Characterization of integrin-tetraspanin adhesion complexes: role of tetraspanins in integrin signaling

Abstract

Tetraspanins (or proteins from the transmembrane 4 superfamily, TM4SF) form membrane complexes with integrin receptors and are implicated in integrin-mediated cell migration. Here we characterized cellular localization, structural composition, and signaling properties of alpha3beta1-TM4SF adhesion complexes. Double-immunofluorescence staining showed that various TM4SF proteins, including CD9, CD63, CD81, CD82, and CD151 are colocalized within dot-like structures that are particularly abundant at the cell periphery. Differential extraction in conjunction with chemical cross-linking indicated that the cell surface fraction of alpha3beta1-TM4SF protein complexes may not be directly linked to the cytoskeleton. However, in cells treated with cytochalasin B alpha3beta1-TM4SF protein complexes are relocated into intracellular vesicles suggesting that actin cytoskeleton plays an important role in the distribution of tetraspanins into adhesion structures. Talin and MARCKS are partially codistributed with TM4SF proteins, whereas vinculin is not detected within the tetraspanin-containing adhesion structures. Attachment of serum-starved cells to the immobilized anti-TM4SF mAbs induced dephosphorylation of focal adhesion kinase (FAK). On the other hand, clustering of tetraspanins in cells attached to collagen enhanced tyrosine phosphorylation of FAK. Furthermore, ectopic expression of CD9 in fibrosarcoma cells affected adhesion-induced tyrosine phosphorylation of FAK, that correlated with the reorganization of the cortical actin cytoskeleton. These results show that tetraspanins can modulate integrin signaling, and point to a mechanism by which TM4SF proteins regulate cell motility.

Figure 1

Distribution of TM4SF proteins in cells plated on laminin-5–containing ECM. MDA-MB-231 cells were plated in serum-free DMEM on glass coverslips coated with laminin-5–containing ECM. After 1 h cells were fixed with 2% paraformaldehyde, and permeabilized with 1% Brij 98. Indirect immunofluorescence staining was carried out using mAb C9-BB to CD9 (A–D), M38 to CD81 (E–H), M104 to CD82 (I–L), and 5C11 to CD151 (M–P). Staining was visualized using FITC-conjugated goat anti–mouse IgG. Serial images were captured by CCD camera at various focal planes: A, E, I, M, at the plane of cell–ECM contacts; B, F, J, N, 1 μM above the plane of the attachment; C, G, K, O, 2 μM above the plane of the attachment; D, H, L, P, 3.5 μM above the plane of the attachment. Captured images were processed using the OpenLab software package and the deconvolution module. Bar, 10 μM.

Figure 2

Distribution of tetraspanins in cells plated on collagen. MDA-MB-231 cells were plated in serum-free DMEM on glass coverslips coated with collagen (A–D) or laminin-5–containing ECM (E–H). After 1 h cells were fixed with 2% paraformaldehyde, and either permeabilized with 1% Brij 98 (A–D) or left untreated (E–H). Indirect immunofluorescence staining was carried out using mAb C9-BB to CD9 (A and E), M38 to CD81 (B and F), M104 to CD82 (C and G), and 5C11 to CD151 (D and H). Staining was visualized using FITC-conjugated goat anti–mouse IgG. Images A–D focused on cell–ECM contacts were captured and processed by using the OpenLab software package. Images E–H were photographed using Nikon photocamera attached to the microscope. Bar, 5 μM.

Figure 3

Colocalization of TM4SF proteins and α3β1 integrin within dot-like adhesion complexes. MDA-MB-231 cells were prepared for double-immunofluorescence staining as described in the legend to Fig. 1. Codistribution of TM4SF proteins was studied by using mAbs C33/anti-CD82 and 5C11/anti-CD151 (A and B, respectively), mAbs 6H1/anti-CD63 and BU16/anti-CD9 (D and E, respectively), and mAbs JS64/anti-CD81 and 5C11/anti-CD151 (G and H, respectively). Localization of α3β1 integrin within TM4SF protein–containing adhesion complexes was studied by using mixtures of anti-α3 (A3-X8/A3-IVA5) and anti-TM4SF (JS64/C33) mAbs (J and K, respectively). Staining was visualized using Rhodamine-conjugated goat anti–mouse IgG1 and FITC-conjugated goat anti–mouse IgG2a antibodies. Images focused on cell–ECM contacts were captured, processed and superimposed (C, F, I, and L) by using the OpenLab software package. Bar, 5 μM.

Figure 4

Localization of α3β1 and CD81 in the peripheral adhesion complexes by immunoelectron microscopy. MDA-MB-231 cells were fixed with 2% paraformaldehyde and stained with mAb P1B5 to α3 integrin subunit (A) and JS64 to CD81 (B). Staining was visualized by using goat anti–mouse IgG Ab coupled to 10-nm gold particles. Bar, 0.2 μM.

Figure 5

Specific association of tetraspanins with α3β1 integrin in MDA-MB-231 cells, and a role of α3β1/tetraspanin protein complexes in cell adhesion. (A) MDA-MB-231 cells were lysed in the buffer containing 1% Brij 96, and protein complexes were immunoprecipitated from the cellular lysate using specific anti-integrin mAbs (A3-X8 to α3β1, A2-7C6 to α2β1, and A5-PUJ2 to α5β1). Immunoprecipitated proteins were separated in 10% PAGE, and transferred to a nitrocellulose membrane. The membrane was probed with anti-tetraspanin mAbs (C9-BB to CD9, JS64 to CD81, C33 to CD82, and 11G1.B4 to CD151), or polyclonal sera to β1 integrin subunit. The mAb P3 was used as a negative control. (B) BCECF-AM–labeled MDA-MB-231 cells were preincubated with anti-integrin (15 μg/ml) or a mixture of anti-TM4SF mAbs (used at 10 μg/ml each) and tested for adhesion to 96-well microtiter plate coated with collagen type I (10 μg/ml) or laminin-5–containing ECM as described in Materials and Methods. The mAbs used were A3-IVA5 to α3; A2-IIE10 to α2; G0H3 to α6; C9-BB, anti-CD9; 6H1, anti-CD63; M38 and JS64, anti-CD81; M104, anti-CD82, 5C11, anti-CD151. The results are presented as a ratio of relative fluorescence values obtained for cells attached in the presence of mAbs to the value of untreated cells.

Figure 5

Specific association of tetraspanins with α3β1 integrin in MDA-MB-231 cells, and a role of α3β1/tetraspanin protein complexes in cell adhesion. (A) MDA-MB-231 cells were lysed in the buffer containing 1% Brij 96, and protein complexes were immunoprecipitated from the cellular lysate using specific anti-integrin mAbs (A3-X8 to α3β1, A2-7C6 to α2β1, and A5-PUJ2 to α5β1). Immunoprecipitated proteins were separated in 10% PAGE, and transferred to a nitrocellulose membrane. The membrane was probed with anti-tetraspanin mAbs (C9-BB to CD9, JS64 to CD81, C33 to CD82, and 11G1.B4 to CD151), or polyclonal sera to β1 integrin subunit. The mAb P3 was used as a negative control. (B) BCECF-AM–labeled MDA-MB-231 cells were preincubated with anti-integrin (15 μg/ml) or a mixture of anti-TM4SF mAbs (used at 10 μg/ml each) and tested for adhesion to 96-well microtiter plate coated with collagen type I (10 μg/ml) or laminin-5–containing ECM as described in Materials and Methods. The mAbs used were A3-IVA5 to α3; A2-IIE10 to α2; G0H3 to α6; C9-BB, anti-CD9; 6H1, anti-CD63; M38 and JS64, anti-CD81; M104, anti-CD82, 5C11, anti-CD151. The results are presented as a ratio of relative fluorescence values obtained for cells attached in the presence of mAbs to the value of untreated cells.

Figure 6

Colocalization of tetraspanins with vinculin and talin. MDA-MB-231 cells were plated in serum-free DMEM (A–C and G–I) or in DMEM supplemented with 10% FCS (D–F) on glass coverslips coated with laminin-5–containing ECM. After 1 h cells were fixed with 2% paraformaldehyde and permeabilized with 1% Brij 98. Codistribution of TM4SF proteins with vinculin (A–C and D–F) was studied in double-immunofluorescence staining experiments by using mAb to vinculin, hVIN-1 (B and E), and a mixture of mAbs to TM4SF proteins, BU16/JS64/C33/RUU.SP.2.28 (A and D). Staining was visualized using Rhodamine-conjugated goat anti–mouse IgG1 and FITC-conjugated goat anti–mouse IgG2a antibodies. Colocalization of TM4SF proteins with talin (G–I) was studied by using mouse mAb to talin, 8d6 (H), and a mixture of mAbs to TM4SF proteins, BU16/JS64/C33/RUU.SP.2.28 (G). Staining was visualized using Rhodamine-conjugated goat anti–mouse IgG1 and FITC-conjugated goat anti–mouse IgG2a and IgG2b antibodies. Images focused on cell–ECM contacts were captured, processed, and superimposed (C, E, I) by using the OpenLab software package. Inset shows digitally enlarged area of the colocalization of TM4SF proteins with talin in punctate adhesion structures (arrows), and absence of tetraspanins from focal adhesions (arrow head). Bar, 5 μM.

Figure 7

TM4SF proteins colocalize with MARCKS but not adaptin. MDA-MB-231 cells were prepared for double-immunofluorescence staining as described in the legend to Fig. 1. Codistribution of TM4SF proteins with MARCKS (A–C) and adaptin (D–F) was studied by using mAb 2F12/MARCKS (B) or 100/2/adaptin (E) in combination with a mixture of mAbs to TM4SF proteins, BU16/JS64/C33 (A and D). Staining was visualized using Rhodamine-conjugated goat anti–mouse IgG1 and FITC-conjugated goat anti–mouse IgG2a antibodies. Images focused on cell–ECM contacts were captured, processed, and superimposed (C and F) by using the OpenLab software package. Inset shows digitally enlarged area of colocalization of TM4SF proteins with MARCKS. Bar, 5 μM.

Figure 8

A role of cytoskeleton in cellular distribution of α3β1–TM4SF protein complexes. MDA-MB-231 cells were first plated in serum-free DMEM on glass coverslips coated with laminin-5–containing ECM for 1 h, and then treated with cytochalasin B at 10 μg/ml (A–C) or nocodazole at 20 μg/ml (D–F) for 30 min. Cells were prepared for double-immunofluorescence experiments as described in the legend to Fig. 1. Cellular localization of α3β1–TM4SF protein complexes was studied with mAb to α3 (B and E) or with a mixture of mAbs to TM4SF proteins (A and D). Staining was visualized using Rhodamine-conjugated goat anti–mouse IgG1 and FITC-conjugated goat anti–mouse IgG2a and anti–mouse IgG2b antibodies. The mAbs were: A3-X8 anti-α3; BU16, anti-CD9; RUU.SP.2.28, anti-CD63; JS64, anti-CD81; C33, anti-CD82. Note, α3β1 and TM4SF proteins are colocalized on intracellular vesicles (C), and in the peripheral adhesion clusters (F). Bar, 2.5 μM.

Figure 9

A role of cytoskeleton in association of TM4SF proteins with α3β1 integrin. MDA-MB-231 cells spread on laminin-5–containing ECM were treated for 30 min with cytochalasin B at 10 μg/ml (CD), and nocodazole at 20 μg/ml (NC), or left untreated (−), and subsequently lysed in buffer containing 1% Brij 98. Protein complexes were immunoprecipitated from cell lysates with mAb to α3 (lanes 1–3) or a mixture of mAbs to TM4SF proteins (lanes 4–6). Proteins were eluted from beads in Laemmli-loading buffer containing 3% β-mercaptoethanol and resolved in 11% PAGE. Presence of α3β1 integrin in the immunoprecipitates was detected by Western immunoblotting with polyclonal Ab to α3 integrin subunit. The mAb A3-X8 was used to immunoprecipitate α3β1 integrin; a mixture of the following mAbs was used to immunoprecipitate TM4SF protein complexes: BU16 and C9-BB, anti-CD9; 6H1, anti-CD63; M38 and JS64, anti-CD81; M104, anti-CD82.

Figure 10

α3β1–TM4SF protein complexes are not directly linked to the cytoskeleton. MDA-MB-231 cells were plated on laminin-5–coated ECM in serum-free DMEM. After 1 h membrane proteins were extracted in buffer containing 0.2% Triton X-100 for 10 min (A) or 1% Tween-20 for 20 min followed by the extraction with 1% Triton X-100 for 30 min (B). Proteins were separated in 10% PAGE and detected using Western immunoblotting with the appropriate Ab. The Abs were as follows: polyclonal sera to the cytoplasmic domain of α3 integrin subunits, mAb C9-BB to CD9, mAb M38 to CD81, mAb C33 to CD82, mAb 11G1.B4 to CD151, mAb F10-44-2 to CD44. (C) MDA-MB-231 cells plated on laminin-5–coated ECM were first pretreated with a membrane-impermeable chemical cross-linker, DTSSP, and then lysed in buffer containing 1% Tween-20. Nonsolubilized material was reextracted with buffer containing both 1% Tween 20 and 1% Triton X-100. Subsequently, Tween- and Tween/Triton-lysates (lanes 1 and 2, respectively) were supplemented with SDS (0.1%), and α3β1–TM4SF protein complexes were immunoprecipitated with a mixture of anti-TM4SF mAbs as described in the legend to Fig. 7. Proteins were eluted from beads in Laemmli loading buffer containing 3% β-mercaptoethanol and resolved in 11% PAGE. The α3β1 integrin in the immunoprecipitates was detected by Western immunoblotting with polyclonal Ab to α3 integrin subunit.

Figure 11

The effect of ligation of TM4SF proteins on FAK tyrosine phosphorylation in MDA-MB-231 cells. Serum-starved MDA-MB-231 cells were detached using EDTA and either left in suspension (lane 1 in A, B, and C) or replated for 1 h on bacteriological dishes coated with: (A) collagen (lane 2), anti-α2 (lane 3), anti-α3 (lane 4), anti-TM4SF (lane 5), anti–MHC class I (lane 6) mAbs; (B) anti-CD9 mAb (lane 2), and anti-CD63 mAb (lane 3), anti-CD81 mAb (lane 4), anti-CD82 mAb (lane 5), anti-CD151 mAb (lane 6); (C) collagen (lane 2), collagen and anti-CD9 mAb (lane 3), collagen and anti-CD63 mAb (lane 4), collagen and anti-CD81 mAb (lane 5), collagen and anti-CD82 mAb (lane 6), collagen and anti-CD151 mAb (lane 7), collagen and anti–MHC class I mAb (lane 8). Cells were lysed in buffer containing 1% Triton X-100, and FAK was immunoprecipitated using polyclonal Ab. The immunoprecipitates were divided into two equal aliquots, and tyrosine phosphorylation of FAK was analyzed by Western immunoblotting with mAb 4G10 (top), and with anti-FAK Ab (bottom). The immobilized mAbs were A2-IIE10 anti-α2 and A3-IVA5 anti-α3. The following anti-TM4SF mAbs were used either as a mixture (A) or separately (B and C): BU16 and C9-BB, anti-CD9; 6H1, anti-CD63; M38 and JS64, anti-CD81; M104, anti-CD82, 5C11-anti-CD151.

Figure 11

The effect of ligation of TM4SF proteins on FAK tyrosine phosphorylation in MDA-MB-231 cells. Serum-starved MDA-MB-231 cells were detached using EDTA and either left in suspension (lane 1 in A, B, and C) or replated for 1 h on bacteriological dishes coated with: (A) collagen (lane 2), anti-α2 (lane 3), anti-α3 (lane 4), anti-TM4SF (lane 5), anti–MHC class I (lane 6) mAbs; (B) anti-CD9 mAb (lane 2), and anti-CD63 mAb (lane 3), anti-CD81 mAb (lane 4), anti-CD82 mAb (lane 5), anti-CD151 mAb (lane 6); (C) collagen (lane 2), collagen and anti-CD9 mAb (lane 3), collagen and anti-CD63 mAb (lane 4), collagen and anti-CD81 mAb (lane 5), collagen and anti-CD82 mAb (lane 6), collagen and anti-CD151 mAb (lane 7), collagen and anti–MHC class I mAb (lane 8). Cells were lysed in buffer containing 1% Triton X-100, and FAK was immunoprecipitated using polyclonal Ab. The immunoprecipitates were divided into two equal aliquots, and tyrosine phosphorylation of FAK was analyzed by Western immunoblotting with mAb 4G10 (top), and with anti-FAK Ab (bottom). The immobilized mAbs were A2-IIE10 anti-α2 and A3-IVA5 anti-α3. The following anti-TM4SF mAbs were used either as a mixture (A) or separately (B and C): BU16 and C9-BB, anti-CD9; 6H1, anti-CD63; M38 and JS64, anti-CD81; M104, anti-CD82, 5C11-anti-CD151.

Figure 11

The effect of ligation of TM4SF proteins on FAK tyrosine phosphorylation in MDA-MB-231 cells. Serum-starved MDA-MB-231 cells were detached using EDTA and either left in suspension (lane 1 in A, B, and C) or replated for 1 h on bacteriological dishes coated with: (A) collagen (lane 2), anti-α2 (lane 3), anti-α3 (lane 4), anti-TM4SF (lane 5), anti–MHC class I (lane 6) mAbs; (B) anti-CD9 mAb (lane 2), and anti-CD63 mAb (lane 3), anti-CD81 mAb (lane 4), anti-CD82 mAb (lane 5), anti-CD151 mAb (lane 6); (C) collagen (lane 2), collagen and anti-CD9 mAb (lane 3), collagen and anti-CD63 mAb (lane 4), collagen and anti-CD81 mAb (lane 5), collagen and anti-CD82 mAb (lane 6), collagen and anti-CD151 mAb (lane 7), collagen and anti–MHC class I mAb (lane 8). Cells were lysed in buffer containing 1% Triton X-100, and FAK was immunoprecipitated using polyclonal Ab. The immunoprecipitates were divided into two equal aliquots, and tyrosine phosphorylation of FAK was analyzed by Western immunoblotting with mAb 4G10 (top), and with anti-FAK Ab (bottom). The immobilized mAbs were A2-IIE10 anti-α2 and A3-IVA5 anti-α3. The following anti-TM4SF mAbs were used either as a mixture (A) or separately (B and C): BU16 and C9-BB, anti-CD9; 6H1, anti-CD63; M38 and JS64, anti-CD81; M104, anti-CD82, 5C11-anti-CD151.

Figure 12

Ectopic expression of CD9 in HT1080 cells does not affect cell adhesion to collagen I and laminin 1. BCECF-AM–labeled cells tested for adhesion to 96-well microtiter plate coated with collagen type I or laminin-1 as described in Materials and Methods. The results are presented as percentage of attached cells.

Figure 13

The effect of overexpression of CD9 on adhesion-dependent phosphorylation of FAK. (A) HT1080/zeo and HT1080/CD9 cells were serum-starved for 4 h in suspension before plating on collagen- (lanes 3 and 4) or laminin-coated (lanes 5 and 6) bacteriological dishes for 2 h at 37°C. Cells were lysed in buffer containing 1% Triton X-100, and FAK was immunoprecipitated using polyclonal Ab. Tyrosine phosphorylation of FAK was analyzed by using Western immunoblotting with mAb 4G10 (top). In the lower panel, the filter was stripped and reprobed with anti-FAK Ab. (B) Cells were prepared for the experiment as above before plating to laminin-coated dishes for various lengths of time. Immunoprecipitation of FAK and subsequent analysis of phosphorylation was carried out as described above.

Figure 13

The effect of overexpression of CD9 on adhesion-dependent phosphorylation of FAK. (A) HT1080/zeo and HT1080/CD9 cells were serum-starved for 4 h in suspension before plating on collagen- (lanes 3 and 4) or laminin-coated (lanes 5 and 6) bacteriological dishes for 2 h at 37°C. Cells were lysed in buffer containing 1% Triton X-100, and FAK was immunoprecipitated using polyclonal Ab. Tyrosine phosphorylation of FAK was analyzed by using Western immunoblotting with mAb 4G10 (top). In the lower panel, the filter was stripped and reprobed with anti-FAK Ab. (B) Cells were prepared for the experiment as above before plating to laminin-coated dishes for various lengths of time. Immunoprecipitation of FAK and subsequent analysis of phosphorylation was carried out as described above.

Figure 14

The effect of overexpression of CD9 on the organization of actin cytoskeleton. Serum-starved HT1080/zeo and HT1080/CD9 cells were plated on collagen- (A and B) or laminin-coated (C and D) glass coverslips for 1 h. Cells were fixed with paraformaldehyde and permeabilized with 0.25% CHAPS. Actin filaments were visualized with TRITC-conjugated phalloidin. Bar, 10 μM.