, these IL-8-related effects were suppressed in STAT3-deficient cells (Fig. 8C).
miR-141 is also under the positive control of p16, which is negatively regulated by STAT3 either directly or indirectly through positive regulation of the p16 repressor AUF1 (18–20). Therefore, we investigated the implication of p16 in IL-8-dependent miR-141 downregulation in luminal NBL-10 cells. As expected, IL-8 treatment reduced the levels of both the CDKN2A mRNA and miR-141, while it enhanced the downstream effectors CD44 and ZEB1 (Fig. 8D). On the other hand, the ectopic expression of p16 strongly enhanced the level of miR-141 and suppressed IL-8-dependent miR-141 downregulation and the consequent upregulation of CD44 and ZEB1 (Fig. 8D). This indicates that the IL-8 effect on luminal cells is mediated through p16 and its downstream effector miR-141.
To further confirm the importance of miR-141 downregulation in the IL-8-dependent dedifferentiation of luminal cells, we ectopically expressed miR-141 in NBL-10 cells using a plasmid bearing pre-miR-141 or an empty vector used as a control (pre-miR-ctrl), and these cells were then either exposed or not to IL-8 (pre-miR-141-IL-8 and pre-miR-ctrl-IL-8). Figure 8E shows the formation of mammospheres only in control cells treated with IL-8 but not in those expressing miR-141. This was confirmed by immunofluorescence and the expression of cytokeratin 19 and cytokeratin 14 (Fig. 8E). Figure 8 also shows that only pre-miR-ctrl-IL-8 cells expressed the myoepithelial marker cytokeratin 14, which shows the importance of miR-141 downregulation in the IL-8-related dedifferentiation of human luminal cells.
To confirm this at the molecular level, we assessed the levels of several EMT- and stemness-related genes and showed that the ectopic expression of pre-miR-141 inhibited the IL-8-dependent promotion of the EMT and stemness processes (Fig. 9A). To confirm the role of miR-141 in the STAT3-related reprogramming of luminal cells upon treatment with IL-8, miR-141 was inhibited with specific miRZip in both STAT3-deficient cells as well as their corresponding controls, and the cells were then exposed to IL-8. Interestingly, while STAT3 siRNA inhibited the IL-8-dependent promotion of EMT/stemness, the inhibition of miR-141 restored the ability of IL-8 to promote EMT and stemness in STAT3-deficient cells (Fig. 9B). These results were confirmed by showing that while STAT3si-ctrl-IL-8 cells do not form spheres and express only cytokeratin 19, STAT3- and miR-141-deficient cells treated with IL-8 formed spheres, which expressed both cytokeratin 19 and cytokeratin 14 (myoepithelium specific) (Fig. 9C).
This indicates that miR-141 indeed acts downstream of STAT3 and that its downregulation is a prerequisite for the IL-8-dependent induction of the EMT and stemness processes in human luminal cells.
IL-8-generated MaSC and breast natural MaSC exhibit similar gene expression patterns.
To further characterize the IL-8-generated MaSC (IL-8-MaSC) in vitro, we decided to compare the gene expression pattern in these MaSC with that of breast natural MaSC isolated from human breast tissue obtained postmammoplasty (21). Both MaSC exhibited similar gene expression patterns, with 95% of genes showing similar expression levels, and only 1.67% were upregulated and 3.29% were downregulated in IL-8-MaSC relative to the controls (Fig. 10A, C, and D). Strikingly, gene expression patterns of luminal and myoepithelial cells obtained after differentiation of IL-8-generated MaSC (postdifferentiated luminal [PD-luminal] and PD-myoepithelial, respectively) were identical to those of the original luminal (NBL-10) and myoepithelial (NBM-10) cells, respectively (Fig. 10A). In addition, we analyzed the effect of the IL-8-related dedifferentiation of luminal cells to MaSC on gene expression. We found that 88.6% of the analyzed genes showed similar expression levels between the original luminal cells (NBL-10) and their corresponding derivative MaSC (IL-8-MaSC) (Fig. 10A, E, and F). The upregulated genes represented 3% of the genes, while 8.4% of the genes were downregulated (Fig. 10E and F). Among the important differentially expressed genes, we identified several EMT- and stemness-related genes, such as ZEB1, vimentin, N-cadherin, E-cadherin, EpCAM, CD44, and CD24 (Table 1). These variations in gene expression were confirmed by RT-quantitative PCR (qRT-PCR). Indeed, striking differences were observed for vimentin, SNAIL1, N-cadherin, TWIST1, ZEB1, and E-cadherin between NBL-10 and its derivative IL-8-MaSC (Fig. 10B). On the other hand, the expression levels of these genes were identical in both lines of luminal cells, the original line (NBL-10) and those obtained postdifferentiation (PD-luminal) (Fig. 10B).
TABLE 1 List of the most highly differentially expressed EMT and stemness genes in NBL-10 cells versus IL-8-MaSC
Reconstituted humanized mammary fat pads produce milk after pregnancy and delivery.
To show the active status of the reconstituted humanized mammary glands with the mammary stem cells generated in vitro with IL-8-treated luminal cells, the females bearing these cells in their mammary fat pads 4 and 5 were mated. When the females were pregnant, all nipples but 4 and 5 were cauterized and removed, and delivery occurred after 3 weeks of pregnancy. Figure 11A shows milk in the mammary gland. Reconstituted fat pads used for lactation were subjected to carmine whole-mount staining, which showed highly developed mammary outgrowth (Fig. 11B). Interestingly, milk spots were observed in the pups’ stomachs the second day after birth (Fig. 11C). To confirm the human nature of the milk, fluorescent immunostaining was performed using specific anti-human casein antibody. Figure 11D shows strong casein staining on the fat pad obtained from the lactating animal compared to that obtained from the virgin one. This confirmed that the generated milk was produced by human cells and that IL-8-generated mammary stem cells can regenerate functional mammary glands in an animal model.
After showing that senescent luminal cells can transactivate breast stromal fibroblasts in an IL-8-dependent manner and that these cells secrete high levels of IL-8 and have procarcinogenic potential (12), we decided to elucidate the effects of IL-8- and senescent luminal cell (SLC)-dependent active breast stromal fibroblasts on their adjacent normal luminal cells. Intriguingly, breast myofibroblasts as well as IL-8 dedifferentiated normal luminal cells to multipotent mammary stem cells with a long-term mammary gland-reconstituting ability. These cells exhibited all the cellular and molecular features of mammary stem cells, including the gene expression pattern; self-renewal; differentiation into luminal, myoepithelial, and bipotent cells; and the formation of branching structures both in vitro and in vivo, when they were injected into cleared mouse mammary fat pads. Furthermore, luminal and myoepithelial cells produced by the differentiation in vitro of the obtained IL-8-related mammospheres exhibited gene expression patterns identical to those of the initial/parental luminal cells and their counterpart myoepithelial cells, respectively. Interestingly, the mammary stem cells generated in vitro reconstituted functional humanized mammary glands, which produced milk postdelivery. This paradigm may be applied to other epithelial cells, and since the process does not need any genetic alteration, the resulting stem cells could be safely utilized for autologous regenerative therapies.
Similarly, Poli et al. have recently shown the role of MYC overexpression in dedifferentiating human telomerase reverse transcriptase (hTERT)-immortalized human mammary epithelial cells into stem cell-like cells (22). Moreover, dedifferentiation of murine mammary luminal cells has been previously obtained upon the transient coexpression of Slug and Sox9, which converted them into long-term-repopulating mammary stem cells (15). In another study, Panciera et al. found that the transient expression of YAP/TAZ converted primary differentiated mouse cells to tissue-specific stem/progenitor cells (23). Furthermore, it was recently shown that transient exposure of primary mouse keratinocytes to senescence-associated secretions enhanced plasticity and stemness in these cells (10). Moreover, Schwitalla et al. showed that constitutive activation of β-catenin dedifferentiates intestinal epithelial cells into crypt stem cells (24). These findings indicate that differentiated luminal cells can be reprogrammed and regain stem cell properties through significant plasticity in the epithelial cell hierarchy.
At the molecular level, we have shown that IL-8-dependent MaSC formation is mediated through STAT3 activation and miR-141 downregulation. In fact, while IL-8 is a well-known activator of STAT3 signaling, the repression of miR-141 could be mediated through AUF1 upregulation and consequent p16 repression. Indeed, we have previously shown that p16, which is negatively regulated by AUF1, positively controls miR-141 (18, 19). Moreover, AUF1 is a direct target of the STAT3 transcription factor (20). Interestingly, our findings indicate that IL-8 downregulates miR-141 through STAT3/AUF1/p16 signaling (Fig. 8 and 9). miR-141 belongs to the miR-200 family, which has been shown to be associated with EMT and stemness through the negative regulation of ZEB1 and ZEB2, two major EMT proteins (25, 26). Furthermore, the stem cell proteins CD44 and EZH2 were shown to be direct and functionally relevant targets of miR-141, which also positively controls the expression of several epithelial genes, while it represses numerous mesenchymal genes (27). In addition to STAT3, IL-8 activates other signaling pathways, such as the phosphatidylinositol 3-kinase (PI3K)/AKT, p38 mitogen-activated protein kinase (MAPK), and RAS/extracellular signal-regulated kinase (ERK) pathways (13), which may also be involved in the IL-8-dependent dedifferentiation of luminal cells.
The present findings also show that IL-8 plays major roles in the cross talk between breast epithelial and stromal cells. In addition to senescence, IL-8 can be activated by several other forms of physiological and physical signaling, including inflammatory signals, reactive oxygen species, radiation, and hypoxia (28). Furthermore, accumulating evidence shows that IL-8 plays a critical role in EMT and is a key regulator of breast cancer stem cell (CSC) activity (29–33). The possible IL-8-dependent dedifferentiation of luminal cells into multipotent stem cells suggests the potential accumulation of mammary stem cells with a proliferative capacity in response to inflammation and stress as well as during aging. As a case in point, Garbe et al. have shown the accumulation of multipotent progenitors during the aging of human mammary epithelia (7). Similarly, Dong et al. have recently shown an age-associated increase in MaSC frequency in mouse mammary glands (6). Both reports described a decline in the function and an increased transformation potential of MaSC with aging. Other groups have also shown that senescence-associated secretions promote in vivo reprogramming and enhance the regenerative capacity (10, 34, 35). This suggests that the accumulation of MaSC, which might be more susceptible to mutagenesis and transformation in a procarcinogenic niche, could constitute an important step during breast carcinogenesis. Furthermore, IL-8 may also be part of a procarcinogenic niche that can reprogram noncancer stem cells to cancer stem cells, which could be a great source of tumor heterogeneity. This plasticity was recently shown in 2 studies, which indicated that the introduction of a PIK3CAH1047R mutation in basally or luminally restricted mammary cells triggered lineage plasticity and cell expansion during tumorigenesis in mice (36, 37). These findings suggest that non-CSC can acquire CSC-like features either as a result of genetic/epigenetic changes or in a microenvironment-dependent manner (38, 39). Thereby, heterogeneity in tumors does not need to arise from the transformation of a cell with multipotent potential but could be a natural consequence of oncogene activation, stress, or mutation, regardless of the cell of origin, which could be a multipotent stem cell, a luminal unipotent stem/progenitor cell, or a mature luminal lineage-restricted cell (36).
In conclusion, our study shows that senescence-related active breast stromal fibroblasts as well as IL-8 promote the dedifferentiation of breast luminal cells into multipotent mammary stem cells capable of generating functional mammary glands. Thereby, these MaSC generated in vitro by IL-8 treatment clearly show the plasticity of these cells and could be of great cosmetic/therapeutic value as well as a precious tool for biomedical research and drug development.
MATERIALS AND METHODS
Normal breast tissue processing and cell preparation.
Breast tissues were collected from subjects who underwent reduction mammoplasty according to the guidelines of the Ethics Committee of the King Faisal Specialist Hospital and Research Center (KFSHRC) (approval RAC 2140017). Informed consent was obtained from all subjects. Luminal cells and fibroblast cells were purified through 2 sorting steps as previously described (12).
Cell culture and reagents.
Primary breast fibroblast cells (NBF-10, NBF-20, NBF-21, and NBF-34) and luminal epithelial cells (NBL-10 and NBL-34) were obtained from females who underwent reduction mammoplasty and then cultured using fibroblast growth medium (M-199 and F-12 media at a 1:1 ratio, supplemented with 10% fetal bovine serum and 1% antibiotic and antimycotic [Gibco, NY]) and luminal growth medium (Dulbecco’s modified Eagle’s medium [DMEM]–F-12 medium supplemented with 2% fetal bovine serum, human mammary epithelial cell (HuMEC) supplement [Gibco], bovine pituitary extract [Gibco], and 1% antibiotic and antimycotic), respectively, as previously described (40). Human full-length IL-8 recombinant protein (P01) (Abnova Corporation, Taoyuan City, Taiwan) (sequence, MTSKLAVALLAAFLISAALCEGAVLPRSAKELRCQCIKTYSKPFHPKFIKELRVIESGPHCANTEIIVKLSDGRELCLDPKENWVQRVVEKFLKRAENS), human truncated IL-8 recombinant protein (GenBank accession number P10145) (Abcam, MA) (sequence, AVLPRSAKELRCQCIKTYSKPFHPKFIKELRVIESGPHCANTEIIVKLSDGRELCLDPKENWVQRVVEKFLKRAENS), and human full-length IL-6 recombinant protein (hBA-184) (Santa Cruz Biotechnology, Inc.) were used.
Single cells were first passed through a 40-μm sieve and plated in ultra-low-attachment 6-well plates at densities of 20,000 viable cells/ml in primary culture and 1,000 cells/ml in passages (to assess the self-renewal capacity). Cells were grown in serum-free mammosphere-specific medium (mammary epithelial cell basal medium [MEBM]) (Lonza, NJ) supplemented with 2% B27, 20 ng/ml epidermal growth factor (EGF), 500 ng/ml hydrocortisone, and 5 μg/ml insulin (all from Gibco). Mammospheres were collected by gentle centrifugation (800 rpm) after 7 to 10 days and dissociated enzymatically (10 min in a 1:1 mixture of 0.05% trypsin, 0.53 mM EDTA, and TrypLE). The cells obtained by dissociation were sieved through a 40-μm sieve and analyzed microscopically for single cellularity.
Reconstitution of human mammary glands in a mouse model.
Animal experiments were approved by the KFSHRC institutional animal care and use committee (ACUC) under RAC proposal number 2090027 and were conducted according to guidelines in the Guide for the Care and Use of Laboratory Animals, 8th edition (41). Animals were individually housed in a controlled environment (pathogen free) in filter-top cages under 12-h light/dark cycles at 22°C ± 2°C and provided with sterile food and water ad libitum. Experiments for the reconstitution of mammary glands were performed according to the protocol of Proia and Kuperwasser, with slight modifications (42). Immortalized fibroblasts (BP28-tert) (125,000 X-ray-irradiated cells and 125,000 nonirradiated cells) were mixed with different numbers of single cells of YLAF, SLAF, ILAF, or IL-8-treated luminal mammospheres in the presence of Matrigel. Twenty blindly chosen 3-week-old NOD.CB17-Prkdcscid/J female mice (Jackson Laboratories, CA) were anesthetized, and the junction of blood vessels near the lymph node between the fourth and fifth sets of nipples was cauterized. The fourth inguinal mammary glands were then cleared from the cauterized area, and the mice were then randomized into 4 groups. The cells were injected under the skin of the cleared gland. After 8 weeks of injection, mice were sacrificed, and the reconstituted mammary gland was excised and utilized for whole-mount analysis and immunohistochemistry. For secondary mammary gland reconstitution, primary mammary ductal outgrowths were cut into very small fragments (1 mm each), and each fragment was then reimplanted into cleared mammary fat pads.
Whole-mount staining of mammary glands.
Mammary gland tissue was spread onto a glass slide, fixed with Carnoy’s fixative (6:3:1 mixture of ethanol-chloroform-acetic acid) for 4 h, and stained with carmine alum stain (carmine, aluminum potassium sulfate) overnight.
Cells were cultured in medium without serum for 48 h, and the media were then collected and passed through a 0.45-μm filter unit. The resulting supernatants were either used immediately or frozen at −80°C until needed.
Telomerase activity assay.
A telomerase activity assay was performed using a TeloTAGGG telomerase PCR enzyme-linked immunosorbent assay (ELISA) according to the manufacturer’s instructions (Roche, NY). Briefly, 20 μg of heat- or sham-treated cell extracts was amplified by PCR (94°C for 30 s, 50°C for 30 s, and 72°C for 90 s). The PCR products were hybridized to anti-digoxigenin-peroxidase, the ELISA reaction was then performed, and the optical density (OD) was measured at 450 nm on a standard ELISA plate reader (x-Mark; Bio-Rad). These experiments were performed in triplicates.
Mammospheres were digested into single cells, sieved through a 40-μm sieve, and then analyzed microscopically for single cellularity. A total of 1,000 single cells were plated in an ultra-low-attachment 96-well plate (Corning) in stem cell-specific medium and counted weekly upon separation by digestion.
Single-cell suspensions were plated on 0.1% gelatin-coated plates at a density of 2,000 viable cells/10-cm-diameter dish. Cells were grown in luminal medium, and after 48 h, cell images were captured using a Floid cell imaging station (Life Technologies, NY). Next, differentiated cells were characterized by immunofluorescence or fluorescence-activated cell sorter (FACS) analysis using luminal and myoepithelial markers.
To test alveolar differentiation, cells were differentiated as described above, and after 5 days, a layer of Matrigel was added along with prolactin (1 μg/ml). The complex acinar and branched acinar structures that formed after 3 days of plating were imaged using the Floid cell imaging station (Life Technologies).
Total RNA was prepared using the miRNeasy minikit (Qiagen, UK) according to the manufacturer’s instructions and treated with RNase-free DNase. RNA quality and yield were assessed using an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA), and samples yielding an RNA integrity number (RIN) of ≥8 were considered for global expression profiling performed in duplicates for each sample using Affymetrix GeneChip human genome U133 plus 2.0 arrays (Affymetrix, Santa Clara, CA) according to the manufacturer’s instructions. Briefly, 200 ng of RNA was reverse transcribed to synthesize first-strand cDNA with the help of oligo(dT) primers containing a T7 flanking sequence. Following second-strand synthesis, the in vitro transcription step reaction was carried out with labeled amplified RNA (aRNA). Following purification and fragmentation of aRNA, the samples were hybridized overnight onto U133 plus 2.0 human genome arrays. Finally, after washing to remove the unbound transcripts, the hybridized microarrays were scanned, and intensity (cel) files with the acquisition and initial quantification of array images were generated using Transcriptome Analysis Console (TAC) (Applied Biosystems, NY). Robust multichip average (RMA) analysis for quality control was used to filter and eliminate genes displaying an averaged intensity inferior to the global array background. Significantly modulated genes were defined as those with an absolute fold change (FC) of >2 and an adjusted P value of <0.05, while one-way analysis of variance (ANOVA) was used to calculate the P values between the two groups. The identification of differentially expressed genes was achieved by using scatterplots and hierarchical clustering analysis. P values of less than 0.05 were considered statistically significant.
Cellular lysate preparation and immunoblotting.
Cellular lysate preparation and immunoblotting were performed as previously described (43). Antibodies are listed in Table 2.
TABLE 2 Antibodies utilized for immunoblot, immunofluorescence, and immunohistochemistry analyses
Cells and mammospheres were fixed with 4% paraformaldehyde for 10 min, permeabilized with 0.2% Triton X-100 for 10 min, and quenched in 100 mM glycine for 10 min at room temperature. Cells were then blocked in 10% fetal calf serum (FCS) and incubated with the primary antibody. Cells were then washed and incubated with Alexa Fluor 594- and 488-conjugated antibodies, respectively. Nuclei were stained with DAPI (4′,6-diamidino-2-phenylindole), and images were captured using the Floid cell imaging station (Life Technologies). Antibodies are listed in Table 2.
Cell migration, invasion, and proliferation.
Cell migration and invasion were assessed according to the manufacturer’s instructions. In brief, 1 × 104 cells in serum-free medium were added to the upper wells of the cell invasion/migration (CIM) plate with (for invasion) or without (for migration) a thin layer of Matrigel basement membrane matrix, and complete medium was added to the lower chamber wells as a chemoattractant. Subsequently, the plates were incubated in the real-time cell analyzer (RTCA) system for 24 h, and the impedance value of each well was automatically monitored by the xCELLigence system and expressed as a cell index (CI) value, which represents cell status based on the measured electrical impedance change divided by the background value. Each assay was performed in triplicate.
For the proliferation assay, exponentially growing cells were seeded in an E-plate (1 × 104 cells) with complete medium according to the manufacturer’s instructions. All data were recorded and analyzed by using RTCA software. The CI was used to measure the change in the electrical impedance divided by the background value to represent cell status. Each assay was performed in triplicate.
Cells were incubated for 30 min on ice with directly labeled antibodies. Cells were then washed with phosphate-buffered saline (PBS), followed by fixation in 0.5% paraformaldehyde in PBS. Data were acquired using an LSR II flow cytometer using BD FACSDiva operating software. Positive staining was considered based on the negativity of an isotype control. A minimum of 10,000 events were recorded for all samples. Antibodies are listed in Table 2.
siRNA transfection/viral infection.
Two different sequences for STAT3 siRNAs were utilized: Signal Silence STAT3 siRNA and control siRNA were obtained from Cell Signaling Technology. Another one was obtained from Ambion (catalog number s745), with the sequence GCA CCU UCC UGC UAA GAU UTT AAU CUU AGC AGG AAG GUG CCT. The transfections were carried out using the High Perfect reagent (Qiagen) as recommended by the manufacturer.
pLKO.1-miRZip141 (inhibitor of miR-141), pCDH-miR-141 (expressing pre-miR-141) (System Biosciences, CA), and the corresponding control plasmids were used at 1 μg/ml each for the transfection of 293FT cells. Lentiviral supernatants were collected at 48 h posttransfection. Culture media were removed from the target cells and replaced with the lentiviral supernatant, and the mixture was incubated for 24 h in the presence of 1 μg/ml Polybrene (Sigma-Aldrich).
RNA purification and qRT-PCR.
Total RNA containing miRNAs was purified using the miRNeasy minikit (Qiagen, UK) according to the manufacturer’s instructions and treated with RNase-free DNase. One microgram of RNA was used to synthesize cDNA by utilizing either an Advantage RT-PCR kit (Clontech Laboratories, Mountain View, CA) or a miScript II RT kit (Qiagen, UK) for mature miRNAs. qRT-PCR was performed in triplicate using 4 μl of cDNA mixed with 2× FastStart essential DNA green quantitative PCR (qPCR) master mix (Roche, New York, NY) and 0.3 μM forward and reverse primers. Amplifications were performed by utilizing the LightCycler 96 real-time PCR detection system (Roche) under the following cycle conditions: 95°C for 10 min (1 cycle) followed by 95°C for 10 s, 59°C for 20 s, and 72°C for 30 s (45 cycles). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) expression levels were used for normalization, and gene expression differences were calculated using the threshold cycle (CT). The obtained values were plotted as means ± standard deviations (SD). Three independent experiments were performed for each reaction. (See Table 3 for the sequences of the primers used.)
TABLE 3 Sequences of the primers utilized for qRT-PCR
An Alvatex scaffold was utilized, according to the manufacturer’s instructions (Amsbio, MA), to allow the formation of multilayered high-density cell populations. NBL-10 cells were seeded at 1 × 106 cells per well prior to exposure to either SFM, SFM containing IL-8 (25 ng), or SFCM from SLAF or ILAF for 24 h. Cultures were maintained in growing medium for 2 weeks. The discs were then fixed in 10% formalin, paraffin embedded, sectioned, and utilized for either hematoxylin and eosin staining or immunofluorescence.
Immunohistochemistry was done on formalin-fixed, paraffin-embedded (FFPE) tissue as previously described (44), using specific antibodies overnight at a dilution of 1:500. Color was developed with 3,3′-diaminobenzidine (DAB), and instant hematoxylin was used for counterstaining.
Statistical analysis was performed using Student’s t test, and two-sided P values of ≤0.05 were considered statistically significant. For microarray experiments, ANOVA was utilized for statistical analysis.
Protein quantification and mammosphere sizes were quantified using ImageJ software, FACS profile analysis was performed using BD FACSDiva operating software, and all statistical analyses were performed using GraphPad Prism v5 software.
We are thankful to the Research Center administration for their continuous help and support.
This work was supported by the King Faisal Specialist Hospital and Research Centre under RAC proposal 2090027.
Colacino JA, Azizi E, Brooks MD, Harouaka R, Fouladdel S, McDermott SP, Lee M, Hill D, Madden J, Boerner J, Cote ML, Sartor MA, Rozek LS, Wicha MS. 2018. Heterogeneity of human breast stem and progenitor cells as revealed by transcriptional profiling. Stem Cell Rep 10:1596–1609.
Dong Q, Gao H, Shi Y, Zhang F, Gu X, Wu A, Wang D, Chen Y, Bandyopadhyay A, Yeh IT, Daniel BJ, Chen Y, Zou Y, Rebel VL, Walter CA, Lu J, Huang C, Sun LZ. 2016. Aging is associated with an expansion of CD49fhi mammary stem cells that show a decline in function and increased transformation potential. Aging 8:2754–2776.
Garbe JC, Pepin F, Pelissier FA, Sputova K, Fridriksdottir AJ, Guo DE, Villadsen R, Park M, Petersen OW, Borowsky AD, Stampfer MR, Labarge MA. 2012. Accumulation of multipotent progenitors with a basal differentiation bias during aging of human mammary epithelia. Cancer Res 72:3687–3701.
Mosteiro L, Pantoja C, Alcazar N, Marión RM, Chondronasiou D, Rovira M, Fernandez-Marcos PJ, Muñoz-Martin M, Blanco-Aparicio C, Pastor J, Gómez-López G, De Martino A, Blasco MA, Abad M, Serrano M. 2016. Tissue damage and senescence provide critical signals for cellular reprogramming in vivo. Science 354:aaf4445.
Guo W, Keckesova Z, Donaher JL, Shibue T, Tischler V, Reinhardt F, Itzkovitz S, Noske A, Zurrer-Hardi U, Bell G, Tam WL, Mani SA, van Oudenaarden A, Weinberg RA. 2012. Slug and Sox9 cooperatively determine the mammary stem cell state. Cell 148:1015–1028.
Mani SA, Guo W, Liao MJ, Eaton EN, Ayyanan A, Zhou AY, Brooks M, Reinhard F, Zhang CC, Shipitsin M, Campbell LL, Polyak K, Brisken C, Yang J, Weinberg RA. 2008. The epithelial-mesenchymal transition generates cells with properties of stem cells. Cell 133:704–715.
Al-Khalaf HH, Mohideen P, Nallar SC, Kalvakolanu DV, Aboussekhra A. 2013. The cyclin-dependent kinase inhibitor p16INK4a physically interacts with transcription factor Sp1 and cyclin-dependent kinase 4 to transactivate microRNA-141 and microRNA-146b-5p spontaneously and in response to ultraviolet light-induced DNA damage. J Biol Chem 288:35511–35525.
Hendrayani SF, Al-Khalaf HH, Aboussekhra A. 2014. The cytokine IL-6 reactivates breast stromal fibroblasts through transcription factor STAT3-dependent up-regulation of the RNA-binding protein AUF1. J Biol Chem 289:30962–30976.
Ghebeh H, Sleiman GM, Manogaran PS, Al-Mazrou A, Barhoush E, Al-Mohanna FH, Tulbah A, Al-Faqeeh K, Adra CN. 2013. Profiling of normal and malignant breast tissue show CD44high/CD24low phenotype as a predominant stem/progenitor marker when used in combination with Ep-CAM/CD49f markers. BMC Cancer 13:289.
Poli V, Fagnocchi L, Fasciani A, Cherubini A, Mazzoleni S, Ferrillo S, Miluzio A, Gaudioso G, Vaira V, Turdo A, Gaggianesi M, Chinnici A, Lipari E, Bicciato S, Bosari S, Todaro M, Zippo A. 2018. MYC-driven epigenetic reprogramming favors the onset of tumorigenesis by inducing a stem cell-like state. Nat Commun 9:1024.
Panciera T, Azzolin L, Fujimura A, Biagio DD, Frasson C, Bresolin S, Soligo S, Basso G, Bicciato S, Rosato A, Cordenonsi M, Piccolo S. 2016. Induction of expandable tissue-specific stem/progenitor cells through transient expression of YAP/TAZ. Cell Stem Cell 19:725–737.
Liu C, Liu R, Zhang D, Deng Q, Liu B, Chao HP, Rycaj K, Takata Y, Lin K, Lu Y, Zhong Y, Krolewski J, Shen J, Tang DG. 2017. MicroRNA-141 suppresses prostate cancer stem cells and metastasis by targeting a cohort of pro-metastasis genes. Nat Commun 8:14270.
Charafe-Jauffret E, Ginestier C, Iovino F, Wicinski J, Cervera N, Finetti P, Hur MH, Diebel ME, Monville F, Dutcher J, Brown M, Viens P, Xerri L, Bertucci F, Stassi G, Dontu G, Birnbaum D, Wicha MS. 2009. Breast cancer cell lines contain functional cancer stem cells with metastatic capacity and a distinct molecular signature. Cancer Res 69:1302–1313.
Chiche A, Roux IL, von Joest M, Sakai H, Aguin SB, Cazin C, Salam R, Fiette L, Alegria O, Flamant P, Tajbakhsh S, Li H. 2017. Injury-induced senescence enables in vivo reprogramming in skeletal muscle. Cell Stem Cell 20:407–414.e4.
Ocampo A, Reddy P, Martinez-Redondo P, Platero-Luengo A, Hatanaka F, Hishida T, Li M, Lam D, Kurita M, Beyret E, Araoka T, Vazquez-Ferrer E, Donoso D, Roman JL, Xu J, Rodriguez Esteban C, Nuñez G, Nuñez Delicado E, Campistol JM, Guillen I, Guillen P, Izpisua Belmonte JC. 2016. In vivo amelioration of age-associated hallmarks by partial reprogramming. Cell 167:1719–1733.e12.
Van Keymeulen A, Lee MY, Ousset M, Brohee S, Rorive S, Giraddi RR, Wuidart A, Bouvencourt G, Dubois C, Salmon I, Sotiriou C, Phillips WA, Blanpain C. 2015. Reactivation of multipotency by oncogenic PIK3CA induces breast tumour heterogeneity. Nature 525:119–123.
Ghebeh H, Tulbah A, Mohammed S, Elkum N, Bin Amer SM, Al-Tweigeri T, Dermime S. 2007. Expression of B7-H1 in breast cancer patients is strongly associated with high proliferative Ki-67-expressing tumor cells. Int J Cancer 121:751–758.
Hawsawi NM, Ghebeh H, Hendrayani SF, Tulbah A, Al-Eid M, Al-Tweigeri T, Ajarim D, Alaiya A, Dermime S, Aboussekhra A. 2008. Breast carcinoma-associated fibroblasts and their counterparts display neoplastic-specific changes. Cancer Res 68:2717–2725.
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.