Cross-talk between integrins alpha1beta1 and alpha2beta1 in renal epithelial cells

Abstract

The collagen-binding integrins alpha1beta1 and alpha2beta1 have profoundly different functions, yet they are often co-expressed in epithelial cells. When both integrins are expressed in the same cell, it has been suggested that alpha1beta1 negatively regulates integrin alpha2beta1-dependent functions. In this study we utilized murine ureteric bud (UB) epithelial cells, which express no functionally detectable levels of endogenous integrins alpha1beta1 and alpha2beta1, to determine the mechanism whereby this regulation occurs. We demonstrate that UB cells expressing integrin alpha2beta1, but not alpha1beta1 adhere, migrate and proliferate on collagen I as well as form cellular cords in 3D collagen I gels. Substitution of the transmembrane domain of the integrin alpha2 subunit with that of alpha1 results in decreased cell adhesion, migration and cord formation. In contrast, substitution of the integrin alpha2 cytoplasmic tail with that of alpha1, decreases cell migration and cord formation, but increases proliferation. When integrin alpha1 and alpha2 subunits are co-expressed in UB cells, the alpha1 subunit negatively regulates integrin alpha2beta1-dependent cord formation, adhesion and migration and this inhibition requires expression of both alpha1 and alpha2 tails. Thus, we provide evidence that the transmembrane and cytoplasmic domains of the alpha2 integrin subunit, as well as the alpha1 integrin subunit, regulate integrin alpha2beta1 cell function.

Figure 1. Integrin α2β1 but not α1β1 mediates UB cell interactions with collagen I

(A) The different UB cell populations were plated in serum free medium on collagen I at the indicated concentrations and cell adhesion was evaluated 1 hour after plating. Values are the mean ± SD of three experiments performed in triplicates. (B) UB cell populations were allowed to adhere to collagen I for an hour after which they were fixed and stained with rhodamine phalloidin to visualize the cytoskeleton. (C) UB cell populations were plated on transwells coated with 2.5 μg/ml collagen I and migration was evaluated after 4 hours. Values are the mean ± SD of three experiments performed in triplicates. (D) The UB cell populations were plated within collagen I gels. After 2 days, cells were treated with ³H-Thymidine and incubated for a further 2 days. ³H-Thymidine incorporation was then determined as described in the Methods. Values are the mean ± SD of three experiments performed in triplicates and are expressed as fold changes over UB cells transfected with vector only. (*) indicates statistically significant differences (p<0.05) between UB cells untransfected or transfected with the integrin α1 or α2 subunits. (E) The UB cell populations were embedded in collagen I gels as described in the Methods. Nine days later, the gels were fixed and stained with rhodamine phalloidin to visualize the cytoskeleton. The number of branches of 40 cellular structures per experiment was quantified and the values represent the mean ± SD of three experiments.

Figure 2. The transmembrane and cytoplasmic domains of the integrin α1 and α2 subunits differentially mediate UB cell function

(A) Schematic representation of the chimeric integrin α1 and α2 mutants. (B) Histogram representation of the expression levels of the cell populations shown relative to UB cells transfected with pcDNA3 vector only. (C–E) cell adhesion (C), spreading (D), migration (E), and proliferation (F) were evaluated and expressed as described in Fig. 1. (*) indicates statistically significant differences (p<0.05) between α2-UB and UB cells expressing the integrin α1/α2 chimeras. (G) The UB cell populations were plated in collagen I gels and branching was evaluated as described in Fig. 1E. The values are the mean ± SD of three experiments are shown.

Figure 3. Integrin α2β1 and the α1/α2 chimeras signaling differently in response to collagen I binding

(A) Serum starved UB cell populations were plated on collagen I (10μg/ml) for 1 hour, after which they were stained with anti-phosphotyrosine antibody (PY99). (B) Serum starved UB cell populations were plated collagen I (20 μg/ml) for the minutes indicated. 20 μg of total cell lysates were then analyzed by western blot for levels of activated and total ERK, p38 MAPK, and Akt. Activated kinase and total kinase bands were quantified by densitometry analysis and expressed as activated kinase/total kinase ratio. Values are the mean ± SD of 3 experiments and represent fold changes relative to UB cells at time 0. (*) indicates significant differences (p<0.05) relative to UB cells at time 0, while (#) denotes significant differences between α2UB-cells and the chimera expressing cells at the same time points. (C, D) Cell migration (C) and proliferation (D) were performed as described in Fig 1 with the exception that cells were either untreated or treated with PD98059 (50 μM), PD169316 (10 μM), LY294002 (5 μM). Values are means ± SD of three experiments performed in triplicate and proliferation is expressed as fold changes relative to UB cells transfected with vector only. (#) denotes significant differences (p<0.05) between untreated or inhibitor treated cells within the same cell populations.

Figure 4. Integrin α1β1 modulates integrin α2β1-mediated cell function

(A) Histogram representation of the expression levels of the integrin α1 constructs in α2+α1-UB (α1) and α2+α1Δcyt-UB (α1Δcyt) cells relative to the α2-UB cell population (control). (B-D) Cell adhesion (B), spreading (C), migration (D) and proliferation (E) were evaluated and quantified as described in Fig. 1. (*) indicates statistically significant differences between α2-UB and α2+α1-UB cells. (F) The UB cell populations were plated in collagen I gels and branching was evaluated as described in Fig. 1E. The values are the mean ± SD of three experiments.

Figure 5. Integrin α1 subunit downregulates integrin α2β1-dependent signaling

(A) Serum starved UB cell populations cells plated on collagen I (10μg/ml) for 1 hour, after which they were stained with anti-phosphotyrosine antibody (PY99). (B) Activation of ERK, p38 MAPK, and Akt was evaluated as described in Fig. 3. Values are the mean ± SD of 3 experiments and represent fold changes relative to α2-UB cells at time 0. (*) indicates significant differences (p<0.05) relative to α2-UB cells at time 0, while (#) denotes significant differences between α2UB-cells and α2+α1-UB cells at the same time points.(C) Activation of ERK, p38 MAPK, and Akt was evaluated in the UB cell populations indicated as described in Fig. 3. Values are the mean ± SD of 3 experiments and represent fold changes relative to α2α2α1-UB (B) or α2α1α2-UB (C) cells at time 0. (*) indicates significant differences (p<0.05) relative to α2α2α1-UB or α2α1α2 -UB cells at time 0, while (#) denotes significant differences between α2α2α1-UB and α2α2α1+α1-UB or α2α1α2-UB and α2α1α2-UB cells at the same time points.