The molecular mechanism of Profilin for its tumour suppressor activity is still unknown. Nuclear transcription factor κB (NF-κB) is known to activate many target genes involved in cell proliferation. In the present study, we provide evidence that supports the involvement of Profilin in regulation of NF-κB, which might repress the tumorigenic response. Profilin overexpressing cells show low basal activity of IκBα kinase (IKK), high amounts of cytoplasmic inhibitory subunit of NF-κB (IκBα) and p65, and low nuclear NF-κB DNA binding activity. Co-localization and co-immunoprecipitation (Co-IP) studies suggest that Profilin interacts with a protein phosphatase, phosphatase and tension homologue (PTEN), and protects it from degradation. In turn, PTEN interacts physically and maintains a low phosphorylated state of the IKK complex and thereby suppresses NF-κB signalling. Thus, Profilin overexpressing cells show a decrease in NF-κB activation mediated by most of the inducers and potentiate cell death by repressing NF-κB-dependent genes involved in cell cycle progression. For the first time, we provide evidence, which suggests that Profilin increases tumour suppressor activity by regulating NF-κB.

INTRODUCTION

Profilin is a key cellular protein that interacts with actin microfilament and regulates cell migration implicated in tumour angiogenesis [1]. It interacts with several proline-rich cellular proteins and regulates multiple functions, like organ development, wound healing, immunity etc. [24]. It has been shown to interact with Hsp70 interacting protein (CHIP), a co-chaperone E3 ligase. The CHIP overexpressing cells show down-regulation of Profilin and enhanced cell migration [5]. In general, the levels of Profilin are decreased in most of the aggressive tumour cells [6,7]. In contrast, Profilin has been shown to induce expression of macrophage chemotactic factor 1 (MCP1), interleukin (IL)-12 and interferon gamma (IFNγ) through the nuclear transcription factor κB (NF-κB) activation pathway [8]. It also interacts with oestrogen receptor α as corepressor and inhibits several oestrogen-induced genes in breast tumours [9].

Most of the cancer cells have high basal activity of nuclear NF-κB that constantly drives its proliferation. NF-κB exists predominantly as p65–p50 heterodimer, remains sequestered with inhibitory subunit of NF-κB (IκBα) and stays inactive in the cytoplasm. Several stimuli activate and phosphorylate the principal kinase–the IκBα kinase (IKK) complex–which consists of two homologous kinase subunits, IKKα and IKKβ, and a regulatory subunit, IKKγ. Phosphorylation of IKKβ at two serine residues (177 and 181) is itself sufficient for the activation of the complex. Activated IKK phosphorylates IκBα, which leads to its ubiquitination and degradation by proteasomes [10]. Thus, released p65–p50 heterodimer gets post-translationally modified in the cytoplasm. This heterodimer, the active NF-κB, translocates to nucleus, binds to its promoter sites and transcribes several genes, which participate in the cell cycle progression. The sustained phosphorylated state of IKK for constitutively activated NF-κB in various solid and haematopoietic malignancies strongly suggests that IKKβ has become the focal point for anti-cancer therapies in recent years [11,12].

Phosphatase and tension homologue (PTEN) is also known as a negative regulator of NF-κB transcriptional activity [13]. PTEN, a well-known tumour suppressor protein was discovered in 1997 as a lipid phosphatase, although its protein phosphatase activity has also been reported [14]. Profilin has been shown to facilitate expression of PTEN and prevents phosphorylation of protein kinase B (AKT) [15]. Cellular levels of PTEN are regulated by its ubiquitin-dependent degradation by proteasome [16]. Thus, restoration of PTEN can prove to be a viable strategy against tumour cell proliferation.

In the present study, we show that Profilin suppresses NF-κB by inhibiting IKK phosphorylation. Inhibition of IKK phosphorylation is preceded by stabilization of PTEN as Profilin interacts with PTEN and protects it from degradation. For the first time, we provide evidence, which suggests that Profilin inhibits NF-κB activation in breast tumour cells and thereby prevents cell cycle progression, leading to cell death.

EXPERIMENTAL

Reagents and antibodies

DMSO, EDTA, EGTA, cycloheximide (CHX), tubulin antibody (anti-mouse), GAPDH antibody (anti-rabbit), Profilin1 antibody (anti-mouse), FLAG antibody (anti-mouse), Myc antibody (anti-rabbbit) and ubiquitin antibody (anti-rabbit) were obtained from Sigma–Aldrich. Penicillin, streptomycin, neomycin, RPMI 1640, Dulbecco's Modified Eagle Medium (DMEM), FBS and Lipofectamine 2000 were obtained from Life Technologies. Antibodies against PTEN (anti-mouse), phospho-AKT (Ser473) (anti-rabbit), AKT (anti-rabbit), phospho-GSK-3β (Ser9) (anti-rabbit), phospho-IKKα/β (Ser177/181) (anti-rabbit), phospho-IκBα (Ser32) (anti-rabbit), and phospho-p65 (Ser276) (anti-rabbit) were obtained from Cell Signaling Technologies whereas antibodies for IKKα/β (anti-rabbit), IκBα (anti-rabbit), p65 (anti-rabbit) were obtained from Santa Cruz Biotechnology. Gel shift oligonucleotides for NF-κB were synthesized from XCelris.

Cell culture and clonal selection

The cell lines used in the present study HuT-78 (human T-cell lymphoma), MDA-MB-231 (designated as MB-231 from now), and MDA-MB-468 (human breast cancer) were obtained from A.T.C.C. Cells were cultured in DMEM containing 10% FBS, penicillin (100 unit/ml) and streptomycin (100 μg/ml). For stable cell generation, MB-231 cells were transfected with pcDNA3.1 (+) Profilin1 using Lipofectamine 2000 reagent. G418 selection (800 μg/ml) started 48 h post-transfection and was maintained for 2 weeks before switching to regular culture medium. Profilin1 expression was examined by Western blot using anti-Profilin1 antibody and further confirmed by reverse transcriptase-PCR (RT-PCR).

Plasmids

Wild-type or H133S mutant of Profilin1 was either FLAG-tagged or un-taggged and cloned in pcDNA3.1 (+). Full length or deletion mutants of PTEN were FLAG- or Myc-tagged and cloned into pcDNA3.1 (+). Constructs of wild and dominant negative (IKKβ-WT and IKKβ-DN) p65 were cloned in pcDNA3 as described previously. The constitutive active mutant of IKKβ, in which two serine residues are mutated to glutamic acid, at position 177 and 181 (referred as IKKβ-EE or IKKβ-CA) was obtained from Prof. Gourisankar Ghosh (University of California, San Diego, USA)

Gel shift assay of NF-κB and AP-1

To determine tumour necrosis factor (TNF)-induced NF-κB or AP-1 activation, EMSA was conducted essentially as described previously [17]. Briefly, 8–10 μg of nuclear extract (NE) proteins were incubated with 32P end-labelled double-stranded NF-κB or AP-1 oligonucleotides for 30 min at 37°C, and the DNA–protein complexes were separated from free oligonucleotides on 6.6% native polyacrylamide gels. The double-stranded NF-κB and AP-1 oligonucleotide used in the experiment are 5′-TTGTTACAAGGGACTTTCCGCTGGGGACTTTCCAGGGA-GGCGTGG-3′ and 5′-CGCTTGATGACTCAGCCGGAA-3′ respectively.

Assay of NF-κB-dependent SEAP reporter gene

Cells were transiently transfected with Lipofectamine 2000 transfection reagent with 0.5 μg of reporter plasmid containing NF-κB-binding site cloned upstream of heat-stable secretory alkaline phosphatase (SEAP) designated as NF-κB-SEAP; 0.5 μg of plasmid DNA of IKKβ, p65, IKKβ-CA or IKKβ-DN; and 0.5 μg of GFP constructs. After different treatments, cell culture-conditioned medium (25 μl) was analysed for SEAP activity essentially as per the Clontech protocol and reported as fold activation with respect to empty SEAP-transfected cells as described previously [18].

Reverse transcriptase-PCR

One microgram of total RNA, isolated by TRIzol method (Gibco BRL) was reverse transcribed into cDNA by One step Access RT-PCR kit (Promega), followed by the amplification of the gene of interest using gene specific primers for Profilin1, intercellular cell adhesion molecule (ICAM)1, cyclooxygenase (Cox)2 and actin. PCR was performed and amplified products were separated by agarose gel electrophoresis (2%) and visualized by ethidium bromide staining [19]. The primer sequence and product size are as follows: Profilin: 117 bp {forward} 5′-ACGCCTACATCGAC-AACCTC-3′, (reverse) 5′-TGATGTTGACGAACGTTTTCC-3′; ICAM1: 406 bp (forward) 5′-AGGCCACCCCAGAGGACAAC-3′, (reverse) 5′-CCCATTATGACTGGGGCTGCTA-3′; Cox2: 305 bp (forward) 5′-TTCAAATGAGATTGTGGGAAAAT-3′, (reverse) 5′-AGATCATCTCTGCCTGAGTATCTT-3′; actin: 616 bp (forward) 5′-CCAACCGTGAAAAGATGACC-3′, (reverse) 5′-GCAGTAATCTCCTTCTGCATCC-3′.

Immunoblotting and immunoprecipitation

Cells were washed with ice-cold PBS, lysed in Tris/HCl lysis buffer and processed for immunoblotting (IB) and immunoprecipitation (IP) as per protocol described. Lysates were kept at −80°C until use. In a typical assay, 20–50 μg total proteins were used for Western blot and 200–500 μg for IP. Cell lysates or IP-products were resolved using standard SDS/PAGE, wet transferred on to PVDF membrane (Bio-Rad Laboratories), blocked in standard TBS–0.15% Tween-20 (TBST) supplemented with 5% non-fat-dry milk or 5% BSA (Sigma). Primary antibody was incubated in TBST at 4°C overnight and followed by horseradish peroxidase (HRP)-conjugated secondary antibody incubation for 1–3 h at room temperature. The membrane was developed using the ECL system (170-5040, Bio-Rad Laboratories). IB band intensity was quantified using ImageJ software. Student's t tests were performed for evaluating density difference of bands.

Immunofluorescence staining

Cells, grown on coverslips were fixed with 3.7% paraformaldehyde solution in PBS for 15 min. After permeabilization with 0.5% Triton X-100 buffer containing 20 mM HEPES at pH 7.4, 50 mM sodium chloride, 3 mM magnesium chloride and 300 mM sucrose at room temperature for 10 min, cells were incubated with a primary antibody for 2 h. After washing with PBS, cells were incubated with fluorescent-conjugated secondary antibody (Alexa Fluor 488 or 594 goat anti-rabbit or mouse) for 30 min. After final wash with PBS, nuclei were counterstained with DAPI containing mounting medium (Vectashield). All the steps were performed at room temperature, unless otherwise stated. Images were acquired by confocal microscopy (Zeiss LSM 510 META).

Cycloheximide chase assay

Cells were treated with CHX (50 μg/ml) and collected at different times (1, 2, 4, 6, 8 and 10 h). The amount of PTEN was determined by Western blot using anti-PTEN antibody.

RNA knockdown/silencing

Retroviral vector containing either scrambled shRNA or pool of PTEN shRNA (shRNA#1-AGGCGCTATGTGTATTATTAT; shRNA#2-CCACAGCTAGAACTTATCAAA; shRNA#3-CCACAAATGAAGGGATATAAA) along with a PcL-Ampho helper plasmid was co-transfected into BOSC23 packing cells. Virus containing supernatant was collected 48 h post-transfection and was used to infect cells in the presence of polybrene.

Ubiquitination assay

Cells were treated with MG132 (10 μM for 6 h) and the whole cell extracts prepared by Tris/HCl lysis buffer were subjected to immunoprecipitation of the anti-ubiquitin antibody. The analysis of ubiquitination was performed by immunoblotting with anti-PTEN antibody.

Cell viability assay by flow cytometry

Cells were transiently transfected with either vector, p65, IKKβ, or IKKβ-DN constructs. After 12 h, cells were treated with a combination of CHX (25 μg/ml) and TNF (5 nM) for 24 h. Cells were washed, trypsinized and then subjected to flow cytometry (FACS Aria, BD Biosciences) using Live-Dead Cytotoxicity assay kit (Invitrogen). Live cells were detected by green fluorescent dye, Calcein-AM; whereas, dead cells were detected by red fluorescent dye, ethidium homodimer. Live compared with dead cells were analysed using FlowJo software.

Molecular docking

Preparation of protein structures for docking

The 3D structure of PTEN (PDB ID: 2PBD, chain P) [20] and Profilin1 (PDB ID: 1D5R, chain A) [21] protein were obtained from protein data bank. All the non-protein molecules were removed by Chimaera [22]. AutoDock tools 1.5.6 [23,24] was used to prepare, and analyse the docking simulations. Protein PTEN was used as receptor and Profilin was used as ligand. Kollman united atom charges and polar hydrogen were added to the receptor protein. The grid box size was set at 58×76×58 Å (x, y and z) with the spacing between grid points at 1 Å (1 Å=0.1 nm) and the centre at 36.253, 82.395 and 31.728 for x, y and z coordinates respectively. AutoDock Vina [25] software was used for docking simulations. Clustering tolerance of 2.0 Å was used to form distinct conformation clusters. Docking results were analysed and visualized by AutoDock tools and PyMOL [23,24].

Statistical analysis

Results were expressed as mean ± S.D. for three independent experiments. Statistical analysis of the samples was done by Student's t test wherever applicable. The P≤0.05 was considered to be significant (*P≤0.05, **P≤0.01, ***P≤0.001 and ****P≤0.001).

RESULTS

Profilin-transfected cells show low amounts of NF-κB

To understand the effect of Profilin on NF-κB, breast tumour cells, MB-231 and T-cell lymphoid cells, HuT-78 were transfected with Profilin1 construct. The DNA binding activity was assayed from NE of non-transfected, vector-transfected and Profilin1-transfected cells followed by stimulation with TNF (1 nM for 30 min). MB-231 cells having low basal activity were unable to induce NF-κB DNA binding activity in the presence or absence of TNF upon Profilin1 overexpression. (Figure 1A, upper panel). On the other hand, HuT-78 cells showed high basal activity of NF-κB DNA binding and TNF was unable to further increase it. Both basal and TNF-induced NF-κB DNA binding activity was completely inhibited upon Profilin1 overexpression (Figure 1B). Further, Profilin1 was able to decrease TNF induced activity of the NF-κB-dependent SEAP when NF-κB-SEAP was transfected along with Profilin1 in MDA-231 cells (Figure 1C). However, HuT-78 cells showed high NF-κB-dependent SEAP activity, in un-stimulated as well as on TNF-stimulation and Profilin1 was able to suppress it in both the conditions (Figure 1D). On the other hand, Profilin1 was unable to inhibit TNF-induced AP-1 activation, as shown by gel shift assay, suggesting that Profilin selectively inhibits NF-κB, but not AP-1 upon TNF stimulation (Figure 1A, middle panel). These data suggest that upon Profilin overexpression, the activation of NF-κB, but not AP-1 is suppressed.

Effect of transient Profilin overexpression on NF-κB activation

Figure 1
Effect of transient Profilin overexpression on NF-κB activation

HuT-78 and MDA-MB-231 cells were transfected with vector or Profilin1 for 6 h, followed by culture for 24 h and then stimulated with TNF (1 nM) for 30 min. The NE were prepared and NF-κB DNA binding was assayed by gel shift assay (A and B, upper panel). NE of MDA-MB-231 cells were also assayed for AP-1 DNA binding by gel shift assay (A, middle panel). Protein expression levels of Profilin1 were also measured from WCE by Western blot, under similar condition (A and B, lower panel). HuT-78 and MDA-MB-231 cells were transfected with vector or Profilin1, in combination with plasmids for NF-κB promoter DNA that had been linked to SEAP (NF-κB-SEAP) and GFP for 6 h. Cells were then cultured for 24 h and treated with 1 nM TNF. The GFP-positive cells were counted, and transfection efficiency was calculated. The culture supernatant was assayed for SEAP activity. Results are represented as fold activation over the empty SEAP-transfected control (C and D). The error bars represents S.D.; Student's t test (*P<0.05, **P<0.01 and ***P<0.001).

Figure 1
Effect of transient Profilin overexpression on NF-κB activation

HuT-78 and MDA-MB-231 cells were transfected with vector or Profilin1 for 6 h, followed by culture for 24 h and then stimulated with TNF (1 nM) for 30 min. The NE were prepared and NF-κB DNA binding was assayed by gel shift assay (A and B, upper panel). NE of MDA-MB-231 cells were also assayed for AP-1 DNA binding by gel shift assay (A, middle panel). Protein expression levels of Profilin1 were also measured from WCE by Western blot, under similar condition (A and B, lower panel). HuT-78 and MDA-MB-231 cells were transfected with vector or Profilin1, in combination with plasmids for NF-κB promoter DNA that had been linked to SEAP (NF-κB-SEAP) and GFP for 6 h. Cells were then cultured for 24 h and treated with 1 nM TNF. The GFP-positive cells were counted, and transfection efficiency was calculated. The culture supernatant was assayed for SEAP activity. Results are represented as fold activation over the empty SEAP-transfected control (C and D). The error bars represents S.D.; Student's t test (*P<0.05, **P<0.01 and ***P<0.001).

Generation of Profilin-stable MB-231 cell lines

Profilin has been shown to be down-regulated in a variety of cancers, especially in breast and hepato-carcinoma. For this, the basal expression of Profilin was determined from whole cell lysates in different cell types such as A375, HuT-78, HEK293, HepG2, Hep3B, MCF-7, MCF-10A, MDA-MB-231, MDA-MB-468, MIA PaCa-2, PC-3, and U-937 cells by Western blot (Supplementary Figure S1A). HepG2 and MB-231 cells were found to have the lowest expression of Profilin. But MB-231 cells were selected as stable cells owing to their high mobility and invasiveness. For generation of stable cells, MB-231 cells were transfected with either empty vector or Profilin1, followed by selection and screening of stable clones, as described in experimental procedures. The stably overexpressing vector and Profilin1 will be referred as parental and Profilin-stable cells respectively. The amount of Profilin was determined by Western blot (Figure 2A, upper panel) and RT-PCR followed by PCR from total RNA (Figure 2A, lower panel). The amount of Profilin expressed in these clones was indicated in fold expression (Figure 2B). The phase contrast image suggests the morphology of parental cells was quite different from the Profilin-stable cells. Profilin-stable cells were less stellate and more spread than parental, which can be correlated to its reduced invasiveness (Supplementary Figure S1B).

Profilin overexpression inhibits activation of NF-κB and its dependent genes

Figure 2
Profilin overexpression inhibits activation of NF-κB and its dependent genes

Stable Profilin overexpressing as well as vector are generated in MB-231 cell lines as per described protocol. Western blot and RT-PCR further confirm Profilin overexpression in these cells (A). Tubulin and β-actin serve as the loading control for Western blot and RT-PCR respectively. Quantification of Profilin band from Western blot was done using ImageJ software and presented in fold (B). Data are representative of three experiments and are normalized with tubulin as loading control. Parental and Profilin-stable cells were stimulated with TNFα (10 pM), IL-1 (10 ng/ml), PMA (25 ng/ml), H2O2 (100 μM), OA (100 nM) and IL-8 (10 ng/ml) for 4 h. NEs were prepared and assayed for NF-κB DNA binding (C1). For NF-κB-dependent reporter gene assay, cells were transfected with NF-κB-SEAP and GFP for 6 h. Cells were then cultured for 24 h and were treated with various inducers, as indicated. GFP-positive cells were counted, and transfection efficiency was calculated. The culture supernatant was assayed for SEAP activity. Results are represented as fold of activation over the empty SEAP-transfected control (C2). Cells were treated with various concentrations of TNF for 1 h as well as with 1 nM TNF for different times, as indicated. NE were prepared and assayed for NF-κB DNA binding activity (D and E). For NF-κB-dependent reporter gene assay, cells were transfected with NF-κB-SEAP and GFP for 6 h. Cells were then cultured for 24 h and were treated with 1 nM TNF for various times, as indicated. GFP-positive cells were counted, and transfection efficiency was calculated. The culture supernatant was assayed for SEAP activity. Results are presented as fold of activation over the empty SEAP-transfected control (F). Total RNA was extracted from parental and Profilin-stable cells, stimulated with various concentrations of TNF for 24 h. Amounts of NF-κB-dependent genes, ICAM1 and Cox2 were measured by RT-PCR along with actin as the loading control (G). The error bars represents S.D.; Student's t test (*P<0.05, **P<0.01 and ***P<0.001).

Figure 2
Profilin overexpression inhibits activation of NF-κB and its dependent genes

Stable Profilin overexpressing as well as vector are generated in MB-231 cell lines as per described protocol. Western blot and RT-PCR further confirm Profilin overexpression in these cells (A). Tubulin and β-actin serve as the loading control for Western blot and RT-PCR respectively. Quantification of Profilin band from Western blot was done using ImageJ software and presented in fold (B). Data are representative of three experiments and are normalized with tubulin as loading control. Parental and Profilin-stable cells were stimulated with TNFα (10 pM), IL-1 (10 ng/ml), PMA (25 ng/ml), H2O2 (100 μM), OA (100 nM) and IL-8 (10 ng/ml) for 4 h. NEs were prepared and assayed for NF-κB DNA binding (C1). For NF-κB-dependent reporter gene assay, cells were transfected with NF-κB-SEAP and GFP for 6 h. Cells were then cultured for 24 h and were treated with various inducers, as indicated. GFP-positive cells were counted, and transfection efficiency was calculated. The culture supernatant was assayed for SEAP activity. Results are represented as fold of activation over the empty SEAP-transfected control (C2). Cells were treated with various concentrations of TNF for 1 h as well as with 1 nM TNF for different times, as indicated. NE were prepared and assayed for NF-κB DNA binding activity (D and E). For NF-κB-dependent reporter gene assay, cells were transfected with NF-κB-SEAP and GFP for 6 h. Cells were then cultured for 24 h and were treated with 1 nM TNF for various times, as indicated. GFP-positive cells were counted, and transfection efficiency was calculated. The culture supernatant was assayed for SEAP activity. Results are presented as fold of activation over the empty SEAP-transfected control (F). Total RNA was extracted from parental and Profilin-stable cells, stimulated with various concentrations of TNF for 24 h. Amounts of NF-κB-dependent genes, ICAM1 and Cox2 were measured by RT-PCR along with actin as the loading control (G). The error bars represents S.D.; Student's t test (*P<0.05, **P<0.01 and ***P<0.001).

Profilin-stable cells inhibit TNF-induced NF-κB activation and NF-κB-dependent gene expression

To check whether the effect of Profilin on NF-κB suppression is a general phenomenon or specific to TNF-mediated response, parental and Profilin-stable cells were treated with different activators of NF-κB. The inducers like TNF, IL-1, PMA, hydrogen peroxide (H2O2), okadaic acid (OA) and IL-8 induced NF-κB DNA binding in parental cells and this activity was partially or completely inhibited in Profilin-stable cells (Figure 2C1) and the same was confirmed by the NF-κB reporter SEAP assay (Figure 2C2). In the parental cells, the NF-κB DNA binding was increased with increasing concentration of TNF as well as with time of TNF treatment and it was completely inhibited in Profilin-stable cells (Figures 2D and 2E). Various subunits of NF-κB such as p65, p50, Rel-B and c-Rel were able to bind to specific sequences in DNA. To determine the specificity of subunits involved, nuclear extract from TNF-treated cells was incubated with antibodies against these subunits, alone or in combination and super-shift was observed by gel shift assay. Antibodies to either p65 or p50 subunit shifted the band to higher molecular mass (Supplementary Figure S2A), thus suggesting the TNF-activated complex consisted of both p50 and p65 subunits. The other antibodies such as anti-c-Rel or anti-Rel B had no effect on the mobility of NF-κB. Further, the complex partially disappeared in the presence of excess of cold NF-κB, whereas, there was no binding by mutant NF-κB oligonucleotides.

The activity of NF-κB-dependent SEAP in the parental and Profilin-stable MB-231 cells upon transfection with NF-κB-SEAP was also determined and found to be consistent with the previous results (Figure 2F). The amount of ICAM1 and Cox2 were increased with increasing concentrations of TNF in parental cells, but not in Profilin-stable cells as determined from whole-cell extract (WCE) by Western blot (Figure 2G) or from total RNA by RT-PCR (Supplementary Figure S2B). These data suggest that Profilin overexpressing cells suppress NF-κB and NF-κB-dependent gene expression.

Profilin-stable cells inhibit TNF-induced IKK phosphorylation and p65 nuclear translocation

As NF-κB was suppressed in Profilin-stable cells, the nuclear translocation of p65, the subunit of NF-κB, was determined. Also, as NF-κB activation and translocation is preceded by phosphorylation and degradation of IκBα, the amounts of IκBα was also measured. The amount of IκBα decreased along with time of TNF stimulation in parental cells, whereas it remained unchanged in Profilin-stable cells, even after 2 h of TNF-stimulation. The level of p65 gradually decreased in cytoplasm and increased in nucleus with time of TNF stimulation in parental cells. In Profilin-stable cells, the level of p65 remains unchanged in cytoplasm, as well as in nucleus (Figure 3A). Since Profilin inhibits p65 nuclear translocation and IκBα degradation, we checked for the phosphorylation status of IKKα/β, IκBα and p65 in TNF-stimulated conditions. The levels of phospho-IKKα/β, phospho-IκBα and phospho-p65 were also increased to 2.13-, 1.83- and 2.02-fold respectively in TNF-stimulated parental cells. But, very low phosphorylation of these molecules was observed in Profilin-stable cells upon TNF stimulation (Figure 3B and Supplementary Figures 3A–3D). TNF-induced parental cells showed the presence of p65 predominantly in the nucleus compared with the un-induced cells. Immunofluorescence data also indicated that p65 remained predominantly in the cytoplasm of TNF-induced cells like un-induced Profilin-stable cells (Figure 3C). These data suggested that Profilin suppressed NF-κB by inhibiting IKK phosphorylation and phosphorylation of all downstream molecules such as IκBα and p65, and subsequently, p65 nuclear translocation and activation.

Profilin overexpression inhibits TNFα induced NF-κB translocation and activation

Figure 3
Profilin overexpression inhibits TNFα induced NF-κB translocation and activation

Parental and Profilin-stable cells were stimulated with 1 nM TNF for different times, as indicated. The amount of p65 was measured from CE and NE; IκBα levels were also measured from CE by Western blot. The blot for NE was reprobed for Lamin B and blots for CE were reprobed for tubulin (A). WCE prepared from TNF (1 nM for 30 min) treated cells, were used to determine the amount of phospho-IKKα/β (Ser177/181), phospho-IκBα (Ser36) and phospho-p65 (Ser276) using specific antibodies by Western blot. Blots were reprobed with IKKα/β, IκBα and p65 respectively, and tubulin as loading control (B). Cells stimulated with 1 nM TNF for 30 min were fixed and subjected to immunocytochemistry to detect NF-κB localization using anti-p65 antibody, as described under the section ‘Experimental’ (C). The graphs summarize the nucleus-to-cytoplasm (Nuc/Cyt) ratio of p65 fluorescence in the cells (number of cells=30). The error bars represents S.E.M; Student's t test (****P<0.0001).

Figure 3
Profilin overexpression inhibits TNFα induced NF-κB translocation and activation

Parental and Profilin-stable cells were stimulated with 1 nM TNF for different times, as indicated. The amount of p65 was measured from CE and NE; IκBα levels were also measured from CE by Western blot. The blot for NE was reprobed for Lamin B and blots for CE were reprobed for tubulin (A). WCE prepared from TNF (1 nM for 30 min) treated cells, were used to determine the amount of phospho-IKKα/β (Ser177/181), phospho-IκBα (Ser36) and phospho-p65 (Ser276) using specific antibodies by Western blot. Blots were reprobed with IKKα/β, IκBα and p65 respectively, and tubulin as loading control (B). Cells stimulated with 1 nM TNF for 30 min were fixed and subjected to immunocytochemistry to detect NF-κB localization using anti-p65 antibody, as described under the section ‘Experimental’ (C). The graphs summarize the nucleus-to-cytoplasm (Nuc/Cyt) ratio of p65 fluorescence in the cells (number of cells=30). The error bars represents S.E.M; Student's t test (****P<0.0001).

p65, not IKK, rescues Profilin-mediated suppression of NF-κB

As Profilin inhibits phosphorylation of IKK, we were interested in determining whether IKK is involved in Profilin-mediated suppression of NF-κB or not. For this, parental and Profilin-stable cells were transfected with the wild-type constructs of p65 or IKKβ as well as with constitutive active or kinase dead mutants of IKKβ (IKKβ-CA and IKKβ-DN respectively). Cells stimulated with TNF and NE were prepared to measure NF-κB DNA binding activity by gel shift assay. The p65 overexpression showed that an increase in the NF-κB DNA binding and TNF marginally increased this activation (as compared with vector alone) in parental cells. Upon p65 overexpression in Profilin-stable cells, NF-κB activation was found to be almost similar to parental cells, with or without TNF stimulation. These data suggest that p65 overexpression rescues Profilin-mediated NF-κB down-regulation and Profilin suppresses NF-κB activation either by inhibiting IKK complex or via an upstream event.

However, when IKKβ was overexpressed, TNF-induced NF-κB activation was significantly low in Profilin-stable cells as compared with parental cells. Overexpression of IKKβ-CA in parental cells showed significant increase in the NF-κB DNA binding and TNF was unable to increase it further. Besides, in Profilin-stable cells IKKβ-CA was able to induce NF-κB comparable to the parental cells, unlike that observed with IKKβ-WT. On the other hand, transfection of IKKβ-DN in parental or Profilin-stable cells completely abrogated NF-κB DNA binding and stimulation with TNF was unable to alter this effect (Figure 4A). This was further confirmed by transfection of NF-κB reporter SEAP (NF-κB-SEAP) along with these constructs and SEAP activity was assayed under similar experimental conditions (Figure 4B). The levels of these ectopically expressed proteins were also measured by Western blot to ensure the uniform expression in parental and Profilin-stable cells (Figures 4C1 and 4C2). Altogether, these results clearly indicate that phosphorylation and activation of IKK complex is essential and Profilin is probably targeting it for NF-κB deactivation.

p65, not IKK, restores TNF-induced NF-κB down-regulation in Profilin overexpressing cells

Figure 4
p65, not IKK, restores TNF-induced NF-κB down-regulation in Profilin overexpressing cells

Parental and Profilin-stable cells were transiently transfected with p65, IKKβ, IKKβ-CA or IKKβ-DN constructs, for 6 h. At 24 h post-transfection, cells were treated with 1 nM TNF for 30 min. NEs were prepared and assayed for NF-κB DNA binding by gel shift assay (A). Cells were also transfected with p65, IKKβ, IKKβ-CA or IKKβ-DN constructs along with reporter NF-κB-SEAP and GFP, for NF-κB-dependent reporter gene assay. The assay was performed as described previously, under similar experimental conditions. Results are presented as fold of activation over the empty SEAP-transfected control (B). The error bars represents S.D.; Student's t test (*P<0.05, **P<0.01 and ***P<0.001). The amounts of p65 and HA-tagged IKKβ were determined by Western blot from the different transfection conditions (C1 and C2).

Figure 4
p65, not IKK, restores TNF-induced NF-κB down-regulation in Profilin overexpressing cells

Parental and Profilin-stable cells were transiently transfected with p65, IKKβ, IKKβ-CA or IKKβ-DN constructs, for 6 h. At 24 h post-transfection, cells were treated with 1 nM TNF for 30 min. NEs were prepared and assayed for NF-κB DNA binding by gel shift assay (A). Cells were also transfected with p65, IKKβ, IKKβ-CA or IKKβ-DN constructs along with reporter NF-κB-SEAP and GFP, for NF-κB-dependent reporter gene assay. The assay was performed as described previously, under similar experimental conditions. Results are presented as fold of activation over the empty SEAP-transfected control (B). The error bars represents S.D.; Student's t test (*P<0.05, **P<0.01 and ***P<0.001). The amounts of p65 and HA-tagged IKKβ were determined by Western blot from the different transfection conditions (C1 and C2).

Profilin interacts with PTEN and stabilizes it

Since, Profilin overexpression inhibits the phosphorylation of IKK, so the physical association among them was tested. For this, co-immunoprecipitation (Co-IP) experiment for Profilin and IKK was performed and no interaction was detected. We advocated indirect involvement of Profilin in regulation of NF-κB and looked for the possible role of phosphatases in dephosphorylation of IKK [13]. Previously, Profilin was shown to down-regulate AKT signalling through up-regulation of a phosphatase, PTEN [15]. To test our hypothesis, MB-231 cells were transiently transfected with FLAG-tagged Profilin1 and immunoprecipitated with anti-IgG or anti-FLAG antibody, followed by immunoblotting with anti-PTEN antibody (Figure 5A). To assess the significance of Profilin–PTEN interaction, we looked for PTEN stability in Profilin overexpressed conditions. Parental or Profilin-stable cells were chased with CHX for regular time intervals and the amount of PTEN was determined by Western blot. Levels of PTEN diminished in the parental cells, but remain unaffected in Profilin-stable cells (Figure 5B). To further prove the Profilin-mediated inhibition of PTEN degradation, cells were treated with proteasome inhibitor, MG132. The amount of PTEN ubiquitination was detected by immunoprecipitation of total ubiquitin and immunoblotting with anti-PTEN antibody. Profilin-stable cells showed less poly-ubiquitination as compared with parental cells (Figure 5C). Immunofluorescence study also indicated the co-localization of Profilin and PTEN on the membrane in Profilin overexpressing cells than in parental cells (Figure 5D). These data further suggest that Profilin inhibits PTEN ubiquitination, and subsequent degradation.

Profilin interacts with PTEN and prevents its degradation

Figure 5
Profilin interacts with PTEN and prevents its degradation

MB-231 cells were transiently transfected with FLAG-tagged Profilin and its interaction with PTEN was detected by immunoprecipitation with anti-IgG or anti-FLAG, followed by immunoblotting with anti-PTEN (A). Parental and Profilin-stable cells were treated with CHX (50 μg/ml) and collected at different times as indicated. The levels of PTEN were determined from WCE, using anti-PTEN by Western blot (B). Cells were also treated with MG132 (10 μM for 6 h) and WCE were prepared. The levels of PTEN ubiquitination were evaluated by immunoprecipitating with anti-PTEN and immunoblotting with anti-ubiquitin. Tubulin was also probed to ensure equal loading of cell lysate (C). Parental and Profilin-stable cells were fixed and subjected to immunofluorescence staining using anti-Profilin (red, left panel) and anti-PTEN (green, middle panel) antibodies. The right panel shows the overlay of Profilin, PTEN and nuclear DAPI (blue) staining in the same field. Scale bar, 10 μm (D). Different truncation mutants of PTEN were Myc-tagged and co-transfected with FLAG-tagged Profilin. The interaction was determined by immunoprecipitation with anti-Myc, followed by immunoblotting with anti-FLAG (E). FLAG-tagged Profilin or its H133S mutant was co-transfected with Myc-tagged PTEN and immunoprecipitated with anti-FLAG and probed with anti-Myc by Western blot (F). MDA-MB-231 cells were transfected with Profilin1 or Profilin-H133S and NF-κB-SEAP for 6 h, followed by culture for 24 h and then stimulated with TNF (1 nM) for 30 min. The NEs were prepared and NF-κB DNA binding was assayed by gel shift assay (G). The culture supernatant was used to determine the activity of SEAP (H). The amount of PTEN was measured from WCE of untreated Profilin1 or Profilin-H133S transfected cells (I).

Figure 5
Profilin interacts with PTEN and prevents its degradation

MB-231 cells were transiently transfected with FLAG-tagged Profilin and its interaction with PTEN was detected by immunoprecipitation with anti-IgG or anti-FLAG, followed by immunoblotting with anti-PTEN (A). Parental and Profilin-stable cells were treated with CHX (50 μg/ml) and collected at different times as indicated. The levels of PTEN were determined from WCE, using anti-PTEN by Western blot (B). Cells were also treated with MG132 (10 μM for 6 h) and WCE were prepared. The levels of PTEN ubiquitination were evaluated by immunoprecipitating with anti-PTEN and immunoblotting with anti-ubiquitin. Tubulin was also probed to ensure equal loading of cell lysate (C). Parental and Profilin-stable cells were fixed and subjected to immunofluorescence staining using anti-Profilin (red, left panel) and anti-PTEN (green, middle panel) antibodies. The right panel shows the overlay of Profilin, PTEN and nuclear DAPI (blue) staining in the same field. Scale bar, 10 μm (D). Different truncation mutants of PTEN were Myc-tagged and co-transfected with FLAG-tagged Profilin. The interaction was determined by immunoprecipitation with anti-Myc, followed by immunoblotting with anti-FLAG (E). FLAG-tagged Profilin or its H133S mutant was co-transfected with Myc-tagged PTEN and immunoprecipitated with anti-FLAG and probed with anti-Myc by Western blot (F). MDA-MB-231 cells were transfected with Profilin1 or Profilin-H133S and NF-κB-SEAP for 6 h, followed by culture for 24 h and then stimulated with TNF (1 nM) for 30 min. The NEs were prepared and NF-κB DNA binding was assayed by gel shift assay (G). The culture supernatant was used to determine the activity of SEAP (H). The amount of PTEN was measured from WCE of untreated Profilin1 or Profilin-H133S transfected cells (I).

Since various domains of PTEN are involved in different functions, a series of Myc-tagged PTEN deletion mutants (D1–D5) that lack different domains were generated in order to map the interaction region of Profilin on PTEN. These deletion mutants along with full-length FLAG-tagged Profilin1 were co-expressed, immunoprecipitated with anti-Myc antibody and immunoblotted with anti-FLAG antibody (Figure 5E). Profilin is known to interact with its target proteins through its poly-L proline (PLP) binding motif and H133S mutation abrogates this interaction. To find the binding site of PTEN on Profilin, Myc-PTEN was co-transfected with either empty FLAG vector or FLAG-tagged Profilin1 or H133S mutant of Profilin1. The complex was immunoprecipitated with anti-FLAG and immunoblotted with anti-Myc antibody (Figure 5F). Co-immunoprecipitation results clearly indicate that Profilin interacts with the phosphatase domain of PTEN through its PLP-binding site.

As Profilin H133S mutant is defective in binding to PTEN, its effect on NF-κB activation was determined. For this, FLAG-Profilin WT or FLAG-Profilin H133S were transiently overexpressed in MDA-MB-231 cells and stimulated with TNF. NE was prepared and NF-κB DNA binding was detected by gel shift assay (Figure 5G). Interestingly, Profilin H133S mutant was unable to inhibit TNF-triggered NF-κB activation and the NF-κB reporter SEAP assay confirmed the same (Figure 5H). Moreover, the endogenous level of PTEN was also increased in Profilin WT-transfected cells, but not in Profilin-H133S transfected cells (Figure 5I). Hence, these results clearly indicate that Profilin attenuates NF-κB signalling via PTEN.

PTEN interacts with IKK and inhibits its phosphorylation

As PTEN, a phosphatase is known to negatively regulate NF-κB signalling and its activity is up-regulated in Profilin-stable cells, we looked for the interaction between PTEN and IKK using Co-IP experiment. For this, MB-231 cells were lysed and immunoprecipitated with anti-IKKα/β or anti-PTEN along with IgG control, followed by Western blotting. PTEN was found to be associated with immunoprecipitated IKK (Figure 6A1). On the other hand, IKK was also found in the immunoprecipitated PTEN (Figure 6A2). This confirmed the association of PTEN with IKK. To assess the functional significance of this association, PTEN was overexpressed in MB-231 cells and phosphorylation of IKKα/β was determined. PTEN overexpressed cells failed to show any IKK phosphorylation on TNF stimulation (Figure 6B). This result suggests that PTEN physically interacts with IKK and prevents its phosphorylation and thus, plays a critical role in Profilin-mediated NF-κB signalling.

PTEN interacts with IKK and prevents its phosphorylation

Figure 6
PTEN interacts with IKK and prevents its phosphorylation

MB-231 cells were cultured and prepared for WCE. The WCE was immunoprecipitated with anti-IgG and anti-FLAG or anti-IKKα/β and probed with anti-FLAG and anti-IKKα/β respectively by Western blot (A1 and A2). Cells were also transfected with empty vector and FLAG-tagged PTEN for 6 h. At 24 h post-transfection, cells were stimulated with 1 nM TNF for 30 min. WCE were used to determine the levels of phospho-IKKα/β (Ser177/181) and IκBα using specific antibodies by Western blot. Blots were reprobed with IKKα/β, and tubulin as loading control (B). MB-231 cells were transiently transfected with FLAG-tagged Profilin along with Myc-PTEN and HA-IKKβ and the interaction was detected by immunoprecipitation with anti-IgG or anti-FLAG, followed by immunoblotting with anti-Myc and anti-HA (C).

Figure 6
PTEN interacts with IKK and prevents its phosphorylation

MB-231 cells were cultured and prepared for WCE. The WCE was immunoprecipitated with anti-IgG and anti-FLAG or anti-IKKα/β and probed with anti-FLAG and anti-IKKα/β respectively by Western blot (A1 and A2). Cells were also transfected with empty vector and FLAG-tagged PTEN for 6 h. At 24 h post-transfection, cells were stimulated with 1 nM TNF for 30 min. WCE were used to determine the levels of phospho-IKKα/β (Ser177/181) and IκBα using specific antibodies by Western blot. Blots were reprobed with IKKα/β, and tubulin as loading control (B). MB-231 cells were transiently transfected with FLAG-tagged Profilin along with Myc-PTEN and HA-IKKβ and the interaction was detected by immunoprecipitation with anti-IgG or anti-FLAG, followed by immunoblotting with anti-Myc and anti-HA (C).

Although, we are able to demonstrate the interaction of Profilin with PTEN and PTEN with IKK, we were unable to show the association of Profilin with IKK endogenously. To answer this, we ectopically expressed the various constructs and looked for the association. Cells were transfected with empty FLAG or FLAG-Profilin along with Myc-PTEN and HA-IKKβ and the complex was immunoprecipitated with FLAG and tested for the anti-Myc and anti-HA by Western blot. Myc-PTEN and HA-IKK were found to be present in the immunoprecipitated FLAG-Profilin, but not in control (Figure 6C), indicating the presence of Profilin–PTEN–IKK complex.

Profilin is unable to inhibit TNF-induced NF-κB activation in PTEN-null cells

In order to address the role of PTEN in down-regulation of NF-κB in Profilin overexpressing cells, PTEN null MB-468 cells were transfected with Profilin1 or PTEN or both and stimulated with TNF. TNF treatment significantly increased NF-κB DNA binding in vector or Profilin1-transfected MB-468 cells, but was unable to induce NF-κB DNA binding when transfected with PTEN alone or in combination with Profilin1 (Figure 7A). NF-κB-dependent reporter SEAP assay further confirmed the results (Figure 7B). The levels of IκBα decreased upon TNF stimulation in vector or Profilin1-transfected MB-468 cells. MB-468 cells transfected with PTEN alone or in combination with Profilin1, were unable to reduce the level of IκBα as the Profilin interacts with PTEN and inhibits IKK activation, and subsequent IκBα degradation (Figure 7C).

Profilin overexpression has no effect on NF-κB signalling in PTEN-null or PTEN knockdown cells

Figure 7
Profilin overexpression has no effect on NF-κB signalling in PTEN-null or PTEN knockdown cells

MB-468 (PTEN null) cells were transfected with empty vector, Profilin and PTEN in various combinations as indicated for 6 h. After 24 h, cells were treated with 1 nM TNF for 30 min. NEs were prepared and used to measure NF-κB DNA binding by gel shift assay (A). Cells were also transfected with NF-κB-SEAP and GFP, along with Profilin and PTEN, as indicated. NF-κB-dependent reporter SEAP assay was performed as described previously (B). The amount of IκBα was determined from WCE by Western blot, under similar treatment conditions. The blot was reprobed for tubulin as loading control (C). MB-468 cells were transfected with empty vector or Myc-PTEN-WT and Myc-PTEN-G129E along with FLAG-Profilin as indicated for 6 h. After 24 h, cells were treated with 1 nM TNF for 30 min. The amounts of IκBα was detected from WCE by Western blot (D). Parental and Profilin-stable cells were transduced with retroviral pool of control and PTEN shRNA for 48 h, followed by stimulation with 1 nM TNF for 30 min. NEs were prepared and NF-κB DNA binding was measured (E). For NF-κB SEAP assay, cells were infected with shRNA, followed by transfection with NF-κB-SEAP and GFP, as indicated. Assay was performed as described previously (F). The WCE were used to detect the amount of IκBα by Western blot, under similar treatment conditions. The blot was reprobed for tubulin as loading control (G). The error bars represents S.D.; Student's t test (*P<0.05, **P<0.01 and ***P<0.001).

Figure 7
Profilin overexpression has no effect on NF-κB signalling in PTEN-null or PTEN knockdown cells

MB-468 (PTEN null) cells were transfected with empty vector, Profilin and PTEN in various combinations as indicated for 6 h. After 24 h, cells were treated with 1 nM TNF for 30 min. NEs were prepared and used to measure NF-κB DNA binding by gel shift assay (A). Cells were also transfected with NF-κB-SEAP and GFP, along with Profilin and PTEN, as indicated. NF-κB-dependent reporter SEAP assay was performed as described previously (B). The amount of IκBα was determined from WCE by Western blot, under similar treatment conditions. The blot was reprobed for tubulin as loading control (C). MB-468 cells were transfected with empty vector or Myc-PTEN-WT and Myc-PTEN-G129E along with FLAG-Profilin as indicated for 6 h. After 24 h, cells were treated with 1 nM TNF for 30 min. The amounts of IκBα was detected from WCE by Western blot (D). Parental and Profilin-stable cells were transduced with retroviral pool of control and PTEN shRNA for 48 h, followed by stimulation with 1 nM TNF for 30 min. NEs were prepared and NF-κB DNA binding was measured (E). For NF-κB SEAP assay, cells were infected with shRNA, followed by transfection with NF-κB-SEAP and GFP, as indicated. Assay was performed as described previously (F). The WCE were used to detect the amount of IκBα by Western blot, under similar treatment conditions. The blot was reprobed for tubulin as loading control (G). The error bars represents S.D.; Student's t test (*P<0.05, **P<0.01 and ***P<0.001).

The major role of PTEN is to regulate intracellular levels of second-messenger phosphatidylinositol 3,4,5-trisphosphate (PIP3), which is required for activation of PI3-kinase signalling. To rule out the effect of Profilin–PTEN on NF-κB regulation via PI3K/AKT pathway, PTEN lipid phosphatase defective mutant G129E was used. Myc-tagged PTEN-WT or G129E mutant was transfected along with FLAG-tagged Profilin in PTEN null MDA-468 cells. Cells were then stimulated with TNF and lysate was used to measure the level of IκBα (Figure 7D). PTEN G129E mutant was unable to induce the degradation of IκBα, similar to the PTEN WT, indicating that PTEN modulates NF-κB independent of AKT.

Knocking down of PTEN restores TNF-induced NF-κB activation in Profilin-stable cells

To confirm the role of PTEN in Profilin-mediated suppression of NF-κB, both parental and Profilin-stable cells were transfected with either scrambled or pool of PTEN shRNA. This effect was assayed by NF-κB DNA binding by gel shift and NF-κB-dependent reporter SEAP activity stimulation of TNF. PTEN knockdown increased TNF induced NF-κB DNA binding and SEAP activity in Profilin-stable cells (Figures 7E and 7F). Further, TNF-stimulated IκBα degradation was also observed in Profilin-stable cells upon PTEN knockdown (Figure 7G). These data suggested that silencing of PTEN results in NF-κB activation even in Profilin overexpressing cells.

Reports have suggested that phosphatases such as PP2A and PP4 are also known to be involved in NF-κB regulation [13]. To rule out possible involvement of these phosphatases in Profilin-mediated suppression of NF-κB, Profilin-stable cells were transfected with either shRNA of PP2A or PP4 alone, or in combination with PTEN shRNA for 12 h, followed by TNF treatment. Neither PP2A nor PP4 knockdown were able to rescue NF-κB down-regulation in Profilin overexpressed cells. On the other hand, PTEN alone or in combination with PP2A or PP4 was able to restore NF-κB signalling in Profilin-stable cells (Supplementary Figure S4). Thus, the results apparently indicate that inhibition of IKK phosphorylation is a direct consequence of PTEN and other phosphatase, such as PP2A and PP4, which are not involved in Profilin-mediated NF-κB down-regulation.

Profilin interacts with PTEN without interfering with the actin-binding site

As we have shown that Profilin protects PTEN degradation, the interaction between them and the possible domain of interaction were determined. In silico analysis showed that the Profilin and PTEN interaction was quite strong and amino acids which interact with Profilin (in green) and PTEN (in blue) are shown (Figures 8A and 8B; Supplementary Figure S5A) as suggested by the binding energy (−17.2 to −15.0 kcal/mol for different clusters) (Figure 8C). The known interacting protein of Profilin, actin showed interact with PTEN in different domain (Supplementary Figures S5B1 and S5B2). Different interacting regions of Profilin (PTEN binding in blue, acting binding in red and proline-rich in green) are indicated (Supplementary Figure S5C). All these data suggest that there is a strong interaction between Profilin and PTEN and the interacting domain of PTEN somewhat overlaps with the PLP binding domain, but is different from the actin binding domain.

In silico analysis of the interaction of Profilin with PTEN

Figure 8
In silico analysis of the interaction of Profilin with PTEN

Detailed interaction involving amino acids of PTEN (blue) with Profilin (green) is depicted (A). Various amino acids of PTEN, Profilin and actin participating in interactions are tabulated (B). Energy values for the interaction of PTEN and Profilin, as determined by AutoDock Vina (C).

Figure 8
In silico analysis of the interaction of Profilin with PTEN

Detailed interaction involving amino acids of PTEN (blue) with Profilin (green) is depicted (A). Various amino acids of PTEN, Profilin and actin participating in interactions are tabulated (B). Energy values for the interaction of PTEN and Profilin, as determined by AutoDock Vina (C).

Profilin sensitizes cell death and IKK or p65 rescues Profilin-mediated cell death

As we have shown that Profilin overexpression down-regulates NF-κB, we further argued whether Profilin overexpression has any effect on cell survival or not. For this, parental and Profilin-stable MB-231 cells were treated with CHX along with TNF (CHX + TNF) for 24 h and cell death was determined by flow cytometry. The data suggested that Profilin enhances breast tumour cell death.

As we have proved that the p65 or IKK mediated restoration of NF-κB down-regulation, the cell death was determined in parental and Profilin-stable cells, transfected with p65 or IKKβ upon CHX + TNF treatment. The amount of cell death was increased in CHX + TNF treated cells and rescued upon p65 or IKKβ transfection in parental cells. In Profilin-stable cells, CHX + TNF treatment showed pronounced (almost 82%) cell death which was reduced in p65 or IKKβ-transfected cells (Figures 9A and 9B). These data suggest that NF-κB down-regulation causes rigorous cell death but NF-κB overexpression (via p65 or IKK transfection) decreases cell death even in Profilin overexpressing cells.

p65 and IKKβ protects Profilin overexpressed cells from cell death

Figure 9
p65 and IKKβ protects Profilin overexpressed cells from cell death

Parental and Profilin-stable cells were transiently transfected with either vector or p65 and IKKβ constructs and treated with a combination of CHX (25 μg/ml) and TNF (5 nM) for 24 h. Cells were then subjected to flow cytometry using Live-Dead Cytotoxicity assay kit (Invitrogen) (A). The percentage of live cells were counted in the flow cytometer and indicated in bar diagram (B). The error bars represents S.D.

Figure 9
p65 and IKKβ protects Profilin overexpressed cells from cell death

Parental and Profilin-stable cells were transiently transfected with either vector or p65 and IKKβ constructs and treated with a combination of CHX (25 μg/ml) and TNF (5 nM) for 24 h. Cells were then subjected to flow cytometry using Live-Dead Cytotoxicity assay kit (Invitrogen) (A). The percentage of live cells were counted in the flow cytometer and indicated in bar diagram (B). The error bars represents S.D.

DISCUSSION

Deregulation of Profilin in most of the tumour cells prompted us to look for its role in tumorigenesis. As NF-κB is considered to be a master regulator whose dependent genes are involved in tumour progression, we have established the role of Profilin in NF-κB regulation. In general, NF-κB acts antagonistically to tumour suppressors due to its ability to promote survival by up-regulating genes involved in cell proliferation. High basal activity of NF-κB is found to be associated with a variety of tumour cell types, such as HuT-78, PC-3 and considered as an important determinant in maintaining a ‘pro-inflammatory microenvironment’ for tumour progression, suppression of which inhibits the growth of tumour cells [2629]. We have found that Profilin overexpression not only down-regulates NF-κB activity in MB-231 cells, but even in constitutively NF-κB expressing HuT-78 cells. AP1 is another transcription factor that functions in close association with NF-κB and is activated upon TNF stimulation. Profilin was unable to suppress AP1 activation, suggesting Profilin inhibits downstream of TNF-mediated NF-κB and AP1 cross-talk. Profilin overexpressing cells completely suppressed TNF-, IL-1- or IL-8-mediated NF-κB activation, whereas PMA, H2O2 or OA were able to inhibit this activation partially. This is probably because PMA, H2O2 or OA might be inducing NF-κB by phosphorylating IκBα through activation of protein tyrosine kinases such as p56 (lck), ZAP-70 etc. [3032]. On the other hand, TNF, IL-1 or IL-8 activate ‘canonical’ serine/threonine kinases that phosphorylate IκBα at Ser32/36 residues, leading to its ubiquitination and degradation, followed by activation of NF-κB [33]. Profilin may be targeting these serine/threonine kinases to inhibit NF-κB activation.

Profilin overexpressing cells showed complete inhibition of IκBα degradation, p65 nuclear translocation, NF-κB DNA binding activity and NF-κB-dependent genes, ICAM1 and Cox2 expression when stimulated with TNF. TNF triggers a rapid phosphorylation of IKKα/β, followed by phosphorylation and degradation of IκBα, but no phosphorylation was found in Profilin overexpressing cells. Thus, Profilin attenuates IKK phosphorylation and thereby inhibits activation of downstream events to induce NF-κB. Phosphorylation of p65 was also found suppressed in these cells. It is known that phosphorylation of p65 is post-translationally regulated by kinases like CK2 and protein kinase A (PKA) [3437]. Whether Profilin also targets these kinases needs to be studied further.

The ectopic expression of p65 or IKKβ-EE in Profilin overexpressing cells completely recovers the TNF-mediated suppression of NF-κB, whereas IKKβ-DN further suppresses NF-κB signalling. However, IKKβ was able to restore NF-κB suppression partially, suggesting that Profilin might be targeting IKK for NF-κB down-regulation. As Profilin failed to show any association with IKK, we searched for another possible target of Profilin that influences the NF-κB pathway. Independent studies have indicated that Profilin overexpression up-regulates PTEN and PTEN antagonizes the NF-κB pathway [15,38]. On this basis, we hypothesized that PTEN might be involved in Profilin-mediated NF-κB suppression. To our surprise, Profilin is found to interact with PTEN and in turn, PTEN interacts with IKK complex. Overexpression of Profilin was unable to inhibit TNF-induced NF-κB activation in PTEN-null MB-468 cells. Knocking down of PTEN in Profilin overexpressing cells failed to prevent TNF-mediated NF-κB suppression. Moreover, overexpression of PTEN prevents the phosphorylation of IKK, when stimulated with TNF, which further suggests that PTEN is an important determinant in Profilin-mediated attenuation of the NF-κB pathway. Whether PTEN acts as a phosphatase to dephosphorylate IKK or simply sequesters it, requires further investigation.

PTEN is essential for normal cellular function and tumour prevention. Expression of PTEN was found to be increase at the level of protein, but not at the level of mRNA, suggesting that Profilin overexpression post-translationally up-regulates PTEN (Supplementary Figure S6A). For this, stability of PTEN was examined and half-life of PTEN was prolonged in the Profilin overexpressing cells. The degradation of PTEN is preceded by its ubiquitination. Profilin overexpression prevents the poly-ubiquitination of PTEN. Thus, Profilin interaction inhibits PTEN poly-ubiquitination, leading to PTEN stabilization and tumour suppression. Hence, stabilization of PTEN upon interaction with Profilin leads to decrease in the phosphorylation of IKK. Domain mapping using various PTEN deletion mutants revealed that Profilin interaction site is present in the phosphatase domain of PTEN. Previous studies have shown that the cellular level of PTEN is regulated by its E3 ligase WW domain containing protein 2 (WWP2) and binds to PTEN in the phosphatase domain [16]. WWP2 belongs to the neural precursor cell expressed developmentally downregulated protein 4 (NEDD4) ubiquitin family, which have a WW domain for substrate recognition and degradation of proteins that generally have a proline-proline-x–tyrosine (PPXY, where x=any amino acid) motif, such as PTEN. As Profilin contains a PLP binding domain that has high affinity towards PPXY motifs [2] and PTEN is shown to bind to this site, we hypothesize that Profilin competes with WWP2 for binding to PTEN at the PPXY motif and Profilin–PTEN association prevents WWP2-mediated PTEN ubiquitination, and subsequent degradation. In silico analysis showed strong binding of Profilin with PTEN. The interacting site of Profilin with PTEN partially overlaps with PLP binding domain, which explains the inability of the PLP binding defective H133S mutant of Profilin to interact with PTEN in Co-IP. Further, the site of interaction is different than the actin-binding site or phosphoinositide-binding site. The phoshoinositide-binding site is present within the actin-binding site and is crucial for membrane association of Profilin [2]. Although PTEN has its own membrane binding region(s), our immuno-fluorescence data not only show the co-localization of Profilin and PTEN, but also suggests their increased localization on the membrane. Thus, we speculate that Profilin interacts with PTEN and assists in its recruitment on membrane to negatively regulate PI3K/AKT signalling (Supplementary Figure S6B).

Although Profilin overexpression did not show any cytolysis as determined by cytosolic marker, lactate dehydrogenase (LDH) assay, it showed enhanced cell death in the presence of CHX and TNF treatment. Overexpression of p65 or IKK in the Profilin overexpressing cells, showed significant protection from apoptosis. However, complete recovery from cell death was not observed, indicating that other signalling pathway such as PI3K/AKT, which is reciprocally regulated by PTEN, must be involved. Moreover, Profilin has been implicated in many other cellular functions such as cell cycle arrest and cytokinesis defects [47], suggesting its PTEN independent role in cell survival. Thus, reduced Profilin expression in breast cancer cells significantly increases the resistance to chemotherapy and conversely, overexpression of Profilin sensitizes cancer cells to apoptosis.

In conclusion, our observations suggest that Profilin exerts its tumour suppressive effects through stabilization of PTEN, which facilitates dephosphorylation of IKK complex followed by suppression of NF-κB. Thus, Profilin inhibits NF-κB-dependent genes involved in tumour cell cycle progression, leading to cell death. For the first time, we provide data that support the role of Profilin in tumour suppression involving NF-κB deregulation.

Mutation in Profilin is known to cause familial amyotrophic lateral sclerosis (ALS), a neurodegenerative disorder resulting from motor neuron death [39]. This suggests that there might be a possibility of alterations in other functions of Profilin, which may contribute to tumorigenesis. As Profilin is down-regulated in human breast tumours and correlates with low PTEN and high NF-κB expression, we propose that mutation or loss of Profilin function may drive mammary cells for tumour progression. Modulating the expression level of Profilin may be useful in mitigating the tumorigenic growth and can be targeted along with other agents that are used against different pathways for effective combination therapy.

AUTHOR CONTRIBUTION

Adeel Zaidi did all the experiments and Sunil Manna conceptualized the project and wrote the manuscript.

We thank Dr M.S. Reddy, Staff Scientist, CDFD for providing the PTEN and its mutants and shRNA for PTEN, PP2A and PP4 and Mr. Binay K. Sahoo, Immunology, CDFD for helping in the docking studies.

FUNDING

This work was supported by the Centre for DNA Fingerprinting and Diagnostics (CDFD) Institute Core Fund; and the fellowship provided by Council for Scientific and Industrial Research (CSIR), Govt. of India to A.H.Z.

Abbreviations

     
  • CHX

    cycloheximide

  •  
  • Co-IP

    co-immunoprecipitation

  •  
  • Cox

    cyclooxygenase

  •  
  • H2O2

    hydrogen peroxide

  •  
  • ICAM

    intercellular cell adhesion molecule

  •  
  • IκBα

    inhibitory subunit of NF-κB

  •  
  • IKK

    IκBα kinase

  •  
  • IL

    interleukin

  •  
  • NE

    nuclear extract

  •  
  • NF-κB

    nuclear transcription factor κB

  •  
  • OA

    okadaic acid

  •  
  • PLP

    poly-L proline

  •  
  • PTEN

    phosphatase and tension homologue

  •  
  • RT-PCR

    reverse transcriptase-PCR

  •  
  • SEAP

    secretory alkaline phosphatase

  •  
  • TNF

    tumour necrosis factor

  •  
  • WWP2

    WW domain containing protein 2

References

References
1
Carlsson
L.
Nystrom
L.E.
Sundkvist
I.
Markey
F.
Lindberg
U.
Actin polymerizability is influenced by Profilin, a low molecular weight protein in non-muscle cells
J. Mol. Biol.
1977
, vol. 
115
 (pg. 
465
-
483
)
[PubMed]
2
Ding
Z.
Gau
D.
Deasy
B.
Wells
A.
Roy
P.
Both actin and polyproline interactions of Profilin-1 are required for migration, invasion and capillary morphogenesis of vascular endothelial cells
Exp. Cell Res.
2009
, vol. 
315
 (pg. 
2963
-
2973
)
[PubMed]
3
Pollard
T.D.
Borisy
G.G.
Cellular motility driven by assembly and disassembly of actin filaments
Cell
2003
, vol. 
112
 (pg. 
453
-
465
)
[PubMed]
4
Ding
Z.
Lambrechts
A.
Parepally
M.
Roy
P.
Silencing Profilin-1 inhibits endothelial cell proliferation, migration and cord morphogenesis
J. Cell Sci.
2006
, vol. 
119
 (pg. 
4127
-
4137
)
[PubMed]
5
Choi
Y.N.
Lee
S.K.
Seo
T.W.
Lee
J.S.
Yoo
S.J.
C-terminus of Hsc70-interacting protein regulates Profilin1 and breast cancer cell migration
Biochem. Biophys. Res. Commun.
2014
, vol. 
446
 (pg. 
1060
-
1066
)
[PubMed]
6
Witke
W.
The role of Profilin complexes in cell motility and other cellular processes
Trends Cell Biol.
2004
, vol. 
14
 (pg. 
461
-
469
)
[PubMed]
7
Ding
Z.
Joy
M.
Gunsaulus
M.
Lakshman
N.
Miron-Mendoza
M.
Petroll
M.
Condeelis
J.
Wells
A.
Roy
P.
Profilin-1 downregulation has contrasting effects on early vs late steps of breast cancer metastasis
Oncogene
2014
, vol. 
33
 (pg. 
2065
-
2074
)
[PubMed]
8
Chen
Q.
Zhu
W.
Liu
Z.
Yan
K.
Zhao
S.
Han
D.
Toll-like receptor 11-initiated innate immune response in male mouse germ cells
Biol. Reprod.
2014
, vol. 
38
 (pg. 
1
-
11
)
9
Kanaujiya
J.K.
Lochab
S.
Kapoor
I.
Pal
P.
Datta
D.
Bhatt
M.L.
Sanyal
S.
Behre
G.
Trivedi
A.K.
Proteomic identification of Profilin1 as a corepressor of estrogen receptor alpha in MCF7 breast cancer cells
Proteomics
2013
, vol. 
13
 (pg. 
2100
-
2112
)
[PubMed]
10
Salt
L.A.
May
M.J.
The IkappaB kinase complex: master regulator of NF-kappaB signalling
Immunol. Res.
2008
, vol. 
42
 (pg. 
3
-
18
)
[PubMed]
11
Escárcega
R.O.
Fuentes-Alexandro
S.
García-Carrasco
M.
Gatica
A.
Zamora
A.
The transcription factor nuclear factor-κB and cancer
Clin. Oncol.
2007
, vol. 
19
 (pg. 
154
-
161
)
12
Hayden
M.S.
Ghosh
S.
NF-κB, the first quarter-century: remarkable progress and outstanding questions
Genes. Dev.
2012
, vol. 
26
 (pg. 
203
-
234
)
[PubMed]
13
Li
S.
Wang
L.
Berman
M.A.
Zhang
Y.
Dorf
M.E.
RNAi screen in mouse astrocytes identifies phosphatases that regulate NF-kappaB signaling
Mol. Cell
2006
, vol. 
17
 (pg. 
497
-
509
)
14
Knobble
C.B.
Lapin
V.
Suzuki
A.
Mak
T.W.
The roles of PTEN in development, physiology and tumourigenesis in mouse models: a tissue-by-tissue survey
Oncogene
2008
, vol. 
27
 (pg. 
5398
-
5415
)
[PubMed]
15
Das
T.
Bae
Y.H.
Wells
A.
Roy
P.
Profilin 1 overexpression upregulates PTEN and suppresses AKT activation in breast cancer cells
J. Cell Physiol.
2009
, vol. 
218
 (pg. 
436
-
443
)
[PubMed]
16
Maddika
S.
Kavela
S.
Rani
N.
Palicharla
V.R.
Pokorny
J.L.
Sarkaria
J.N.
Chen
J.
WWP2 is an E3 ubiquitin ligase for PTEN
Nat. Cell Biol.
2011
, vol. 
13
 (pg. 
728
-
733
)
[PubMed]
17
Raviprakash
N.
Manna
S.K.
Short-term exposure to oleandrin enhances responses to IL-8 by increasing cell surface IL-8 receptors
Br. J. Pharmacol.
2014
, vol. 
171
 (pg. 
3339
-
3351
)
[PubMed]
18
Manna
S.K.
Ramesh
G.T.
Interleukin-8 induces nuclear transcription factor-kappaB through TRAF6-dependent pathway
J. Biol. Chem.
2005
, vol. 
280
 (pg. 
7010
-
7021
)
[PubMed]
19
Manna
S.K.
Babajan
B.
Raghavendra
P.B.
Raviprakash
N.
Kumar
C.S.
Inhibiting TRAF2-mediated activation of NF-kappaB facilitates induction of AP-1
J. Biol. Chem.
2010
, vol. 
285
 (pg. 
11617
-
11627
)
[PubMed]
20
Ferron
F.
Rebowski
G.
Lee
S.H.
Dominguez
R.
Structural basis for the recruitment of profilin-actin complexes during filament elongation by Ena/VASP
EMBO J.
2007
, vol. 
26
 (pg. 
4597
-
4606
)
[PubMed]
21
Lee
J.O.
Yang
H.
Georgescu
M.M.
Di-Cristofano
A.
Maehama
T.
Shi
Y.
Dixon
J.E.
Pandolfi
P.
Pavletich
N.P.
Crystal structure of the PTEN tumour suppressor: implications for its phosphoinositide phosphatase activity and membrane association
Cell
1999
, vol. 
99
 (pg. 
323
-
334
)
[PubMed]
22
Pettersen
E.F.
Goddard
T.D.
Huang
C.C.
Couch
G.S.
Greenblatt
D.M.
Meng
E.C.
Ferrin
T.E.
UCSF Chimera-a visualisation system for exploratory research and analysis
J. Comput. Chem.
2014
, vol. 
25
 (pg. 
1605
-
1612
)
23
Sanner
M.F.
Python: a programming language for software integration and development
J. Mol. Graph. Mod.
1999
, vol. 
17
 (pg. 
57
-
61
)
[PubMed]
24
Morris
G.M.
Huey
R.
Lindstrom
W.
Sanner
M.F.
Belew
R.K.
Goodsell
D.S.
Olson
A.J.
Autodock4 and AutoDockTools4: automated docking with selective receptor flexibility
J. Comput. Chem.
2009
, vol. 
16
 (pg. 
2785
-
2791
)
25
Trott
O.
Olson
A.J.
AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimisation and multithreading
J. Comput. Chem.
2010
, vol. 
31
 (pg. 
455
-
461
)
[PubMed]
26
Manna
S.K.
Double-edged sword effect of biochanin to inhibit nuclear factor kappaB: suppression of serine/threonine and tyrosine kinases
Biochem. Pharmacol.
2012
, vol. 
83
 (pg. 
1383
-
1392
)
[PubMed]
27
Mukhopadhyay
A.
Manna
S.K.
Aggarwal
B.B.
Pervanadate-induced nuclear factor-kappaB activation requires tyrosine phosphorylation and degradation of IkappaBalpha. Comparison with tumour necrosis factor-alpha
J. Biol. Chem.
2000
, vol. 
275
 (pg. 
8549
-
8555
)
[PubMed]
28
Livolsi
A.
Busuttil
V.
Imbert
V.
Abraham
R.T.
Peyron
J.F.
Tyrosine phosphorylation-dependent activation of NF-kappa B. Requirement for p56 LCK and ZAP-70 protein tyrosine kinases
Eur. J. Biochem.
2001
, vol. 
268
 (pg. 
1508
-
1515
)
[PubMed]
29
Hinz
M.
Scheidereit
C.
The IκB kinase complex in NF-κB regulation and beyond
EMBO Rep.
2014
, vol. 
15
 (pg. 
46
-
61
)
[PubMed]
30
Karin
M.
Greten
F.R.
NF-kappaB: linking inflammation and immunity to cancer development and progression
Nat. Rev. Immunol.
2005
, vol. 
5
 (pg. 
749
-
759
)
[PubMed]
31
Manna
S.K.
Aggarwal
B.B.
Lipopolysaccharide inhibits TNF-induced apoptosis: role of nuclear factor-kappaB activation and reactive oxygen intermediates
J. Immunol.
1999
, vol. 
162
 (pg. 
1510
-
1518
)
[PubMed]
32
Sreenivasan
Y.
Sarkar
A.
Manna
S.K.
Mechanism of cytosine arabinoside-mediated apoptosis: role of Rel A (p65) dephosphorylation
Oncogene
2003
, vol. 
22
 (pg. 
4356
-
4369
)
[PubMed]
33
Bhat-Nakshatri
P.
Sweeney
C.J.
Nakshatri
H.
Identification of signal transduction pathways involved in constitutive NF-kappaB activation in breast cancer cells
Oncogene
2002
, vol. 
21
 (pg. 
2066
-
2078
)
[PubMed]
34
Izeradjene
K.
Douglas
L.
Delaney
A.
Houghton
J.A.
Influence of casein kinase II in tumour necrosis factor-related apoptosis-inducing ligand-induced apoptosis in human rhabdomyosarcoma cells
Clin. Cancer Res.
2005
, vol. 
10
 (pg. 
6650
-
6660
)
35
Zhong
H.
Voll
R.E.
Ghosh
S.
Phosphorylation of NF-kappa B p65 by PKA stimulates transcriptional activity by promoting a novel bivalent interaction with the coactivator CBP/p300
Mol. Cell
1998
, vol. 
1
 (pg. 
661
-
671
)
[PubMed]
36
Manna
S.K.
Gangadharan
C.
Decrease in RelA phosphorylation by inhibiting protein kinase A induces cell death in NF-kappaB-expressing and drug-resistant tumour cells
Mol. Immunol.
2009
, vol. 
46
 (pg. 
1340
-
1350
)
[PubMed]
37
Vermeulen
L.
De Wilde
G.
Van Damme
P.
Vanden Berghe
W.
Haegeman
G.
Transcriptional activation of the NF-kappaB p65 subunit by mitogen- and stress-activated protein kinase-1 (MSK1)
EMBO J.
2003
, vol. 
22
 (pg. 
1313
-
1324
)
[PubMed]
38
Gustin
J.A.
Maehama
T.
Dixon
J.E.
Donner
D.B.
The PTEN tumour suppressor protein inhibits tumour necrosis factor-induced nuclear factor κB activity
J. Biol. Chem.
2001
, vol. 
276
 (pg. 
27740
-
27744
)
[PubMed]
39
Wu
C.H.
Fallini
C.
Ticozzi
N.
Keagle
P.J.
Sapp
P.C.
Piotrowska
K.
Lowe
P.
Koppers
M.
McKenna-Yasek
D.
Baron
D.M.
, et al. 
Mutations in the Profilin 1 gene cause familial amyotrophic lateral sclerosis
Nature
2012
, vol. 
488
 (pg. 
499
-
450
)
[PubMed]

Supplementary data