2022 年 45 巻 5 号 p. 649-658
Growing evidence suggests that cancer originates from cancer stem cells (CSCs), which can be identified by aldehyde dehydrogenase (ALDH) activity-based flow cytometry. However, the regulation of CSC growth is not fully understood. In the present study, we investigated the effects of Transforming Growth Factor-β (TGFβ) in breast CSC expansion. Stimulation with TGFβ increased the ALDH-positive breast CSC population via the phosphorylation of sphingosine kinase 1 (SphK1), a sphingosine-1-phosphate (S1P)-producing enzyme, and subsequent S1P-mediated S1P receptor 3 (S1PR3) activation. These data suggest that TGFβ promotes breast CSC expansion via the ALK5/SphK1/S1P/S1PR3 signaling pathway. Our findings provide new insights into the role of TGFβ in the regulation of CSCs.
Growing evidence suggests that various types of cancers, such as breast cancer and lung cancer, are initiated from a small population of cancer stem cells (CSCs; also called tumor-initiating cells).1) This minor population is considered to produce the bulk of cancer cells by self-renewal and differentiation, which result in cancer heterogeneity. It is essential to elucidate the signaling pathways that are involved in CSC regulation and to identify novel therapeutic approaches targeting CSCs.
Transforming Growth Factor-β (TGFβ) has been shown to be a bifunctional regulator during tumorigenesis, which involves a tumor suppression in early stages and a tumor progression via CSC regulation and epithelial–mesenchymal transition (EMT) in later stages of cancer.2) TGFβ exerts its cellular effects via TGFβ type I receptor kinase (ALK5).3) TGFβ–ALK5 signaling functions via Smad-dependent canonical pathways. TGFβ–ALK5 is involved in the phosphorylation and activation of Smad2/3, which subsequently bind to Smad4 to form complexes that are translocated to the nucleus and regulate transcription.4) In addition, TGFβ–ALK5 induces the activation of Smad-independent noncanonical pathways, such as the extracellular signal-regulated kinase (ERK) and RhoA–Rho-associated protein kinase 1 (Rock1). Mitogen-activated protein kinase pathways have direct effects on TGFβ-induced CSC properties, such as migration and invasion.5) RhoA–Rock1 interactions with TGFβ are required for the EMT.6) However, the molecular mechanism by which the TGFβ–ALK5 pathway regulates CSCs has not been fully elucidated.
TGFβ crosstalk with the bioactive lipid mediator sphingosine-1-phosphate (S1P) via ERK-sphingosine kinase 1 (SphK1), an S1P-producing enzyme from sphingosine (Sph), is involved in the migration and invasion of esophageal cancer cells.7) S1P is known to regulate various biological processes, including proliferation, survival, and cytoskeletal rearrangement, via G protein-coupled receptors 1 to 5 (S1PR1 to S1PR5) in many cell types.8)
We have previously demonstrated that S1P regulates the expansion of CSCs via S1PR3 in several types of cancer, including breast cancer.9) We also found that the self-renewal signaling pathway Notch is essential for the S1P-induced proliferation of CSCs via an ADAM17-dependent shedding mechanism.
In this study, we evaluated the contribution of crosstalk between TGFβ and S1P signaling to breast CSC regulation. Our findings suggest that Notch activation mediates the TGFβ-induced proliferation of breast CSCs via ALK5–ERK–SphK1–S1P–S1PR3–Notch. Thus, these results suggest that the TGFβ–ALK5-induced S1P signaling pathway is a therapeutic target in breast cancer.
Antibodies to ERK, phospho-ERK, and N1ICD were obtained from Cell Signaling Technology (Danvers, MA, U.S.A.). An antibody to SphK1 was obtained from Abgent (San Diego, CA, U.S.A.). An antibody to phospho-SphK1 was obtained from ECM Biosciences (Versailles, KY, U.S.A.). Antibodies to hemagglutinin (HA), FLAG (M2), myc, and β-actin were obtained from Sigma-Aldrich (St. Louis, MO, U.S.A.). An antibody to SphK2 was kindly provided by Dr. Okada (Kobe University). Antibodies to fluorescein isothiocyanate (FITC) mouse anti-human CD44, PE mouse anti-human CD24, FITC mouse immunoglobulin G (IgG)2b κ isotype control, and PE mouse IgG2a κ isotype control were obtained from BD Biosciences (Franklin Lakes, NJ, U.S.A.). Antibodies to S1PR2 and S1PR3 were obtained from Santa Cruz Biotechnology (Santa Cruz, CA, U.S.A.). Antibodies against SphK1 and ABCC1 were purchased from Abcam (Cambridge, U.K.). TGFβ was obtained from PEPROTECH (Rocky Hill, NJ, U.S.A.). SB431542 was purchased from STEMGENT (Cambridge, MA, U.S.A.). U0126 and DAPT were obtained from Enzo Life Sciences (Farmingdale, NY, U.S.A.). DMS, CAY10444, JTE013, and MK571 were obtained from Cayman Chemicals (Ann Arbor, MI, U.S.A.). SKI-II was obtained from Sigma-Aldrich. Pertussis toxin (PTX) was obtained from Wako Pure Chemical Corporation (Osaka, Japan). Double-strand RNA oligonucleotides (small interfering RNAs (siRNAs)) against SphK1, SphK2, ABCC1 and appropriate control scrambled siRNA were obtained from Invitrogen. The siRNAs against S1PR3 and S1PR2 were obtained from Santa Cruz Biotechnology. All other reagents were of analytical grade and were obtained from commercial sources.
Cell CultureMCF-7 cells (American Type Culture Collection, Manassas, VA, U.S.A.) were cultured in Dulbecco’s modified Eagle’s medium (DMEM; Sigma-Aldrich) supplemented with 10% heat-inactivated fetal bovine serum (FBS; Hyclone, Logan, UT, U.S.A.).
PlasmidPlasmids encoding ALK5TD, ALK5KR, Smad were previously reported.10) Plasmids encoding DN-Sphk1 were kindly provided by Dr. Stuart M. Pitson (University of Adelaide). Plasmids encoding DN-SphK1 were kindly provided by Taro Okada (Kobe University). Plasmids encoding DN-ADAM17 were previously reported.9)
Aldehyde Dehydrogenase (ALDH) AssaysCSCs was detected as ALDH-positive cells using ALDEFLUOR Kit (Stem Cell Technologies, Vancouver, Canada), as previously reported.11)
Mammosphere Formation AssaysMammosphere formation assay was performed as previously described.11) Formation of mammospheres were counted using Olympus IX71 (Olympus, Tokyo, Japan).
CD44+/CD24− Cell PopulationA CD44+/CD24− cell population assay was performed as previously described.9) The antibodies were used at dilution level as follows; FITC-conjugated mouse anti-human CD44 (1 : 5), PE-conjugated mouse anti-human CD24 (1 : 5), FITC-conjugated mouse IgG2b κ isotype control (1 : 5) and PE-conjugated mouse IgG2a κ isotype control (1 : 5). CD44+/CD24− cell population was detected using a FACS Aria II Cell Sorter (BD Biosciences).
Quantitative PCR (qPCR)The qPCR was performed using a QuantiTect SYBR Green RT-PCR Kit (QIAGEN, Hilden, Germany) and an QuantStudio 7 Flex Real-Time PCR System (Applied Biosystems, Foster City, CA, U.S.A.) as described previously.12,13) The relative expression of each gene was calculated using glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA levels as internal control. The primer sequences were shown in Table 1.
| Gene | Forward | Reverse |
|---|---|---|
| PAI-1 | AATGTGTCATTTCCGGCTGCTGTG | ACATCCATCTTTGTGCCCTACCCT |
| SphK1 | TCCTGGCACTGCTGCACTC | TAACCATCAATTCCCCATCCAC |
| SphK2 | GCTCAACTGCTCACTGTTGC | GCAGGTCAGACACAGAACGA |
| Hes1 | AGCGGGCGCAGATGAC | CGTTCATGCACTCGCTGAA |
| LPAR5 | CGCCATCTTCCAGATGAAC | TAGCGGTCCACGTTGATG |
| LPAR6 | TCTGGCAATTGTCTACCCATT | TCAAAGCAGGCTTCTGAGG |
| Gli1 | GTGCAAGTCAAGCCAGAACA | ATAGGGGCCTGACTGGAGAT |
| Dkk1 | GGGCGGGAATAAGTACCAG | CATAGCGTGACGCATGCAG |
| S1PR3 | ATCCTGCCCCTCTACTCCAAGA | GTGCGTAGAGGATCACGATGGT |
| S1PR2 | ATTACTTTAACTGGTAGGGAACG | AAGACATCTCTCGGTTTAATTGC |
| GAPDH | GTCTCCTCTGACTTCAACAGCG | ACCACCCTGTTGCTGTAGCCAA |
An immunoblot analysis was performed, as previously reported.9) The primary antibodies were used at dilution level as follows; ERK (1 : 1000), phospho-ERK (1 : 1000), SphK1 (1 : 500), phosphor-SphK1 (1 : 2000), FLAG (1 : 1000), HA (1 : 1000), β-actin (1 : 10000), SphK2 (1 : 3000), S1PR3 (1 : 2000), S1PR2 (1 : 1000), ABCC1 (1 : 1000), NICD (1 : 1000), and myc (1 : 5000).
Analysis of Enzymatic ActivitiesAnalysis of SphK, ADAM17 and γ-secretase activity was performed as previously described.9) Sphk activities were measured using omega-(7-nitro-2-1,3-benzoxadiazol-4-yl)-D-erythro-sphingosine (NBD)-labeled fluorescent substrate (NBD-sphingosine, Avanti Polar Lipids, Alabaster, AL, U.S.A.). ADAM17 activity was measured using a SensoLyte 520 ADAM17 Activity Assay Kit (ANASPEC, Fremont, CA, U.S.A.). γ-Secretase activity was measured using a fluorogenic γ-secretase substrate (Peptide Inc., Osaka, Japan).
Statistical AnalysisResults are shown as means ± standard deviation (S.D.). Student’s t-tests were used to analyze data. p < 0.05 was considered significant.
Many cancer cell lines, including MCF-7 cells, contain an ALDH-positive cell population.11) We previously confirmed that ALDH-positive MCF-7 cells possess CSC-like properties, as determined by the expression of stem cell markers, drug resistance, and tumorigenicity.9) Since TGFβ has been considered to have dual effects in cancer, we investigated the effect of TGFβ on CSC regulation of MCF-7 cells. Stimulation with TGFβ increased the proportion of ALDH-positive MCF-7 cells in a dose-dependent manner, with a maximal response at 5 ng/mL (Fig. 1A). Moreover, the increase in CSCs by TGFβ was confirmed by analyses of mammosphere formation efficiency (Fig. 1B) and the CD44+/CD24− population (Fig. 1C). Furthermore, a selective inhibitor of ALK5 (a TGFβ receptor), SB431542, blocked the effect of TGFβ (Fig. 1D). To evaluate downstream signaling mechanisms, we next examined whether Smad mediates the TGFβ-induced proliferation of breast CSCs. TGFβ induced the downstream target gene Plasminogen-activating inhibitor 1 (PAI-1) via ALK5, which was inhibited by SB431542 (Fig. 1E). The overexpression of constitutively active mutants of ALK5 (ALK5TD) increased the proportion of ALDH-positive cells, whereas the overexpression of Smad2, 3, or 4 did not induce proliferation (Fig. 1F). We confirmed that the overexpression of Smad2, 3, 4, and ALK5TD induced PAI-1 expression (Fig. 1G). In addition, TGFβ did not induce ALDH-positive cell proliferation in dominant-negative ALK5 (ALK5KR)-expressing cells (Fig. 1H). These data suggest that TGFβ promotes the expansion of breast CSCs via ALK5 in a Smad-independent manner.
(A) Dose-dependent effect of TGFβ on the proportion of ALDH-positive cells. (B) Effect of TGFβ (5 ng/mL) on mammosphere-formation efficiency (%) in MCF-7 cells. Scale bar indicates 100 µm. (C) Effect of TGFβ (5 ng/mL) on the CD44+/CD24− population in MCF-7 cells. (D) Effect of an ALK5 inhibitor (SB431542, 10 µM) on the TGFβ-induced increase in the ALDH-positive cell population. (E) After stimulation with TGFβ and/or SB431542 for 24 h, the TGFβ target gene PAI-1 was evaluated by real-time RT-PCR. (F) Effect of the overexpression of ALK5TD, Smad2, Smad3, or Smad4 on the ALDH-positive cell population. (G) Effect of the overexpression of ALK5TD, Smad2, Smad3, or Smad4 on PAI-1 expression. (H) Effect of the overexpression of ALK5KR on TGFβ-induced increase in the ALDH-positive cell population. Data are presented as means ± S.D. (n = 3). * p < 0.05.
TGFβ–ALK5 has been reported to induce the activation of Smad-independent noncanonical pathways, including ERK.5) Since ERK is involved in TGFβ-induced cancer metastasis,5) we focused on the role of ERK signaling in the TGFβ-mediated pathway in CSC expansion. We found that TGFβ increased basal ERK phosphorylation levels, which were diminished by the administration of the ERK inhibitor U0126 (Fig. 2A). ERK phosphorylation was also increased by ALK5TD overexpression. TGFβ-induced ALDH-positive cell proliferation was suppressed by U0126 (Fig. 2B). To further investigate the crosstalk between TGFβ–ALK5–ERK and S1P signaling, we examined SphK activity after TGFβ stimulation. We found that TGFβ upregulated SphK1 activity, whereas SphK2 was not affected (Fig. 2C). Since TGFβ did not induce SphK1 gene expression (Fig. 2D), we investigated other SphK1 activation mechanism. Both SB431542 and U0126 abolished the upregulation of SphK1 activity by TGFβ (Fig. 2E). The overexpression of ALK5TD upregulated SphK1 activity, whereas Smad2, 3, and 4 had no effect (Fig. 2F). SB431542 and U0126 abolished the upregulation of SphK1 activity via ALK5TD (Fig. 2G). To confirm the upregulation of SphK1 activity by TGFβ signaling, we examined the phosphorylation of SphK1 by immunoblotting. Stimulation with TGFβ or the overexpression of ALK5TD increased SphK1 phosphorylation levels, which were diminished by U0126 (Figs. 2H–J). These data suggest that TGFβ–ALK5 signaling activates SphK1 via ERK phosphorylation.
(A) After stimulation with TGFβ and/or an ERK inhibitor (U0126, 5 µM) for 24 h or the overexpression of ALK5TD, ERK phosphorylation was examined by immunoblotting. (B) Effect of U0126 on the TGFβ-induced increase in the ALDH-positive cell population. (C) After stimulation with TGFβ for 24 h, SphK1 and SphK2 levels were measured. (D) Effect of TGFβ on SphK1 expression. (E) Effect of SB431542 or U0126 on the TGFβ-induced upregulation of SphK1 activity. (F) After ALK5TD, Smad2, Smad3, or Smad4 overexpression, SphK1 activity was measured. (G) Effect of SB431542 or U0126 on the ALK5-mediated upregulation of SphK1 activity. (H) After stimulation with TGFβ for 120 min, SphK1 phosphorylation was examined by immunoblotting. (I) Effect of U0126 on the TGFβ-induced upregulation of SphK1 phosphorylation. (J) After the overexpression of ALK5TD and/or stimulation with U0126 for 24 h, SphK1 phosphorylation was examined by immunoblotting. Data represent the mean ± S.D. (n = 3). * p < 0.05.
To investigate whether SphK1 is involved in TGFβ-induced breast CSC expansion, we used two SphK1 inhibitors, DMS and SKI-II. Both inhibitors abolished TGFβ-induced ALDH-positive cells (Fig. 3A). We examined the effects of dominant-negative SphK (DN-SphK) using MCF7 cells stably transfected with FLAG-DN-SphK1 and HA-DN-SphK2 (Fig. 3B). The overexpression of DN-SphK1, but not DN-SphK2, abolished the TGFβ-induced increase in ALDH-positive cells (Fig. 3C). Consistent with these findings, SphK1 knockdown by siRNA inhibited TGFβ-induced ALDH-positive cell proliferation (Figs. 3D, E). In contrast, SphK2 knockdown had little effect. These data suggest that SphK1 activation selectively mediates TGFβ-induced breast CSC expansion.
(A) Effect of SphK1 inhibitors (DMS, 10 µM; SKI-II, 10 µM) on TGFβ-induced increase in the ALDH-positive cell population. (B) Overexpression of FLAG-DN-SphK1 or HA-DN-SphK2 in MCF-7 cells was confirmed by immunoblotting. (C) Effect of the overexpression of DN-SphK1 or DN-SphK2 on the TGFβ-induced increase in the ALDH-positive cell population. (D) After transfection with siRNA, expression levels of SphK1 and SphK2 were examined by real-time PCR and immunoblotting. (E) Effects of siRNAs against SphK1 on the TGFβ-induced increase in the ALDH-positive cell population. Data are presented as means ± S.D. (n = 3). * p < 0.05.
We have previously reported that S1P increases CSCs in MCF-7 cells via S1PR3.9) To investigate whether TGFβ-induced breast CSC expansion via SphK1 involves S1PR3 activation, we inhibited S1PR3 using a pharmacological antagonist and RNA interference. The effect of TGFβ was blocked by the S1PR3 antagonist CAY10444 (Fig. 4A). In contrast, the S1PR2 antagonist JTE013 had little effect. Similar results were obtained by siRNA-mediated knockdown (Figs. 4B, C). Since it has been reported that TGFβ upregulated S1PR3 expression via Smad3 in lung epithelial cells HBEC2-KT,14) we examined whether TGFβ induces S1PR3 in MCF-7. TGFβ downregulated S1PR3 expression and slightly upregulated S1PR2 expression in MCF-7 (Fig. 4D). These results confirmed that TGFβ induces the proliferation of breast CSCs in Smad-independent manner. PTX, which inactivates Gi protein, also inhibited the effect of TGFβ, suggesting the importance of SiPR3-Gi (Fig. 4E). Recent studies have suggested that S1P is synthesized intracellularly, secreted by ABC transporters, and binds to its receptor from outside of cells. The ABCC1 inhibitor MK571 inhibited the TGFβ-induced increase of ALDH-positive cells (Fig. 4F). Similar results were obtained using siRNA against ABCC1 (Figs. 4G, H). These data suggest that S1P–S1PR3 signaling is involved in TGFβ-induced breast CSC expansion.
(A) Effects of an S1PR3 antagonist (CAY10444, 10 µM) and S1PR2 antagonist (JTE013, 10 µM) on the TGFβ-induced increase in the ALDH-positive cell population. (B) After transfection with siRNA, S1PR2 and S1PR3 levels were examined by immunoblotting. (C) Effects of siRNAs against S1PR3 and S1PR2 on the TGFβ-induced increase in the ALDH-positive cell population. (D) Effects of TGFβ on S1PR3 and S1PR2 expression. (E) Effects of toxin (PTX, 0.1 µg/mL) on the TGFβ-induced increase in the ALDH-positive cell population. (F) Effects of an ABCC1 inhibitor (MK571, 5 µM) on the TGFβ-induced increase in the ALDH-positive cell population. (G) After transfection with siRNA, ABCC1 levels were examined by immunoblotting. (H) Effects of siRNA against ABCC1 on the TGFβ-induced increase in the ALDH-positive cell population. Data are presented as means ± S.D. (n = 3). * p < 0.05.
There are many similarities between embryonic stem cells and CSCs. We have previously reported that S1P enhances Notch signaling via S1PR3 in MCF-7 cells.9) Thus, we focused on the Notch pathway as a downstream candidate in the TGFβ–ALK5–ERK–SphK1–S1PR3 axis. Stimulation with TGFβ induced the expression of the Notch target gene Hes1 in MCF-7 cells (Fig. 5A). In contrast, the Hedgehog target gene Gli1 and Wnt target gene Dkk1 were not induced. The induction of Hes1 was also observed by ALK5TD overexpression (Fig. 5B). In contrast, the overexpression of Smad2, 3 and 4 had little effect. To determine whether TGFβ has the ability to activate Notch, we examined the cleavage of Notch in MCF-7 cells. Stimulation with TGFβ resulted in Notch intracellular domain (NICD) production (Fig. 5C). TGFβ-induced Hes1 expression and NICD production were suppressed by SB431542 and U0126, suggesting that Notch signaling acts downstream of TGFβ-ALK5-ERK (Figs. 5C, D). TGFβ-induced Hes1 expression was further suppressed by DMS, SKI-II, MK571, CAY10444, and PTX (Figs. 5E, G, I). The overexpression of DN-SphK1, SphK1 (siSphK1) siRNA, and S1PR3 (siS1PR3) siRNA confirmed the involvement of Notch signaling in the TGFβ–ALK5–ERK–SphK1–S1P–S1PR3 axis (Figs. 5F, H). The Notch inhibitor DAPT abolished TGFβ-induced ALDH-positive cell proliferation (Fig. 5J). These data suggest that Notch signaling mediates the effect of TGFβ via ERK-SphK1-S1P-S1PR3.
(A) After stimulation with TGFβ for 24 h, expression levels of a Notch target gene (Hes1), Hedgehog target gene (Gli1) and Wnt target gene (Dkk1) were measured by real-time PCR. (B) After stimulation with TGFβ for 24 h, Hes1, Gli1, and Dkk1 expression levels were measured by real-time PCR. (C) Effects of TGFβ, SB431542, and/or U0126 on N1ICD production by immunoblotting. (D) Effects of SB431542 or U0126 on the TGFβ-induced upregulation of Hes1 expression. (E) Effects of DMS or SKI-II on the TGFβ-induced upregulation of Hes1 expression. (F) Effects of DN-SphK1 overexpression on the TGFβ-induced upregulation of Hes1. (G) Effects of MK571 on the TGFβ-induced upregulation of Hes1. (H) Effects of siRNAs against SphK1 and S1PR3 on the TGFβ-induced up-regulation of Hes1. (I) Effects of CAY10444 or PTX on the TGFβ-induced upregulation of Hes1. (J) Effects of a Notch inhibitor (DAPT, 5 µM) on the TGFβ-induced increase in the ALDH-positive cell population. Data are presented as means ± S.D. (n = 3). * p < 0.05.
We further investigated the crosstalk between TGFβ and Notch in MCF-7 cells. We have previously shown that S1P induces Notch activation, with cleavage by ADAM17 and γ-secretase, in MCF-7 cells.9) Thus, we investigated whether ADAM17 and γ-secretase are also responsible for TGFβ-induced Notch activation. We found that TGFβ increased ADAM17 and γ-secretase activities, which were inhibited by SKI-II, CAY10444, and PTX (Figs. 6A, B). Furthermore, DN-ADAM17 (E406A; with a point mutation in the metalloprotease domain) inhibited the TGFβ-induced upregulation of Hes1 expression and increase in ALDH-positive cells (Figs. 6C–E). These data suggest that ADAM17 and γ-secretase are involved in TGFβ-induced breast CSC proliferation via Notch activation.
After stimulation with TGFβ and/or several inhibitors (SKI-II, CAY10444, and PTX) for 24 h, levels of ADAM17 (A) and γ-secretase (B) were measured. (C) Overexpression of myc-DN-ADAM17 in MCF-7 cells was confirmed by immunoblotting. (D) Effects of DN-ADAM17 overexpression on the TGFβ-induced upregulation of Hes1. (E) Effects of DN-ADAM17 overexpression on the TGFβ-induced increase in the ALDH-positive cell population. Data are presented as means ± S.D. (n = 3). * p < 0.05.
In the present study, we provide evidence for crosstalk between TGFβ and S1P signaling via SphK1 activation and the subsequent proliferation of breast CSCs (Fig. 7). Notch activation is essential for the TGFβ-induced proliferation of breast CSCs via ALK5–SphK1–S1P–S1PR3. These findings suggest that the TGFβ–ALK5-induced S1P signaling pathway is an important therapeutic target in breast CSCs.
Crosstalk between TGFβ and S1P plays a role in breast CSC regulation. Our findings suggest that the TGFβ-induced proliferation of breast CSCs is involved in Notch activation via an ALK5/ERK/ADAM17-dependent pathway. The abbreviations are as follow: TGFβ (transforming growth factor-β), ALK5 (activin receptor-like kinase 5), ERK (extracellular signal-regulated kinase), SphK1 (sphingosine kinase 1), sph (sphingosine), S1P (sphingosine-1-phosphate), ABCC1 (ATP-binding cassette subfamily C member 1), S1PR3 (S1P receptor 3), ADAM17 (A disintegrin and metalloprotease 17), MAML (mastermind like transcriptional coactivator), CSL (CBF1/Su(H)/Lag2 protein).
Crosstalk between TGFβ and S1P had a critical role in breast CSC expansion. TGFβ crosstalk with S1P signaling via SphK1 activation contributes to esophageal cancer progression and lung myofibroblast differentiation.7,15) However, crosstalk between these factors has not been implicated in the regulation of stem cells, including CSCs. S1P might have self-renewal properties and play a key role in stem cell regulation downstream of TGFβ signaling.
TGFβ induced breast CSC expansion via S1P–S1PR3–ADAM17-mediated Notch activation. TGFβ increases breast CSCs by promoting EMT.16) Concerning EMT regulation, TGFβ also upregulates the Notch-ligand Jagged1, followed by Notch activation and the induction of EMT.17) We observed a role of TGFβ-induced Notch activation without the Notch ligand in CSC regulation. We have previously shown that S1P-S1PR3 signaling promotes breast CSC expansion via ligand-independent Notch activation by ADAM17.9) Thus, the crosstalk between TGFβ and S1P plays a novel role in TGFβ-induced breast CSC expansion via ligand-independent Notch activation. Further studies using tissue samples or primary CSCs derived from patients with breast cancer are needed to establish the clinical relevance of the crosstalk.
High levels of TGFβ have been detected in the majority of malignant breast cancer tissues.18) Plasma TGFβ levels are also elevated in patients with breast cancer and are correlated with tumor stage.19) SphK1 and Notch are upregulated in patients with breast cancer,20) and their expression levels are correlated with cancer progression and poor prognosis.21) In addition, S1PR3 is the most highly expressed S1PR in breast cancer cells.22) Thus, our data provide a potential explanation for the significance of the TGFβ–S1P–Notch axis in breast tumor formation.
Several pathway components evaluated in this study have been tested as clinical targets. ALK5 is a candidate target, and several inhibitors have been tested in pre-clinical studies.23) LY2157299, the first ALK5 inhibitor, has been used in clinical trials of pancreatic, lung, and liver cancer.24) Further studies should evaluate the effectiveness of LY2157299 against breast cancer. S1PR3 is another promising therapeutic target for various cancers. We previously showed that TY52156, an S1PR3 antagonist, inhibits the tumorigenicity of SphK1-overexpressing breast CSCs.9) Several previously reported S1PR antagonists remain to be tested in clinical trials. Targeting Notch is expected to facilitate tumor regression,25) and several types of Notch inhibitors have been shown to be quite effective in pre-clinical models.26,27) Our results suggest that efforts to explore other approaches for targeting the TGFβ–S1P–Notch axis may be promising for breast cancer therapy.
In summary, our results demonstrate a novel mechanism underlying the expansion of breast CSCs. Further investigations should focus on the effects of TGFβ signaling on the proliferation of other CSC types. TGFβ-induced proliferation is likely caused by crosstalk with S1P signaling. It has been reported that the crosstalk between signaling pathways contributes to tumor growth of various cancers.28) Thus, it might be possible to establish specific treatments that reduce tumorigenesis by targeting crosstalk between key signaling pathways.
This work was funded by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology, Japan (#17K19503 and #21H02634 to Y. K.), the Research on Regulatory Harmonization and Evaluation of Pharmaceuticals, Medical Devices, Regenerative and Cellular Therapy Products, Gene Therapy Products, and Cosmetics from Japan Agency for Medical Research and Development, AMED (JP21mk0101189 to Y.K.), and a grant from the Smoking Research Foundation (Y.K.).
The authors declare no conflict of interest.
