Introduction
Integrins are a family of transmembrane (TM) signaling receptors that are found in many species ranging from sponges, nematodes, to mammals, including humans [
1]. Integrins comprise two distinct and non-covalently associated subunits, a and b, each containing a large extracellular domain, a single pass TM domain, and a generally small cytoplasmic tail. Integrin aIIbb3, also known as glycoprotein GPIIb/IIIa (CD41/CD61) complex, is a dominant integrin on the platelet surface and is critical for platelet functions [
2]. In resting platelets, integrin aIIbb3 stays in a low-affinity state for its ligand, soluble fibrinogen. A process known as inside-out signaling induces a conformational change in the extracellular domain of aIIbb3 when platelets are activated by various soluble agonists, such as epinephrine, adenosine diphosphate (ADP), thromboxane A2 (TXA2), and thrombin, or by immobilized agonists, such as von Willebrand factor (vWF) and collagen. This conformational change leads to the conversion of the integrin aIIbb3 receptor to a high-affinity state for binding soluble fibrinogen. A cascade of outside-in signaling occurs once the fibrinogen binds to the extracellular domain of integrin aIIbb3. The signals are transmitted from the extracellular domain of aIIbb3 to the cytoplasmic tail through the TM domain [
3]. The outside-in signaling not only mediates the stable adhesion, spreading, irreversible aggregation of platelets and subsequent thrombus growth, but also regulates the secretion of a- and dense granules, which in turn increases the affinity of the extracellular domain to its ligand [
4].
The interaction between the TM and cytoplasmic tail of integrin aIIb and that of b3 is important in maintaining a resting conformation of integrin aIIbb3. Previous studies suggested that the helix–helix interaction motif of the aIIbb3 TM domain is sufficient to prevent aIIbb3 activation [
5,
6]. The disruption of this interaction by either the binding of some cytoplasmic signaling proteins to the aIIb or b3 cytoplasmic tails, or mutations in aIIb or b3 TM domain and cytoplasmic tail triggers inside-out signaling [
3,
7]. Several cytoplasmic proteins are involved in the bidirectional signaling by directly or indirectly interacting with the aIIbb3 cytoplasmic tails, particularly the b3 tail [
8]. In platelets, talin and kindlin-3 bind to the evolutionary conserved NPLY and NITY motifs of the b3 cytoplasmic tails respectively, and play important roles in inside-out signaling [
9]. Meanwhile, c-Src [
10,
11] and Ga13 [
12,
13] are involved in outside-in signaling by interacting with the RGT sequence and the EEE motif of the b3 cytoplasmic tail, respectively. Xiang
et al. recently reported that the proximal domain of integrin b3, residues 716–730, is required for VPS33B binding and integrin aIIbb3 outside-in signaling [
14]. Several amino acid mutations in the b3 cytoplasmic tail also affect aIIbb3 bidirectional signaling. The phenylalanine substitution for the tyrosines in the b3 tail exhibits a defect in aIIbb3 outside-in signaling [
15]. The bidirectional signals in platelets bearing a b3 Y747A mutation are disrupted. Meanwhile, the outside-in signals in those bearing a T762A mutation are selectively disrupted [
16]. The b3 cytoplasmic tail is modified during bidirectional signaling. For example, T753, Y747, and Y759 are phosphorylated by protein kinases [
17]. The unphosphorylated b3 cytoplasmic tail is cleaved by calpain at the C-terminal side of T741, Y747, F754, and Y759 [
18]. Calpain regulates platelet signaling by suppressing the tyrosine dephosphorylation of phosphatase 1B [
19]. The cleavage is prevented by the phosphorylation of the b3 cytoplasmic tyrosines [
18]. The platelets from
calpain-1−/− mice show an enhanced spreading on collagen- and fibrinogen-coated surfaces [
20]. The results from the mouse models indicate that the aIIbb3-mediated outside-in signaling is completely abolished by the deletion of b3 R760GT762 corresponding to the most C-terminal calpain cleavage, while the inside-out signaling is only partially disrupted [
21]. However, the roles and detailed mechanisms of the calpain cleavage of the b3 cytoplasmic domain in bidirectional signaling need to be further elucidated.
A major difficulty in studying integrin aIIbb3 signaling is that the platelets are short-lived, anuclear cells that cannot be cultured in medium. The Chinese hamster ovary (CHO) cell line has been extensively used as a surrogate for platelets to circumvent these obstacles and explore the mechanism of integrin aIIbb3 signaling by expressing integrin aIIbb3 on its surface [
22]. Gu
et al. previously showed that the GPIb-IX-mediated integrin activation is reconstituted in 123 CHO (CHO cell line stably expressing GPIb-IX and integrin aIIbb3) cells, in which bidirectional signaling remained intact [
23]. The model cells expressing the b3 mutants with truncations at the T741, Y747, F754, and Y759 sites, instead of the wild-type b3, all exhibit an impaired outside-in signaling. The truncations at F741, Y747, or T754, but not Y759, abolish integrin aIIbb3 inside-out signaling [
24]. The calpain cleavage on the b3 cytoplasmic tail is thought to affect signaling. This effect may most likely result from the interruption of the interaction of the b3 tail with the cytoplasmic proteins, especially those of importance in signaling, such as talin, kindlins, and c-Src. However, the effect may also be attributed to the truncation-induced structural/conformational changes of the b3 integrin. To test this hypothesis, we employed a trans-dominant inhibition (dominant-negative) model to investigate the roles of the calpain cleavage of the b3 cytoplasmic domain in integrin signaling. In this model, the signals can be transduced through the wild-type aIIbb3, and the chimeric Tac-b3 tail can modulate the signal transduction only by sequestrating the cytoplasmic proteins. For this purpose, we chose to use the Tac-chimeras as a trans-dominant inhibition model, which has been well established in previous publications, to study the integrin signal transduction [
25–
27].
In the present study, we constructed a series of cDNAs encoding chimeric Tac-b3 proteins that consisted of Tac and wild-type b3 cytoplasmic tail or those with truncations at the C-terminal T741, Y747, F754, and Y759 or deletion of the NITY motif. The chimeric cDNAs were stably expressed in 123 CHO cells. These cells were used to test the integrin aIIbb3 signal transduction through interacting with soluble or immobilized fibrinogen. The application of this trans-dominant inhibition model will enable a comprehensive evaluation of the effect of the calpain cleavage of the b3 tail on signal transduction free of direct influence from the b3 conformational modifications.
Materials and methods
Antibodies and reagents
Mouse anti-GPIb (SZ-2), anti-b3 (SZ-21), and anti-aIIb (SZ-22) monoclonal antibodies were kindly provided by Ruan C. (Jiangsu Institute of Hematology, Suzhou, China). Anti-b3 cytoplasmic tail monoclonal antibody (C3a) was a gift from Kieffer N. (Sino-French Research Center for Life Sciences and Genomics, Shanghai, China). Anti-human interleukin-2 (IL-2) receptor a (anti-Tac, 7G7B6) antibody was purchased from R&D Systems. Purified human fibrinogen and Alexa Fluor 488-conjugated human fibrinogen were acquired from Enzyme Research Laboratories and Invitrogen, respectively. All other reagents were obtained from Sigma-Aldrich.
cDNA constructs
The cDNA encoding chimeric Tac-b3, consisting of extracellular and TM domains of Tac and integrin b3 cytoplasmic tail, into the CMV-IL2R vector was generated as previously described [
25]. The cDNA constructs encoding chimeric Tac-b3 truncation at 741, 747, 754, and 759 or deletion of NITY motif were created using standard molecular biological techniques to introduce stop codons into the integrin b3 cDNA at sites corresponding to the carboxyl side of amino acid residues 742, 748, 755, and 760 or to delete the integrin b3 cDNA sequence encoding NITY. All constructs were verified by DNA sequencing.
Cell lines and transfection
The 123 CHO cells were cultured in an F12 medium supplemented with 10% fetal bovine serum (FBS), glutamine, 1% non-essential amino acids, and 0.1% penicillin–streptomycin in a humidified atmosphere of 5% CO
2 at 37 °C as previously described by Xi
et al. [
24]. Each cDNA encoding chimeric Tac-b3 or Tac-b3 truncation mutants was co-transfected together with pcDNA3.1/Hygro(+) plasmid at a ratio of 3:1 into 123 CHO cells using Lipofectamine 2000. Stably transfected cells were selected using a hygromycin selection medium (0.4 mg/ml). The Tac chimeric protein positive cells were isolated and analyzed by flow cytometry using an anti-Tac monoclonal antibody, 7G7B6. These positive cell populations were enriched by cell sorting, and then subcloned by limiting dilution.
Cell adhesion assays
Cell adhesion experiments were performed as previously described [
28]. Wells of 96-well plates were briefly coated overnight with 20 mg/ml human fibrinogen in 0.1 mol/L NaHCO
3, pH 8.3 at 4 °C, then blocked for 2 h at room temperature with 2% bovine serum albumin in phosphate-buffered saline (PBS). The cells (1 × 10
5) were added into wells in triplicate and incubated at 37 °C for 90 min in the CO
2 incubator. The non-adherent cells were gently washed with PBS, while the adherent cells were observed under an inverted microscope (40× objective lens). In the quantitative assays, 50 ml of 0.3%
p-nitrophenyl phosphate (1 mg/ml dissolved in 0.1 mol/L sodium citrate, 0.1% Triton X-100, pH 5.4) was added into the 96 wells and incubated at 37 °C for 60 min. The reaction was then stopped with 1 mol/L NaOH. The optical density was measured at 405 nm.
Soluble fibrinogen binding
The transfected CHO cells were resuspended at 1 × 106 cells/100 µl in modified Tyrode’s solution with 15 µg/ml Alexa Fluor 488-conjugated fibrinogen. The suspension was incubated with or without purified human vWF (25 mg/ml) and ristocetin (1 mg/ml) at 22 °C for 30 min. The suspension was analyzed by flow cytometry. The non-specific fibrinogen binding was estimated in the presence of an integrin antagonist, RGDS (1 mmol/L).
Western blot
Preparation of cell lysates, sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE), and Western blot were all performed according to previously published procedures [
11,
29]. In brief, cultured cells were detached, washed with ice-cold PBS, and lysed at 4 °C for 30 min in ice-cold Nonidet P-40 lysis buffer (1% Nonidet P-40, 150 mmol/L NaCl, 50 mmol/L Tris, pH 7.2, 1 mmol/L EDTA, 1 mmol/L sodium vanadate, and 1 mmol/L phenylmethylsulfonyl fluoride) containing the protease inhibitor mixture. Protein concentrations were measured using the BCA protein assay kit. The cell lysates were centrifuged for 5 min at 4 °C at 12 000 r/min. Accordingly, 100 ml aliquots of each cell lysate supernatant contained equal amounts of protein (ranging from 400 to 500 mg between the experiments). Protein samples were denatured in an SDS loading buffer and loaded on 8% or 12% sodium dodecyl sulfate-polyacrylamide gels. Electrophoresis was performed, and the proteins were transferred onto the polyvinylidene fluoride membranes. The target proteins can be detected using specific primary and HRP-conjugated secondary antibodies as indicated. Immunoreactive bands were visualized by enhanced chemiluminescence with reaction times ranging from 30 s to 5 min.
Statistics
Differences between groups were assessed using the Student’s t-test. We used ANOVA (SPSS software 18) when three or more groups need to be compared. Data are presented as mean and standard deviation (SD).
Results
Establishment of the trans-dominant inhibition models in 123 CHO cells by expressing Tac chimeras with wild-type or mutant b3 tails
We generated cDNAs encoding Tac-b3 and five mutant proteins (i.e., Tac-b3D759, Tac-b3DNITY, Tac-b3D754, Tac-b3D747, and Tac-b3D741) (Fig. 1A). The truncations at the 759, 754, 747, and 741 sites were consistent with the previously identified calpain cleavage sites on b3 [
30]. In addition, a mutant with a deletion of the NITY motif was also established.
The expression of the Tac-b3 chimeras and the wild-type integrin b3 on the surface of 123 CHO cells was assessed by flow cytometry using 7G7B6 and SZ21 antibodies, respectively. Fig. 1B shows that similar expression levels of different constructs, also of integrin aIIb, b3, or GPIb-IX complex, were exhibited in the cell lines. However, the cytometric data did not enable us to determine the relative expression levels between the Tac-b3 chimeras and the wild-type integrin b3 in the same CHO cell line because of the possible difference in affinity of the 7G7B6 and SZ-21 antibodies. To address this issue, the cells were analyzed by Western blot using an anti-b3 cytoplasmic tail monoclonal antibody (C3a) that equally recognizes Tac-b3 and wild-type b3 because both possess an intact b3 cytoplasmic tail. Fig. 1C shows that the expression levels of chimeric Tac-b3 (bottom band) are significantly higher than those of wild-type b3 (upper band), thereby suggesting that the Tac-b3 chimeras were expressed at a higher level over the wild-type b3, which ensured the trans-dominant inhibition effect. These results, along with the similar flow cytometry data (Fig. 1B) and with respect to the expression of the chimeric Tac-b3 proteins, actually validated all these Tac-b3 chimeras in terms of trans-dominant effectiveness.
Trans-dominant model cells adhered to and spread on immobilized fibrinogen
An important role of integrin aIIbb3 outside-in signaling is mediating cell stable adhesion to immobilized multivalent ligands and spreading. We examined the ability of the trans-dominant inhibition model cells to adhere to and spread on immobilized fibrinogen. The 123 CHO cells with different Tac-b3 chimeras were allowed to adhere to and spread on the fibrinogen-coated coverslips for 2 h. The 123 CHO cells underwent complete adhesion and spreading. Representative microscope images (Fig. 2A), quantitative analysis of the adhesive phenotype (Fig. 2B), and surface coverage (Fig. 2C) of the cells plated on immobilized fibrinogen revealed that the Tac-b3, but not Tac-b3D759, Tac-b3D754, Tac-b3D747, or Tac-b3D741, caused a substantial reduction in spreading on immobilized fibrinogen. The spreading of the 123 CHO cells was partially inhibited by the expression of the chimeric Tac-b3DNITY proteins. Similar results were obtained from the disodium 4-nitrophenyl phosphate (PNPP) assays (Fig. 2D), which quantitatively represented the effect of the Tac-chimeras on cell adhesion.
Soluble fibrinogen binding to the 123 CHO cells expressing Tac-chimeras
aIIbb3 is known to be a specific receptor for soluble fibrinogen with a high-affinity state in response to inside-out signaling. Soluble fibrinogen binding to integrin aIIbb3 was assayed to investigate whether the Tac-b3 chimeras affect integrin aIIbb3 inside-out signaling. Binding of vWF to GPIb-IX in the presence of ristocetin induces integrin aIIbb3 inside-out signaling and results in the activation of the fibrinogen binding function of aIIbb3 in the CHO cell models [
18]. The flow cytometry results (Fig. 3A) and the quantitative fluorescence intensity (Fig. 3B) of soluble fibrinogen binding revealed that Tac-b3, Tac-b3D759, Tac-b3DNITY, and Tac-b3D754 affected the binding capacity of cells upon activation by ristocetin in the presence of vWF, whereas Tac-b3D747 or Tac-b3D741 did not.
Discussion
The bidirectional signaling through aIIbb3 is important for platelet activation resulting in the blood loss prevention at vascular injuries by forming occlusive clots. However, inappropriate activation leads to thrombosis, which is a principal trigger for heart attack and ischemic stroke. The integrin aIIbb3 bidirectional signaling is thought to be mediated by complex and dynamic associations between many cytoplasmic proteins (e.g., kinases, cytoskeletal proteins, and adaptor proteins) and the short cytoplasmic tails of integrin aIIbb3, particularly that of b3 [
8]. Fig. 4 demonstrates the relevant binding sites of the b3 tail with some cytoplasmic signaling proteins.
However, some debates have been raised as to the function of CHO cells as a surrogate for platelets because they do not contain any a-granules or dense granules as platelets do. Once expressing integrin aIIbb3, they adhere to and spread on immobilized fibrinogen, and bind soluble fibrinogen in response to stimuli. Therefore, the CHO cell model is extensively applied to investigate the function of recombinant integrin aIIbb3 when platelets are not amenable to genetic manipulations [
23,
24]. vWF binding to GPIb-IX expressed in the 123 CHO cells in the presence of ristocetin triggers integrin signaling. The truncation mutations of the b3 cytoplasmic tail at sites of T741, Y747, F754, and Y759 co-expressed with GPIb-IX and integrin aIIb in CHO cells show that inside-out and outside-in signals can be differentially regulated upon the calpain cleavage as demonstrated by our previous studies [
24]. In these studies, the signals are actually transduced through integrin aIIbb3 with the truncations and the possible contributions of the truncation-induced conformational modifications to the cellular functions cannot be ruled out even if the regulatory effects of the b3 truncations on integrin aIIbb3 signaling are attributed to the disruption of the interaction between b3 and cytoplasmic proteins.
We employed herein a trans-dominant inhibition model to address this issue, in which the structure/conformation of the b3 subunit of the integrin aIIbb3 is completely unaffected when it transduced signals. The flow cytometry (Fig. 1B) and Western blot (Fig. 1C) results demonstrate that the expression levels of the Tac-chimeric proteins were significantly higher than those of the wild-type b3 in the 123 CHO cells. The efficacy of this trans-dominant inhibition model has also been validated by a competitive enzyme-linked immunosorbent assay in our previous study [
31]. The Tac-chimeras cannot transmit biological signals [
32], but they can affect signaling through a competition for cytoplasmic signaling proteins with the wild-type b3 [
33]. Indeed, the expression of Tac-b3 chimeric protein in the 123 CHO cells inhibited stable adhesion, spreading and soluble fibrinogen binding, which indicates that this chimeric protein was able to regulate integrin aIIbb3 signaling by recruiting intracellular signaling proteins. We further found that the chimeric Tac-b3 and Tac-b3DNITY, but not Tac-b3D759, Tac-b3D754, Tac-b3D747, or Tac-b3D741, inhibited stable adhesion and cell spreading (Fig. 2). This finding clearly indicates that the RGT sequence was important for outside-in signaling because both Tac-b3 and Tac-b3DNITY proteins contained the RGT sequence. Accordingly, both proteins were able to compete with the wild-type b3 for c-Src binding. In cases where Tac-b3D759, Tac-b3D754, Tac-b3D747, and Tac-b3D741 all devoid of the RGT sequence are expressed, the loss of the c-Src binding site allowed the intracellular c-Src to bind to the wild-type b3 and mediate outside-in signaling. These results point to a mechanism in which the intracellular proteins interacting with the C-terminal sequence of b3 play important roles in outside-in signaling. Among them, c-Src may be pivotal, judging from the failure of Tac-b3D759 to inhibit stable adhesion and cell spreading as in this cell model. All the b3 binding intracellular proteins except for c-Src were supposed to be sequestrated by Tac-b3D759. In other words, outside-in signaling would be achieved once c-Src, even alone, associated with b3. These data are consistent with previous reports, in which the RGT sequence interacts with c-Src and mediates outside-in signaling [
10,
11,
34]. Notably, the expression of chimeric Tac-b3DNITY protein only partially inhibited stable adhesion and spreading of the 123 CHO cells (Fig. 2). This finding suggests that the outside-in signaling partially remained. Several intracellular proteins (e.g., CD98, ICAP1, Shc, b3 endonexin, kindlin, filamin, etc.) interact with the NITY motif of the b3 tail [
35]. These interactions may somehow be responsible for the outside-in signaling that partially remained upon the absence of c-Src binding. Furthermore, the deletion of NITY alone in the trans-dominant inhibition model directly led to a connection of T755 to the RGT sequence, which may affect the conformation of Tac-b3DNITY for c-Src binding and decrease the trans-dominant inhibition effects.
Interesting data were also acquired in soluble fibrinogen binding assays, where the expression of Tac-b3, Tac-b3D759, Tac-b3DNITY, and Tac-b3D754, but not Tac-b3D747 or Tac-b3D741, impaired the binding (Fig. 3). The present results suggest that the binding between the NITY motif of b3 and relevant cytoplasmic proteins is not sufficient for inside-out signaling based on the fact that the soluble fibrinogen binding was impaired in the 123/Tac-b3D754 and 123/Tac-b3DNITY cells, even though both lack the NITY motif and are, therefore, unable to recruit the NITY-binding proteins. The NITY and TST753 sequences were speculated to be critical for kindlin binding and kindlin-mediated inside-out signaling [
36,
37]. Therefore, both the 123/Tac-b3D754 and 123/Tac-b3DNITY cells were supposed to be capable of transducing inside-out signals. However, it seemed not to be the case in the observations on these cells. The data suggest that some other proteins binding to the N-terminal sequence to the 754 sites may participate in inside-out signaling. Moser
et al. showed that talin and kindlin interact with the conserved NPLY and NITY motifs of the b3 cytoplasmic tails, respectively, and play essential roles in inside-out signaling [
2]. The results of Tac-b3D754 and Tac-b3DNITY suggest that the maintenance of inside-out signaling requires a mutual regulation by talin and kindlin that was supported by the data from 123/Tac-b3D741 cells by showing an intact fibrinogen binding. However, the molecular details concerning how these two proteins coordinately work to regulate integrin activation remain unclear. Furthermore, our conclusion would be challenged by the observations with Tac-b3D747, which contains NPLY motif and is supposed to compete for talin binding, thereby inhibiting inside-out signaling. Unexpectedly, inside-out signaling occurs in both 123/Tac-b3D747 and 123/Tac-b3D741 cells. Therefore, we reason that the deletion of the C-terminal neighboring amino acids may alter the conformational structure of the NPLY motif for talin binding. However, the precise mechanisms need further elucidation with detailed protein‒protein interaction data.
The application of proteomic approaches would theoretically be able to address this issue and we have performed co-immunoprecipitation, 2D electrophoresis, and mass spectrometry assays. However, the mechanisms would not be exactly deciphered before the appropriate Chinese hamster protein databank becomes available. Further work is required to identify the intracellular signaling proteins responsible for the trans-dominant inhibition. The HEK293 cells, which are originally derived from the human embryonic kidney cells, may be a suitable cell model for this purpose [
38,
39]. In addition, innovative chimeric designs may be considered by fusing Tac with an N-terminally shorter b3 cytoplasmic domain to provide more precise data without possible interruption on intracellular proteins binding to the membrane-proximal sequence of the b3 cytoplasmic domain.
Calpain activation in platelets is involved in integrin aIIbb3 signaling and platelet activation [
40]. Calpain is activated upon the elevation of intracellular Ca
2+ once fibrinogen binds to integrin aIIbb3 [
20]. The b3 cytoplasmic tail has been established to be progressively cleaved by calpain from the C-terminal side of b3 in platelets. The sites have been identified as Y759, F754, T747, and Y741. The present work suggests that the integrin aIIbb3-mediated bidirectional signaling is regulated by the interaction of the b3 tail during platelet activation with the cytoplasmic proteins rather than the truncation-induced structural/conformational changes of the b3 integrin because the structure/conformation of the integrin aIIbb3, which is responsible for the signal transduction, was actually unaltered in the trans-dominant inhibition model. This concept is reinforced by the data showing that the T741, Y747, F754, and Y759 truncations or the NITY deletion were able to affect the signaling mediated by intact integrin aIIbb3. Our data further suggest that the interaction of the RGT sequence of the b3 tail with c-Src could play a critical role in regulating the integrin aIIbb3-mediated outside-in signaling. This regulation may be primarily achieved by the b3/c-Src association evidenced by the substantial spreading of the Tac-b3D759 cells, in which only c-Src is supposed to escape from the sequestration by the chimeric protein. Compared with outside-in signaling, inside-out signaling is regulated by a rather complex molecular mechanism. The b3 cytoplasmic tail needs to recruit talin, kindlin, and perhaps other cytoplasmic proteins for integrin aIIbb3 activation. The unaffected inside-out signaling may still enable the platelets to adhere almost normally at low shear rates after the b3 cytoplasmic tail is cleaved by calpain at the C-terminal side of Y759. However, the platelets scarcely adhered at high shear rates because of the disruption of the bidirectional signaling once Y754, Y747, or Y741 was cleaved [
10,
41]. With this trans-dominant inhibition model, we may better understand the mechanisms of integrin aIIbb3 bidirectional signaling regulated by calpain cleavage and their relevance to important physio- or pathological processes.
Higher Education Press and Springer-Verlag Berlin Heidelberg
PDF
(366KB)
