Cell adhesion molecule IGPR-1 activates AMPK connecting cell adhesion to autophagy

Autophagy plays critical roles in the maintenance of endothelial cells in response to cellular stress caused by blood flow. There is growing evidence that both cell adhesion and cell detachment can modulate autophagy, but the mechanisms responsible for this regulation remain unclear. Immunoglobulin and proline-rich receptor-1 (IGPR-1) is a cell adhesion molecule that regulates angiogenesis and endothelial barrier function. In this study, using various biochemical and cellular assays, we demonstrate that IGPR-1 is activated by autophagy-inducing stimuli, such as amino acid starvation, nutrient deprivation, rapamycin, and lipopolysaccharide. Manipulating the IκB kinase β activity coupled with in vivo and in vitro kinase assays demonstrated that IκB kinase β is a key serine/threonine kinase activated by autophagy stimuli and that it catalyzes phosphorylation of IGPR-1 at Ser220. The subsequent activation of IGPR-1, in turn, stimulates phosphorylation of AMP-activated protein kinase, which leads to phosphorylation of the major pro-autophagy proteins ULK1 and Beclin-1 (BECN1), increased LC3-II levels, and accumulation of LC3 punctum. Thus, our data demonstrate that IGPR-1 is activated by autophagy-inducing stimuli and in response regulates autophagy, connecting cell adhesion to autophagy. These findings may have important significance for autophagy-driven pathologies such cardiovascular diseases and cancer and suggest that IGPR-1 may serve as a promising therapeutic target.

regulator of autophagy in response to nutrient availability. In the presence of amino acids, mTORC1 is activated and suppresses autophagy through phosphorylation of ULK1 and ATG13. However, upon nutrient deprivation, mTORC1 activity is inhibited, leading to the activation of ULK1 that induces the autophagy program (8,9). Suppression of mTORC1 activity by AMP-activated protein kinase (AMPK) is central to the regulation of autophagy. AMPK inactivates mTORC1 through phosphorylation of RAPTOR, a key protein present within the mTORC1 complex, and more importantly directly phosphorylates ULK1 at multiple serine residues and activates it (10,11).
Commonly known autophagy-inducing conditions or agents such as nutrient withdrawal including amino acid and serum starvation, immunosuppressant rapamycin/Sirolimus, and LPS all activate several key kinases such as the IkB kinase (IKK) complex (1). The IKK complex is composed of at least three proteins, including two catalytic subunits (IKKa and IKKb) and the scaffold protein NF-kB essential modulator (NEMO; also called IKKg) (12). In addition to its pivotal role in mediating phosphorylation of IkB (12), IKKb can regulate autophagy in IkB-independent manner (13,14) by mechanisms that are not fully understood.
IGPR-1 was identified as a novel cell adhesion molecule expressed in various human cell types, including endothelial and epithelial cells, and it mediates cell-cell adhesion (15). IGPR-1 regulates angiogenesis and endothelial barrier function (15,16), decreases sensitivity of tumor cells to the genotoxic agent doxorubicin, and supports tumor cell survival in response to anoikis (17). IGPR-1 is localized to adherens junctions and is activated through transhomophilic dimerization (16). Additionally, IGPR-1 responds to various cellular stresses, because its phosphorylation (i.e. Ser 220 ) is significantly increased by flow shear stress (18) and exposure to doxorubicin (17,18). Curiously, both shear stress (19) and doxorubicin (20,21) are well-known potent inducers of autophagy, raising a possibility for the involvement of IGPR-1 in autophagy. In this study, we demonstrate that upon the induction of autophagy, IGPR-1 is phosphorylated at Ser 220 via a mechanism that involves activation of IKKb. IKKbdependent phosphorylation of IGPR-1 stimulates phosphorylation of AMPK, leading to activation of BECN1 and ULK1, connecting cell adhesion and energy sensing to autophagy. as doxorubicin (17) and flow shear stress (18) also simulate phosphorylation of IGPR-1at Ser 220 . Because both shear stress and genotoxic agents are well-known for their roles in autophagy, we asked whether IGPR-1 is activated in response to autophagy. We used human embryonic kidney epithelial-293 (HEK-293) cells ectopically expressing IGPR-1 as a model system to study the role of IGPR-1 in autophagy, because these cells do not express IGPR-1 endogenously at a detectable level (16). To this end, we first, tested whether amino acid starvation, the best-known inducer of autophagy, can stimulate phosphorylation of IGPR-1. The cells were lysed, and whole cell lysates were subjected to Western blotting analysis followed by immunoblotting with pSer 220 and total IGPR-1 antibodies. Phosphorylation of IGPR-1 at Ser 220 was significantly increased by brief amino acid starvation of HEK-293 cells. The increase in phosphorylation of Ser 220 peaked after 1 min with amino acid starvation and remained highly phosphorylated until 15 min (Fig. 1A).
Furthermore, in an additional set of experiments we subjected HEK-293 cells expressing IGPR-1 to various other autophagy-inducing conditions or factors such as serum-starvation, rapamycin, and LPS treatments and measured phosphorylation of Ser 220 . Both rapamycin and LPS are known to induce autophagy (1,22,23). Phosphorylation of IGPR-1 at Ser 220 was significantly increased by brief serum starvation of HEK-293 cells (Fig. 1B). Furthermore, both rapamycin and LPS treat-ments of HEK-293 cells stimulated phosphorylation of IGPR-1 at Ser 220 (Fig. 1, C and D). To demonstrate whether IGPR-1 is phosphorylated by autophagy in biologically relevant human endothelial cells in which IGPR-1 is expressed endogenously, we subjected primary human microvascular endothelial cells (HMVECs) to amino acid starvation. The result showed that IGPR-1 is phosphorylated at Ser 220 in HMVECs (Fig. 1E). Taken together, the data demonstrate that IGPR-1 is activated by autophagy in HEK-293 cells and human primary endothelial cells.

IKKb is activated by autophagy and phosphorylates IGPR-1
Activation of serine/threonine kinases represents a salient mechanistic feature of autophagy. Particularly, activation of IKKb by nutrient deprivation (24), LPS (25,26), and rapamycin (27) is known to play a central role in autophagy. Therefore, we asked whether activation of IKKb by serum starvation, LPS, or rapamycin can mediate phosphorylation of IGPR-1 at Ser 220 . To this end, we overexpressed WT IKKb or kinase inactive IKKb (IKKb-A44) in IGPR-1/HEK-293 cells. After 48 h of transfection, the cells were either kept in 10% FBS or serumstarved for 15, 30, or 60 min. The cells were lysed, and whole cell lysates were immunoblotted for pSer 220 , total IGPR-1, and phospho-IKKb. Expression of WT IKKb in IGPR-1/HEK-293 cells resulted in a robust phosphorylation of IGPR-1 in the Figure 1. IGPR-1 is activated by autophagy stimuli. A, IGPR-1/HEK-293 cells were kept in 10% FBS or in amino acid-free medium for indicated times. The whole cell lysates were prepared and subjected to Western blotting analysis and blotted with anti-phosphoserine 220 IGPR-1 antibody (pSer 220 ), anti-IGPR-1 antibody, and anti-GAPDH antibody for protein loading control. The graph is representative of three independent experiments. Individual data sets are shown (). p , 0.05. B, IGPR-1/HEK-293 cells were seeded for 48 h in 10% FBS followed by serum starvation for 15 and 30 min by replacing the 10% DMEM with serum-free DMEM or cells left in 10% FBS DMEM as a control (0). The cells were lysed, and the whole cell lysates were blotted with the same antibodies as A. The graph is representative of three independent experiments. Individual data sets are shown (). p , 0.05. C, IGPR-1/HEK-293 cells were treated with vehicle (0) or with rapamycin (10 nM) for 1 h. The whole cell lysates were prepared and subjected to Western blotting analysis using the same antibodies as in A. The graph is representative of three independent experiments. Individual data sets are shown (). p = 0.021. D, IGPR-1/HEK-293 cells were treated with different concentrations of LPS for 1 h as indicated. The whole cell lysates were prepared and subjected to Western blotting analysis and blotted with using the same antibodies as in A. The graph is representative of three independent experiments. Individual data sets are shown (). p , 0.05. E, HMVECs were kept in 10% FBS or in amino acid-free medium for indicated times. The whole cell lysates were prepared, subjected to Western blotting analysis, and blotted using the same antibodies as in A. Individual data sets are shown (). P , 0.05.

IGPR-1 is a substrate for IKKb and is phosphorylated by IKKb in vitro and in vivo
Ser 220 and the surrounding amino acids in IGPR-1 are strongly conserved both in human and nonhuman primates (Fig. 3A), suggesting an evolutionarily conserved mechanism for the phosphorylation of Ser 220 . IKKb phosphorylates peptides with aromatic residues at the 22 position, hydrophobic residues at the 11 position, and acidic residues at the 13 posi-tion (29), suggesting that IKKb is a likely candidate kinase involved in the phosphorylation of IGPR-1 at Ser 220 (Fig. 3B). Therefore, we asked whether IKKb can phosphorylate IGPR-1 at Ser 220 independent of the autophagy-inducing factors like serum starvation or LPS and rapamycin. We overexpressed WT IKKb or kinase inactive IKKb-A44 in IGPR-1/HEK-293, and 48 h after transfection, the cells were lysed, and phosphorylation of IGPR-1 was determined by Western blotting analysis. Overexpression of WT IKKb increased phosphorylation of IGPR-1, whereas kinase inactive IKKb-A44 inhibited phosphorylation of IGPR-1 at Ser 220 (Fig. 3C). Similarly, IKKb inhibitor inhibited both phosphorylation of IGPR-1 and IKKb (Fig. 3D).
In an additional approach, we knocked out IKKb via CRISPR-Cas9 system and examined the effect of loss of IKKb in IGPR-1 phosphorylation. To this end, cells were either treated with a control vehicle or AMPK activator, oligomycin and cells were lysed, and phosphorylation of IGPR-1 at Ser 220 was determined. Stimulation of IGPR-1/Ctr.sgRNA/HEK-293 cells with oligomycin stimulated AMPK activation and phosphorylation of IGPR-1 at Ser 220 (Fig. 3E). However, in IGPR-1/ IKKb.sgRNA/HEK-293 cells, phosphorylation of Ser 220 was not detected, and treatment with oligomycin also did not stimulate phosphorylation of IGPR-1 at Ser 220 (Fig. 3E). IKKb. sgRNA-mediated knockout of IKKb is shown (Fig. 3E). The cells were lysed, and the whole cell lysates were immunoblotted for pSer 220 , total IGPR-1, phospho-IKKb, or total IKKb or blotted for GAPDH for protein loading control. The graph is representative of three independent experiments. Individual data sets are shown (). p , 0.05. B, IGPR-1/HEK-293 cells were treated with LPS (5 mg/ml) alone or cotreated with LPS and IKK inhibitor III/BMS-345541 (100 hM) for 60 min. The cells were lysed, and the whole cell lysates were subjected to Western blotting analysis and probed for pSer 220 , total IGPR-1, phospho-IKKb, total IKKb, and GAPDH for protein loading control. The graph is representative of three independent experiments. Individual data sets are shown (). C, IGPR-1/HEK-293 cells were transfected with EV (2) or constitutively active IKKb (IKKb-S177E/S181E). After 48 h the cells were either left untreated (2) or treated with LPS (5 mg/ml) for 60 min (1). The whole cell lysates were subjected to Western blotting analysis and probed for pSer 220 , total IGPR-1, total IKKb, and GAPDH for protein loading control. The graph is representative of three independent experiments. Individual data sets are shown (). p = 0.033.

IGPR-1 is activated by and regulates autophagy
We next asked whether IKKb can directly phosphorylates IGPR-1 at Ser 220 . To examine the direct involvement of IKKb in catalyzing the phosphorylation of IGPR-1, we carried out an in vitro kinase assay using a purified recombinant GST-IGPR-1 protein that only encompasses the cytoplasmic domain of IGPR-1 and demonstrated that IKKb phosphorylates IGPR-1 at Ser 220 (Fig. 3F). Taken together, our data identify IGPR-1 as a novel substrate of IKKb. Furthermore, we carried additional experiments to demonstrate the selectively of IGPR-1 phosphorylation at Ser 220 by IKKb. To this end, we immunoprecipitated IGPR-1 proteins from HEK-293 cells expressing either WT IGPR-1, A220-IGPR-1, or D220-IGPR-1. The immunoprecipitated proteins were subjected to calf intestinal alkaline phosphatase treatment, which removes phosphorylation. The removal of phosphorylation on IGPR-1 was confirmed by Western blotting analysis using pSer 220 specific antibody (Fig. S1A). The dephosphorylated proteins were subjected to an in vitro kinase assay using a recombinant IKKb. The result showed that IKKb selectively phosphorylates IGPR-1 at Ser 220 (Fig. S1B).
We next asked whether IGPR-1 induced AMPK activation also stimulates phosphorylation of BECN1 and ULK1, which are the best-known substrates of AMPK and play central roles in autophagy (32). Similar to AMPK activation, phosphorylation of BECN1 at Ser 93 was significantly increased in IGPR-1/HEK-293 cells compared with EV/HEK-293 or A220-IGPR-1 cells (Fig. 4A), indicating that IGPR-1-induced activation of AMPK in HEK-293 cells also induced phosphorylation of BECN1 (Fig.  4A) as previously reported (33). Furthermore, phosphorylation of ULK1 at Ser 555 was also increased in IGPR-1/HEK-293 cells compared with control EV/HEK-293 cells. Interestingly, ULK1 phosphorylation was significantly reduced in A220-IGPR-1 cells, indicating that phosphorylation of Ser 220 , in part, is required for phosphorylation of ULK-1 (Fig. 4A). The data indicate that phosphorylation of Ser 220 on IGPR-1 plays an important role in the phosphorylation of AMPK, BECN1, and ULK1.

IGPR-1 induces autophagy in HEK-293 cells
We next examined the role of IGPR-1 in the autophagosome formation by measuring expression of LC3-phosphatidylethanolamine conjugate (LC3-II) and p62 endogenously expressed in HEK-293 cells, which are required for autophagosome development during autophagy (34), by Western blotting analysis. HEK-293 cells expressing EV or IGPR-1 were either kept at 10% FBS or serum-starved for overnight. The cells were lysed, and expression of LC3 and p62 levels was determined by immunoblotting with LC3 and p62 antibodies. Although expression of LC3II was increased in response to serum starvation in EV/ HEK-293 cells, expression of LC3II was significantly higher both in 10% FBS and in serum-starved conditions in IGPR-1/ HEK-293 cells (Fig. 5A), indicating that IGPR-1 through increased in expression of LC3II regulates autophagosome formation. Additionally, as expected, p62 level was markedly decreased in response to serum starvation (Fig. 5A).

Discussion
Previous studies have shown that the activation of IKKb regulates autophagy through mechanisms that involve expression IGPR-1 is activated by and regulates autophagy of pro-autophagic genes via the NF-kB-independent pathway and phosphorylation of the p85 subunit of PI3K, which leads to inhibition of mTOR (13,24). We revealed the existence of a previously unidentified pathway in autophagy that involves IKKb-dependent activation of IGPR-1. We provide a mechanistic link between activation of IKKb and phosphorylation of IGPR-1 at Ser 220 . IKKb is activated by autophagy, leading to phosphorylation of IGPR-1 at Ser 220 both in vivo and in vitro. Mechanistically, IKKb-mediated phosphorylation of IGPR-1 at Ser 220 leads to activation of AMPK, which plays a central role in autophagy. Activation of IKKb plays an essential role in autophagy because both the loss of function of IKKb in mice and cell culture blocked autophagy (13). Likewise, IKKb null cells are deficient in their ability to undergo autophagy in response to cellular starvation (14), which further underscores the critical role of IKKb in autophagy.
AMPK is believed to exert its effect in autophagy by multiple mechanisms including, inactivating mTORC1 through phosphorylation of RAPTOR, a key protein present within the mTORC1, phosphorylating ULK1 at multiple serine residues, including Ser 555 , that lead to its activation (10,11) and more importantly phosphorylating BECN1 at Ser 93 and Ser 96 (33). IGPR-1-mediated activation of AMPK in HEK-293 cells increased phosphorylation of ULK1 at Ser555 and BECN1 at Ser93. Activation of AMPK and phosphorylation of BECN1 requires phosphorylation of IGPR-1 at Ser 220 . Activation of ULK1 and phosphorylation of BECN1 both play central roles in autophagy. ULK1 phosphorylation enables ULK1 to form a complex with ATG13 and FIP200 (focal adhesion kinase family interacting protein of 200 kDa) that leads to its translocation to autophagy initiation sites and subsequent recruitment of the class III phosphatidylinositol 3-kinase, VPS34 (vacuolar protein sorting 34) complex consisting of BECN1 and multiple other autophagy-related proteins leading to the phagophore formation (6). Phosphorylation of BECN1 plays a key role in the initial steps in the assembly of autophagosomes from preautophagic structures, which is the recruitment and activation of VPS34 complex (32).
IGPR-1 is a cell adhesion molecule that mediates cell-cell adhesion, and its activation regulates cell morphology and actin stress fiber alignment (15,16). The finding that IGPR-1 is activated by and regulates autophagy by stimulating activation of AMPK not only suggests a significant role for IGPR-1 in autophagy but also links cell-cell adhesion to energy sensing and autophagy. Recent studies illuminated the key roles of autophagy in endothelial cells in response to various metabolic, blood flow-induced stresses and angiogenesis (37), the same cellular events are also regulated by IGPR-1 (15,16,18).
Additionally, cellular stress induced by flow shear stress (19,38) and exposure of cells to the chemotherapeutic agent doxorubicin (20,21) are linked to induction of autophagy, the conditions in which IGPR-1 also is activated (17, 18). Moreover, Figure 5. IGPR-1 mediates serum starvation-induced autophagy. A, HEK-293 cells expressing EV, IGPR-1 were either kept in 10% FBS or serum-starved for 12 h. The cells were lysed and immunoblotted for LC3 and p26, total IGPR-1, and GAPDH. The graph is representative of three experiments. B and C, HEK-293 cells expressing GFP-LC3-RFP/HEK-293 alone or coexpressing GFP-LC3-RFP and IGPR-1 either were kept in 10% FBS DMEM (B) or in serum-free DMEM for 6 h (C). The cells were fixed and stained with 49,6-diamino-2-phenylindole (DAPI, nucleus) and viewed under a fluorescence microscope, and pictures were taken. Scale bar, 10 mM. D, the graph is representative of three independent experiments. ImageJ was used to quantify images (n = 6-10 images for group). Individual data sets are shown (). p , 0.05. E, expression and phosphorylation of IGPR-1, A220-IGPR-1, or D220-IGPR-1 in HEK-293 cells. F, HEK-293 cells coexpressing GFP-LC3-RFP with IGPR-1, A220-IGPR-1, or D220-IGPR-1 were kept in serum-free DMEM for 6 h. The cells were fixed and stained with as in A and viewed under a fluorescence microscope, and representative pictures were taken. Scale bar, 10 mM. G, ImageJ was used to quantify images (n = 6-10 images for group). Individual data sets are shown (). p , 0.05. autophagy is associated with therapeutic resistance to chemotherapeutic agents (e.g. cisplatin, doxorubicin, temozolomide, and etoposide), metabolic stresses, and small molecule inhibitors, suggesting a pro-tumor function for autophagy (39,40). Curiously, IGPR-1 is strongly phosphorylated by doxorubicin and regulates sensitivity of tumor cells to doxorubicin (17), indicating that IGPR-1 through induction of autophagy program could contribute to the development of resistance in cancer cells.
Taken together, the data presented here suggest a significant role for IGPR-1 in autophagy and autophagy-associated diseases such as cancer and cardiovascular diseases. We propose IGPR-1 as a pro-autophagy cell adhesion molecule that upon activation stimulates AMPK activation, leading to phosphorylation of BECN1 and ULK1, key proteins involved in autophagy (Fig. 6), linking cell adhesion to autophagy, a finding that has important significance for autophagy-driven pathologies such cardiovascular diseases and cancer.
Cell culture assays HEK-293 cells expressing EV, IGPR-1 or A220-IGPR-1 were maintained in DMEM supplemented with 10% fetal bovine serum and penicillin/streptomycin. To measure phosphorylation of IGPR-1 in response to serum starvation, the cells were plated in 60-cm plates with 10% FBS DMEM overnight with approximate 80-90% confluency. The cells were washed twice with PBS, and the cells were starved for 15 and 30 min or as described in the figure legends. The cells were lysed, and the whole cell lysates were mixed with sample buffer (53) and boiled for 5 min. The whole cell lysates were subjected to Western blotting analysis and immunoblotted with antibody of the interest as described in the figure legends. In some experiments, the cells were treated with a specific chemical inhibitor or transfected with a particular construct as indicated in the figure legends. Human microvascular endothelial cells were purchased from Cell Applications, Inc. (San Diego, CA, USA) and were grown in endothelial cell medium.

Recombinant GST-fusion protein production
The generation of GST-fusion cytoplasmic domain of IGPR-1 cloned into pGEX-2T vector as previously described (15). The purified GST-fusion IGPR-1 protein was subsequently used to measure the ability of IKKb to phosphorylate IGPR-1 in an in vitro kinase.

In vitro kinase assay
To detect phosphorylation of IGPR-1 at Ser 220 , the purified recombinant GST-IGPR-1 encompassing the cytoplasmic domain of IGPR-1 was mixed with WT or kinase inactive IKKb expressed in HEK-293 cells in 13 kinase buffer plus 0.2 mM ATP and incubated at 30°C for 15 min. The samples were mixed in 23 sample buffer and after boiling at 95°C for 5 min were resolved on 12% SDS-PAGE followed by Western blotting analysis using anti-pSer 220 antibody.

Western blotting analysis
The cells were prepared as described in the figure legends and lysed, and the whole cell lysates were subjected to Western blotting analysis. Normalized whole cell lysates were subjected Figure 6. Proposed role of IGPR-1 in autophagy. Upon activation by autophagy stimuli, IGPR-1 is phosphorylated at Ser 220 via IKKb. IGPR-1 acts as a pro-autophagy signaling receptor leading to activation of AMPK. AMPK catalyzes phosphorylation of BECN1 and ULK1, key proteins involved in autophagy. IGPR-1 is a dimeric protein and undergoes homophilic transdimerization in a cell density-dependent manner (17) (not shown). Activation of IKKb was previously thought to regulate autophagy by activation of AMPK, PI3K, and induction of NF-kB.

IGPR-1 is activated by and regulates autophagy
to Western blotting analysis using IGPR-1 antibody, pSer 220 antibodies, or an appropriate antibody as indicated in the figure legends. The proteins were visualized using streptavidin-horseradish peroxidase-conjugated secondary antibody via chemiluminescence system. For each blot, the films were exposed multiple times, and the films that showed within the linear range detection of protein bands were selected, scanned, and subsequently used for quantification. Blots from at least three independent experiments were used for quantification purposes, and representative data are shown. ImageJ software, an open source image-processing program, was used to quantify the blots.

Immunofluorescence microscopy
The cells expressing IGPR-1 or other constructs were seeded (1.5 3 10 6 cells) onto coverslips and grown overnight in 60-mm plates to 90-100% confluence. The coverslips were mounted in Vectashield mounting medium with 49,6-diamino-2-phenylindole onto glass microscope slides. The slides were examined using a fluorescence microscope.

Statistical analyses
The experimental data were subjected to Student's t test or one-way analysis of variance analysis where appropriate with representative of at least three independent experiments. p , 0.05 was considered significant or as indicated in the figure legends.

Data availability
The data and reagents are available from the corresponding author upon request. Funding and additional information-This work was supported in part by NCI, National Institutes of Health Grants R21CA191970 and R21CA193958 and Clinical and Translational Science Institute Grant 1UL1TR001430 (to N. R.). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Conflict of interest-The authors declare that they have no conflicts of interest with the contents of this article.