Prevention of Akt phosphorylation is a key to targeting cancer stem‑like cells by mTOR inhibition

Shyuichiro Matsubara1,2 · Koichiro Tsukasa1,2 · Taisaku Kuwahata1,2 · Sonshin Takao1,2,3

Received: 1 May 2020 / Accepted: 13 August 2020 © Japan Human Cell Society 2020

CD133 expression in pancreatic cancer correlates with poor prognosis and increased metastasis. CD133+ pancreatic cancer cells exhibit cancer stem cell (CSC)-like properties. We established a CD133+ cell-rich subline from Capan-1 pancreatic cancer cells as a pancreatic CSC model and compared the effects of KU-0063794, a dual mTORC1/mTORC2 inhibitor, against those of mTORC1-specific rapamycin. We found that KU-0063794 prevents sphere formation, a self-renewal index, at high concentrations. Rapamycin inhibited sphere formation but to a lesser degree. In the present study, we aimed to deter- mine the mechanistic roles of mTOR complex 2 (mTORC2) in maintaining CSC-like properties. By examining the PI3K/
Akt/mTOR signaling pathway, we observed lower Akt phosphorylation in KU-0063794-treated cells. Phosphorylation of mTORC1 downstream effectors was inhibited by both inhibitors. Thus, mTORC2 activates Akt and modulate stem-like properties, whereas mTORC1 downstream signaling correlates directly with stem-like properties.

Keywords Pancreatic cancer stem cells · CD133 · Mechanistic/mammalian target of rapamycin (mTOR) · mTOR complex 1 (mTORC1) · mTOR complex 2 (mTORC2)


Pancreatic cancer is an aggressive and fatal malignancy with a 5-year survival rate of approximately 8% [1]. Furthermore, the mortality rate continues to increase. Currently, it is the fourth leading cause of cancer-related deaths in Japan and the United States [1, 2]. The clinical course of the disease is disastrous and innovative new treatments are needed.
The cancer stem cell (CSC) theory has been established and the discovery of CSCs in solid tumor types has been
reported [3–5]. CSCs have the capacity for self-renewal, and for “differentiating” into non-self-renewing bulk tumor cells [3–5]. CSCs can therefore initiate and repopulate an entire tumor. They are believed to be responsible for metas- tasis, and for chemical and radiation resistance. Removal of pancreatic CSCs will abolish tumor initiation and result in effective inhibition of cancer recurrence. Moreover, signal pathways maintaining CSCs are good targets for innovative new treatments.
CD133 [3] and a combination of CD44/CD24/EpCAM [4] have been reported as pancreatic CSC markers. In our

Electronic supplementary material The online version of this article (https://doi.org/10.1007/s13577-020-00416-9) contains supplementary material, which is available to authorized users.
study, CD133 expression in clinical pancreatic cancer cor- relates with poor prognosis and increased metastasis [6]. The CD133+ population of Capan-1 pancreatic cancer cells exhibits CSC-like properties [7]. Establishment of a high

[email protected]; [email protected]
mortality subline with a high CD133+ cell content from this cell line and the knockdown of CD133 in the subline enabled

1Division of Cancer and Regenerative Medicine, Kagoshima University Graduate School of Medical and Dental Sciences, 8-35-1 Sakuragaoka, Kagoshima 890-8520, Japan
2Center for Advanced Biomedical Science and Swine Research, Kagoshima University, 8-35-1 Sakuragaoka, Kagoshima 890-8520, Japan
3Tanegashima Medical Center, 7463, Nishino-omote 891-3198, Japan
us to determine that CD133 facilitates epithelial-mesenchy- mal transition (EMT) [8, 9]. This association of EMT with CSCs in cancer has been previously established [10–12].
We investigated signaling pathways important for main- taining stem cell-like properties using CD133+ pancreatic cancer cells as a model for pancreatic CSCs. We found that


mechanistic/mammalian target of rapamycin (mTOR) and GLI are key molecules in these pathways [13, 14].
Rapamycin, an mTOR inhibitor, reduces the viability of CD133+ cells and stem cell-like properties of pancre- atic cancer cells [13]. mTOR is a component of the PI3K/
Akt/mTOR signaling pathway. This pathway is complex and is an important activation pathway for mTOR, a serine/
threonine protein kinase, depending on upstream signals. mTOR functions as two different kinase complexes that act on unique substrates [15]. The two are called mTOR complex 1 (mTORC1) and 2 (mTORC2). mTORC1 con- tains mTOR, mTOR regulatory related protein (Raptor), and other proteins (mLST8/GβL, PRAS40, and DEPTOR). This complex is inhibited allosterically by rapamycin. In contrast, mTORC2 contains mTOR, rapamycin-insensitive companion of mTOR (Rictor), and other proteins (mLST8/
GβL, mSIN1, DEPTOR, and Protor1/2) [15, 16]. Rapamy- cin does not directly inhibit mTORC2 activity although it prevents mTORC2 assembly in a subset of cancer cells [15].
mTORC1-specific inhibitor rapamycin showed the effi- cient suppression of CSC-related properties and tumor growth [13]. In addition using KU-0063794, an ATP-com- petitive inhibitor of both mTORC1 and mTORC2, mTORC2 was suggested to play some roles in maintaining pancreatic CSC-like properties.
In the present paper, we aimed to determine the mecha- nistic role of mTORC2 in maintaining CSC-like properties. We found differences in the signaling output following treat- ment with KU-0063794 versus the mTORC1-specific inhibi- tor, rapamycin, which suggests that these two complexes have different functions.

Materials and methods

Cell culture and reagents

Capan-1M9 cells were established as described previously [8]. PANC-1 cells were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). The cells were cultured as reported previously (10% FCS + DMEM/
F12, 37 °C, 5% CO2) [13]. After being plated at a density of 2 × 105 cells in 100 mm dishes and allowed to recover for 48 h, the cells treated with inhibitors for 72 h. Rapamycin and KU-0063794 were purchased from Sigma-Aldrich (St. Louis, MO, USA). GSK2110183 was purchased from Sell- eck Chemicals (Houston, TX, USA). Fetal calf serum (FCS) was purchased from Gibco (Waltham, MA, USA).

Sphere formation assay

Spheres were cultured in stem cell medium, namely DMEM/
F12 serum-free medium supplemented with epidermal growth factor (20 ng/ml; PeproTech EC Ltd., London, UK), basic fibroblast growth factor (10 ng/ml; PeproTech EC Ltd.), and B27 supplement (1:50; Gibco, Grand Island, NY, USA). Single cell suspension was prepared (7 cells/ml) and cells were plated (100 µl suspension/well) in 96-well ultra- low attachment plates (Corning, NY, USA) before being cultured for 10 days.

MTT cell viability assay

Single cells were resuspended in fresh medium at a concen- tration of 2 × 103 cells/100 μl and seeded in a 96-well plate. The cells were incubated for 48 h at 37 °C, and the medium was replaced with fresh medium containing inhibitor(s). The plate was incubated subsequently at 37 °C for another 72 h. Cell viability was subsequently measured by an MTT assay.

Flow cytometric analysis

CD133 mAb [either R-phycoerythrin (PE)-conjugated or allophycocyanin (APC)-conjugated] and CD44 mAb [PE- conjugated] were purchased from Miltenyi Biotec (GmbH, Bergisch Gladbach, Germany). Antibody-stained cells were analysed with a FACSAria II flow cytometer (Becton Dick- inson, Franklin Lakes, NJ, USA). Precise methods were described previously [13].

Anchorage‑independent colony formation assay

Each dish (35 mm) was coated with 1 ml bottom agar- medium mixture (DMEM/F12, 10% FCS, 0.5% agar). After the bottom layer was solidified, 1 ml of top agar-medium mixture (DMEM/F12, 10% FCS, 0.33% agar) containing
4× 103 cells was added and the dishes were incubated at 37 °C for 3 weeks. The numbers of colonies were counted under a microscope.


Following drug treatment, cells were washed with ice-cold PBS, mixed with SDS sample buffer, lysed, and boiled for
5min. Proteins were separated by SDS-PAGE (5–20% poly- acrylamide) and transferred to nitrocellulose membranes for detection with antibodies. Anti-phospho-p70/p85 S6 kinase (S6K, T389), anti-p70/p85 S6 kinase (S6K), anti-phospho- S6 ribosomal protein (S240/244), anti-S6 ribosomal protein, anti-phospho-Akt (T308), anti-phospho-Akt (S473), and

anti-Akt antibodies were all purchased from Cell Signal- ing Technology (Beverly, MA, USA). Precise methods were described previously [13].

Statistical analyses

Results are expressed as mean ± standard deviation (SD). Comparisons between the averages of two groups were per- formed using a two-tailed Student’s t test; p < 0.05 was con- sidered statistically significant. Results mTORC1 downstream signaling correlates with CSC‑like properties, whereas mTORC2 activates Akt upstream of the PI3K/Akt/mTOR cascade The pancreatic cancer cell line Capan-1 subclone Capan- 1M9 [8] shows elevated CD133 expression, with approxi- mately 90% of the cell population expressing CD133. We used this subclone as a model for pancreatic CSCs. Previous studies [13] have shown that the dual mTORC1/mTORC2 inhibitor effectively reduced the cell viability and sphere for- mation of CD133+ pancreatic cancer cells. The reduction by rapamycin reaches plateau around 60%. Similar results were a phospho-S6 kinase (T389) total S6 kinase phospho-S6 (S240/S244) total S6 b phospho-Akt (T308) phospho-Akt (S473) total Akt KU-0063794 KU-0063794 obtained with another pancreatic cancer cell line PANC-1 (Supplementary Fig. S1). We determined the effects of KU-0063794 on PI3K/Akt/ mTOR signaling by immunoblotting. KU-0063794 inhib- ited the phosphorylation of two effectors downstream of mTORC1, namely S6 kinase (at site T389) and S6 ribosomal protein (at sites S240/244), thereby inhibiting the activation of downstream signaling (Fig. 1a). This result is the same as that treated with rapamycin [13] (also see Supplementary Fig. S2a). Previous studies [13] have shown that Akt phosphoryla- tion increased following rapamycin treatment (also see Sup- plementary Fig. S2b). In contrast, Akt phosphorylation at either T308 or S473 did not increase following KU-0063794 treatment (Fig. 1b). These two are the major regulatory residues whose phosphorylation is required for full activa- tion [17]. Therefore, Akt activation status was dramatically affected by mTORC2 inhibition. To clarify the nature of this type of inhibition, mTORC1 inhibition without Akt activation, we further analyzed the effect of KU-0063794 on CD133+ cell properties. KU-0063794 reduced Capan-1M9 cell growth measured by direct cell counting, in a concentration dependent manner (Fig. 2a). KU-0063794 did not significantly affect the per- centage of CD133+ cells, similarly to rapamycin treatment (Fig. 2b). Colony formation in soft agar, which is an index Fig. 1 Changes in the phosphorylation and activation status of Akt/ mTOR signaling pathway after KU-0063794 treatment. a Phos- phorylation of mTORC1 downstream effectors in Capan-1M9 cells treated with KU-0063794. b Phosphorylation of Akt in Capan-1M9 cells treated with KU-0063794. Cells were treated with the inhibi- tor, and the resulting cell lysates were analyzed by immunoblotting. Arrowhead indicates the signal shown in the left end of the blot is nonspecific band in the adjacent marker lane, therefore it is not related to the vehicle-treated cell sample of malignancy and linked to CSC property, was reduced by KU-0063794 (Fig. 2c, d). There was no plateau observed. Because both KU-0063794 and rapamycin reduced the CSC-like properties, the signaling output downstream of mTORC1 correlates with the suppression of CSC-like properties, but not from upstream Akt. On the other hand, prevention of Akt phosphorylation is necessary for effective suppression of CSC-like properties. Akt inhibition overcome the plateau after rapamycin treatment GSK2110183 is a potent pan-Akt inhibitor with Ki of 0.08 nM, 2 nM, and 2.6 nM for Akt1, Akt2, and Akt3, respectively. After 3 days treatment with 10 nM or 100 nM Fig. 2 The mTORC1 and mTORC2 inhibitor KU-0063794 reduces the growth of, and colony formation in soft agar by, CD133+ pancre- atic cancer cells. a KU-0063794 suppresses the growth of Capan-1M9 pancreatic cancer cells. Cells were seeded at an initial density of 5 × 104 cells per 35 mm dish. KU-0063794 was administered 2 days after plating and cell number was determined using trypsin/ EDTA treatment for 5 days. The vertical axis is displayed on a logarithmic scale. b Cells were treated with KU-0063794 for 3 days and CD133+ cell per- centage was determined by flow cytometry. c Single cells in soft agar were seeded on a 35 mm dish and cultured for 3 weeks. The number of colonies formed in one field of the microscope is indicated as mean ± SD from three dishes in a representative experiment of two. d Represent- ative photographs of colonies in soft agar in the absence or presence of KU-0063794 (0.1–10 µM). Scale bar: 200 μm rapamycin, cell viability was reduced, but remained at about half, indicating a plateau [13]. When treated with rapamycin in the same manner in combination with 3 µM GSK2110183, the viability of Capan-1M9 cells decreased linearly to near 10% (Fig. 3a). The combined inhibition of mTORC1 and Akt effectively reduced the cell viability of CD133+ pancreatic cancer cells. To examine the self-renewal activity, we performed a sphere formation assay. Treatment with 10 nM rapamycin or 3 μM GSK2110183 reduced sphere counts to 65% and 75%, respectively. The combination of these two showed a marked reduction of sphere number, similar to dual inhibi- tor KU-0063794 treatment. The results of similar experi- ment using PANC-1 cells are shown in Supplementary Figure S3. The sphere formation of PANC-1 cells was effectively reduced by the combination treatment. The combination of KU-0063794 and GSK2110183 also reduced the viability and sphere formation of Capan- 1M9 cells, but the reduction in sphere formation appears to be relatively small. Discussion KRAS mutations have been observed in almost all cases of pancreatic cancer, and this mutation is well known as a critical driver of pancreatic cancer [18]. mTOR func- tions in downstream RAS signaling, and has important roles in stem and progenitor cells [16]. In a previous study, we reported that mTOR has a role in the mainte- nance of pancreatic CSCs, which we discovered using the mTOR inhibitor, rapamycin [13]. The inhibition of mTOR (mTORC1) reduced the viability of CD133+ pancreatic cancer cells, sphere formation, and anchorage-independ- ent colony formation, indicating that mTOR has a role in maintaining CSC-like properties [13]. We investigated the role of mTORC2 in maintaining the CSC-like properties of pancreatic cancer cells because rapamycin does not inhibit mTORC2 [15, 16]. Both inhibitors, namely the dual mTORC1/mTORC2 inhibitor, KU-0063794, and the mTORC1-specific Fig. 3 Inhibition of Akt in combination with rapamycin effectively reduces the stem- like properties of pancreatic cancer cells. a The Akt inhibitor GSK2110183 in combination with the mTORC1 inhibitor rapamycin effectively reduced the viability of Capan-1M9 cells. The results are presented as percentages of control values in untreated cells, showing the mean ± SD of four replicates obtained in one representative experiment out of three. b The Akt inhibitor GSK2110183 in combination with the mTORC1 inhibitor rapamycin effectively reduced the sphere formation of Capan-1M9 cells. *p < 0.05 inhibitor rapamycin caused significant inhibition of cell growth and CSC-like properties. However, the patterns of inhibition were different. The inhibition caused by KU-0063794 was linearly related to concentration, but that by rapamycin had a plateau. mTORC2 inhibition must overcome the plateau observed in rapamycin treat- ment. Therefore, we conclude that mTORC2 plays a role in maintaining CSC-like properties. If mTORC2 was not involved, no additional change would occur after treatment with the dual inhibitor [13]. To investigate the mechanistic roles of mTORC1 and mTORC2 in the maintenance of CSC-like properties, we analyzed the phosphorylation of components in the PI3K/ Akt/mTOR signaling pathway. In a previous paper [13], we detected two signaling changes in the pathway following rapamycin treatment. The phosphorylation of downstream effectors of mTOR, 4E-BP1, S6 kinase, and S6 ribosomal protein, were all inhibited by rapamycin. In contrast, Akt phosphorylation was increased by rapamycin treatment. Akt functions upstream of mTORC1 in the PI3K/Akt/mTOR pathway. Thus, there were two possible mechanisms by which rapamycin may suppress the properties of CSCs: (1) inhibition of effectors downstream of mTORC1 and (2) the activation of Akt. Following dual mTORC1/mTORC2 inhibition by KU-0063794, the phosphorylation of effectors downstream of mTORC1 was blocked, whereas an increase in Akt phos- phorylation was not detected. Both inhibitors significantly reduced CD133+ cell viability and sphere formation. Hence, the phosphorylation of mTORC1 effectors correlates with CSC-like properties. In other words, mTORC1 downstream signaling, but not Akt, directly affects the properties of CSCs. The new finding of the present study provided the experimental proof to support the critical role of mTORC1 downstream signaling in the maintenance of pancreatic CSCs.
mTORC2 phosphorylates S473 of Akt [16, 19], which may indicate that mTORC2 activity was involved in the feedback activation of Akt in rapamycin-treated cells. This hypothesis is supported by the fact that the inhibitory effect of rapamycin reaches a plateau, but that of KU-0063794 does not. The increase in Akt phosphorylation following

Fig.Akt 4 Summary: mTORC1 downstream signaling correlates directly with CSC-like properties, whereas mTORC2 modulates this by activating

rapamycin treatment has a negative feedback effect on CSC suppression. Experiments with the Akt inhibitor GSK2110183 confirmed the involvement of Akt activity in the plateau formation after mTORC1 inhibition. However, upstream regulation of mTORC2-Akt signaling has not yet been elucidated.
Recent data suggest the significance of additional signals from networks other than the core RAS pathway, even in KRAS-driven tumors [20–22]. In this relation, elucidation of the mTORC2-Akt activation mechanism, as well as any downstream mechanisms, involved in maintaining CSC-like properties, is an important issue for future study.
The findings of our current study are summarized sche- matically in Fig. 4. We detected a mechanistic role mTORC2 in relation to the maintenance of pancreatic CSCs. mTORC1 downstream signaling directly correlated with the stem-like property of CSCs; whereas mTORC2 can modulate this sig- nal through upstream Akt activity. Therefore, prevention of Akt phosphorylation is a key to targeting pancreatic CSCs by mTORC1 inhibition.

Acknowledgements We wish to thank T. Obara and Y. Miyazaki for their cooperation in a part of this study, and Y. Setogawa for her expert secretarial assistance. This work was supported by JSPS Grant-in-Aid for Scientific Research (B) Grant Number 25293288 and (C) Grant Number 26462069 from the Ministry of Education, Culture, Sports, Science and Technology, Japan.

Compliance with ethical standards

Conflict of interest The authors declare no competing interests.

1.Siegel RL, Miller KD, Jemal A. Cancer statistics, 2018. CA Cancer J Clin. 2018;68:7–30.
2.Cancer Statistics in Japan ’18. Cancer Information Service, National Cancer Center, Japan. https://ganjoho.jp/en/profession al/statistics/brochure/2018_en.html. Accessed 14 May 2019.
3.Hermann PC, Huber SL, Herrler T, et al. Distinct populations of cancer stem cells determine tumor growth and metastatic activ- ity in human pancreatic cancer. Cell Stem Cell. 2007;1:313–23.
4.Li C, Heidt DG, Dalerba P, et al. Identification of pancreatic cancer stem cells. Can Res. 2007;67:1030–7.
5.Takao S, Ding Q, Matsubara S. Pancreatic cancer stem cells: regulatory networks in the tumor microenvironment and tar- geted therapy. J Hepatobiliary Pancreat Sci. 2012;6:614–20.
6.Maeda S, Shinchi H, Kurahara H, et al. CD133 expression is correlated with lymph node metastasis and vascular endothelial growth factor-C expression in pancreatic cancer. Int J Mol Sci. 2008;98:1389–97.
7.Hayashi T, Ding Q, Kuwahata T, et al. Interferon-alpha modu- lates the chemosensitivity of CD133-expressing pancreatic can- cer cells to gemcitabine. Cancer Sci. 2012;103:889–96.
8.Ding Q, Yoshimitsu M, Kuwahata T, et al. Establishment of a highly migratory subclone reveals that CD133 contributes to migration and invasion through epithelial-mesenchymal transi- tion in pancreatic cancer. Hum Cell. 2012;25:1–8.
9.Ding Q, Miyazaki Y, Tsukasa K, Matsubara S, Yoshimitsu M, Takao S. CD133 facilitates epithelial-mesenchymal transition through interaction with the ERK pathway in pancreatic cancer metastasis. Mol Cancer. 2014;13:15.
10.Mani SA, Guo W, Liao MJ, et al. The epithelial-mesenchymal transition generates cells with properties of stem cells. Cell. 2008;133:704–15.
11.Morel AP, Lievre M, Thomas C, Hinkal G, Ansieau S, Puisieux A. Generation of breast cancer stem cells through epithelial- mesenchymal transition. PLoS ONE. 2008;3:e2888.

12.Zhou P, Li B, Liu F, et al. The epithelial to mesenchymal transition (EMT) and cancer stem cells: implication for treatment resistance in pancreatic cancer treatment resistance. Mol Cancer. 2017;1:52.
13.Matsubara S, Ding Q, Miyazaki Y, Kuwahata T, Tsukasa K, Takao S. mTOR plays critical roles in pancreatic cancer stem cells through specific and stemness-related functions. Sci Rep. 2013;3:3230.
14.Miyazaki Y, Matsubara S, Ding Q, et al. Efficient elimination of pancreatic cancer stem cells by hedgehog/GLI inhibitor GANT61 in combination with mTOR inhibition. Mol Cancer. 2016;15:49.
15.Guertin DA, Stevens DM, Saitoh M, et al. mTOR complex 2 is required for the development of prostate cancer induced by Pten loss in mice. Cancer Cell. 2009;15:148–59. https://doi. org/10.1016/j.ccr.2008.12.017.
16.Manning BD, Toker A. AKT/PKB signaling: navigating the net- work. Cell. 2017;169:381–405.
17.Waters AM, Der CJ. KRAS: the critical driver and therapeutic tar- get for pancreatic cancer. Cold Spring Harb Perspect Med. 2018. https://doi.org/10.1101/cshperspect.a031435(pii: a03143518).
18.Meng D, Frank AR, Jewell JL. mTOR signaling in stem and progenitor cells. Development. 2018. https://doi.org/10.1242/
dev.152595(pii: dev15259519).

19.Sarbassov DD, Guertin DA, Ali SM, Sabatini DM. Phosphoryla- tion and regulation of Akt/PKB by the rictor-mTOR complex. Science. 2005;307:1098–101.
20.Kruspig B, Monteverde T, Neidler S, et al. The ERBB network facilitates KRAS-driven lung tumorigenesis. Sci Trans Med. 2018. https://doi.org/10.1126/scitranslmed.aao2565(pii: eaao2565).
21.Moll HP, Pranz K, Musteanu M, et al. Afatinib restrains K-RAS- driven lung tumorigenesis. Sci Trans Med. 2018. https://doi. org/10.1126/scitranslmed.aao2301(pii: eaao2301).
22.Ruess DA, Heynen GJ, Ciecielski KJ, et al. Mutant KRAS- driven cancers depend on PTPN11/SHP2 phosphatase. Nat Med. 2018;24:954–60.
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.