Presentation of BMP‐2 from a Soft Biopolymeric Film Unveils its Activity on Cell Adhesion and Migration

Abstract

The dramatic effect of morphogenetic protein 2 (BMP-2), presented to cells in a “matrix bound” manner is described. BMP-2 is either delivered in solution or bound to a thin film made of poly(lysine) and hyaluronan. In this later case only, a striking effect of BMP-2 on cell adhesion and migration is observed. Cell biologists and biomaterials scientists are used to presenting cell adhesive ligands as they are in vivo: that is, in their insoluble form and in proximity to the cell adhesion site.1, 2 Intriguingly, this approach has barely been adopted for growth factors, usually presented to the cell in solution or as a releasable molecule, whereas growth factors in vivo are known to strongly interact and to bind with the components of the extracellular matrix.3, 4 In addition to these biochemical signals, the mechanical properties of the cell’s microenvironment are known to greatly impact cell behavior. To the best of our knowledge, the interplay between the mechanical properties of a biomaterial and the presentation of growth factor by the same material has not been investigated, to date. Over the years, growth factors have been shown to govern numerous cellular processes and are now recognized as promising therapeutic tools.5 Many tissue-engineering efforts to date have focused on modifying surfaces with extracellular matrix (ECM) proteins or adhesive peptides. Growth factors have been mixed with biomaterials to be released by diffusion.6, 7 In fundamental cell-biology research, there is also a large body of research dedicated to understanding the signalization pathways underlying cell stimulation by growth factors. Until now, in these cell biology studies, growth factors have mostly been added in solution to the culture medium of cells grown on stiff materials such as glass or tissue-culture polystyrene.8 The effects of such stiff materials on cellular processes can be quite subtle. Often, harsh serum starvation for about one day is necessary to render the cell quiescent, if any consequence of growth factor treatments on cellular behavior is to be observed.8, 9 Interestingly, an interplay between cell adhesion-receptor (e.g., integrin) signaling and growth-factor signaling has also been evidenced10 by plating cells on a given type of ECM protein to induce a specific signal while providing the growth factor in solution.11 The concept of presenting a growth factor “in the solid state”, which we will, here, refer to as “matrix-bound”, has been proposed in the pioneering work by Griffith et al.12 on tethered epidermal growth factor (EGF), but it has, surprisingly, been neglected for a long time. However, this mode of delivery is closer to physiological conditions, since most growth factors in ECM are bound to proteins and glycosaminoglycans.4 Thus, they are thus presented to cells in a matrix-bound manner. Only recently, this concept has been extended to other types of growth factor, including the vascular endothelial growth factor (VEGF)13 and the insulin-like growth factor 1.14 These recent experimental studies showed that the efficacy of the growth factors was improved, compared to the soluble form, and that their biological function was enhanced.13-15 The study of the family of bone morphogenetic proteins (BMPs) is an intensive field of research toward tissue-engineering applications16 and for fundamental insight into cell biology17 due to its physiological importance.18 BMPs play a crucial role in morphogenesis, tissue patterning, and regeneration after tissue damage, including cell differentiation.18, 19 In particular, BMP-2 is a highly potent morphogen that induces muscle precursors and mesenchymal stem-cell differentiation in bone cells.20, 21 Cell biologists have already noticed that BMP-2 added in solution to serum-starved cells plays also a role in early adhesive events, including adhesion and migration through actin reorganization.8, 22 The delivery of matrix-bound BMP-2 in a bioactive form is still a major challenge in the field of tissue engineering:23 one of the ultimate goals is to control cell differentiation in contact with a carrier material. New types of materials, such as materials mimicing natural ECM including fibrin films,24 polypeptide, polysaccharides-based layer-by-layer films,25, 26 and hyaluronan hydrogels,27 appear particularly interesting. We have recently demonstrated that crosslinked poly(L-lysine)/hyaluronan (PLL/HA) polyelectrolyte multilayer films can effectively retain high and tunable quantities of human recombinant (rh)BMP-2.26 The effective bioactivity of BMP-2 was confirmed by measuring the osteogenic marker alkaline phosphatase that is expressed during the differentiation of muscle-derived C2C12 cells, which was directly related to the BMP-2 amount. We have also shown that myoblasts needed direct contact with the bound rhBMP-2 to start the differentiation process into bone cells, thus showing that BMP-2 was delivered in an immobilized fashion.26 In addition to chemical stimuli, the mechanical microenvironment of the cells is now recognized to play a key role in various aspects of cell processes, including adhesion, migration, and differentiation.28 Currently, synthetic polymers are the most popular tool for adjusting both biochemical and mechanical properties.29 However, natural materials are needed, again, and are emerging to mimic extracellular matrices by recapitulating several cues from the physiological cell microenvironment.30Using poly(L-lysine)/hyaluronan (PLL/HA) polyelectrolyte multi­layer films of controlled stiffness, we recently demonstrated that film stiffness induces dramatic changes in C2C12 myoblast cell adhesion and spreading,31 and that myoblasts can differentiate into myotubes, with morphological differences depending on the stiffness of the film.31 Here, we advantageously combine two functionalities of the biomimetic (PLL/HA) films, namely (i) the ability to modulate the film stiffness and (ii) the influence on the bioactivity of the film by presenting BMP-2 in a matrix-bound fashion (Figure 1) to investigate early cellular events. We focus on cell adhesion and migration, which are key steps prior to differentiation. Indeed, these processes are of upmost importance in embryonic morphogenesis, tissue regeneration, and tumor metastasis. We demonstrate that, by modulating both the presentation of the growth factor (soluble versus matrix-bound) and the mechanical properties of the material, it is possible to provide a permissive niche and unravel early cellular events related to confinement of BMP-2 at the ventral cell surface. A) Biomimetic thin film combining physical and biochemical cues. 1) A polyelectrolyte multilayer film (PEM) is attached to a substrate by alternating deposition of two biocompatible polyelectrolytes: poly(L-lysine) and hyaluronan. 2) The PEM film can be covalently crosslinked to different extents using a water soluble carbodiimide, thus allowing control over the film stiffness. 3) Finally, an additional functionality is provided to the film by loading it with the growth factor BMP-2; BMP-2 is trapped within the film and is subsequently delivered to the cells in a “matrix-bound” manner. B) The luminescence of C2C12 cells transfected with p(BRE)4-luciferase was measured after cell culture for 24 h in the presence of soluble or bound BMP-2. C) Western blot analysis using an anti-phospho Smad antibody to show that phosphorylation of Smad is present. For a loading control, the membrane was incubated with anti β-actin antibody. C’) Corresponding intensity analysis (ratio of phospho-Smad/β-actin) obtained by Western blot for the different conditions. The design principle of the biomimetic nanoassembly is depicted in Figure 1. First, we deposited a layer-by-layer film of PLL and HA composed of 12 layer pairs (≈1.3 μm in thickness) on a supporting material (glass substrate) (Figure 1A). Second, to control the film stiffness, we covalently crosslinked the films by using carbodiimide chemistry, as previously described.26 Two crosslinker concentrations were used, resulting in films with different Young moduli (Table S1 in the Supporting Information (SI)). Hereafter, these films will be referred to hereafter soft and stiff films. Finally, we loaded the films with rhBMP-2 by post-diffusion in acidic condition, as previously described.26 Using this strategy, tunable amounts of BMP-2 can be loaded in the films, depending on the initial concentration of BMP-2 and the film thickness. We have previously shown that, after a slight release during the first hours, the amount of BMP-2 remained stable without any further loss of BMP-2 during the next four days. Thus, BMP-2 can be considered as matrix-bound.26 We verified that the amount of rhBMP-2 loaded by post-diffusion into these films did not vary significantly in the two types of film at our working concentration (20 μg mL−1 for the initial rhBMP-2 concentration in solution, see Figure SI1 in the SI). Furthermore, we verified that BMP-2 loading did not affect the film’s overall morphology (Figure SI2), roughness, mechanical properties, or wettability (Table S1). This simple design strategy enables us to use a well-characterized biomimetic film that allows the variation of two major properties simultaneously: its stiffness and the presentation of matrix-bound BMP-2. For the cell culture experiments, we chose C2C12 myoblasts as a working model. These cells constitute an acknowledged in vitro model system to study the ability of BMPs to alter cell lineage from the myogenic to the osteogenic phenotype.21, 32, 33 In a low concentration of serum, C2C12 cells differentiate into multi­nucleated myotubes. In contrary, BMP-2 treatment inhibits myotube formation and induces osteoblastic differentiation, as characterized by alkaline phosphatase activity or osteocalcin production.21 Importantly, these cells are also already known to be sensitive to the mechanical properties of the underlying matrix in terms of adhesion, spreading, and differentiation in myotubes.31, 34, 35 C2C12 myoblasts were cultured on a stiff control surface (polystyrene culture plates and glass) and on soft and stiff BMP-2-loaded (PLL/HA) films containing bound BMP-2 (bBMP-2). Importantly, in the present work, the serum-starvation step was suppressed in order to test the ability of the bioactive films to induce specific cellular events in a complete medium and to investigate whether this effect is sustained. Significant differences in the quantities of BMP-2 available to the cells might cast doubt on the conclusions regarding the biological importance of soluble versus bound BMP-2 presentations. Although it is very difficult to ensure that similar quantities are accessible by the cells in the case of soluble versus bound BMP-2, we ensure that the number of BMP-2 molecules surpassed the number of BMP-2 receptors in both conditions. We measured a bBMP-2 concentration of ≈700 ng cm−2 in the films (Figure S1), which corresponds to 3 × 105 molecules per μm2. Theoretically, this is enough to saturate the endogenous BMP-2 receptors (estimated at less than 100 per μm2 cell surface area for cells typically expressing BMP2R at 10 000 receptors per cell).36 When delivered in solution (sBMP-2), a high concentration of BMP-2 was used (600 ng mL−1) allowing the assumtion that the receptors were saturated. This was supported by the fact that alkaline phosphatase expression showed a plateau for concentrations equal or higher than 300 ng mL−1 (Figure S3). We first verified that bBMP-2 is effectively active when bound to soft and stiff films by performing two complementary assays. We measured the luciferase activity of C2C12 cells transfected with a BMP-responsive element (Id1) fused to the firefly luciferase reporter gene.37 (Figure 1B). We that the luciferase expression was significantly in the presence of soluble and bound BMP-2 for conditions. The luciferase signal of bBMP-2 on stiff films and was but it was significantly higher for bBMP-2 on soft films. In we the phosphorylation of a protein known to play a key role in the from BMP-2 receptors to the The expression of the form of Smad was measured by Western blot (Figure and The of Smad phosphorylation in the presence of and bBMP-2 on stiff was but a higher phosphorylation was for bBMP-2 on soft films. this that BMP-2 is bioactive in conditions, with a similar or Smad signal when it is delivered in a matrix-bound manner. In the next we measured the effect of or bBMP-2 on the initial cell adhesion and (Figure In the first 4 h after cell the cells on stiff tissue culture material μm (Figure and on stiff films μm (Figure cells already very on stiff in the of BMP-2, it was not to that soluble or matrix-bound BMP-2 did not to a on cell on these stiff In the were significantly in the case of soft (PLL/HA) films presenting bBMP-2 μm Figure whereas cells remained and in the of BMP-2. Interestingly, cells remained and in the presence of (Figure This specific of on soft films by bBMP-2 was also with a in the number of cells after 4 h on soft films (Figure Cell for bBMP-2 presented on soft (PLL/HA) films thus that initial myoblast adhesion and spreading, whereas this effect was in the case of stiff films. that is not in our culture conditions, of the the effect of bBMP-2 is specific and is not related to the presence of serum in the culture since the striking effect of bBMP-2 on soft films was still in a medium (Figure Furthermore, the effect of bBMP-2 be to the changes of film stiffness addition of bBMP-2 (Table S1), as these changes are not Indeed, they were not to induce any in cell for other cells, such as (Figure BMP-2 in soft films early cell adhesion and cell to 4 is greatly on soft films by bBMP-2 and Cell over the first 4 h is on and on stiff multilayer films the culture or whereas cells remained on soft films, in the presence of soluble BMP-2 bBMP-2 on soft films significantly cell B) 4 h of adhesion, effects were for cells grown on and stiff films but the number of cells was significantly on soft films in the presence of our analysis of cell adhesion, we whether the effect of BMP-2 was for several We thus cell morphology using after the actin and measured the cell for experimental at a after h of culture (Figure initial adhesion, the remained the the control tissue culture surface cells both with and without (Figure the cells on the stiff films a high area that was not by bBMP-2 (Figure Only cells cultured on soft films with bBMP-2 the effect of BMP-2 on the of cell adhesion and (Figure We also that the adhesion protein a for cells on soft films with of adhesion formation (Figure of with the of cells grown on films. This that are and during the cell to matrix-bound BMP-2. effect of matrix-bound BMP-2 on cell Cell morphology is after plating the and of C2C12 cells a morphology on the glass control substrate as as on stiff (PLL/HA) films in the presence or in the of BMP-2, of the presentation mode of BMP-2 versus for soft (PLL/HA) films, did not induce any effect on cell but bBMP-2 a striking in cell The is The cell surface area after of culture is for the different conditions. This the in cell in to bBMP-2 on soft films is after initial is to that our are not specific to C2C12 myoblast Indeed, we that cells, an acknowledged cellular model to study differentiation in bone and which also adhesion and were using bBMP-2 on soft films (Figure Importantly, this by BMP-2 is specific to cells that to BMP-2 signals, since we did a very effect on cells, known to BMP-2 but known not to BMP-2 (Figure of a on one and BMP-2 on the other have already been shown to play a key role in In BMP-2 has been to cell migration in several cell we further whether the cell migration on the different surfaces was also by or bBMP-2 (Figure again, in our culture medium and starvation step to effect on cell migration on any types of surface. In bBMP-2 significantly cell migration on stiff films and in an on soft films. In our that bBMP-2 a and effect on the of the actin in C2C12 cells, which can be to the presentation of the growth factor by the soft as to the soluble presentation of BMP-2. BMP-2 cell of C2C12 cells cultured on different tissue culture and either stiff or soft films. bBMP-2 significantly the cell on stiff films and significantly on soft films. The soluble form of BMP-2 effect compared to the BMP-2 is on a key can this effect of bound BMP-2 be We that these dramatic changes in cell and migration are of which are due to the of from BMP-2 presentation in to its presentation in This confinement its as it is bound to the film and not its that is, when a BMP-2 is is available in and an interplay of BMP-2 receptors with other types of such as The differences between matrix-bound and solution presentation of BMP-2 are in Figure BMP-2 presentation from the ECM film allows confinement of and BMP-2 at the ventral of the This the major differences between matrix-bound and soluble delivery of BMP-2 to the When BMP-2 is bound to the film of the it is and its is In the of BMP-2 receptors is with a possible formation and and is not by high number of ligands is available in proximity of the Furthermore, due to the proximity of growth factor receptors and adhesion a between these two types of is Thus, between BMP-2 signaling and adhesion which can induce might the striking effects for cells on soft films with effect be when BMP-2 is presented in solution of the BMP-2 can in and has a low due to the between receptors and Furthermore, in this BMP-2 receptors are at the membrane and are not at the of adhesion This confinement is for the of the BMP-2 Indeed, BMP-2 is in and can bind BMP-2 Furthermore, BMP-2 receptors which are composed of type and type can form the cell of of the of type and type to et the of different receptors with BMP-2, of the are low which most whereas of the high Thus, the of the low receptors be greatly in the presence of a large amount of bBMP-2 which are characterized by both a presentation and This between and strongly the of the BMP-2 Furthermore, the confinement of bBMP-2 at the ventral cell surface and the striking effect of BMP-2 on cell and between growth factor receptors and adhesion between adhesion receptors and growth factor receptors has been for BMP-2 are which are known Indeed, it has already been shown that the mechanical between and the ECM can be by from growth factors and other is known to the phosphorylation of other growth factor receptors or directly the growth factor by to its one that the or BMP-2 might with in a signaling that both BMP2R and or it might through its to BMP-2 BMP-2 thus as to has been for which modulate to the of the ligands they This is by recent work that that tethered can in with adhesion ligands on soft synthetic Finally, bBMP-2 be from and this signaling events compared to These to be further in the This “matrix-bound” mode of delivery thus the importance of of and their of and possible at cell The adhesive and that BMP-2 to of receptors in the of the and it the for of two receptors factor and adhesion to a specific signaling these that the materials used for cell culture be with when the effects of growth factors on cells, given that the properties of the material on the cellular behavior are from materials can induce cellular differentiation, and most materials are to cell adhesion and is a to materials the of a an in which the cells are and not to a specific due to the underlying stiff We a highly tunable material that cell from the while cell and their for adhesion and migration when the is A of be cells, and to and can thus be obtained when cells are grown on soft films. These cells can be of use for the effects of various chemical or physical cues on cell adhesion and These biomimetic films have us to the effects of biochemical by matrix-bound BMP-2, when cells are on a stiff In we have a biomaterial the of a permissive For the first we have been to using C2C12 myoblasts as model BMP-2 cells, that cells when BMP-2 is presented in matrix-bound form on a soft than in its soluble This was for early cellular events. The for these differences is and further studies of the of BMP-2 signaling pathways versus are BMP-2 biochemical signaling differentiation, and the interplay between matrix mechanical properties and growth factor that signaling can be by the of different types of receptors and that is necessary for the of a cell We that this concept of presenting matrix-bound growth factors and bioactive molecules from films of tunable stiffness will be used to fundamental biological on growth-factor and that they will to the of a of of and of HA 2 × 105 was from and PLL × was from PLL mL−1) and HA mL−1) were in a (20 For the cell experiments, the films were as previously with a on glass For the cell the films were in plates with a first layer of × at 3 polyelectrolyte were deposited in to for with solution of for The was the of was is the number of layer The films were crosslinked the previously using carbodiimide at or 100 mL−1 and at mL−1 from The BMP-2 was into films for in the medium in which BMP-2 was A of for plates or for plates of BMP-2 was deposited on to the films and to at of were added to and at for The were for 7 h in in order to matrix-bound for The were at in with per in Cell C2C12 cells were in in a and cultured in a medium medium with serum containing 10 mL−1 and 10 μg mL−1 were prior to 2 For experiments, C2C12 cells on films at × cells cm−2 in growth medium were to for one of Cell and The cell were in The cell were after 4 h of adhesion using a In the cells were with and at the cells at a of the and was and the of the plates was measured using the For the cells were in in for and for 4 in containing were in containing for and incubated with in with for For adhesion in with were incubated for antibody was incubated for the were with and with using or were with using a from To cell adhesion and spreading, were with the to cell and cell area Smad Using and Western were transfected with an expression containing a BMP-responsive element fused to the firefly luciferase reporter gene.37 were cultured in same as C2C12 one day of culture on the cell and luciferase were to the luciferase were to the of as measured by the of by Western was to and at for the membrane was incubated with at of was by after with a control, of actin was also were over a h on either culture soft or stiff films in the presence of or BMP-2 in growth medium at in The were using a in For cellular the cells were deposited and cell surface were measured using the For migration, the was for after 4 h of initial adhesion and were measured using the and and from For both at cells were for were at and analysis was using the analysis of test to the differences or or between and Supporting Information is available from the or from the We are to for providing the transfected cells as a and to and for of the and This work was supported by from and the and to is to the for The of and to the of importance to are as are but not or are made available as by the The is not for the or functionality of any supporting by the than be to the for the