- Email: [email protected]
Neural stem cell differentiation is mediated by integrin β4 in vitro
The International Journal of Biochemistry & Cell Biology 41 (2009) 916–924
Contents lists available at ScienceDirect
The International Journal of Biochemistry & Cell Biology journal homepage: www.elsevier.com/locate/biocel
Neural stem cell differentiation is mediated by integrin 4 in vitro Le Su a,b , Xin Lv a,b , JiPing Xu c , DeLing Yin d , HaiYan Zhang a,b , Yi Li d , Jing Zhao a,b , ShangLi Zhang a,b , JunYing Miao a,b,∗ a
Institute of Developmental Biology, School of Life Science, Shandong University, Jinan 250100, China The Key Laboratory of Cardiovascular Remodeling and Function Research, Chinese Ministry of Education and Chinese Ministry of Health, Shandong University Qilu Hospital, Jinan 250012, China c Department of Neural Medicine, The Second Hospital of Shandong University, Jinan 250100, China d Department of Internal Medicine, East Tennessee State University, Box 70622, Johnson City, TN 37604, USA b
a r t i c l e
i n f o
Article history: Received 18 July 2008 Received in revised form 5 September 2008 Accepted 6 September 2008 Available online 12 September 2008 Keywords: Integrin 4 Neural stem cell Differentiation RNA interference Fibroblast growth factor receptor
a b s t r a c t Neural stem cells are capable of differentiating into three major neural cell types, but the underlying molecular mechanisms remain unclear. Here, we investigated the mechanism by which integrin 4 modulates mouse neural stem cell differentiation in vitro. Inhibition of endogenous integrin 4 by RNA interference inhibited the cell differentiation and the expression of fibroblast growth factor receptor 2 but not fibroblast growth factor receptor 1 or fibroblast growth factor receptor 3. Overexpression of integrin 4 in neural stem cells promoted neural stem cell differentiation. Furthermore, integrin 4-induced differentiation of neural stem cells was attenuated by SU5402, the inhibitor of fibroblast growth factor receptors. Finally, we investigated the role of integrin 4 in neural stem cell survival: knockdown of integrin 4 did not affect survival or apoptosis of neural stem cells. These data provide evidence that integrin 4 promotes differentiation of mouse neural stem cells in vitro possibly through fibroblast growth factor receptor 2. © 2008 Elsevier Ltd. All rights reserved.
1. Introduction The central nervous system appears to have only limited potential to generate new neurons and gliacytes. Neural transplantation of stem cells is used in clinical trials of many degenerative neurological diseases, including Huntington’s disease, Parkinson’s disease, and strokes (Kwon, 2002), so determining the key factors that regulate the fate of neural stem cells (NSCs) is important. Integrins, a large family of heterodimeric receptors, of ␣ and  subunits, play fundamental roles in cellular signaling (Tarone et al., 2000). In contrast to other  subunits, integrin 4 has an exceptionally large cytoplasmic domain, which is associated with cytoskeletal and signaling molecules (Hogervorst et al., 1990). It participates in the initiation of myelin sheath formation in Schwann cells and anchors astrocytes to the vascular basal lamina to increase the resistance of the blood brain barrier (Su et al., 2008). Recently, the expression of integrin 4 was found to induce monocytic and pancreatic carcinoma cell differentiation (Stagge et al., 2001; Morena et al., 2002). Additionally, in our previous study, knock-
∗ Corresponding author at: Institute of Developmental Biology, School of Life Science, Shandong University, Jinan 250100, China. Tel.: +86 531 88364929; fax: +86 531 88565610. E-mail address: [email protected] (J. Miao). 1357-2725/$ – see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.biocel.2008.09.001
down of integrin 4 could induce neuronal apoptosis (Lv et al., 2008). However, the roles of integrin 4 in NSC differentiation and survival are not clear. One proposed mechanism critical to NSC biology involves the activity of fibroblast growth factor receptors (FGFRs). Recent studies showed that coactivation of FGFR1 and FGFR3 promoted selfrenewal of NSCs, whereas developmental up-regulation of FGFR2 expression was associated with a shift of NSCs into a multipotential state or apoptosis (Maric et al., 2007). Furthermore, FGFR1 could associate with ␣v3 integrin to form an integrin–FGFR signaling complex (Tanghetti et al., 2002). However, whether integrin 4 regulates NSC differentiation through FGFRs is not clear. Therefore, we investigated the relation between integrin 4 and FGFRs in mouse NSC differentiation in vitro. 2. Materials and methods 2.1. Primary culture of cells Mouse NSC neurospheres were obtained from embryo forebrains on day 14 of gestation under sterile conditions as described (the mice were purchased from Shandong University Laboratory Animal Center) (Su et al., 2007a). Briefly, pregnant mice were decapitated, the brains were removed and placed in artificial cerebrospinal fluid, then brains were cut into small pieces and
L. Su et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 916–924
917
Fig. 1. Integrin 4 level in neurospheres, primary cultured neurons and differentiated NSCs. (A) Level of integrin 4 in neurospheres obtained from the embryo forebrains on day 14 of gestation. (B) Level of integrin 4 in differentiated cells induced by 10% FBS. (C) Relative quantity of integrin 4 in (A) and (B). (D) Level of integrin 4 in NSCs from the embryo forebrains on day 14 of gestation. (E) Content of integrin 4 in primary cultured mouse neurons obtained from the embryo forebrains on day 14 of gestation. (F) Quantity of integrin 4 in (D) and (E). Values in (C) and (F) represent the relative fluorescent intensity per cell by laser scanning confocal microcopy. (**P < 0.01 vs. # , n = 3). (G–J) Micrographs observed on transmission microscopy corresponding to (A, B, D and E).
digested with use of papain (Worthington Biochemical Lakewood, NJ) at 37 ◦ C for 45 min. The cells, collected by centrifugation, were resuspended, then the material was passed through a 70-m cell strainer to remove undissociated large pieces of tissue. After the material was centrifuged at 300 × g for 10 min, the cells were resuspended in the supplemented culture medium with 0.1 mM l-glutamine, 4 mM d-glucose, 0.4 mM HEPES, 2% B27 and 20 ng/mL recombinant human basic fibroblast growth factor (EssexBio Group, China). Cells were grown at 37 ◦ C in 5% CO2 and 95% air with saturated humidity. The cell concentration obtained was approximately 1 × 106 cells/mL. Seven days after plating, neurospheres were mechanically dissociated and subcultured as described above. Primary cultured mouse neurons were obtained from embryo forebrains on day 14 of gestation under sterile conditions as described (Su et al., 2007b). To induce neurosphere differentiation, 50–100 neurospheres were plated onto poly-lysine (poly-l-lysine, Sigma)-coated plastic dishes then cultured in medium supplemented with 10% fetal bovine serum (FBS) for 72 h to induce NSC differentiation (Vescovi et al., 1993). The neurospheres were observed under a phase-contrast microscope (Nikon, Japan) for cell morphology analysis. The cell viability was evaluated by trypan blue dye exclusion test as described
(Altman et al., 1993): the cell suspension was mixed with the same volume of 0.4% trypan blue solution, and cells were counted under a light microscope. Cell viability was estimated by the proportion of unstained cells. 2.2. RNA interference (RNAi) To knock down the expression of integrin 4 protein, RNAi was performed as described (Walczak et al., 2005; Wen et al., 2005) with the use of the specific integrin 4 small interfering RNA (siRNA), a pool of 3 target-specific, 20- to 25-nt siRNAs (sc-35679; Santa Cruz Biotechnology, Santa Cruz, CA). RNAi experiments followed the manufacturer’s protocols. An amount of 6 × 105 cells was suspended in siPORTTM siRNA Electroporation Buffer (Ambion, USA) in sterile 0.4-cm-gap electroporation cuvettes containing 2 g integrin 4 siRNA following the manufacturer’s protocols. Cells were electroporated by use of a Gene Pulser Xcell Electroporation system (Bio-rad, USA). One pulse lasted for about 8 ms at 160 V. Cells were left in the cuvette holder for 10 min, then moved into 24well plates in the CO2 incubator. To evaluate siRNA-mediated gene silencing, scramble siRNA (Santa Cruz Biotechnology) was used as a control.
918
L. Su et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 916–924
2.3. Overexpression of integrin ˇ4 in NSCs The cDNA constructs encoding wild-type integrin 4 subunit (pCMV-4) (kindly provided by Dr. Filippo Giancotti, New York University Medical Center) were transfected by electroporation as for RNAi above (Morena et al., 2002). Transfection controls used the same amount of empty vector (pRC-CMV) alone. We monitored the effect of overexpression 48 h after electroporation using immunofluorescence staining combined with laser scanning confocal microscopy (Spinardi et al., 1995). We calculated the overexpression rate by counting the cells that expressed a high level of integrin 4 and total cells in the screen. 2.4. Flow cytometry The expression level of integrin 4 was analyzed by flow cytometry analysis of stained cells (anti-mouse CD104 4 chain, 553745, 554016; BD Biosciences, USA) (Morena et al., 2002). Cells were washed twice with cold PBS containing 0.002% ethylenediaminetetraacetic acid. Samples of 1 × 106 untransfected or transfected cells were incubated for 30 min at 4 ◦ C with primary antibody. Cells were
then washed three times with washing buffer (PBS containing 0.5% bovine serum albumin) and incubated for 30 min at 4 ◦ C with FITCconjugated secondary antibody. After extensive washes, 2 × 104 cells were acquired for each sample, and cell fluorescence intensity was measured by flow cytometry (FACSCalibur, BD Biosciences) and analyzed by use of the CELL-QUEST software. Cells incubated without primary antibody were used as a negative control. 2.5. Immunocytochemical staining Immunocytochemical staining was performed as described (Hu et al., 2005). Primary antibodies (rabbit anti-mouse neuron-specific enolase [NSE] IgG, sc-31859; Synapsin IgG, sc-20780; glial fibrillary acidic protein [GFAP] IgG, sc-9065; integrin 4 IgG, sc-9090; FGFR1 IgG, sc-121; FGFR2 IgG, sc-122; FGFR3 IgG, sc-123; Santa Cruz Biotechnology) were 1:100, and secondary antibody labeled with goat anti-rabbit FITC was diluted 1:200 (sc-65218; Santa Cruz Biotechnology). Cell nuclei were counterstained with propidium iodide (10 g/mL) for 15 min at room temperature. Samples were evaluated by laser scanning confocal microscopy (Leica, Germany). We randomly selected the region of interest and then zoomed in
Fig. 2. siRNA-mediated knockdown of integrin 4 in NSCs analyzed by flow cytometry. (a) Negative control, NSCs incubated without primary antibody. (b) Untransfected, NSCs incubated with primary and secondary antibodies. (c) Control, NSCs transfected with scramble control siRNA incubated with primary and secondary antibodies for 48 h. (d) 4 siRNA, NSCs transfected with integrin 4 siRNA incubated with primary and secondary antibodies for 48 h.
L. Su et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 916–924
919
the same frames. The relative fluorescent intensity per cell was the total value of the sample in the zoom scans divided by the total number of cells (at least 200 cells) in the same scan.
et al., 1997; Bernard-Pierrot et al., 2004; Mansukhani et al., 2005). SU5402 (10 M) was added into the medium when NSCs were transfected with pCMV-4.
2.6. Inhibition of FGFR2
2.7. Cell death assay
To inhibit the activity of FGFR2, NSCs were treated with the inhibitor SU5402 (Calbiochem, San Diego, CA), a member of a new class of FGFR antagonists that block the tyrosine kinase activity of the receptor by interacting with the catalytic domain (Mohammadi
For cell death assay, we used terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) assay according to the manufacturer’s protocol (Promega). Cells were observed under a laser scanning confocal microscope (Leica,
Fig. 3. Integrin 4 inhibition attenuates NSC differentiation. (A–C) Phase-contrast micrographs (200×). (D–L) Fluorescent micrographs. (A, D, G and J) Neurospheres obtained from the embryo forebrains on day 14 of gestation. (B, E, H and K) NSCs transfected with scramble control siRNA, then treated with 10% FBS for 72 h. (C, F, I and L) NSCs transfected with integrin 4 siRNA, then treated with 10% FBS for 72 h. (D–F) Expression of NSE in NSCs. G-I: Expression of synapsin in NSCs. (J–L) Expression of GFAP in NSCs.
920
L. Su et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 916–924
Germany). The proportion of apoptotic cells was quantified by the TUNEL-positive rate. Cells were treated with integrin 4 or scramble siRNA for 48 h and 72 h, then underwent lactate dehydrogenase (LDH) release analysis by use of an LDH kit according to the manufacturer’s protocol (ZhongSheng, China) (Lv et al., 2008). LDH catalyzes the reduction of pyruvate to l-lactate, with concomitant oxidation of NADH2 to NAD. Since the oxidation of NADH2 is directly proportional to the reduction of pyruvate in equimolar amounts, the LDH activity can be calculated from the rate of decrease in absorbance at 340 nm. Light absorption was measured by use of a model Cintra 5 UV–vis spectrometer (GBC, Australia), and LDH activity was calculated as follows: LDH(U/L) = (Asample/ min −Ablank/ min) × F, where F = 1000 × V total/(V sample × extinction coefficient); A = change in absorbance; V = volume in ml; extinction coefficient = mmol absorptivity of NADH at 340 nm was 6.3. Cells were fixed in 4% formaldehyde for 10 min, then incubated with Hoechst 33258 (2 g/ml) (Sigma, St. Louis, MO, USA) for stain-
ing for 60 min at 37 ◦ C. Stained cells were washed with PBS twice, then viewed under an Olympus inverted fluorescence microscope. Cells were scored as apoptotic if their nuclei were much brighter or exhibited condensation of chromatin and nuclear fragmentation. 2.8. Statistical analysis Data are expressed as mean ± S.E. An average of more than 10 spheres was analyzed per treatment, per experiment. SPSS 11.5 software (SPSS Inc., Chicago, IL) was used for statistical calculations. One-way ANOVA (followed by Scheffé F-test for post hoc analysis) was used for statistical analysis. A P < 0.05 was considered statistically significant. 3. Results 3.1. Level of integrin ˇ4 was increased during NSC differentiation The level of integrin 4 was much higher in differentiated NSCs induced by 10% FBS than in untreated neurospheres (Fig. 1A–C).
Fig. 4. RNAi knockdown of integrin 4 inhibits the expression of FGFR2. (A) Untransfected neurospheres obtained from mouse embryo forebrains on day 14 of gestation. Control, NSCs transfected with scramble control siRNA, then plated onto poly-lysine-coated plastic dishes and treated with 10% FBS for 72 h. 4 siRNA, NSCs transfected with integrin 4 siRNA, then plated onto poly-lysine-coated plastic dishes and treated with 10% FBS for 72 h. Narrow arrows, cells underwent differentiation in integrin 4 siRNA group. Wide arrow, cells with differentiation inhibited. (B) Micrographs from transmission microscopy corresponding to (A). (C) Quantity of FGFR levels in the three groups mentioned above. Value represents the relative fluorescent intensity per cell determined by laser scanning confocal microcopy. **P < 0.01 vs. control, n = 3.
L. Su et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 916–924
However, the level of integrin 4 was higher in primary cultured neurons than in mouse NSCs obtained from the same embryo forebrain on day 14 of gestation (Fig. 1D–F). The negative control lacking primary antibody showed no staining, which demonstrated that the staining was specific (data not shown). The level of integrin 4 was increased linearly during NSC differentiation. 3.2. siRNA-mediated knockdown of integrin ˇ4 in NSCs Control siRNA had no effect on level of integrin 4 in transfected NSCs (Fig. 2c), and untransfected and control groups did not differ in fluorescence intensity (Fig. 2b and c). However, the expression of integrin 4 was downregulated significantly after treatment with the specific siRNA for 48 h (Fig. 2d), with fluorescence intensity 2.2% compared to 57.3% in control group.
921
3.4. RNAi knockdown of integrin ˇ4 inhibits the expression of FGFR2 Next, we measured the levels of FGFR1-3 by immunofluorescence combined with laser scanning confocal microscopy for semi-quantitative evaluation of protein level (Dinser et al., 2001; Lv et al., 2008). The integrin 4 siRNA group and control group did not differ in level of FGFR1 (Fig. 4A, a–c) or FGFR3 (Fig. 4A, g–i) (P > 0.05), but the level of FGFR2 (Fig. 4A, d–f) was remarkably reduced (P < 0.01) (Fig. 4C). Few cells underwent differentiation and NSCs showed little neurite outgrowth (Fig. 4B, f; narrow arrows). However, most cells transfected with integrin 4 siRNA still showed the NSC shape and did not extend neurites (wide arrow). Thus, the level of FGFR2 was decreased with knockdown of integrin 4 in NSCs. 3.5. Overexpression of integrin ˇ4
3.3. Integrin ˇ4 inhibition attenuates NSC differentiation Since integrin 4 level was increased during NSC differentiation (Fig. 1), we investigated whether the NSC differentiation could be inhibited by knockdown of integrin 4. After transfection with control siRNA and integrin 4 siRNA for 72 h, the cells in the control group underwent differentiation and the adjacent neurospheres formed a network by neurites (Fig. 3B) that was similar to the untransfected group (Fig. 3A). However, in the integrin 4 siRNA group, only a few cells underwent differentiation and NSCs showed little neurite outgrowth (Fig. 3C). To confirm the NSC differentiation, we examined by immunofluorescence staining the expression of specific markers for neurons and astrocytes, including NSE (Fig. 3D–F) and synapsin (Fig. 3G–I) for neurons and GFAP (Fig. 3J–L) for astrocytes. Most of the untransfected (Fig. 3D, G, and J) and control-group cells (Fig. 3E, H, and K) were positive for neuron and astrocyte markers. However, the integrin 4 siRNA group showed a markedly reduced quantity of marker-positive cells (Fig. 3F, I, L), which suggests that NSC differentiation was blocked by knockdown of integrin 4.
NSCs transfected with control cDNA (pRC-CMV) showed no effect on level of integrin 4 (Fig. 5A, d–f), with no difference in the level between untransfected and control cells (Fig. 5A, a–f). However, with integrin 4 overexpression, the level of integrin 4 was elevated by about 25% (Fig. 5A, g–i; B). This transfection rate is consistent with other results in NSCs (Zwaka and Thomson, 2003). 3.6. Integrin ˇ4 overexpression promoted NSC differentiation in vitro After cells were transfected with pCMV-4 and pRC-CMV for 3 days, phase contrast microscope revealed few cells undergoing differentiation in the control group (Fig. 6A, b) and NSCs showed little neurite outgrowth, a finding similar to that in the untransfected group (Fig. 6A, a). However, in the integrin 4 overexpression group, most cells differentiated, and the adjacent neurospheres formed a network by neurites (Fig. 6A, c). Immunofluorescent double staining of overexpressed NSCs for NSE (Fig. 6B) and GFAP (Fig. 6C) in the untransfected (Fig. 6B
Fig. 5. Overexpression of integrin 4 in NSCs. (A) Immunofluorescent double staining of overexpression of integrin 4 in NSCs. Untransfected, neurospheres obtained from mouse embryo forebrains on day 14 of gestation. Control, NSCs transfected with empty vectors (pRC-CMV) for 48 h. Overexpression, NSCs transfected with wild-type integrin 4 cDNA (pCMV-4) for 48 h. (a, d, and g) Fluorescence photomicrographs of NSCs expressing integrin 4. (b, e, and h) Cell nuclei counterstained with PI. (c, f, and i) Overlay photographs of integrin 4 and PI. (B) The rate integrin 4 overexpression in NSCs was about 25% (n = 3).
922
L. Su et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 916–924
Fig. 6. Overexpression of integrin 4 stimulates NSC differentiation. Untransfected, the neurospheres obtained from mouse embryo forebrains on day 14 of gestation. Control, NSCs transfected with empty vector (pRC-CMV) for 3 days. Overexpression, NSCs transfected with wild-type integrin 4 cDNA (pCMV-4) for 3 days. SU5402, NSCs transfected with pCMV-4 and treated with 10 M SU5402 for 3 days. (A) Overexpression of integrin 4 promoted NSC differentiation as seen on phase-contrast microscopy. (B) (a, d, g, and j) Fluorescence photomicrographs of NSCs stained with NSE. (b, e, h, and k) Cell nuclei counterstained with PI. (c, f, i, and l) Overlay photographs of NSE and PI. (C) (a, d, g, and j) Fluorescence photomicrographs of NSCs expressing GFAP. (b, e, h, and k) Cell nuclei counterstained with PI. (c, f, i, and l) Overlay photographs of GFAP and PI. (D) The percentages of the positive cells in untransfected, control, overexpression and SU5402 groups (*P < 0.05, n = 3, **P < 0.01, n = 3).
and C, a–c) and control groups (Fig. 6B and C, d–f) revealed few cells positive for neuron and astrocyte markers. However, in the integrin 4-overexpression group, the quantity of markerpositive cells was remarkably increased (Fig. 6B and C, g–i). Immunofluorescent staining revealed NSCs overexpressing integrin 4 with significantly increased viability to differentiate (Fig. 6D). Immunofluorescent double staining revealed that treatment with SU5402, a specific inhibitor of FGFR, reduced the number of positive neurons and astrocytes (Fig. 6B and C, j–l). The NSCs treated with SU5402 showed a significantly lower proportion of
neurons (P < 0.01) and astrocytes (P < 0.01) than cells transfected with pCMV-4 (Fig. 6D). 3.7. Suppression of integrin ˇ4 does not induce NSC apoptosis and necrosis Finally, we examined the effect of integrin 4 on NSC apoptosis. The results from morphology and trypan blue exclusion showed that knockdown of integrin 4 had no effect on NSC survival (Fig. 7A and D). After knockdown with integrin 4 siRNA, the proportion of apoptotic cells did not differ between control and integrin 4 siRNA
L. Su et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 916–924
923
Fig. 7. Suppression of integrin 4 does not induce NSC apoptosis and necrosis. (A) The morphology of untransfected NSCs (a), NSCs transfected with scramble control siRNA for 72 h (b) and NSCs transfected with integrin 4 siRNA for 72 h (c) under a phase contrast microscope (200×). (B) Quantification of apoptotic cells by TUNEL assay. The treatments correspond to those of (A). Scramble control siRNA and integrin 4 siRNA groups (d, & P > 0.05, n = 3). (C) Hoechst 33258 nuclear staining of untransfected NSCs (a), NSCs transfected with scramble control siRNA for 72 h (b) and NSCs transfected with integrin 4 siRNA for 72 h (c). The arrows indicate apoptotic cells. (D) Quantification of living cells by trypan blue staining. Cells were transfected with integrin 4 siRNA and scramble control siRNA for 24 h, 48 h and 72 h. (& P > 0.05 vs. control groups at the same time, n = 4). We assumed the viability of untransfected cells as 100%. (E) Effect of integrin 4 siRNA on lactate dehydrogenase (LDH) release from NSCs. Cells were treated with integrin 4 siRNA or scramble control siRNA for 48 h and 72 h. Light absorption was analyzed at 340 nm with use of a model Cintra 5 UV–vis spectrometer (& P > 0.05, n = 3).
groups at 48 h and 72 h (P > 0.05; Fig. 7B). The results from Hoechst 33258 staining showed that nuclear condensation was similar in the control and integrin 4 siRNA groups (Fig. 7C), which suggests that knockdown of integrin 4 did not trigger NSC apoptosis. To understand whether knockdown of integrin 4 induces NSC necrosis, we measured LDH release from NSCs treated with integrin 4 siRNA for 48 h and 72 h. LDH release did not differ between control and siRNA-treated groups (P > 0.05) (Fig. 7E), which suggests that knockdown of integrin 4 did not cause necrosis in NSCs. 4. Discussion Although many reports have described NSC proliferation and differentiation, the exact mechanisms of NSC differentiation are not well understood. Integrin 4 is expressed in astrocytes, Schwann cells and neurons (Su et al., 2008). Recently, human neurospheres were found to lack integrin 4 mRNA (Hall et al., 2006), but another report showed that laminin matrices enhanced human NSC differentiation and elongation of neurites from NSPC-derived neurons (Flanagan et al., 2006). Flow cytometry revealed that human neural stem/precursor cells (NSPCs) express the laminin-binding integrins ␣3, ␣6, ␣7, 1, and 4, and function-blocking antibodies to ␣6 subunit confirmed a role for integrins in laminin-dependent migration of human NSPCs (Flanagan et al., 2006). These data suggest that
laminin and its integrin receptors, including integrin 4, might be key regulators of human NSPCs. However, no direct evidence has shown the function of integrin 4 in NSC differentiation and survival. In this study, we found integrin 4 protein expressed in mouse NSCs, which is consistent with previous results (Flanagan et al., 2006). Moreover, we observed the level of integrin 4 increased during NSC differentiation, which encouraged the exploration of the role of integrin 4 in NSC differentiation and survival. In vitro, NSCs grown in suspension form floating clusters termed neurospheres. Because neurospheres are clonally derived from neural stem cells, they provide a good experimental system for studying neural cell differentiation, not only to obtain phenotypes of potential therapeutic interest but also to define the basic mechanisms that regulate the acquisition of a specific cell fate in the nervous system. So we examined the role of integrin 4 on the generation of neurons and gliacytes from NSCs by using neurospheres. Increasing evidence indicates that integrin 4 induces cell differentiation (Stagge et al., 2001). Additionally, the cytoplasmic domain of the 4 subunit is directly involved in signaling events that generate monocytic differentiation (Morena et al., 2002). Differentiated enterocytes express a full-length 205-kDa 4 subunit, whereas undifferentiated crypt cells express a 4 subunit that does not contain the COOH-terminal segment of the cytoplasmic domain (4 ctd-integrin). Furthermore, 4 ctd-integrin expressed
924
L. Su et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 916–924
by the undifferentiated crypt cells is not functional for adhesion to laminin-5 (Basora et al., 1999). In this study, we showed that knockdown of integrin 4 by its specific siRNA that targets the cytoplasmic tail of this integrin inhibited NSC differentiation, and overexpression of the full-length integrin 4 promoted NSC differentiation. As well, differentiation of not only astrocytes but also neurons was promoted by overexpression of integrin 4. The data provide new evidence for the role of the integrin 4 cytoplasmic tail in cell differentiation. One recent report suggested that co-activation of FGFR1 and FGFR3 promoted symmetrical division of NSCs (self-renewal), whereas inactivation of either receptor triggered asymmetrical division and neurogenesis from these cells. Developmental upregulation of FGFR2 expression was associated with a shift of NSCs into a multipotential state or apoptosis (Maric et al., 2007). As well, FGFR1 could be associated with ␣v3 integrin forming an integrin–FGFR signalling complex to potentiate FGF2 signaling (Tanghetti et al., 2002). However, the relation between FGFRs and integrin 4 is not clear. Here, we found that knockdown of integrin 4 could inhibit NSC differentiation, and during this differentiation inhibitory process, the levels of FGFR1 and FGFR3 were not altered, but the level of FGFR2 was significantly depressed. These results support that FGFR2 but not FGFR1 or FGFR3 might be involved in the NSC differentiation mediated by integrin 4. Furthermore, use of the specific inhibitor of FGFR, SU5402, validated the involvement of FGFR2 in the increased differentiation ability of cells transfected with pCMV-integrin 4. These results provide new insights into the roles of FGFRs and integrin 4 in diversification of NSC properties and initiation of neural lineage-restricted differentiation. Laminin-5 is an established ligand of the alpha6beta4-integrin (Giancotti, 2007) and integrin–laminin matrices can enhance NSPC differentiation (Flanagan et al., 2006). In our study, we found that integrin 4 and FGFR2 might form a complex to induce NSC differentiation. Whether this process depends on the binding of the alpha6beta4-integrin to laminin-5 is a subject for further study. In a previous study, knockdown of integrin 4 could induce apoptosis in primary cultured mouse neurons (Lv et al., 2008), which is similar to the function of integrin 4 in epithelial cells (Werner et al., 2007). In the current study, knockdown of integrin 4 did not induce apoptosis but did inhibit differentiation of mouse NSCs. The data further suggest the cell type-specific function of this integrin subunit. In summary, blocking integrin 4 by its specific siRNA inhibited differentiation but did not induce apoptosis in mouse NSCs. As well, the level of FGFR2 was depressed significantly. Furthermore, overexpression of integrin 4 in NSCs promoted neuron and astrocyte differentiation in vitro, and this promotion was suppressed by the specific inhibitor of FGFR, SU5402. These results show that integrin 4 is a key factor in NSC differentiation, and this integrin subunit might perform its action through FGFR2 in the differentiation process. Acknowledgements This work was supported by the National Natural Science Foundation of China (No. 30470404) and the National 973 Research Project (No. 2006CB503803). The authors thank Dr. Filippo Giancotti for the generous gift of wild-type integrin 4 cDNA. References Altman SA, Randers L, Rao G. Comparison of trypan blue dye exclusion and fluorometric assays for mammalian cell viability determinations. Biotechnology Progress 1993;9:671–4.
Basora N, Herring-Gillam FE, Boudreau F, Perreault N, Pageot LP, Simoneau M, et al. Expression of functionally distinct variants of the beta(4)A integrin subunit in relation to the differentiation state in human intestinal cells. The Journal of Biological Chemistry 1999;274:29819–25. Bernard-Pierrot I, Ricol D, Cassidy A, Graham A, Elvin P, Caillault A, et al. Inhibition of human bladder tumour cell growth by fibroblast growth factor receptor 2b is independent of its kinase activity. Involvement of the carboxy-terminal region of the receptor. Oncogene 2004;23:9201–11. Dinser R, Kreppel F, Zaucke F, Blank C, Paulsson M, Kochanek S, et al. Comparison of long-term transgenic expression after nonviral and adenoviral gene transfer into primary articular chondrocytes. Histochemistry and Cell Biology 2001;116:69–77. Flanagan LA, Rebaza LM, Derzic S, Schwartz PH, Monuki ES. Regulation of human neural precursor cells by laminin and integrins. Journal of Neuroscience Research 2006;83:845–56. Giancotti FG. Targeting integrin beta4 for cancer and anti-angiogenic therapy. Trends in Pharmacological Sciences 2007;28:506–11. Hall PE, Lathia JD, Miller NG, Caldwell MA, Ffrench-Constant C. Integrins are markers of human neural stem cells. Stem Cells 2006;24:2078–84. Hogervorst F, Kuikman I, Borne A, Sonnenberg A. Cloning and sequence analysis of beta-4 cDNA: an integrin subunit that contains a unique 118 kd cytoplasmic domain. The EMBO Journal 1990;9:756–70. Hu Z, Dai Y, Prentki M, Chohnan S, Lane MD. A role for hypothalamic malonylCoA in the control of food intake. The Journal of Biological Chemistry 2005;280:39681–3. Kwon YK. Effect of neurotrophic factors on neuronal stem cell death. Journal of Biochemistry and Molecular Biology 2002;35:87–93. Lv X, Su L, Yin D, Sun C, Zhao J, Zhang S, et al. Knockdown of integrin beta4 in primary cultured mouse neurons blocks survival and induces apoptosis by elevating NADPH oxidase activity and reactive oxygen species level. The International Journal of Biochemistry & Cell Biology 2008;40:689–99. Mansukhani A, Ambrosetti D, Holmes G, Cornivelli L, Basilico C. Sox2 induction by FGF and FGFR2 activating mutations inhibits Wnt signaling and osteoblast differentiation. The International of Cell Biology 2005;168:1065–76. Maric D, Fiorio Pla A, Chang YH, Barker JL. Self-renewing and differentiating properties of cortical neural stem cells are selectively regulated by basic fibroblast growth factor (FGF) signaling via specific FGF receptors. The Journal of Neuroscience 2007;27:1836–52. Mohammadi M, McMahon G, Sun L, Tang C, Hirth P, Yeh BK, et al. Structures of the tyrosine kinase domain of fibroblast growth factor receptor in complex with inhibitors. Science 1997;276:955–60. Morena A, Riccioni S, Marchetti A, Polcini AT, Mercurio AM, Blandino G, et al. Expression of the beta 4 integrin subunit induces monocytic differentiation of 32D/v-Abl cells. Blood 2002;100:96–106. Spinardi L, Einheber S, Cullen T, Milner TA, Giancotti FG. A recombinant tailless integrin beta 4 subunit disrupts hemidesmosomes, but does not suppress alpha 6 beta 4-mediated cell adhesion to laminins. The Journal of Cell Biology 1995;129:473–87. Stagge V, Seufferlein T, Dürschmied D, Menke A, Adler G, Beil M. Integrin-mediated differentiation of a pancreatic carcinoma cell line is independent of FAK or MAPK activation levels. Pancreas 2001;23:236–45. Su L, Zhao B, Lv X, Zhao J, Zhang S, Miao J. Safrole oxide is a useful tool for investigating the effect of apoptosis in vascular endothelial cells on neural stem cell survival and differentiation in vitro. Bioorganic & Medicinal Chemistry Letters 2007a;17:3167–71. Su L, Zhao B, Lv X, Wang N, Zhao J, Zhang S, et al. Safrole oxide induces neuronal apoptosis through inhibition of integrin beta4/SOD activity and elevation of ROS/NADPH oxidase activity. Life Sciences 2007b;80:999–1006. Su L, Lv X, Miao JY. Integrin 4 in neural cells. Neuromolecular Medicine 2008 [Epub ahead of print]. Tanghetti E, Ria R, Dell’Era P, Urbinati C, Rusnati M, Ennas MG, et al. Biological activity of substrate-bound basic fibroblast growth factor (FGF2): recruitment of FGF receptor-1 in endothelial cell adhesion contacts. Oncogene 2002;21: 3889–97. Tarone G, Hirsch E, Brancaccio M, De Acetis M, Barberis L, Balzac F, et al. Integrin function and regulation in development. The International Journal of Developmental Biology 2000;44:725–31. Vescovi AL, Reynolds BA, Fraser DD, Weiss S. bFGF regulates the proliferative fate of unipotent (neuronal) and bipotent (neuronal/astroglial) EGF-generated CNS progenitor cells. Neuron 1993;11:951–66. Walczak P, Kedziorek DA, Gilad AA, Lin S, Bulte JW. Instant MR labeling of stem cells using magnetoelectroporation. Magnetic Resonance in Medicine 2005;54:769–74. Wen T, Li H, Song H, Chen F, Zhao C, Lu W, et al. Down-regulation of specific gene expression by double-strand RNA induces neural stem cell differentiation in vitro. Molecular and Cellular Biochemistry 2005;275:215–21. Werner ME, Chen F, Moyano JV, Yehiely F, Jones JC, Cryns VL, et al. Caspase proteolysis of the integrin beta4 subunit disrupts hemidesmosome assembly, promotes apoptosis, and inhibits cell migration. The Journal of Biological Chemistry 2007;282:5560–9. Zwaka TP, Thomson JA. Homologous recombination in human embryonic stem cells. Nature Biotechnology 2003;21:319–21.
Contents lists available at ScienceDirect
The International Journal of Biochemistry & Cell Biology journal homepage: www.elsevier.com/locate/biocel
Neural stem cell differentiation is mediated by integrin 4 in vitro Le Su a,b , Xin Lv a,b , JiPing Xu c , DeLing Yin d , HaiYan Zhang a,b , Yi Li d , Jing Zhao a,b , ShangLi Zhang a,b , JunYing Miao a,b,∗ a
Institute of Developmental Biology, School of Life Science, Shandong University, Jinan 250100, China The Key Laboratory of Cardiovascular Remodeling and Function Research, Chinese Ministry of Education and Chinese Ministry of Health, Shandong University Qilu Hospital, Jinan 250012, China c Department of Neural Medicine, The Second Hospital of Shandong University, Jinan 250100, China d Department of Internal Medicine, East Tennessee State University, Box 70622, Johnson City, TN 37604, USA b
a r t i c l e
i n f o
Article history: Received 18 July 2008 Received in revised form 5 September 2008 Accepted 6 September 2008 Available online 12 September 2008 Keywords: Integrin 4 Neural stem cell Differentiation RNA interference Fibroblast growth factor receptor
a b s t r a c t Neural stem cells are capable of differentiating into three major neural cell types, but the underlying molecular mechanisms remain unclear. Here, we investigated the mechanism by which integrin 4 modulates mouse neural stem cell differentiation in vitro. Inhibition of endogenous integrin 4 by RNA interference inhibited the cell differentiation and the expression of fibroblast growth factor receptor 2 but not fibroblast growth factor receptor 1 or fibroblast growth factor receptor 3. Overexpression of integrin 4 in neural stem cells promoted neural stem cell differentiation. Furthermore, integrin 4-induced differentiation of neural stem cells was attenuated by SU5402, the inhibitor of fibroblast growth factor receptors. Finally, we investigated the role of integrin 4 in neural stem cell survival: knockdown of integrin 4 did not affect survival or apoptosis of neural stem cells. These data provide evidence that integrin 4 promotes differentiation of mouse neural stem cells in vitro possibly through fibroblast growth factor receptor 2. © 2008 Elsevier Ltd. All rights reserved.
1. Introduction The central nervous system appears to have only limited potential to generate new neurons and gliacytes. Neural transplantation of stem cells is used in clinical trials of many degenerative neurological diseases, including Huntington’s disease, Parkinson’s disease, and strokes (Kwon, 2002), so determining the key factors that regulate the fate of neural stem cells (NSCs) is important. Integrins, a large family of heterodimeric receptors, of ␣ and  subunits, play fundamental roles in cellular signaling (Tarone et al., 2000). In contrast to other  subunits, integrin 4 has an exceptionally large cytoplasmic domain, which is associated with cytoskeletal and signaling molecules (Hogervorst et al., 1990). It participates in the initiation of myelin sheath formation in Schwann cells and anchors astrocytes to the vascular basal lamina to increase the resistance of the blood brain barrier (Su et al., 2008). Recently, the expression of integrin 4 was found to induce monocytic and pancreatic carcinoma cell differentiation (Stagge et al., 2001; Morena et al., 2002). Additionally, in our previous study, knock-
∗ Corresponding author at: Institute of Developmental Biology, School of Life Science, Shandong University, Jinan 250100, China. Tel.: +86 531 88364929; fax: +86 531 88565610. E-mail address: [email protected] (J. Miao). 1357-2725/$ – see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.biocel.2008.09.001
down of integrin 4 could induce neuronal apoptosis (Lv et al., 2008). However, the roles of integrin 4 in NSC differentiation and survival are not clear. One proposed mechanism critical to NSC biology involves the activity of fibroblast growth factor receptors (FGFRs). Recent studies showed that coactivation of FGFR1 and FGFR3 promoted selfrenewal of NSCs, whereas developmental up-regulation of FGFR2 expression was associated with a shift of NSCs into a multipotential state or apoptosis (Maric et al., 2007). Furthermore, FGFR1 could associate with ␣v3 integrin to form an integrin–FGFR signaling complex (Tanghetti et al., 2002). However, whether integrin 4 regulates NSC differentiation through FGFRs is not clear. Therefore, we investigated the relation between integrin 4 and FGFRs in mouse NSC differentiation in vitro. 2. Materials and methods 2.1. Primary culture of cells Mouse NSC neurospheres were obtained from embryo forebrains on day 14 of gestation under sterile conditions as described (the mice were purchased from Shandong University Laboratory Animal Center) (Su et al., 2007a). Briefly, pregnant mice were decapitated, the brains were removed and placed in artificial cerebrospinal fluid, then brains were cut into small pieces and
L. Su et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 916–924
917
Fig. 1. Integrin 4 level in neurospheres, primary cultured neurons and differentiated NSCs. (A) Level of integrin 4 in neurospheres obtained from the embryo forebrains on day 14 of gestation. (B) Level of integrin 4 in differentiated cells induced by 10% FBS. (C) Relative quantity of integrin 4 in (A) and (B). (D) Level of integrin 4 in NSCs from the embryo forebrains on day 14 of gestation. (E) Content of integrin 4 in primary cultured mouse neurons obtained from the embryo forebrains on day 14 of gestation. (F) Quantity of integrin 4 in (D) and (E). Values in (C) and (F) represent the relative fluorescent intensity per cell by laser scanning confocal microcopy. (**P < 0.01 vs. # , n = 3). (G–J) Micrographs observed on transmission microscopy corresponding to (A, B, D and E).
digested with use of papain (Worthington Biochemical Lakewood, NJ) at 37 ◦ C for 45 min. The cells, collected by centrifugation, were resuspended, then the material was passed through a 70-m cell strainer to remove undissociated large pieces of tissue. After the material was centrifuged at 300 × g for 10 min, the cells were resuspended in the supplemented culture medium with 0.1 mM l-glutamine, 4 mM d-glucose, 0.4 mM HEPES, 2% B27 and 20 ng/mL recombinant human basic fibroblast growth factor (EssexBio Group, China). Cells were grown at 37 ◦ C in 5% CO2 and 95% air with saturated humidity. The cell concentration obtained was approximately 1 × 106 cells/mL. Seven days after plating, neurospheres were mechanically dissociated and subcultured as described above. Primary cultured mouse neurons were obtained from embryo forebrains on day 14 of gestation under sterile conditions as described (Su et al., 2007b). To induce neurosphere differentiation, 50–100 neurospheres were plated onto poly-lysine (poly-l-lysine, Sigma)-coated plastic dishes then cultured in medium supplemented with 10% fetal bovine serum (FBS) for 72 h to induce NSC differentiation (Vescovi et al., 1993). The neurospheres were observed under a phase-contrast microscope (Nikon, Japan) for cell morphology analysis. The cell viability was evaluated by trypan blue dye exclusion test as described
(Altman et al., 1993): the cell suspension was mixed with the same volume of 0.4% trypan blue solution, and cells were counted under a light microscope. Cell viability was estimated by the proportion of unstained cells. 2.2. RNA interference (RNAi) To knock down the expression of integrin 4 protein, RNAi was performed as described (Walczak et al., 2005; Wen et al., 2005) with the use of the specific integrin 4 small interfering RNA (siRNA), a pool of 3 target-specific, 20- to 25-nt siRNAs (sc-35679; Santa Cruz Biotechnology, Santa Cruz, CA). RNAi experiments followed the manufacturer’s protocols. An amount of 6 × 105 cells was suspended in siPORTTM siRNA Electroporation Buffer (Ambion, USA) in sterile 0.4-cm-gap electroporation cuvettes containing 2 g integrin 4 siRNA following the manufacturer’s protocols. Cells were electroporated by use of a Gene Pulser Xcell Electroporation system (Bio-rad, USA). One pulse lasted for about 8 ms at 160 V. Cells were left in the cuvette holder for 10 min, then moved into 24well plates in the CO2 incubator. To evaluate siRNA-mediated gene silencing, scramble siRNA (Santa Cruz Biotechnology) was used as a control.
918
L. Su et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 916–924
2.3. Overexpression of integrin ˇ4 in NSCs The cDNA constructs encoding wild-type integrin 4 subunit (pCMV-4) (kindly provided by Dr. Filippo Giancotti, New York University Medical Center) were transfected by electroporation as for RNAi above (Morena et al., 2002). Transfection controls used the same amount of empty vector (pRC-CMV) alone. We monitored the effect of overexpression 48 h after electroporation using immunofluorescence staining combined with laser scanning confocal microscopy (Spinardi et al., 1995). We calculated the overexpression rate by counting the cells that expressed a high level of integrin 4 and total cells in the screen. 2.4. Flow cytometry The expression level of integrin 4 was analyzed by flow cytometry analysis of stained cells (anti-mouse CD104 4 chain, 553745, 554016; BD Biosciences, USA) (Morena et al., 2002). Cells were washed twice with cold PBS containing 0.002% ethylenediaminetetraacetic acid. Samples of 1 × 106 untransfected or transfected cells were incubated for 30 min at 4 ◦ C with primary antibody. Cells were
then washed three times with washing buffer (PBS containing 0.5% bovine serum albumin) and incubated for 30 min at 4 ◦ C with FITCconjugated secondary antibody. After extensive washes, 2 × 104 cells were acquired for each sample, and cell fluorescence intensity was measured by flow cytometry (FACSCalibur, BD Biosciences) and analyzed by use of the CELL-QUEST software. Cells incubated without primary antibody were used as a negative control. 2.5. Immunocytochemical staining Immunocytochemical staining was performed as described (Hu et al., 2005). Primary antibodies (rabbit anti-mouse neuron-specific enolase [NSE] IgG, sc-31859; Synapsin IgG, sc-20780; glial fibrillary acidic protein [GFAP] IgG, sc-9065; integrin 4 IgG, sc-9090; FGFR1 IgG, sc-121; FGFR2 IgG, sc-122; FGFR3 IgG, sc-123; Santa Cruz Biotechnology) were 1:100, and secondary antibody labeled with goat anti-rabbit FITC was diluted 1:200 (sc-65218; Santa Cruz Biotechnology). Cell nuclei were counterstained with propidium iodide (10 g/mL) for 15 min at room temperature. Samples were evaluated by laser scanning confocal microscopy (Leica, Germany). We randomly selected the region of interest and then zoomed in
Fig. 2. siRNA-mediated knockdown of integrin 4 in NSCs analyzed by flow cytometry. (a) Negative control, NSCs incubated without primary antibody. (b) Untransfected, NSCs incubated with primary and secondary antibodies. (c) Control, NSCs transfected with scramble control siRNA incubated with primary and secondary antibodies for 48 h. (d) 4 siRNA, NSCs transfected with integrin 4 siRNA incubated with primary and secondary antibodies for 48 h.
L. Su et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 916–924
919
the same frames. The relative fluorescent intensity per cell was the total value of the sample in the zoom scans divided by the total number of cells (at least 200 cells) in the same scan.
et al., 1997; Bernard-Pierrot et al., 2004; Mansukhani et al., 2005). SU5402 (10 M) was added into the medium when NSCs were transfected with pCMV-4.
2.6. Inhibition of FGFR2
2.7. Cell death assay
To inhibit the activity of FGFR2, NSCs were treated with the inhibitor SU5402 (Calbiochem, San Diego, CA), a member of a new class of FGFR antagonists that block the tyrosine kinase activity of the receptor by interacting with the catalytic domain (Mohammadi
For cell death assay, we used terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) assay according to the manufacturer’s protocol (Promega). Cells were observed under a laser scanning confocal microscope (Leica,
Fig. 3. Integrin 4 inhibition attenuates NSC differentiation. (A–C) Phase-contrast micrographs (200×). (D–L) Fluorescent micrographs. (A, D, G and J) Neurospheres obtained from the embryo forebrains on day 14 of gestation. (B, E, H and K) NSCs transfected with scramble control siRNA, then treated with 10% FBS for 72 h. (C, F, I and L) NSCs transfected with integrin 4 siRNA, then treated with 10% FBS for 72 h. (D–F) Expression of NSE in NSCs. G-I: Expression of synapsin in NSCs. (J–L) Expression of GFAP in NSCs.
920
L. Su et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 916–924
Germany). The proportion of apoptotic cells was quantified by the TUNEL-positive rate. Cells were treated with integrin 4 or scramble siRNA for 48 h and 72 h, then underwent lactate dehydrogenase (LDH) release analysis by use of an LDH kit according to the manufacturer’s protocol (ZhongSheng, China) (Lv et al., 2008). LDH catalyzes the reduction of pyruvate to l-lactate, with concomitant oxidation of NADH2 to NAD. Since the oxidation of NADH2 is directly proportional to the reduction of pyruvate in equimolar amounts, the LDH activity can be calculated from the rate of decrease in absorbance at 340 nm. Light absorption was measured by use of a model Cintra 5 UV–vis spectrometer (GBC, Australia), and LDH activity was calculated as follows: LDH(U/L) = (Asample/ min −Ablank/ min) × F, where F = 1000 × V total/(V sample × extinction coefficient); A = change in absorbance; V = volume in ml; extinction coefficient = mmol absorptivity of NADH at 340 nm was 6.3. Cells were fixed in 4% formaldehyde for 10 min, then incubated with Hoechst 33258 (2 g/ml) (Sigma, St. Louis, MO, USA) for stain-
ing for 60 min at 37 ◦ C. Stained cells were washed with PBS twice, then viewed under an Olympus inverted fluorescence microscope. Cells were scored as apoptotic if their nuclei were much brighter or exhibited condensation of chromatin and nuclear fragmentation. 2.8. Statistical analysis Data are expressed as mean ± S.E. An average of more than 10 spheres was analyzed per treatment, per experiment. SPSS 11.5 software (SPSS Inc., Chicago, IL) was used for statistical calculations. One-way ANOVA (followed by Scheffé F-test for post hoc analysis) was used for statistical analysis. A P < 0.05 was considered statistically significant. 3. Results 3.1. Level of integrin ˇ4 was increased during NSC differentiation The level of integrin 4 was much higher in differentiated NSCs induced by 10% FBS than in untreated neurospheres (Fig. 1A–C).
Fig. 4. RNAi knockdown of integrin 4 inhibits the expression of FGFR2. (A) Untransfected neurospheres obtained from mouse embryo forebrains on day 14 of gestation. Control, NSCs transfected with scramble control siRNA, then plated onto poly-lysine-coated plastic dishes and treated with 10% FBS for 72 h. 4 siRNA, NSCs transfected with integrin 4 siRNA, then plated onto poly-lysine-coated plastic dishes and treated with 10% FBS for 72 h. Narrow arrows, cells underwent differentiation in integrin 4 siRNA group. Wide arrow, cells with differentiation inhibited. (B) Micrographs from transmission microscopy corresponding to (A). (C) Quantity of FGFR levels in the three groups mentioned above. Value represents the relative fluorescent intensity per cell determined by laser scanning confocal microcopy. **P < 0.01 vs. control, n = 3.
L. Su et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 916–924
However, the level of integrin 4 was higher in primary cultured neurons than in mouse NSCs obtained from the same embryo forebrain on day 14 of gestation (Fig. 1D–F). The negative control lacking primary antibody showed no staining, which demonstrated that the staining was specific (data not shown). The level of integrin 4 was increased linearly during NSC differentiation. 3.2. siRNA-mediated knockdown of integrin ˇ4 in NSCs Control siRNA had no effect on level of integrin 4 in transfected NSCs (Fig. 2c), and untransfected and control groups did not differ in fluorescence intensity (Fig. 2b and c). However, the expression of integrin 4 was downregulated significantly after treatment with the specific siRNA for 48 h (Fig. 2d), with fluorescence intensity 2.2% compared to 57.3% in control group.
921
3.4. RNAi knockdown of integrin ˇ4 inhibits the expression of FGFR2 Next, we measured the levels of FGFR1-3 by immunofluorescence combined with laser scanning confocal microscopy for semi-quantitative evaluation of protein level (Dinser et al., 2001; Lv et al., 2008). The integrin 4 siRNA group and control group did not differ in level of FGFR1 (Fig. 4A, a–c) or FGFR3 (Fig. 4A, g–i) (P > 0.05), but the level of FGFR2 (Fig. 4A, d–f) was remarkably reduced (P < 0.01) (Fig. 4C). Few cells underwent differentiation and NSCs showed little neurite outgrowth (Fig. 4B, f; narrow arrows). However, most cells transfected with integrin 4 siRNA still showed the NSC shape and did not extend neurites (wide arrow). Thus, the level of FGFR2 was decreased with knockdown of integrin 4 in NSCs. 3.5. Overexpression of integrin ˇ4
3.3. Integrin ˇ4 inhibition attenuates NSC differentiation Since integrin 4 level was increased during NSC differentiation (Fig. 1), we investigated whether the NSC differentiation could be inhibited by knockdown of integrin 4. After transfection with control siRNA and integrin 4 siRNA for 72 h, the cells in the control group underwent differentiation and the adjacent neurospheres formed a network by neurites (Fig. 3B) that was similar to the untransfected group (Fig. 3A). However, in the integrin 4 siRNA group, only a few cells underwent differentiation and NSCs showed little neurite outgrowth (Fig. 3C). To confirm the NSC differentiation, we examined by immunofluorescence staining the expression of specific markers for neurons and astrocytes, including NSE (Fig. 3D–F) and synapsin (Fig. 3G–I) for neurons and GFAP (Fig. 3J–L) for astrocytes. Most of the untransfected (Fig. 3D, G, and J) and control-group cells (Fig. 3E, H, and K) were positive for neuron and astrocyte markers. However, the integrin 4 siRNA group showed a markedly reduced quantity of marker-positive cells (Fig. 3F, I, L), which suggests that NSC differentiation was blocked by knockdown of integrin 4.
NSCs transfected with control cDNA (pRC-CMV) showed no effect on level of integrin 4 (Fig. 5A, d–f), with no difference in the level between untransfected and control cells (Fig. 5A, a–f). However, with integrin 4 overexpression, the level of integrin 4 was elevated by about 25% (Fig. 5A, g–i; B). This transfection rate is consistent with other results in NSCs (Zwaka and Thomson, 2003). 3.6. Integrin ˇ4 overexpression promoted NSC differentiation in vitro After cells were transfected with pCMV-4 and pRC-CMV for 3 days, phase contrast microscope revealed few cells undergoing differentiation in the control group (Fig. 6A, b) and NSCs showed little neurite outgrowth, a finding similar to that in the untransfected group (Fig. 6A, a). However, in the integrin 4 overexpression group, most cells differentiated, and the adjacent neurospheres formed a network by neurites (Fig. 6A, c). Immunofluorescent double staining of overexpressed NSCs for NSE (Fig. 6B) and GFAP (Fig. 6C) in the untransfected (Fig. 6B
Fig. 5. Overexpression of integrin 4 in NSCs. (A) Immunofluorescent double staining of overexpression of integrin 4 in NSCs. Untransfected, neurospheres obtained from mouse embryo forebrains on day 14 of gestation. Control, NSCs transfected with empty vectors (pRC-CMV) for 48 h. Overexpression, NSCs transfected with wild-type integrin 4 cDNA (pCMV-4) for 48 h. (a, d, and g) Fluorescence photomicrographs of NSCs expressing integrin 4. (b, e, and h) Cell nuclei counterstained with PI. (c, f, and i) Overlay photographs of integrin 4 and PI. (B) The rate integrin 4 overexpression in NSCs was about 25% (n = 3).
922
L. Su et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 916–924
Fig. 6. Overexpression of integrin 4 stimulates NSC differentiation. Untransfected, the neurospheres obtained from mouse embryo forebrains on day 14 of gestation. Control, NSCs transfected with empty vector (pRC-CMV) for 3 days. Overexpression, NSCs transfected with wild-type integrin 4 cDNA (pCMV-4) for 3 days. SU5402, NSCs transfected with pCMV-4 and treated with 10 M SU5402 for 3 days. (A) Overexpression of integrin 4 promoted NSC differentiation as seen on phase-contrast microscopy. (B) (a, d, g, and j) Fluorescence photomicrographs of NSCs stained with NSE. (b, e, h, and k) Cell nuclei counterstained with PI. (c, f, i, and l) Overlay photographs of NSE and PI. (C) (a, d, g, and j) Fluorescence photomicrographs of NSCs expressing GFAP. (b, e, h, and k) Cell nuclei counterstained with PI. (c, f, i, and l) Overlay photographs of GFAP and PI. (D) The percentages of the positive cells in untransfected, control, overexpression and SU5402 groups (*P < 0.05, n = 3, **P < 0.01, n = 3).
and C, a–c) and control groups (Fig. 6B and C, d–f) revealed few cells positive for neuron and astrocyte markers. However, in the integrin 4-overexpression group, the quantity of markerpositive cells was remarkably increased (Fig. 6B and C, g–i). Immunofluorescent staining revealed NSCs overexpressing integrin 4 with significantly increased viability to differentiate (Fig. 6D). Immunofluorescent double staining revealed that treatment with SU5402, a specific inhibitor of FGFR, reduced the number of positive neurons and astrocytes (Fig. 6B and C, j–l). The NSCs treated with SU5402 showed a significantly lower proportion of
neurons (P < 0.01) and astrocytes (P < 0.01) than cells transfected with pCMV-4 (Fig. 6D). 3.7. Suppression of integrin ˇ4 does not induce NSC apoptosis and necrosis Finally, we examined the effect of integrin 4 on NSC apoptosis. The results from morphology and trypan blue exclusion showed that knockdown of integrin 4 had no effect on NSC survival (Fig. 7A and D). After knockdown with integrin 4 siRNA, the proportion of apoptotic cells did not differ between control and integrin 4 siRNA
L. Su et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 916–924
923
Fig. 7. Suppression of integrin 4 does not induce NSC apoptosis and necrosis. (A) The morphology of untransfected NSCs (a), NSCs transfected with scramble control siRNA for 72 h (b) and NSCs transfected with integrin 4 siRNA for 72 h (c) under a phase contrast microscope (200×). (B) Quantification of apoptotic cells by TUNEL assay. The treatments correspond to those of (A). Scramble control siRNA and integrin 4 siRNA groups (d, & P > 0.05, n = 3). (C) Hoechst 33258 nuclear staining of untransfected NSCs (a), NSCs transfected with scramble control siRNA for 72 h (b) and NSCs transfected with integrin 4 siRNA for 72 h (c). The arrows indicate apoptotic cells. (D) Quantification of living cells by trypan blue staining. Cells were transfected with integrin 4 siRNA and scramble control siRNA for 24 h, 48 h and 72 h. (& P > 0.05 vs. control groups at the same time, n = 4). We assumed the viability of untransfected cells as 100%. (E) Effect of integrin 4 siRNA on lactate dehydrogenase (LDH) release from NSCs. Cells were treated with integrin 4 siRNA or scramble control siRNA for 48 h and 72 h. Light absorption was analyzed at 340 nm with use of a model Cintra 5 UV–vis spectrometer (& P > 0.05, n = 3).
groups at 48 h and 72 h (P > 0.05; Fig. 7B). The results from Hoechst 33258 staining showed that nuclear condensation was similar in the control and integrin 4 siRNA groups (Fig. 7C), which suggests that knockdown of integrin 4 did not trigger NSC apoptosis. To understand whether knockdown of integrin 4 induces NSC necrosis, we measured LDH release from NSCs treated with integrin 4 siRNA for 48 h and 72 h. LDH release did not differ between control and siRNA-treated groups (P > 0.05) (Fig. 7E), which suggests that knockdown of integrin 4 did not cause necrosis in NSCs. 4. Discussion Although many reports have described NSC proliferation and differentiation, the exact mechanisms of NSC differentiation are not well understood. Integrin 4 is expressed in astrocytes, Schwann cells and neurons (Su et al., 2008). Recently, human neurospheres were found to lack integrin 4 mRNA (Hall et al., 2006), but another report showed that laminin matrices enhanced human NSC differentiation and elongation of neurites from NSPC-derived neurons (Flanagan et al., 2006). Flow cytometry revealed that human neural stem/precursor cells (NSPCs) express the laminin-binding integrins ␣3, ␣6, ␣7, 1, and 4, and function-blocking antibodies to ␣6 subunit confirmed a role for integrins in laminin-dependent migration of human NSPCs (Flanagan et al., 2006). These data suggest that
laminin and its integrin receptors, including integrin 4, might be key regulators of human NSPCs. However, no direct evidence has shown the function of integrin 4 in NSC differentiation and survival. In this study, we found integrin 4 protein expressed in mouse NSCs, which is consistent with previous results (Flanagan et al., 2006). Moreover, we observed the level of integrin 4 increased during NSC differentiation, which encouraged the exploration of the role of integrin 4 in NSC differentiation and survival. In vitro, NSCs grown in suspension form floating clusters termed neurospheres. Because neurospheres are clonally derived from neural stem cells, they provide a good experimental system for studying neural cell differentiation, not only to obtain phenotypes of potential therapeutic interest but also to define the basic mechanisms that regulate the acquisition of a specific cell fate in the nervous system. So we examined the role of integrin 4 on the generation of neurons and gliacytes from NSCs by using neurospheres. Increasing evidence indicates that integrin 4 induces cell differentiation (Stagge et al., 2001). Additionally, the cytoplasmic domain of the 4 subunit is directly involved in signaling events that generate monocytic differentiation (Morena et al., 2002). Differentiated enterocytes express a full-length 205-kDa 4 subunit, whereas undifferentiated crypt cells express a 4 subunit that does not contain the COOH-terminal segment of the cytoplasmic domain (4 ctd-integrin). Furthermore, 4 ctd-integrin expressed
924
L. Su et al. / The International Journal of Biochemistry & Cell Biology 41 (2009) 916–924
by the undifferentiated crypt cells is not functional for adhesion to laminin-5 (Basora et al., 1999). In this study, we showed that knockdown of integrin 4 by its specific siRNA that targets the cytoplasmic tail of this integrin inhibited NSC differentiation, and overexpression of the full-length integrin 4 promoted NSC differentiation. As well, differentiation of not only astrocytes but also neurons was promoted by overexpression of integrin 4. The data provide new evidence for the role of the integrin 4 cytoplasmic tail in cell differentiation. One recent report suggested that co-activation of FGFR1 and FGFR3 promoted symmetrical division of NSCs (self-renewal), whereas inactivation of either receptor triggered asymmetrical division and neurogenesis from these cells. Developmental upregulation of FGFR2 expression was associated with a shift of NSCs into a multipotential state or apoptosis (Maric et al., 2007). As well, FGFR1 could be associated with ␣v3 integrin forming an integrin–FGFR signalling complex to potentiate FGF2 signaling (Tanghetti et al., 2002). However, the relation between FGFRs and integrin 4 is not clear. Here, we found that knockdown of integrin 4 could inhibit NSC differentiation, and during this differentiation inhibitory process, the levels of FGFR1 and FGFR3 were not altered, but the level of FGFR2 was significantly depressed. These results support that FGFR2 but not FGFR1 or FGFR3 might be involved in the NSC differentiation mediated by integrin 4. Furthermore, use of the specific inhibitor of FGFR, SU5402, validated the involvement of FGFR2 in the increased differentiation ability of cells transfected with pCMV-integrin 4. These results provide new insights into the roles of FGFRs and integrin 4 in diversification of NSC properties and initiation of neural lineage-restricted differentiation. Laminin-5 is an established ligand of the alpha6beta4-integrin (Giancotti, 2007) and integrin–laminin matrices can enhance NSPC differentiation (Flanagan et al., 2006). In our study, we found that integrin 4 and FGFR2 might form a complex to induce NSC differentiation. Whether this process depends on the binding of the alpha6beta4-integrin to laminin-5 is a subject for further study. In a previous study, knockdown of integrin 4 could induce apoptosis in primary cultured mouse neurons (Lv et al., 2008), which is similar to the function of integrin 4 in epithelial cells (Werner et al., 2007). In the current study, knockdown of integrin 4 did not induce apoptosis but did inhibit differentiation of mouse NSCs. The data further suggest the cell type-specific function of this integrin subunit. In summary, blocking integrin 4 by its specific siRNA inhibited differentiation but did not induce apoptosis in mouse NSCs. As well, the level of FGFR2 was depressed significantly. Furthermore, overexpression of integrin 4 in NSCs promoted neuron and astrocyte differentiation in vitro, and this promotion was suppressed by the specific inhibitor of FGFR, SU5402. These results show that integrin 4 is a key factor in NSC differentiation, and this integrin subunit might perform its action through FGFR2 in the differentiation process. Acknowledgements This work was supported by the National Natural Science Foundation of China (No. 30470404) and the National 973 Research Project (No. 2006CB503803). The authors thank Dr. Filippo Giancotti for the generous gift of wild-type integrin 4 cDNA. References Altman SA, Randers L, Rao G. Comparison of trypan blue dye exclusion and fluorometric assays for mammalian cell viability determinations. Biotechnology Progress 1993;9:671–4.
Basora N, Herring-Gillam FE, Boudreau F, Perreault N, Pageot LP, Simoneau M, et al. Expression of functionally distinct variants of the beta(4)A integrin subunit in relation to the differentiation state in human intestinal cells. The Journal of Biological Chemistry 1999;274:29819–25. Bernard-Pierrot I, Ricol D, Cassidy A, Graham A, Elvin P, Caillault A, et al. Inhibition of human bladder tumour cell growth by fibroblast growth factor receptor 2b is independent of its kinase activity. Involvement of the carboxy-terminal region of the receptor. Oncogene 2004;23:9201–11. Dinser R, Kreppel F, Zaucke F, Blank C, Paulsson M, Kochanek S, et al. Comparison of long-term transgenic expression after nonviral and adenoviral gene transfer into primary articular chondrocytes. Histochemistry and Cell Biology 2001;116:69–77. Flanagan LA, Rebaza LM, Derzic S, Schwartz PH, Monuki ES. Regulation of human neural precursor cells by laminin and integrins. Journal of Neuroscience Research 2006;83:845–56. Giancotti FG. Targeting integrin beta4 for cancer and anti-angiogenic therapy. Trends in Pharmacological Sciences 2007;28:506–11. Hall PE, Lathia JD, Miller NG, Caldwell MA, Ffrench-Constant C. Integrins are markers of human neural stem cells. Stem Cells 2006;24:2078–84. Hogervorst F, Kuikman I, Borne A, Sonnenberg A. Cloning and sequence analysis of beta-4 cDNA: an integrin subunit that contains a unique 118 kd cytoplasmic domain. The EMBO Journal 1990;9:756–70. Hu Z, Dai Y, Prentki M, Chohnan S, Lane MD. A role for hypothalamic malonylCoA in the control of food intake. The Journal of Biological Chemistry 2005;280:39681–3. Kwon YK. Effect of neurotrophic factors on neuronal stem cell death. Journal of Biochemistry and Molecular Biology 2002;35:87–93. Lv X, Su L, Yin D, Sun C, Zhao J, Zhang S, et al. Knockdown of integrin beta4 in primary cultured mouse neurons blocks survival and induces apoptosis by elevating NADPH oxidase activity and reactive oxygen species level. The International Journal of Biochemistry & Cell Biology 2008;40:689–99. Mansukhani A, Ambrosetti D, Holmes G, Cornivelli L, Basilico C. Sox2 induction by FGF and FGFR2 activating mutations inhibits Wnt signaling and osteoblast differentiation. The International of Cell Biology 2005;168:1065–76. Maric D, Fiorio Pla A, Chang YH, Barker JL. Self-renewing and differentiating properties of cortical neural stem cells are selectively regulated by basic fibroblast growth factor (FGF) signaling via specific FGF receptors. The Journal of Neuroscience 2007;27:1836–52. Mohammadi M, McMahon G, Sun L, Tang C, Hirth P, Yeh BK, et al. Structures of the tyrosine kinase domain of fibroblast growth factor receptor in complex with inhibitors. Science 1997;276:955–60. Morena A, Riccioni S, Marchetti A, Polcini AT, Mercurio AM, Blandino G, et al. Expression of the beta 4 integrin subunit induces monocytic differentiation of 32D/v-Abl cells. Blood 2002;100:96–106. Spinardi L, Einheber S, Cullen T, Milner TA, Giancotti FG. A recombinant tailless integrin beta 4 subunit disrupts hemidesmosomes, but does not suppress alpha 6 beta 4-mediated cell adhesion to laminins. The Journal of Cell Biology 1995;129:473–87. Stagge V, Seufferlein T, Dürschmied D, Menke A, Adler G, Beil M. Integrin-mediated differentiation of a pancreatic carcinoma cell line is independent of FAK or MAPK activation levels. Pancreas 2001;23:236–45. Su L, Zhao B, Lv X, Zhao J, Zhang S, Miao J. Safrole oxide is a useful tool for investigating the effect of apoptosis in vascular endothelial cells on neural stem cell survival and differentiation in vitro. Bioorganic & Medicinal Chemistry Letters 2007a;17:3167–71. Su L, Zhao B, Lv X, Wang N, Zhao J, Zhang S, et al. Safrole oxide induces neuronal apoptosis through inhibition of integrin beta4/SOD activity and elevation of ROS/NADPH oxidase activity. Life Sciences 2007b;80:999–1006. Su L, Lv X, Miao JY. Integrin 4 in neural cells. Neuromolecular Medicine 2008 [Epub ahead of print]. Tanghetti E, Ria R, Dell’Era P, Urbinati C, Rusnati M, Ennas MG, et al. Biological activity of substrate-bound basic fibroblast growth factor (FGF2): recruitment of FGF receptor-1 in endothelial cell adhesion contacts. Oncogene 2002;21: 3889–97. Tarone G, Hirsch E, Brancaccio M, De Acetis M, Barberis L, Balzac F, et al. Integrin function and regulation in development. The International Journal of Developmental Biology 2000;44:725–31. Vescovi AL, Reynolds BA, Fraser DD, Weiss S. bFGF regulates the proliferative fate of unipotent (neuronal) and bipotent (neuronal/astroglial) EGF-generated CNS progenitor cells. Neuron 1993;11:951–66. Walczak P, Kedziorek DA, Gilad AA, Lin S, Bulte JW. Instant MR labeling of stem cells using magnetoelectroporation. Magnetic Resonance in Medicine 2005;54:769–74. Wen T, Li H, Song H, Chen F, Zhao C, Lu W, et al. Down-regulation of specific gene expression by double-strand RNA induces neural stem cell differentiation in vitro. Molecular and Cellular Biochemistry 2005;275:215–21. Werner ME, Chen F, Moyano JV, Yehiely F, Jones JC, Cryns VL, et al. Caspase proteolysis of the integrin beta4 subunit disrupts hemidesmosome assembly, promotes apoptosis, and inhibits cell migration. The Journal of Biological Chemistry 2007;282:5560–9. Zwaka TP, Thomson JA. Homologous recombination in human embryonic stem cells. Nature Biotechnology 2003;21:319–21.