Cellular Signalling 27 (2015) 1208–1213
Contents lists available at ScienceDirect
Cellular Signalling journal homepage: www.elsevier.com/locate/cellsig
SEMA4B inhibits growth of non-small cell lung cancer in vitro and in vivo Hong Jian a, Yi Zhao a, Bin Liu b, Shun Lu a,⁎ a b
Shanghai Lung Cancer Center, Shanghai Chest Hospital Affiliated to Shanghai Jiaotong University, Shanghai, China Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Shanghai, China
a r t i c l e
i n f o
Article history: Received 10 December 2014 Received in revised form 17 February 2015 Accepted 26 February 2015 Available online 4 March 2015 Keywords: SEMA4B Non-small cell lung cancer (NSCLC) PI3K FoxO1 P21
a b s t r a c t We have recently shown that Semaphorin 4B (SEMA4B) inhibits the invasion of non-small cell lung cancer (NSCLC) through PI3K-dependent suppression of MMP9 activation. In the current study, we evaluated whether SEMA4B may also affect the growth of NSCLC. We thus used two human NSCLC lines, A549 and Calu-3, to examine our hypothesis. We found that overexpression of SEMA4B significantly decreased NSCLC cell growth, while SEMA4B inhibition significantly increased NSCLC cell growth, both in vitro and in vivo in an implanted NSCLC model. Adaptation of SEMA4B in NSCLC cells did not alter cell apoptosis, but changed the cell proliferation. Further analyses show that SEMA4B may induce FoxO1 nuclear retention through suppressing PI3K/Akt signaling pathway, which subsequently inhibited cell growth through the direct nuclear target of FoxO1, p21. Our study thus demonstrate a role of SEMA4B in suppressing NSCLC growth, besides its role in inhibiting cell metastasis, and highlights SEMA4B as a promising therapeutic target for NSCLC. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Non-small cell lung cancer (NSCLC) is a lung cancer with a high incidence and mainly consists three subtypes: squamous cell carcinoma, large cell carcinoma, and adenocarcinoma [1,2]. NSCLCs are generally insensitive to chemotherapy and radiation therapy, and often appear to grow very fast [1–3]. Thus, understanding of the mechanism controlling the growth of NSCLC is extremely important for its therapy. The semaphorins are a family of proteins that regulate cell migration, angiogenesis and immune response [4]. The semaphorin family has been divided into 7 subclasses. The class 4 semaphorins has 7 members, among which SEMA4B has recently been shown to play a critical role in the tumorigenesis of NSCLC, by us [5,6], and by others [7]. Specifically, the downstream signal transduction of semaphorins has been studied in different cell types, and has been found to involve different pathways including the phosphatidylinositol 3-kinase (PI3K) pathway, the extracellular-related kinase/mitogen-activated protein kinase (ERK/ MAPK) pathway, and the Jun N-terminal kinase (JNK) pathway [4,5, 7–14]. However, we recently reported that only PI3K signaling pathway is affected by SEMA4B in NSCLC [6].
The PI3K pathway is the major signaling pathway to regulate cell proliferation. Activated PI3K generates several phosphoinositols, leading to Akt activation by phosphorylation at Thr308 and Ser473, which further modulate downstream DNA-binding molecules, to trigger DNA synthesis [15–19]. Forkhead box protein O1 (FoxO1) is a key target of Akt, and the nuclear form of FoxO1 is a DNA binding protein to control the expression of many genes. When the phosphorylation status of FoxO1 protein is altered by phosphorylated Akt, FoxO1 will be excluded from the nuclei to cytoplasm in a process called nuclear exclusion, resulting in the adaptation of expressions of FoxO1-binding genes in the nuclei. On the other hand, FoxO1 nuclear translocation and retention have exactly the adverse effect in this context [3,20–25]. In most occasions, nuclear FoxO1 binds to the promoter of cell-cycle-inhibitor p21, which suppresses cell proliferation [3,20–25]. In our previous study, we have reported a significant decrease in SEMA4B levels and a significant increase in MMP9 levels in NSCLC from the patients. We have further shown that SEMA4B negatively regulated MMP9 levels, a key factor that induces NSCLC metastasis, exclusively through PI3K signaling pathway [6]. Here we aimed to figure out whether SEMA4B may affect NSCLC growth through its inhibition on PI3K downstream signals. 2. Materials and methods
⁎ Corresponding author at: Shanghai Lung Cancer Center, Shanghai Chest Hospital Affiliated to Shanghai Jiaotong University, 241 Huaihai West Road, Shanghai 200030, China. Tel.: +86 2162821990; fax: +86 2162804970. E-mail address: [email protected] (S. Lu).
http://dx.doi.org/10.1016/j.cellsig.2015.02.027 0898-6568/© 2015 Elsevier Inc. All rights reserved.
2.1. Cell lines and reagents A549 and Calu-3 are two human NSCLC lines purchased from ATCC, and were cultured in Dulbecco's Modified Eagle's medium
H. Jian et al. / Cellular Signalling 27 (2015) 1208–1213
B
*
20 10 2
*
1 0
NS
A5
A5
A5
49
49
-S
-N
EM
2
0
NS
Ca
lu
-3
lu
-3
-s
l
-s
EM
A4
b
nF
ox
hS
EM
b
b-
-3
hS
ul
A4
Ca
lu
-N
EM
A4
b
Ca
lu
-S
EM
A4
Ca
-3
hS
EM
b
*
1
-s
hS
l
*
10
49
-s
ul
A4
A5
49
20
Fold change of SEMA4b transcript (vs control)
Fold change of SEMA4b transcript (vs control)
A
1209
A4
b-
nF
ox
O
O
1
1
D
C SEMA4b α-tubulin
A5
49
-S
ul
l
A4
50 kDa
α-tubulin
b
92 kDa 50 kDa
Ca
49
-N
EM
SEMA4b
A5
A5
49
92 kDa
Ca
lu
lu
-S
-3
-3
ul
EM
EM
A4
b
lu
-N
-S
hS
Ca
A4
l
b
-3
-S
hS
EM
A4 b
Fig. 1. Modification of SEMA4B levels in NSCLC lines. (A–D) We either overexpressed SEMA4B in A549 and Calu-3 cells (A549-SEMA4B or Calu-3-SEMA4B cells), or inhibited SEMA4B in those cells (A549-shSEMA4B or Calu-3-shSEMA4B cells). A549 and Calu-3 cells transduced with scrambled sequence were used as controls (A549-Null and Calu-3-Null cells). A549shSEMA4B and Calu-3-shSEMA4B cells were further transduced by a constitutive nuclear FoxO1, resulting in A549-shSEMA4B-nFoxO1 and Calu-3-shSEMA4B-nFoxO1 cells. SMAD4b levels in these cells were examined by RT-qPCR (A, B) and by Western blot (C, D). α-Tubulin is a protein loading control. *: p b 0.05. NS: non-significant.
(DMEM) supplemented with 20% fetal bovine serum (Invitrogen, Carlsbad, CA, USA). The A549 cell line was first developed in 1972 by Dr. Giard through the removal and culturing of cancerous lung tissue in the explanted tumor a of 58-year-old Caucasian male [26]. Calu-3 cell line was developed by Dr. Fogh in 1975 [27].
2.3. Cell proliferation assay
2.2. Transduction of the cell lines
2.4. Apoptosis assay
SEMA4B-adapted plasmids have been previously described [6]. Briefly, SEMA4B construct was generated by sub-cloning PCRamplified full-length human SEMA4B cDNA into pcDNA3.1-EGFP-luciferase. For inhibition of SEMA4B, corresponding human short hairpin small interference RNA (ShRNA) sequence targeting “CTCTTCACCTTCCACATTATC” was cloned into pcDNA3.1-EGFP-luciferase to generate the shRNA construct. Scrambled sequence (null) was cloned into pcDNA3.1-EGFP-luciferase to generate control construct. These constructs were used to prepare lentiviral vectors to stably transduce the cells. Transfected cells expressing SEMA4B or ShSEMA4B or Null were selected by flow cytometry based on GFP, which is also included in the plasmid construct. The presence of luciferase allows in vivo tracing of the cells. For the production of cells with constitutively nuclear FoxO1, A549-shSEMA4B and Calu-3-shSEMA4B cells were further transduced with a constitutively nuclear FoxO1-expressing lentivirus (kindly provided by Shengzi Wang, Fudan University, China), as has been previously described [24,28]. The generated A549-shSEMA4B-nFoxO1 and Calu-3shSEMA4B-nFoxO1 cells express high levels of nuclear FoxO1, which cannot be phosphorylated due to a defect on the phosphorylation site.
Cells were labeled with annexin V-FITC and propidium iodide (PI), and then examined with an apoptosis detecting kit (Invitrogen, USA) for apoptosis. Samples were analyzed by flow cytometry and the results were analyzed by CellQuest software (Becton Dickinson, San Jose, CA) as has been described before [29].
For the assay of cell proliferation, pretreated cells were seeded into a 96 well-plate at 4000 cells per well and subjected to a cell proliferation kit (MTT, Roche, USA), according to the instruction from the manufacturer.
2.5. In vivo implanted NSCLC cancer model and imaging of the implanted cancer by bioluminescence Luciferase-carrying A549 cells (106) were injected through the tail vein into 12 weeks of age male NOD/SCID mice, as has been described before [30]. After 2 weeks, only the mice that have developed detectable luminescence were used for this study. At this stage, the size of the tumor is pretty small and comparable among all mice. These mice with a developed tumor were kept for another 4 weeks, and then the tumor growth was quantified by luminescence levels. Bioluminescence was measured with the IVIS imaging system (Xenogen Corp., Alameda, CA). All of the images were taken 10 min after intraperitoneal injection of luciferin (Sigma) of 150 mg/kg body weight, as a 60-second acquisition and 10 of binning. During image acquisition, mice were sedated
1210
H. Jian et al. / Cellular Signalling 27 (2015) 1208–1213
A
*
4
A549-shSEMA4b
O.D. 570 nm
3
2
*
A549-shSEMA4b-nFoxO1
NS
A549-Null NS
1
probed with a primary antibody. After incubation with a horseradish peroxidase-conjugated second antibody, autoradiograms were prepared using the enhanced chemiluminescent system to visualize the protein antigen. The signals were recorded using an X-ray film. The primary antibodies were anti-SEMA4B (Sigma), anti-FoxO1, anti-p21, anti-Bcl-2, anti-caspase-9, anti-LC3, anti-LaminB1 and anti-αtubulin (all from Cell Signaling, USA). The secondary antibody is an HRP-conjugated anti-rabbit (Jackson Labs). Figure images were representative from 5 repeats. α-Tubulin (for total protein or cytoplasmic protein, CY) and LaminB1 (for nuclear protein, NU) were used as protein loading controls.
A549-SEMA4b
*
0
*
1
0
2
2.8. Statistical analysis Time (day)
All statistical analyses were carried out using the SPSS 19.0 statistical software package. All values are depicted as the mean ± standard deviation from 5 individuals and are considered significant if p b 0.05. All data were statistically analyzed using one-way ANOVA with a Bonferroni correction.
Calu-3-shSEMA4b
3. Result
B 4
*
O.D. 570 nm
3
3.1. SEMA4B inhibited NSCLC growth in vitro 2
*
NS
Calu-3-shSEMA4b-nFoxO1 Calu-3-Null
NS
1
Calu-3-SEMA4b
*
0 0
1
* 2
Time (day)
Fig. 2. An MTT assay for cell growth. Cell growth was quantified in an MTT assay. *: p b 0.05. NS: non-significant.
continuously via inhalation of 3% isoflurane. Image analysis and bioluminescent quantification were performed using Living Image software (Xenogen Corp). 2.6. RT-qPCR RNA was extracted from the cultured NSCLC cells with RNeasy (Qiagen, Hilden, Germany) and then used for cDNA synthesis. RTqPCR was performed in duplicates with QuantiTect SYBR Green PCR kit (Qiagen). All primers were purchased from Qiagen. Values of genes were normalized against α-tubulin and then compared to controls. 2.7. Western blot For analysis of total protein, the protein was extracted from the cultured cells, which was homogenized in a RIPA lysis buffer (1% NP40, 0.1% SDS, 100 μg/mL phenylmethylsulfonyl fluoride, 0.5% sodium deoxycholate, in PBS) on ice. The supernatants were collected after centrifugation at 12,000 ×g at 4 °C for 20 min. Protein concentration was determined using a BCA protein assay kit (Bio-Rad, China), and whole lysates were mixed with 4 × SDS loading buffer (125 mmol/L Tris–HCl, 4% SDS, 20% glycerol, 100 mmol/L DTT, and 0.2% bromophenol blue) at a ratio of 1:3. Nuclear and cytoplasmic proteins were isolated with a nuclear and cytoplasmic extraction kit (Thermo Scientific, USA). Protein samples were heated at 100 °C for 5 min and were separated on SDS-polyacrylamide gels. The separated proteins were then transferred to a PVDF membrane. The membrane blots were first
In our previous study, we have reported a significant decrease in SEMA4B levels and a significant increase in the MMP9 levels in NSCLC from the patients. We have further shown that SEMA4B negatively regulated MMP9 levels, a key factor that induces NSCLC metastasis, exclusively through the PI3K signaling pathway [6]. Here we aimed to figure out whether SEMA4B may affect NSCLC growth through its inhibition on the PI3K downstream signals. Thus, we used two human NSCLC lines, A549 and Calu-3, for our study. We either overexpressed SEMA4B in A549 and Calu-3 cells, or inhibited SEMA4B in those cells. The overexpression and inhibition of SEMA4B in these NSCLC cells were confirmed by RT-qPCR (Fig. 1A– B), and by Western blot (Fig. 1C–D). We found that SEMA4B overexpression significantly decreased cell proliferation in both NSCLC lines in a MTT assay, while SEMA4B inhibition significantly increased cell proliferation in both NSCLC lines (Fig. 2A–B). The apoptosis of the cells were then evaluated by flow cytometry, using an apoptosis detecting kit. We found that adaptation of SEMA4B levels in either cell line did not affect the apoptosis of these NSCLC cells (Fig. 3A–B). Microtubule-associated protein 1A/1B-light chain 3 (LC3) is a soluble cellular protein. During autophagy, a cytosolic form of LC3 (LC3-I) is conjugated to phosphatidylethanolamine to form LC3-phosphatidylethanolamine conjugate (LC3-II), which is then recruited to autophagosomes to form autolysosomes. Thus, the autophagosomal marker LC3-II well reflects the autophagic activity [31–33]. We did not detect changes in LC3-II by SEMA4B modifications (Fig. 3B), suggesting that SEMA4B may not change autophagy in NSCLC cells. Thus, the effect of SEMA4B on the NSCLC growth is mainly through alteration in cell proliferation. Taken together, these data suggest that SEMA4B inhibits the growth of NSCLC in vitro. 3.2. SEMA4B inhibited NSCLC growth in vivo The A549-SEMA4B, A549-Null and A549-shSEMA4B cells also had a luciferase reporter for in vivo tracing. These A549 cells (106) were injected through the tail vein into 12 weeks of age male NOD/SCID mice. Only the mice that have developed detectable luminescence for implanted cancer cells in the lung region at two weeks were used for this study. These mice were kept for another 4 weeks, and then the tumor growth was quantified by luminescence levels, showing significantly greater tumors in mice receiving SEMA4Binhibited NSCLC cells, and significantly smaller tumors in mice
H. Jian et al. / Cellular Signalling 27 (2015) 1208–1213
A Fold change in DCF-DA levels (vs control)
1.5
NS
1211
NS
1.0
0.5
0
A5
49
-S
49
-N
EM
A5
A5
A5
49
hS
l
A4
A4
b
-3
-S
lu
-3
-s
l
-s h EM SEM A4 A4 b b-
b
b-
nF
ox
-3
hS
ul
A4
A4
b
Ca
lu
-N
EM
EM
Ca
lu
-3
hS
EM
Ca
lu
-s
-s
ul
Ca
49
nF
ox
O
O
1
1
B Caspase9
46 kDa
Bcl-2
26 kDa
LC3
16 kDa
α-tubulin
50 kDa A5
49
49
-S
49
-N
EM
hS
EM
b
Ca
-3
A4
lu
-N
EM
A4
Ca
lu
-3
-S
ul
-3
-s
hS
l
A4
EM
A4
A4
b-
b
nF
ox
hS
EM
b
b-
b
-3
-S
EM
Ca
lu
lu
-s
hS
l
Ca
49
-S
ul
A4
A5
A5
A5
nF
ox
O
O
1
1
Fig. 3. Analyses of cell apoptosis and autophagy. (A) Cell apoptosis was quantified with an apoptosis detecting kit. (B) Protein levels were examined by Western blot, shown by representative images. α-Tubulin is a protein loading control. NS: non-significant.
receiving SEMA4B-overexpressing NSCLC cells, compared to controls, by presentative images (Fig. 4A), and by quantification
B Fold change in luminescence (vs control)
A
(Fig. 4B). These data suggest that SEMA4B inhibits NSCLC growth in vivo.
A5
49
49
-s
hS
l
A4
b
1 0
* A5
EM
A4
A5
A5
49
49
-S
49
-N
ul
l
b
-S
ul
EM
2
A4
49
-N
10
EM
A5
A5
*
20
-s
hS
EM
A4
b
b
Fig. 4. Quantification of implanted NSCLC growth in living animals. (A–B) The A549-SEMA4B, A549-Null and A549-shSEMA4B cells (106) were injected through the tail vein into 12 weeks of age male NOD/SCID mice. The mice that have developed detectable luminescence for implanted cancer cells were kept for another 4 weeks, and then the tumor growth was quantified by luminescence levels, shown by representative images (A), and by quantification (B). *: p b 0.05.
1212
H. Jian et al. / Cellular Signalling 27 (2015) 1208–1213
A FoxO1
70 kDa
p21
21 kDa
LaminB1
67 kDa
α-tubulin
50 kDa CY
Nu
CY
A5
Nu
CY
49
-s
hS
hS
l
A4
49
-s
ul
EM
Nu
A5
49
-N
-S
CY
A5
A5
49
Nu
EM
EM
A4
b
b
A4
b-
nF
ox
O
C Fold change in luminescence (vs control)
B
20 10 2
0
A4
b
hS
EM
b
EM
EM
-s
-s
A4
hS
hS
49
49
EM
-s
-s
A5
A5
hS
49
49
*
1
A5
A5
1
A4
b-
nF
ox
A4
O
b-
nF
ox
1
O
1
Fig. 5. Constitutive nuclear FoxO1 abolished the effect of SEMA4B on cell growth. (A) Nuclear (NU) and cytoplasmic (CY) proteins from SEMA4B-modified A549 cells were isolated for Western blot. LaminB1 and α-tubulin were used to confirm the purity of the fractions. Representative Western blot images were shown. (B) A549-shSEMA4B and A549-shSEMA4BnFoxO1 cells (106) were injected through the tail vein into 12 weeks of age male NOD/SCID mice. The mice that have developed detectable luminescence for implanted cancer cells were kept for another 4 weeks, and then the tumor growth was quantified by luminescence levels, shown by representative images (A), and by quantification (B). *: p b 0.05.
3.3. SEMA4B may inhibit NSCLC growth through PI3K-regulated FoxO1 cellular location We recently reported that only PI3K signaling pathway is affected by SEMA4B in NSCLC [6], and the PI3K/Akt pathway is the major signaling pathway to regulate cell proliferation. Akt phosphorylation induces nuclear exclusion of FoxO1, resulting in the adaptation of p21, which suppresses cell proliferation. Therefore, we isolated nuclear vs cytoplasmic proteins from SEMA4B-modified A549 cells and performed Western blot with extracted proteins. LaminB1 and α-tubulin were used to assure the purity of the fractions, as has been applied before [3,23–25]. We found that overexpression of SEMA4B significantly increased the ratio of nuclear vs cytoplasmic FoxO1, while inhibition of SEMA4B significantly decreased this ratio by Western blot (Fig. 5A). Moreover, the nuclear vs cytoplasmic FoxO1 levels directly affected the nuclear vs cytoplasmic p21 levels (Fig. 5A). These data suggest that SEMA4B may inhibit NSCLC growth through PI3K/Akt-regulated cellular location of FoxO1 and its target p21. To prove this hypothesis, we further transduced the SEMA4B-depleted A549 or Calu-3 cells with a sustained nuclear FoxO1, as has been previously described [3,23,24]. The transduced cells expressed high levels of nuclear FoxO1, which cannot be phosphorylated due
to a defect on phosphorylation site (Fig. 5A). Importantly, the expression of sustained nuclear FoxO1 did not alter transcription levels of SEMA4B (Fig. 1A–B). Next, we examined the effect of sustained FoxO1 nuclear retention on the growth of NSCLC cells. Our data showed that sustained nuclear FoxO1 completely abolished the effects of SEMA4B inhibition in increasing the growth of NSCLC cells (Fig. 2A–B), without altering cell apoptosis (Fig. 3), suggesting that cellular location of FoxO1 is responsible for the effect of SEMA4B on the NSCLC cell growth. Moreover, sustained nuclear FoxO1 similarly abolished the effects of SEMA4B inhibition on the growth of the NSCLC cells in vivo (Fig. 5B–C). Taken together, based on the findings in the current study and our previous work, we propose a model in which SEMA4B inhibits PI3K/ Akt signaling in NSCLC, which not only regulates MMP9 to control metastasis, but also regulates the cellular location of FoxO1 and its target p21 to control cell growth (Fig. 6). 4. Discussion SEMA4B has been shown to play an important role in lung cancer invasion. However, whether SEMA4B may also regulate the cancer cell growth in lung cancer is not determined yet. Recently, we have shown
H. Jian et al. / Cellular Signalling 27 (2015) 1208–1213
NSCLC SEMA4b
MMP9
PI3K/Akt
Metastasis
FoxO1 FoxO1
1213
cellular location of FoxO1 is critical for its function, since only the nuclear form of FoxO1 can bind to DNA and affect target gene expression. Thus, in the current study, we simplified the evaluation on FoxO1, and used its cellular location for its biological activities. Moreover, our data that support a similar NSCLC-growthsuppressing role for SEMA4B are important, since it suggests that the effect of autologous SEMA4B on the cell growth in NSCLC works independently and is not affected by systematic signals. To summarize, based on the findings in the current study and our previous work, we propose a model in which SEMA4B inhibits PI3K/ Akt signaling in NSCLC, which not only regulates MMP9 to control metastasis, but also regulates cellular location of FoxO1 and its target p21 to control cell growth. These series of studies on the role of SEMA4B in NSCLC thus shed light on a novel target for NSCCL therapy, and suggest that genetic or pharmaceutical overexpression of SEMA4B in NSCLC in the patients may create a promising therapy for NSCLC.
Cell proliferation
Conflict of interest The authors have declared that no competing interests exist. Fig. 6. Schematic of the model of SEMA4B in NSCLC. SEMA4B inhibits PI3K/Akt signaling in NSCLC, which not only regulates MMP9 to control metastasis, but also regulates cellular location of FoxO1 and its target p21 to control cell growth.
that hypoxia-inducible factor 1 (HIF-1) could downregulate the expression of SEMA4B in human NSCLC lines, and then abolished its suppressive effect on NSCLC metastasis [5]. Moreover, we also reported a significant decrease in SEMA4B levels and a significant increase in the MMP9 levels in NSCLC from the patients and further showed that SEMA4B negatively regulated the MMP9 levels, a key factor that induces NSCLC metastasis, exclusively through the PI3K signaling pathway [6]. Thus, the inhibitory effect of SEMA4B may result from its active participation into the signaling pathways and interactions with other molecules that regulate angiogenic factors and matrix proteinases. Since we have shown that SEMA4B significantly inhibited PI3K/Akt pathway activities in the NSCLC cells, and since PI3K/Akt is a wellknown cell-cycle regulator, we hypothesize that SEMA4B may not only affect cancer cell invasiveness and metastasis, but also cancer cell growth in NSCLC. Thus, we used two human NSCLC lines, A549 and Calu-3, for the current study as we have done in our previous work. Using these two cell lines substantially decreased the possibility of drawing a conclusion based on cell-line-dependent data. For in vivo experiment, although we only showed data from A549 cells, we also analyzed Calu-3 cells in the same context and had similar results. In a series of gain-of-function and loss-of-function experiments, performed both in vitro and in vivo, we were able to show that SEMA4B may inhibit the growth of NSCLC, through its direct effect on PI3K/Akt-mediated changes in FoxO1 cellular location. FoxO1 proteins are transcriptional regulators that control cell cycle progression, DNA repair, defense against oxidative damage and apoptosis. Here we did not examine the detail of FoxO1 modification, since it is very complicated. Indeed, divergent functions of FoxO1 proteins are regulated by a variety of signal-induced, post-translational modifications. For example, the phosphorylation of cytoplasmic FoxO1 at specific sites by JNK initiates translocation into the nucleus. Acetylation and deacetylation of nuclear FoxO affect FoxO1-dependent transcriptional programs. The phosphorylation of Akt results in the phosphorylation of nuclear FoxO1 at specific sites followed by additional phosphorylations mediated by other kinases. Akt-dependent phosphorylation reduces the DNA-binding activity of FoxO1, inactivates the FoxO nuclear translocation signal and results in FoxO1 nuclear exclusion, in a CRM1and 14-3-3-dependent process. However, whatever regulated, the
Acknowledgment This work is financially supported by Wujieping Medical Foundation No: 320.6750.14060, and Shanghai Chest Hospital Science and Technology Development Fund (2014 YZDC10600). References [1] G. Caramori, P. Casolari, G.N. Cavallesco, S. Giuffre, I. Adcock, A. Papi, Int J Biochem Cell Biol 43 (2011) 1030–1044. [2] R.C. Buttery, R.C. Rintoul, T. Sethi, Int J Biochem Cell Biol 36 (2004) 1154–1160. [3] J. Pei, Y. Lou, R. Zhong, B. Han, Tumour Biol 35 (2014) 6673–6678. [4] A. Ahmed, B.J. Eickholt, Adv Exp Med Biol 600 (2007) 24–37. [5] H. Jian, B. Liu, J. Zhang, Tumour Biol 35 (2014) 4949–4955. [6] H. Jian, Y. Zhao, B. Liu, S. Lu, Tumour Biol 35 (2014) 11051–11056. [7] H. Nagai, N. Sugito, H. Matsubara, Y. Tatematsu, T. Hida, Y. Sekido, et al., Oncogene 26 (2007) 4025–4031. [8] J.R. Basile, T. Afkhami, J.S. Gutkind, Mol Cell Biol 25 (2005) 6889–6898. [9] H. Aghajanian, C. Choi, V.C. Ho, M. Gupta, M.K. Singh, J.A. Epstein, J Biol Chem 289 (2014) 17971–17979. [10] G. Pan, Z. Zhu, J. Huang, C. Yang, Y. Yang, Y. Wang, et al., Dig Dis Sci 58 (2013) 2197–2204. [11] G. Gallo, Dev Neurobiol 68 (2008) 926–933. [12] J.K. Atwal, K.K. Singh, M. Tessier-Lavigne, F.D. Miller, D.R. Kaplan, J Neurosci 23 (2003) 7602–7609. [13] H. Wen, Y. Lei, S.Y. Eun, J.P. Ting, J Exp Med 207 (2010) 2943–2957. [14] C. Burkhardt, M. Muller, A. Badde, C.C. Garner, E.D. Gundelfinger, A.W. Puschel, FEBS Lett 579 (2005) 3821–3828. [15] S. Kim, J.H. Choi, H.I. Lim, S.K. Lee, W.W. Kim, S. Cho, et al., Cell Signal 21 (2009) 892–898. [16] M.R. Schneider, E. Wolf, J Cell Physiol 218 (2009) 460–466. [17] F.R. Hirsch, P.A. Janne, W.E. Eberhardt, F. Cappuzzo, N. Thatcher, R. Pirker, et al., J Thorac Oncol 8 (2013) 373–384. [18] K. Kobayashi, K. Hagiwara, Target Oncol 8 (2013) 27–33. [19] T. Sasaki, K. Hiroki, Y. Yamashita, Biomed Res Int 2013 (2013) 546318. [20] S.B. Prasad, S.S. Yadav, M. Das, H.B. Govardhan, L.K. Pandey, S. Singh, et al., J Cancer 5 (2014) 655–662. [21] P. Liu, T.P. Kao, H. Huang, Oncogene 27 (2008) 4733–4744. [22] M. Fosbrink, F. Niculescu, V. Rus, M.L. Shin, H. Rus, J Biol Chem 281 (2006) 19009–19018. [23] J. Chen, Q. Huang, F. Wang, Tumour Biol 35 (2014) 7195–7200. [24] H. Ding, Y. Zhu, T. Chu, S. Wang, Tumour Biol 35 (2014) 9987–9992. [25] X. Xiao, I. Gaffar, P. Guo, J. Wiersch, S. Fischbach, L. Peirish, et al., Proc Natl Acad Sci U S A 111 (2014) E1211–E1220. [26] D.J. Giard, S.A. Aaronson, G.J. Todaro, P. Arnstein, J.H. Kersey, H. Dosik, et al., J Natl Cancer Inst 51 (1973) 1417–1423. [27] J. Fogh, J.M. Fogh, T. Orfeo, J Natl Cancer Inst 59 (1977) 221–226. [28] W.H. Biggs III, J. Meisenhelder, T. Hunter, W.K. Cavenee, K.C. Arden, Proc Natl Acad Sci U S A 96 (1999) 7421–7426. [29] Q. Mei, F. Li, H. Quan, Y. Liu, H. Xu, Cancer Sci 105 (2014) 755–762. [30] W. Long, C.E. Foulds, J. Qin, J. Liu, C. Ding, D.M. Lonard, et al., J Clin Invest 122 (2012) 1869–1880. [31] D.R. Green, B. Levine, Cell 157 (2014) 65–75. [32] J.Y. Guo, B. Xia, E. White, Cell 155 (2013) 1216–1219. [33] E. White, Nat Rev Cancer 12 (2012) 401–410.
Contents lists available at ScienceDirect
Cellular Signalling journal homepage: www.elsevier.com/locate/cellsig
SEMA4B inhibits growth of non-small cell lung cancer in vitro and in vivo Hong Jian a, Yi Zhao a, Bin Liu b, Shun Lu a,⁎ a b
Shanghai Lung Cancer Center, Shanghai Chest Hospital Affiliated to Shanghai Jiaotong University, Shanghai, China Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Shanghai, China
a r t i c l e
i n f o
Article history: Received 10 December 2014 Received in revised form 17 February 2015 Accepted 26 February 2015 Available online 4 March 2015 Keywords: SEMA4B Non-small cell lung cancer (NSCLC) PI3K FoxO1 P21
a b s t r a c t We have recently shown that Semaphorin 4B (SEMA4B) inhibits the invasion of non-small cell lung cancer (NSCLC) through PI3K-dependent suppression of MMP9 activation. In the current study, we evaluated whether SEMA4B may also affect the growth of NSCLC. We thus used two human NSCLC lines, A549 and Calu-3, to examine our hypothesis. We found that overexpression of SEMA4B significantly decreased NSCLC cell growth, while SEMA4B inhibition significantly increased NSCLC cell growth, both in vitro and in vivo in an implanted NSCLC model. Adaptation of SEMA4B in NSCLC cells did not alter cell apoptosis, but changed the cell proliferation. Further analyses show that SEMA4B may induce FoxO1 nuclear retention through suppressing PI3K/Akt signaling pathway, which subsequently inhibited cell growth through the direct nuclear target of FoxO1, p21. Our study thus demonstrate a role of SEMA4B in suppressing NSCLC growth, besides its role in inhibiting cell metastasis, and highlights SEMA4B as a promising therapeutic target for NSCLC. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Non-small cell lung cancer (NSCLC) is a lung cancer with a high incidence and mainly consists three subtypes: squamous cell carcinoma, large cell carcinoma, and adenocarcinoma [1,2]. NSCLCs are generally insensitive to chemotherapy and radiation therapy, and often appear to grow very fast [1–3]. Thus, understanding of the mechanism controlling the growth of NSCLC is extremely important for its therapy. The semaphorins are a family of proteins that regulate cell migration, angiogenesis and immune response [4]. The semaphorin family has been divided into 7 subclasses. The class 4 semaphorins has 7 members, among which SEMA4B has recently been shown to play a critical role in the tumorigenesis of NSCLC, by us [5,6], and by others [7]. Specifically, the downstream signal transduction of semaphorins has been studied in different cell types, and has been found to involve different pathways including the phosphatidylinositol 3-kinase (PI3K) pathway, the extracellular-related kinase/mitogen-activated protein kinase (ERK/ MAPK) pathway, and the Jun N-terminal kinase (JNK) pathway [4,5, 7–14]. However, we recently reported that only PI3K signaling pathway is affected by SEMA4B in NSCLC [6].
The PI3K pathway is the major signaling pathway to regulate cell proliferation. Activated PI3K generates several phosphoinositols, leading to Akt activation by phosphorylation at Thr308 and Ser473, which further modulate downstream DNA-binding molecules, to trigger DNA synthesis [15–19]. Forkhead box protein O1 (FoxO1) is a key target of Akt, and the nuclear form of FoxO1 is a DNA binding protein to control the expression of many genes. When the phosphorylation status of FoxO1 protein is altered by phosphorylated Akt, FoxO1 will be excluded from the nuclei to cytoplasm in a process called nuclear exclusion, resulting in the adaptation of expressions of FoxO1-binding genes in the nuclei. On the other hand, FoxO1 nuclear translocation and retention have exactly the adverse effect in this context [3,20–25]. In most occasions, nuclear FoxO1 binds to the promoter of cell-cycle-inhibitor p21, which suppresses cell proliferation [3,20–25]. In our previous study, we have reported a significant decrease in SEMA4B levels and a significant increase in MMP9 levels in NSCLC from the patients. We have further shown that SEMA4B negatively regulated MMP9 levels, a key factor that induces NSCLC metastasis, exclusively through PI3K signaling pathway [6]. Here we aimed to figure out whether SEMA4B may affect NSCLC growth through its inhibition on PI3K downstream signals. 2. Materials and methods
⁎ Corresponding author at: Shanghai Lung Cancer Center, Shanghai Chest Hospital Affiliated to Shanghai Jiaotong University, 241 Huaihai West Road, Shanghai 200030, China. Tel.: +86 2162821990; fax: +86 2162804970. E-mail address: [email protected] (S. Lu).
http://dx.doi.org/10.1016/j.cellsig.2015.02.027 0898-6568/© 2015 Elsevier Inc. All rights reserved.
2.1. Cell lines and reagents A549 and Calu-3 are two human NSCLC lines purchased from ATCC, and were cultured in Dulbecco's Modified Eagle's medium
H. Jian et al. / Cellular Signalling 27 (2015) 1208–1213
B
*
20 10 2
*
1 0
NS
A5
A5
A5
49
49
-S
-N
EM
2
0
NS
Ca
lu
-3
lu
-3
-s
l
-s
EM
A4
b
nF
ox
hS
EM
b
b-
-3
hS
ul
A4
Ca
lu
-N
EM
A4
b
Ca
lu
-S
EM
A4
Ca
-3
hS
EM
b
*
1
-s
hS
l
*
10
49
-s
ul
A4
A5
49
20
Fold change of SEMA4b transcript (vs control)
Fold change of SEMA4b transcript (vs control)
A
1209
A4
b-
nF
ox
O
O
1
1
D
C SEMA4b α-tubulin
A5
49
-S
ul
l
A4
50 kDa
α-tubulin
b
92 kDa 50 kDa
Ca
49
-N
EM
SEMA4b
A5
A5
49
92 kDa
Ca
lu
lu
-S
-3
-3
ul
EM
EM
A4
b
lu
-N
-S
hS
Ca
A4
l
b
-3
-S
hS
EM
A4 b
Fig. 1. Modification of SEMA4B levels in NSCLC lines. (A–D) We either overexpressed SEMA4B in A549 and Calu-3 cells (A549-SEMA4B or Calu-3-SEMA4B cells), or inhibited SEMA4B in those cells (A549-shSEMA4B or Calu-3-shSEMA4B cells). A549 and Calu-3 cells transduced with scrambled sequence were used as controls (A549-Null and Calu-3-Null cells). A549shSEMA4B and Calu-3-shSEMA4B cells were further transduced by a constitutive nuclear FoxO1, resulting in A549-shSEMA4B-nFoxO1 and Calu-3-shSEMA4B-nFoxO1 cells. SMAD4b levels in these cells were examined by RT-qPCR (A, B) and by Western blot (C, D). α-Tubulin is a protein loading control. *: p b 0.05. NS: non-significant.
(DMEM) supplemented with 20% fetal bovine serum (Invitrogen, Carlsbad, CA, USA). The A549 cell line was first developed in 1972 by Dr. Giard through the removal and culturing of cancerous lung tissue in the explanted tumor a of 58-year-old Caucasian male [26]. Calu-3 cell line was developed by Dr. Fogh in 1975 [27].
2.3. Cell proliferation assay
2.2. Transduction of the cell lines
2.4. Apoptosis assay
SEMA4B-adapted plasmids have been previously described [6]. Briefly, SEMA4B construct was generated by sub-cloning PCRamplified full-length human SEMA4B cDNA into pcDNA3.1-EGFP-luciferase. For inhibition of SEMA4B, corresponding human short hairpin small interference RNA (ShRNA) sequence targeting “CTCTTCACCTTCCACATTATC” was cloned into pcDNA3.1-EGFP-luciferase to generate the shRNA construct. Scrambled sequence (null) was cloned into pcDNA3.1-EGFP-luciferase to generate control construct. These constructs were used to prepare lentiviral vectors to stably transduce the cells. Transfected cells expressing SEMA4B or ShSEMA4B or Null were selected by flow cytometry based on GFP, which is also included in the plasmid construct. The presence of luciferase allows in vivo tracing of the cells. For the production of cells with constitutively nuclear FoxO1, A549-shSEMA4B and Calu-3-shSEMA4B cells were further transduced with a constitutively nuclear FoxO1-expressing lentivirus (kindly provided by Shengzi Wang, Fudan University, China), as has been previously described [24,28]. The generated A549-shSEMA4B-nFoxO1 and Calu-3shSEMA4B-nFoxO1 cells express high levels of nuclear FoxO1, which cannot be phosphorylated due to a defect on the phosphorylation site.
Cells were labeled with annexin V-FITC and propidium iodide (PI), and then examined with an apoptosis detecting kit (Invitrogen, USA) for apoptosis. Samples were analyzed by flow cytometry and the results were analyzed by CellQuest software (Becton Dickinson, San Jose, CA) as has been described before [29].
For the assay of cell proliferation, pretreated cells were seeded into a 96 well-plate at 4000 cells per well and subjected to a cell proliferation kit (MTT, Roche, USA), according to the instruction from the manufacturer.
2.5. In vivo implanted NSCLC cancer model and imaging of the implanted cancer by bioluminescence Luciferase-carrying A549 cells (106) were injected through the tail vein into 12 weeks of age male NOD/SCID mice, as has been described before [30]. After 2 weeks, only the mice that have developed detectable luminescence were used for this study. At this stage, the size of the tumor is pretty small and comparable among all mice. These mice with a developed tumor were kept for another 4 weeks, and then the tumor growth was quantified by luminescence levels. Bioluminescence was measured with the IVIS imaging system (Xenogen Corp., Alameda, CA). All of the images were taken 10 min after intraperitoneal injection of luciferin (Sigma) of 150 mg/kg body weight, as a 60-second acquisition and 10 of binning. During image acquisition, mice were sedated
1210
H. Jian et al. / Cellular Signalling 27 (2015) 1208–1213
A
*
4
A549-shSEMA4b
O.D. 570 nm
3
2
*
A549-shSEMA4b-nFoxO1
NS
A549-Null NS
1
probed with a primary antibody. After incubation with a horseradish peroxidase-conjugated second antibody, autoradiograms were prepared using the enhanced chemiluminescent system to visualize the protein antigen. The signals were recorded using an X-ray film. The primary antibodies were anti-SEMA4B (Sigma), anti-FoxO1, anti-p21, anti-Bcl-2, anti-caspase-9, anti-LC3, anti-LaminB1 and anti-αtubulin (all from Cell Signaling, USA). The secondary antibody is an HRP-conjugated anti-rabbit (Jackson Labs). Figure images were representative from 5 repeats. α-Tubulin (for total protein or cytoplasmic protein, CY) and LaminB1 (for nuclear protein, NU) were used as protein loading controls.
A549-SEMA4b
*
0
*
1
0
2
2.8. Statistical analysis Time (day)
All statistical analyses were carried out using the SPSS 19.0 statistical software package. All values are depicted as the mean ± standard deviation from 5 individuals and are considered significant if p b 0.05. All data were statistically analyzed using one-way ANOVA with a Bonferroni correction.
Calu-3-shSEMA4b
3. Result
B 4
*
O.D. 570 nm
3
3.1. SEMA4B inhibited NSCLC growth in vitro 2
*
NS
Calu-3-shSEMA4b-nFoxO1 Calu-3-Null
NS
1
Calu-3-SEMA4b
*
0 0
1
* 2
Time (day)
Fig. 2. An MTT assay for cell growth. Cell growth was quantified in an MTT assay. *: p b 0.05. NS: non-significant.
continuously via inhalation of 3% isoflurane. Image analysis and bioluminescent quantification were performed using Living Image software (Xenogen Corp). 2.6. RT-qPCR RNA was extracted from the cultured NSCLC cells with RNeasy (Qiagen, Hilden, Germany) and then used for cDNA synthesis. RTqPCR was performed in duplicates with QuantiTect SYBR Green PCR kit (Qiagen). All primers were purchased from Qiagen. Values of genes were normalized against α-tubulin and then compared to controls. 2.7. Western blot For analysis of total protein, the protein was extracted from the cultured cells, which was homogenized in a RIPA lysis buffer (1% NP40, 0.1% SDS, 100 μg/mL phenylmethylsulfonyl fluoride, 0.5% sodium deoxycholate, in PBS) on ice. The supernatants were collected after centrifugation at 12,000 ×g at 4 °C for 20 min. Protein concentration was determined using a BCA protein assay kit (Bio-Rad, China), and whole lysates were mixed with 4 × SDS loading buffer (125 mmol/L Tris–HCl, 4% SDS, 20% glycerol, 100 mmol/L DTT, and 0.2% bromophenol blue) at a ratio of 1:3. Nuclear and cytoplasmic proteins were isolated with a nuclear and cytoplasmic extraction kit (Thermo Scientific, USA). Protein samples were heated at 100 °C for 5 min and were separated on SDS-polyacrylamide gels. The separated proteins were then transferred to a PVDF membrane. The membrane blots were first
In our previous study, we have reported a significant decrease in SEMA4B levels and a significant increase in the MMP9 levels in NSCLC from the patients. We have further shown that SEMA4B negatively regulated MMP9 levels, a key factor that induces NSCLC metastasis, exclusively through the PI3K signaling pathway [6]. Here we aimed to figure out whether SEMA4B may affect NSCLC growth through its inhibition on the PI3K downstream signals. Thus, we used two human NSCLC lines, A549 and Calu-3, for our study. We either overexpressed SEMA4B in A549 and Calu-3 cells, or inhibited SEMA4B in those cells. The overexpression and inhibition of SEMA4B in these NSCLC cells were confirmed by RT-qPCR (Fig. 1A– B), and by Western blot (Fig. 1C–D). We found that SEMA4B overexpression significantly decreased cell proliferation in both NSCLC lines in a MTT assay, while SEMA4B inhibition significantly increased cell proliferation in both NSCLC lines (Fig. 2A–B). The apoptosis of the cells were then evaluated by flow cytometry, using an apoptosis detecting kit. We found that adaptation of SEMA4B levels in either cell line did not affect the apoptosis of these NSCLC cells (Fig. 3A–B). Microtubule-associated protein 1A/1B-light chain 3 (LC3) is a soluble cellular protein. During autophagy, a cytosolic form of LC3 (LC3-I) is conjugated to phosphatidylethanolamine to form LC3-phosphatidylethanolamine conjugate (LC3-II), which is then recruited to autophagosomes to form autolysosomes. Thus, the autophagosomal marker LC3-II well reflects the autophagic activity [31–33]. We did not detect changes in LC3-II by SEMA4B modifications (Fig. 3B), suggesting that SEMA4B may not change autophagy in NSCLC cells. Thus, the effect of SEMA4B on the NSCLC growth is mainly through alteration in cell proliferation. Taken together, these data suggest that SEMA4B inhibits the growth of NSCLC in vitro. 3.2. SEMA4B inhibited NSCLC growth in vivo The A549-SEMA4B, A549-Null and A549-shSEMA4B cells also had a luciferase reporter for in vivo tracing. These A549 cells (106) were injected through the tail vein into 12 weeks of age male NOD/SCID mice. Only the mice that have developed detectable luminescence for implanted cancer cells in the lung region at two weeks were used for this study. These mice were kept for another 4 weeks, and then the tumor growth was quantified by luminescence levels, showing significantly greater tumors in mice receiving SEMA4Binhibited NSCLC cells, and significantly smaller tumors in mice
H. Jian et al. / Cellular Signalling 27 (2015) 1208–1213
A Fold change in DCF-DA levels (vs control)
1.5
NS
1211
NS
1.0
0.5
0
A5
49
-S
49
-N
EM
A5
A5
A5
49
hS
l
A4
A4
b
-3
-S
lu
-3
-s
l
-s h EM SEM A4 A4 b b-
b
b-
nF
ox
-3
hS
ul
A4
A4
b
Ca
lu
-N
EM
EM
Ca
lu
-3
hS
EM
Ca
lu
-s
-s
ul
Ca
49
nF
ox
O
O
1
1
B Caspase9
46 kDa
Bcl-2
26 kDa
LC3
16 kDa
α-tubulin
50 kDa A5
49
49
-S
49
-N
EM
hS
EM
b
Ca
-3
A4
lu
-N
EM
A4
Ca
lu
-3
-S
ul
-3
-s
hS
l
A4
EM
A4
A4
b-
b
nF
ox
hS
EM
b
b-
b
-3
-S
EM
Ca
lu
lu
-s
hS
l
Ca
49
-S
ul
A4
A5
A5
A5
nF
ox
O
O
1
1
Fig. 3. Analyses of cell apoptosis and autophagy. (A) Cell apoptosis was quantified with an apoptosis detecting kit. (B) Protein levels were examined by Western blot, shown by representative images. α-Tubulin is a protein loading control. NS: non-significant.
receiving SEMA4B-overexpressing NSCLC cells, compared to controls, by presentative images (Fig. 4A), and by quantification
B Fold change in luminescence (vs control)
A
(Fig. 4B). These data suggest that SEMA4B inhibits NSCLC growth in vivo.
A5
49
49
-s
hS
l
A4
b
1 0
* A5
EM
A4
A5
A5
49
49
-S
49
-N
ul
l
b
-S
ul
EM
2
A4
49
-N
10
EM
A5
A5
*
20
-s
hS
EM
A4
b
b
Fig. 4. Quantification of implanted NSCLC growth in living animals. (A–B) The A549-SEMA4B, A549-Null and A549-shSEMA4B cells (106) were injected through the tail vein into 12 weeks of age male NOD/SCID mice. The mice that have developed detectable luminescence for implanted cancer cells were kept for another 4 weeks, and then the tumor growth was quantified by luminescence levels, shown by representative images (A), and by quantification (B). *: p b 0.05.
1212
H. Jian et al. / Cellular Signalling 27 (2015) 1208–1213
A FoxO1
70 kDa
p21
21 kDa
LaminB1
67 kDa
α-tubulin
50 kDa CY
Nu
CY
A5
Nu
CY
49
-s
hS
hS
l
A4
49
-s
ul
EM
Nu
A5
49
-N
-S
CY
A5
A5
49
Nu
EM
EM
A4
b
b
A4
b-
nF
ox
O
C Fold change in luminescence (vs control)
B
20 10 2
0
A4
b
hS
EM
b
EM
EM
-s
-s
A4
hS
hS
49
49
EM
-s
-s
A5
A5
hS
49
49
*
1
A5
A5
1
A4
b-
nF
ox
A4
O
b-
nF
ox
1
O
1
Fig. 5. Constitutive nuclear FoxO1 abolished the effect of SEMA4B on cell growth. (A) Nuclear (NU) and cytoplasmic (CY) proteins from SEMA4B-modified A549 cells were isolated for Western blot. LaminB1 and α-tubulin were used to confirm the purity of the fractions. Representative Western blot images were shown. (B) A549-shSEMA4B and A549-shSEMA4BnFoxO1 cells (106) were injected through the tail vein into 12 weeks of age male NOD/SCID mice. The mice that have developed detectable luminescence for implanted cancer cells were kept for another 4 weeks, and then the tumor growth was quantified by luminescence levels, shown by representative images (A), and by quantification (B). *: p b 0.05.
3.3. SEMA4B may inhibit NSCLC growth through PI3K-regulated FoxO1 cellular location We recently reported that only PI3K signaling pathway is affected by SEMA4B in NSCLC [6], and the PI3K/Akt pathway is the major signaling pathway to regulate cell proliferation. Akt phosphorylation induces nuclear exclusion of FoxO1, resulting in the adaptation of p21, which suppresses cell proliferation. Therefore, we isolated nuclear vs cytoplasmic proteins from SEMA4B-modified A549 cells and performed Western blot with extracted proteins. LaminB1 and α-tubulin were used to assure the purity of the fractions, as has been applied before [3,23–25]. We found that overexpression of SEMA4B significantly increased the ratio of nuclear vs cytoplasmic FoxO1, while inhibition of SEMA4B significantly decreased this ratio by Western blot (Fig. 5A). Moreover, the nuclear vs cytoplasmic FoxO1 levels directly affected the nuclear vs cytoplasmic p21 levels (Fig. 5A). These data suggest that SEMA4B may inhibit NSCLC growth through PI3K/Akt-regulated cellular location of FoxO1 and its target p21. To prove this hypothesis, we further transduced the SEMA4B-depleted A549 or Calu-3 cells with a sustained nuclear FoxO1, as has been previously described [3,23,24]. The transduced cells expressed high levels of nuclear FoxO1, which cannot be phosphorylated due
to a defect on phosphorylation site (Fig. 5A). Importantly, the expression of sustained nuclear FoxO1 did not alter transcription levels of SEMA4B (Fig. 1A–B). Next, we examined the effect of sustained FoxO1 nuclear retention on the growth of NSCLC cells. Our data showed that sustained nuclear FoxO1 completely abolished the effects of SEMA4B inhibition in increasing the growth of NSCLC cells (Fig. 2A–B), without altering cell apoptosis (Fig. 3), suggesting that cellular location of FoxO1 is responsible for the effect of SEMA4B on the NSCLC cell growth. Moreover, sustained nuclear FoxO1 similarly abolished the effects of SEMA4B inhibition on the growth of the NSCLC cells in vivo (Fig. 5B–C). Taken together, based on the findings in the current study and our previous work, we propose a model in which SEMA4B inhibits PI3K/ Akt signaling in NSCLC, which not only regulates MMP9 to control metastasis, but also regulates the cellular location of FoxO1 and its target p21 to control cell growth (Fig. 6). 4. Discussion SEMA4B has been shown to play an important role in lung cancer invasion. However, whether SEMA4B may also regulate the cancer cell growth in lung cancer is not determined yet. Recently, we have shown
H. Jian et al. / Cellular Signalling 27 (2015) 1208–1213
NSCLC SEMA4b
MMP9
PI3K/Akt
Metastasis
FoxO1 FoxO1
1213
cellular location of FoxO1 is critical for its function, since only the nuclear form of FoxO1 can bind to DNA and affect target gene expression. Thus, in the current study, we simplified the evaluation on FoxO1, and used its cellular location for its biological activities. Moreover, our data that support a similar NSCLC-growthsuppressing role for SEMA4B are important, since it suggests that the effect of autologous SEMA4B on the cell growth in NSCLC works independently and is not affected by systematic signals. To summarize, based on the findings in the current study and our previous work, we propose a model in which SEMA4B inhibits PI3K/ Akt signaling in NSCLC, which not only regulates MMP9 to control metastasis, but also regulates cellular location of FoxO1 and its target p21 to control cell growth. These series of studies on the role of SEMA4B in NSCLC thus shed light on a novel target for NSCCL therapy, and suggest that genetic or pharmaceutical overexpression of SEMA4B in NSCLC in the patients may create a promising therapy for NSCLC.
Cell proliferation
Conflict of interest The authors have declared that no competing interests exist. Fig. 6. Schematic of the model of SEMA4B in NSCLC. SEMA4B inhibits PI3K/Akt signaling in NSCLC, which not only regulates MMP9 to control metastasis, but also regulates cellular location of FoxO1 and its target p21 to control cell growth.
that hypoxia-inducible factor 1 (HIF-1) could downregulate the expression of SEMA4B in human NSCLC lines, and then abolished its suppressive effect on NSCLC metastasis [5]. Moreover, we also reported a significant decrease in SEMA4B levels and a significant increase in the MMP9 levels in NSCLC from the patients and further showed that SEMA4B negatively regulated the MMP9 levels, a key factor that induces NSCLC metastasis, exclusively through the PI3K signaling pathway [6]. Thus, the inhibitory effect of SEMA4B may result from its active participation into the signaling pathways and interactions with other molecules that regulate angiogenic factors and matrix proteinases. Since we have shown that SEMA4B significantly inhibited PI3K/Akt pathway activities in the NSCLC cells, and since PI3K/Akt is a wellknown cell-cycle regulator, we hypothesize that SEMA4B may not only affect cancer cell invasiveness and metastasis, but also cancer cell growth in NSCLC. Thus, we used two human NSCLC lines, A549 and Calu-3, for the current study as we have done in our previous work. Using these two cell lines substantially decreased the possibility of drawing a conclusion based on cell-line-dependent data. For in vivo experiment, although we only showed data from A549 cells, we also analyzed Calu-3 cells in the same context and had similar results. In a series of gain-of-function and loss-of-function experiments, performed both in vitro and in vivo, we were able to show that SEMA4B may inhibit the growth of NSCLC, through its direct effect on PI3K/Akt-mediated changes in FoxO1 cellular location. FoxO1 proteins are transcriptional regulators that control cell cycle progression, DNA repair, defense against oxidative damage and apoptosis. Here we did not examine the detail of FoxO1 modification, since it is very complicated. Indeed, divergent functions of FoxO1 proteins are regulated by a variety of signal-induced, post-translational modifications. For example, the phosphorylation of cytoplasmic FoxO1 at specific sites by JNK initiates translocation into the nucleus. Acetylation and deacetylation of nuclear FoxO affect FoxO1-dependent transcriptional programs. The phosphorylation of Akt results in the phosphorylation of nuclear FoxO1 at specific sites followed by additional phosphorylations mediated by other kinases. Akt-dependent phosphorylation reduces the DNA-binding activity of FoxO1, inactivates the FoxO nuclear translocation signal and results in FoxO1 nuclear exclusion, in a CRM1and 14-3-3-dependent process. However, whatever regulated, the
Acknowledgment This work is financially supported by Wujieping Medical Foundation No: 320.6750.14060, and Shanghai Chest Hospital Science and Technology Development Fund (2014 YZDC10600). References [1] G. Caramori, P. Casolari, G.N. Cavallesco, S. Giuffre, I. Adcock, A. Papi, Int J Biochem Cell Biol 43 (2011) 1030–1044. [2] R.C. Buttery, R.C. Rintoul, T. Sethi, Int J Biochem Cell Biol 36 (2004) 1154–1160. [3] J. Pei, Y. Lou, R. Zhong, B. Han, Tumour Biol 35 (2014) 6673–6678. [4] A. Ahmed, B.J. Eickholt, Adv Exp Med Biol 600 (2007) 24–37. [5] H. Jian, B. Liu, J. Zhang, Tumour Biol 35 (2014) 4949–4955. [6] H. Jian, Y. Zhao, B. Liu, S. Lu, Tumour Biol 35 (2014) 11051–11056. [7] H. Nagai, N. Sugito, H. Matsubara, Y. Tatematsu, T. Hida, Y. Sekido, et al., Oncogene 26 (2007) 4025–4031. [8] J.R. Basile, T. Afkhami, J.S. Gutkind, Mol Cell Biol 25 (2005) 6889–6898. [9] H. Aghajanian, C. Choi, V.C. Ho, M. Gupta, M.K. Singh, J.A. Epstein, J Biol Chem 289 (2014) 17971–17979. [10] G. Pan, Z. Zhu, J. Huang, C. Yang, Y. Yang, Y. Wang, et al., Dig Dis Sci 58 (2013) 2197–2204. [11] G. Gallo, Dev Neurobiol 68 (2008) 926–933. [12] J.K. Atwal, K.K. Singh, M. Tessier-Lavigne, F.D. Miller, D.R. Kaplan, J Neurosci 23 (2003) 7602–7609. [13] H. Wen, Y. Lei, S.Y. Eun, J.P. Ting, J Exp Med 207 (2010) 2943–2957. [14] C. Burkhardt, M. Muller, A. Badde, C.C. Garner, E.D. Gundelfinger, A.W. Puschel, FEBS Lett 579 (2005) 3821–3828. [15] S. Kim, J.H. Choi, H.I. Lim, S.K. Lee, W.W. Kim, S. Cho, et al., Cell Signal 21 (2009) 892–898. [16] M.R. Schneider, E. Wolf, J Cell Physiol 218 (2009) 460–466. [17] F.R. Hirsch, P.A. Janne, W.E. Eberhardt, F. Cappuzzo, N. Thatcher, R. Pirker, et al., J Thorac Oncol 8 (2013) 373–384. [18] K. Kobayashi, K. Hagiwara, Target Oncol 8 (2013) 27–33. [19] T. Sasaki, K. Hiroki, Y. Yamashita, Biomed Res Int 2013 (2013) 546318. [20] S.B. Prasad, S.S. Yadav, M. Das, H.B. Govardhan, L.K. Pandey, S. Singh, et al., J Cancer 5 (2014) 655–662. [21] P. Liu, T.P. Kao, H. Huang, Oncogene 27 (2008) 4733–4744. [22] M. Fosbrink, F. Niculescu, V. Rus, M.L. Shin, H. Rus, J Biol Chem 281 (2006) 19009–19018. [23] J. Chen, Q. Huang, F. Wang, Tumour Biol 35 (2014) 7195–7200. [24] H. Ding, Y. Zhu, T. Chu, S. Wang, Tumour Biol 35 (2014) 9987–9992. [25] X. Xiao, I. Gaffar, P. Guo, J. Wiersch, S. Fischbach, L. Peirish, et al., Proc Natl Acad Sci U S A 111 (2014) E1211–E1220. [26] D.J. Giard, S.A. Aaronson, G.J. Todaro, P. Arnstein, J.H. Kersey, H. Dosik, et al., J Natl Cancer Inst 51 (1973) 1417–1423. [27] J. Fogh, J.M. Fogh, T. Orfeo, J Natl Cancer Inst 59 (1977) 221–226. [28] W.H. Biggs III, J. Meisenhelder, T. Hunter, W.K. Cavenee, K.C. Arden, Proc Natl Acad Sci U S A 96 (1999) 7421–7426. [29] Q. Mei, F. Li, H. Quan, Y. Liu, H. Xu, Cancer Sci 105 (2014) 755–762. [30] W. Long, C.E. Foulds, J. Qin, J. Liu, C. Ding, D.M. Lonard, et al., J Clin Invest 122 (2012) 1869–1880. [31] D.R. Green, B. Levine, Cell 157 (2014) 65–75. [32] J.Y. Guo, B. Xia, E. White, Cell 155 (2013) 1216–1219. [33] E. White, Nat Rev Cancer 12 (2012) 401–410.
