Abstract

   The regulation of inflammatory response at the site of injury and
   macrophage immunotherapy is critical for tissue repair. Chiral
   self‐assemblies are one of the most ubiquitous life cues, which is
   closely related to biological functions, life processes, and even the
   pathogenesis of diseases. However, the role of supramolecular chiral
   self‐assemblies in the regulation of immune functions in the internal
   environment of tissues has not been fully explored yet. Herein, 3D
   supramolecular chiral self‐assembling matrixes are prepared to regulate
   the polarization of macrophages and further enhance the repair of
   myocardial infarction (MI). Experiments studies show that M‐type
   (left‐handed) self‐assembling matrixes significantly inhibit
   inflammation and promote damaged myocardium repair by upregulating M2
   macrophage polarization and downstream immune signaling compared with
   P‐type (right‐handed), and R‐type (non‐chirality) self‐assembling
   matrixes. Classical molecular dynamics (MD) simulation demonstrates
   that M‐type self‐assembling matrixes display higher stereo‐affinity to
   cellular binding, which enhances the clustering of mechanosensitive
   integrin β1 (Itgβ1) and activates focal adhesion kinase (FAK) and
   Rho‐associated protein kinase (ROCK), as well as downstream
   PI3K/Akt1/mTOR signaling axes to promote M2 polarization. This study of
   designing a 3D chiral self‐assembling matrixes microenvironment
   suitable for regulating the polarization of macrophages will provide
   devise basis for immunotherapy with biomimetic materials.

   Keywords: chiral self‐assembling matrixes, immunological response,
   macrophages, myocardial infarction
     __________________________________________________________________

   A study of designing 3D chiral self‐assembling matrixes suitable for
   regulating the polarization of macrophages to inhibit inflammation and
   promote the repair of damaged myocardium.

   graphic file with name ADVS-10-2304627-g006.jpg

1. Introduction

   Myocardial infarction (MI) usually occurs in coronary artery occlusion,
   resulting in myocardial necrosis caused by persistent ischemia and
   hypoxia, which presents high morbidity and mortality.^[ [40]^1 ^] After
   MI, innate immunity is rapidly activated to produce a strong and
   transient inflammatory response, clearing the infarct region of dead
   cells and extracellular matrix debris, followed by a proliferation
   phase. During the proliferative phase, monocytes and macrophage
   subpopulations secrete growth factors that recruit and activate
   mesenchymal repair cells (mainly myoblasts and blood vessel cells).
   Myofibroblasts secrete large amounts of extracellular matrix proteins,
   which preserve the intact structure of the left ventricle and
   ultimately lead to the formation of scar collagen fibers. Repair of
   damaged tissue relies on timely suppression of inflammation, and
   inadequate suppression of cardiac inflammation can lead to catastrophic
   outcomes, resulting in loss of cardiomyocytes, suppression of systolic
   function, enlargement of the ventricles, loss of ventricular wall
   integrity, and rupture of the heart.^[ [41]^2 ^] The efficacy of
   traditional cardiac repairing strategies including drug and cell
   therapies for MI treatment is greatly limited by the cardiac
   microenvironment, which might lead to cardiac remodeling after MI and
   ultimately result in heart failure.^[ [42]^3 ^] Macrophage
   immunotherapy targeting the MI microenvironment is an emerging
   treatment approach that modulates the immune system to promote the
   repair of MI.^[ [43]^4 ^] Macrophages are one of the main responder
   cells of the immune system, infiltrating the infarcted myocardium for
   repairing MI, which can be roughly divided into early pro‐inflammatory
   macrophage phenotype I (M1) and later anti‐inflammatory macrophage
   phenotype II (M2).^[ [44]^5 ^] Macrophage behavior of various tissues
   during healing is closely associated with several biochemical and
   biophysical properties of the extracellular matrix (ECM), including
   chemical species, supramolecular self‐assemblies, mechanical, and
   chiral properties.^[ [45]^6 ^] Among these, chiral supramolecular
   self‐assemblies as basic structural units of living organisms are
   crucial for the good implementation of sophisticated functions involved
   in the life process.^[ [46]^7 ^] However, the strategy by which
   biologically bionic ECM based on 3D chiral supramolecular
   self‐assemblies manipulates the regulation of macrophages’ immune
   response has rarely been explored.

   The ECM network plays a critical role in cardiac homeostasis by
   providing structural support, facilitating force transmission, and
   transducing key signals to cardiomyocytes, vascular cells, and
   interstitial cells.^[ [47]^8 ^] The insoluble matrix components in ECM
   could offer various 3D biophysical cues to the tissue microenvironment
   at the nanometer to micron scales.^[ [48]^9 ^] Macrophages can
   dynamically interact with ECM‐mediated microenvironment, regulating
   their adhesion and polarization, mediated by the dynamic binding of
   integrin and ligand motifs in vivo, such as the RGD tripeptide sequence
   (Arg‐Gly‐Asp).^[ [49]^10 ^] RGD plays an important role in the innate
   immune system in the adhesion of myeloid cells such as granulocytes and
   macrophages. Both soluble and surface‐functionalized RGD polypeptides
   bind to myeloid cells via integrin.^[ [50]^11 ^] Blocking integrin
   binding with soluble RGD polypeptide significantly reduces TNFα and
   IL‐6 production.^[ [51]^12 ^] Macrophages also exhibit altered
   immunomodulatory effects due to RGD interactions, including changes in
   phenotype, cytokine production, and phagocytosis.^[ [52]^5 , [53]^13 ^]
   Accordingly, exploring the interactions between RGD peptides and
   biomimetic ECM to regulate macrophage polarization and then obtain the
   expected host immune response is becoming a novel method for repairing
   MI.

   Supramolecular chiral self‐assemblies (e.g., double helix in DNA,
   α‐helix, and β‐sheet in proteins) arising from the asymmetric spatial
   stacking of molecular blockings are involved in various essential
   biological processes, including the maintenance of cell morphology and
   functional changes in the tissue microenvironment.^[ [54]^14 ^]
   Pioneering research revealed that molecular chirality modified on
   traditional 2D matrixes could affect cellular asymmetrical motions that
   result in the different behaviors of cell morphology.^[ [55]^15 ^]
   However, exploring how artificial 3D supramolecular chiral matrixes
   influence cell behaviors in the ECM is particularly important since 3D
   supramolecular chiral matrixes extensively exist in biophysical ECM,
   which is required for relevant biological events.^[ [56]^7 , [57]^16 ^]
   Recently, the effect of the physical properties of ECM (such as tension
   and mechanical properties) on the phenotype and behavior of macrophages
   has received extensive attention.^[ [58]^17 ^] However, the potential
   importance of 3D chiral supramolecular self‐assemblies to affect the
   adsorption of adhesion proteins or peptides on the surface of materials
   and thus on the behavior of the polarization of macrophages is seldom
   explored.

   Herein, enantiomeric L/D‐phenylalanine gelators (L‐BA and D‐BA) and
   racemic DL‐phenylalanine gelators (R‐BA) are designed to construct
   cell‐supporting 3D chiral supramolecular nanofibers matrixes by a
   self‐assembly strategy, respectively, to regulating macrophage
   phenotype in the MI region, thereby facilitating MI repair (Scheme
   [59]1 ). Experiment studies investigate that M‐type supramolecular
   nanofibers matrixes significantly inhibit inflammation and promote
   damaged myocardium repair by upregulating M2 macrophage polarization
   and downstream immune signaling in contrast to corresponding P‐type and
   R‐type supramolecular nanofibers matrixes. The initial enantioselective
   mechanism between different chiral molecules and peptides was
   theoretically performed by classical molecular dynamics (MD)
   simulation. This work unveils the interaction between the biomimetic 3D
   supramolecular chiral matrixes and the macrophages’ immune system,
   which inspires alternatively sophisticated for developing
   immunotherapeutics.

Scheme 1.

   Scheme 1
   [60]Open in a new tab

   A) A schematic representation of biologically 3D self‐assembling
   matrixes with tunable supramolecular chirality. B) Schematic diagram
   showing 3D chiral self‐assembling matrixes regulating the polarization
   of macrophages and enhancing repair of myocardial infarction (MI).

2. Results and Discussion

2.1. Preparation, Morphology, and Supramolecular Chirality of 3D
Self‐Assembling Matrixes

   Enantiomeric D/L‐phenylalanine gelator (D‐BA and L‐BA) and racemic
   DL‐phenylalanine gelator (R‐BA) are employed to mimic the biophysical
   environment of natural ECM (synthesis and characterization of gelators
   see Schemes [61]S1,S2 and Figures [62]S1–S6, Supporting Information).
   D‐BA and L‐BA molecules can self‐assemble into M‐type nanofibers and
   P‐type nanofibers, respectively, while R‐BA molecules can self‐assemble
   into nanofibers without distinct chirality. These high‐aspect‐ratio
   nanofibers template through hydrogen bonding along the long axis of
   nanofibers and further physically entangle to self‐supportive 3D
   hydrogels networks by a supramolecular gelation process
   (Scheme [63]1A).^[ [64]^18 ^] The critical gel concentration (CGC) for
   L‐BA, D‐BA, and R‐BA is 1.0 wt.%, and hydrogel formation was judged by
   the conventional tube inversion method (Figure [65]S7 and Table [66]S1,
   Supporting Information). The water content of these three hydrogels was
   measured by the method of drying and weighing was almost above 98%
   (Figure [67]S8A, Supporting Information). Thermal analysis results
   demonstrated that the denaturation temperatures of L‐BA, D‐BA, and R‐BA
   were all higher than 37 °C, indicating their good thermal stabilities.
   (Figure [68]S8B, Supporting Information). Atomic force microscopy (AFM)
   and scanning electron microscopy (SEM) were used to investigate the
   morphology of the supramolecular hydrogels. The self‐assembly of D‐BA
   formed exclusively left‐handed (M‐type) helical nanofibers, while the
   self‐assembly of L‐BA produced uniform right‐handed (P‐type) helical
   nanofibers. We have made distribution statistics of the diameter (100 ±
   28 nm) and helical pitch (500 ± 20 nm) of these fiber nanostructures,
   they are almost the same except presented different chiral
   nanostructures (Figure [69]1A–D; Figure [70]S9, Supporting
   Information). Additionally, R‐BA was self‐assembled into uniform
   nanofibers without distinct chiral features (Figures [71]S10,S11,
   Supporting Information).

Figure 1.

   Figure 1
   [72]Open in a new tab

   AFM images of helical nanofibers: A) left‐handed (M‐type) helical
   nanofibers obtained from the self‐assembly of D‐BA and B) right‐handed
   (P‐type) helical nanofibers obtained from the self‐assembly of L‐BA.
   SEM images of helical nanofibers: C) M‐type helical nanofibers obtained
   from the self‐assembly of D‐BA and D) P‐type helical nanofibers
   obtained from the self‐assembly of L‐BA. E) Circular dichroism (CD)
   spectroscopy of M‐type and P‐type self‐assembling matrixes. F)
   Corresponding g [abs] and UV–vis spectroscopy of M‐type and P‐type
   self‐assembling matrixes. G) Temperature‐dependent CD spectroscopy of
   M‐type and P‐type self‐assembling matrixes. H) FTIR spectra of D‐BA and
   M‐type self‐assembling matrixes. I, J) Strain‐dependent oscillatory
   shear rheology of M‐type and P‐type hydrogels, respectively. K, L)
   Dynamic frequency sweep of M‐type and P‐type self‐assembling matrixes,
   respectively.

   The supramolecular chirality of the M‐type, P‐type, and R‐type
   supramolecular hydrogels was investigated by circular dichroism (CD)
   spectroscopy, respectively. The CD spectrum of M‐type hydrogels showed
   a relatively weak positive Cotton effect at 230 nm and a strong
   negative Cotton effect at 306 nm (Figure [73]1E, yellow curve), whereas
   P‐type hydrogels showed the mirror‐imaged CD spectrum of M‐type
   hydrogels (Figure [74]1E, gray curve). The M‐type and P‐type hydrogels
   showed the absorptive dissymmetry factor g [abs], which is defined by
   the ratio of CD intensity to the corresponding absorption (shown in
   Supporting Information), were approximately equal in intensity and
   handedness opposite in chirality (Figure [75]1F), which indicated that
   the molecular chirality of the gelators is well transferred to the
   supramolecular chirality.^[ [76]^14 , [77]^18 ^] With the increasing
   testing temperature, the intensity of the CD signals for both M‐type
   and P‐type self‐assembling hydrogels gradually decreased due to
   disassembly by heating. At ≈95 °C, the CD signals at 230 and 306 nm
   both almost disappeared suggesting the disintegration of M‐type and
   P‐type self‐assembled nanofibers, respectively. Two new weak CD peaks
   at 218 and 232 nm were ascribed to the molecular chirality from
   phenylalanine in D‐BA and L‐BA, respectively (Figure [78]1G). These
   data demonstrate that the chirality of the 3D hydrogels can be
   attributed to the self‐assembled fibrous aggregates and not to the
   individual monomers.^[ [79]^7 , [80]^18 ^] In addition, no CD signals
   existed for R‐type hydrogels due to their natural racemic
   characteristics (Figure [81]S12, Supporting Information).

2.2. Mechanism and Mechanical Properties of 3D Self‐Assembling Matrixes

   The underlying mechanism of the self‐assembly was further confirmed by
   fourier transform infrared (FTIR) spectroscopy measurement. For D‐BA as
   an example, the monomeric state of C═O stretching vibration of the
   peripheral amide and the central amide close to the benzene core at
   1674 and 1652 cm^−1 shifted to 1658 and 1631 cm^−1, respectively, after
   self‐assembly. FTIR results suggest hydrogen bonding interactions are
   stronger due to the intramolecular H─bonds with a small bond length
   formed from the amino acid diamides (Figure [82]1H). The mechanical
   properties of supramolecular self‐assembling matrixes (M‐type, P‐type,
   and R‐type hydrogels) were measured by a rotary rheometer with dynamic
   frequency sweep at a strain of 0.01%−100%, respectively, the values of
   elastic modulus (G′) of three hydrogels were higher than those of the
   viscous modulus values (G″), indicating three supramolecular
   self‐assembling matrixes formed solid hydrogels. (Figure [83]1I–L;
   Figure [84]S13, Supporting Information). Hence, these supramolecular
   self‐assembling matrixes provide an adequate platform to study the
   effect of chiral nanostructure on immunological response.

2.3. Effect of 3D Chiral Self‐Assembling Matrixes on Cytocompatibility

   Supramolecular self‐assembling chiral nanofibers have good
   biocompatibility in cell adhesion and proliferation as demonstrated in
   our previous works.^[ [85]^7 , [86]^18 ^] To verify the heterogenic of
   different 3D chiral self‐assembling matrixes in cardiac cells, we first
   observed the chiral self‐assembling matrixes have cytocompatibility in
   cell adhesion and proliferation. We cultured macrophages, fibroblasts,
   and cardiomyocytes in M‐type, R‐type, and P‐type self‐assembling
   matrixes for 24, 48, and 72 h, respectively. A cell counting kit‐8
   (CCK‐8) assay showed different activities of macrophages, fibroblasts,
   and cardiomyocytes in different cultural chiral self‐assembling
   matrixes environments, separately. Compared with the P‐type and R‐type
   self‐assembling matrixes, the M‐type self‐assembling matrixes reduced
   the apoptosis rate of macrophages, fibroblasts, and cardiomyocytes
   after 72 h of culture (Figure [87]S14, Supporting Information). These
   results indicate that M‐type self‐assembling matrixes have better
   cytocompatibility compared to P‐type and R‐type self‐assembling
   matrixes.

2.4. Effect of 3D Chiral Self‐Assembling Matrixes on Polarization of
Macrophages

   The inflammatory response in the early stage of trauma is critical for
   healing the damaged tissue.^[ [88]^5 , [89]^19 ^] In the initial stage
   of trauma development, naive macrophages (Mϕ) are responsible for
   phagocytosis, releasing pro‐inflammatory mediators, clearing senescent
   cells, and participating in tissue repair processes. Still, during the
   process of the damaged healing, macrophages initially present a
   classically activated M1 phenotype, high surface markers or
   intracellular iNOS can be used to identify M1 macrophages that release
   inflammatory cytokines (such as IL‐1β, IL‐6, IL‐12, and TNFα).^[ [90]^5
   , [91]^20 ^] Following the inflammatory phase, macrophages
   differentiate into alternatively activated M2 types. M2 macrophages can
   secrete anti‐inflammatory mediators such as surface markers of
   scavenger receptors, IL‐10, intracellular Arginase‐1, found in
   inflammatory zone 1(Fizz1), and Chitinase‐like 3 (Chil3/Ym1)
   expression. In this research, novel 3D supramolecular hydrogel systems
   with chiral nanofibrous structures were developed for studying
   supramolecular chirality‐dependent immunological responses (Figure
   [92]2A). To reveal supramolecular chirality regulating the polarization
   of macrophages, RT‐qPCR results showed the expression of the M1‐related
   inflammatory genes IL‐1β, IL‐6 (Figure [93]2B), iNOS, and TNF‐α
   (Figure [94]S15, Supporting Information) in M‐type matrixes were
   significantly lower than those in P‐type and R‐type matrixes,
   respectively. The expression of M2‐related genes including Ym1, Fizz1
   (Figure [95]2B), and TLR9 (Figure [96]S15, Supporting Information) in
   M‐type matrixes were significantly increased compared to P‐type and
   R‐type matrixes, respectively. We also quantified four key
   pro‐inflammatory cytokines released by macrophages in the control group
   and different supramolecular self‐assembling matrixes groups after 48 h
   of culture, respectively. Herein, lipopolysaccharide (LPS) and
   interferon γ (IFN‐γ) induced macrophages as a positive control group
   (denoted as Control). The levels of secretion of TNF‐α, IL‐6, IL‐12,
   and IL‐1β in the Control were significantly more than those in the
   M‐type, R‐type, and P‐type self‐assembling matrixes groups (P < 0.001)
   (Figure [97]2C). In addition, the expression of TNF‐α, IL‐6, IL‐12, and
   IL‐1β in M‐type self‐assembling matrixes was significantly decreased
   compared to those in R‐type and P‐type self‐assembling matrixes. These
   results indicated that macrophages recognized the difference of chiral
   self‐assembling matrixes thus polarized to divergent phenotypes, more
   importantly, our findings showed the effect of M‐type chirality is
   stronger than that of P‐type chirality.

Figure 2.

   Figure 2
   [98]Open in a new tab

   A) Schematic depicting the process of chirality‐dependent macrophage
   induction of polarization. B) RT‐qPCR results showed the expression of
   M1‐related proinflammatory factors IL‐1β and IL‐6 and anti‐inflammatory
   factor Ym1 and M2 marker Fizz1 in the Control group, P‐type, R‐type,
   and M‐type matrixes. C) ELISA results showed that M1‐related
   pro‐inflammatory factors TNF‐α, IL‐1β, IL‐6, and IL‐12 were expressed
   in Control group, P‐type, R‐type, and M‐type matrixes. D) Western blot
   showed the expression of M2 surface markers in M‐type, P‐type and
   R‐type self‐assembled matrixes. E, F) The expression of CD206 and CD86
   in Control group, P‐type, R‐type and M‐type matrixes was detected by
   flow cytometry. G) Quantitative analysis showed the results of flow
   cytometry.

   The pro‐inflammatory macrophages (M1 macrophages) express high levels
   of iNOS while the anti‐inflammatory macrophages (M2 macrophages)
   express high levels of CD163.^[ [99]^5 , [100]^21 ^] Immunofluorescence
   was used to analyze the changes in cell phenotype and the expression of
   membrane protein after the polarization of macrophages.
   Immunofluorescence staining showed the number of macrophages with the
   M1 surface marker iNOS in the Control group was much higher than those
   in the three supramolecular self‐assembling matrixes groups. In
   contrast, the number of macrophages with the M2 surface marker CD163 in
   three supramolecular self‐assembling matrixes groups was much higher
   than those in the Control group (Figure [101]S16, Supporting
   Information). Western blot proved that the proportion of M2 surface
   markers in M‐type self‐assembling matrixes was much higher than that in
   the R‐type and P‐type self‐assembling matrixes, respectively
   (Figure [102]2D). Moreover, the flow cytometry assay confirmed that the
   expression of CD163 and CD206 were significantly upregulated in M‐type
   self‐assembling matrixes than in R‐type and P‐type self‐assembling
   matrixes. (Figure [103]2F; Figure [104]S17, Supporting Information).
   The expression of CD86 in M‐type self‐assembling matrixes was
   significantly down‐regulated compared with that in R‐type and P‐type
   self‐assembling matrixes (Figure [105]2E,F). The above data indicate
   the M‐type self‐assembling matrixes can up‐regulate M2‐related membrane
   protein expression and promote M2 polarization in macrophages more than
   the P‐type and R‐type matrixes.

   Reduced release of pro‐inflammatory cytokines can be regarded as one of
   the types of evidence for the phenotypic transformation of
   macrophages.^[ [106]^5a ^] To further investigate the difference in the
   immune response produced by the chirality‐dependent macrophage
   polarization, gene expression profiling analysis was performed. Cluster
   analysis displayed that macrophages had overexpressed and inhibitory
   genes in all groups (Figure [107]3A). The expression of adhesion genes
   and M2‐related genes in M‐type, R‐type, and P‐type self‐assembling
   matrixes were up‐regulated while the expression of most of the
   M1‐related genes was down‐regulated. Compared with the P‐type and
   R‐type self‐assembling matrixes, the M‐type self‐assembling matrixes
   had a greater degree of genetic difference (Figure [108]S18, Supporting
   Information). The results of gene ontology analysis and pathway
   enrichment analysis showed that the TNF signaling pathway, mTOR
   signaling pathway, NF‐kappa B signaling pathway, and Cytokine‐Cytokine
   receptor interaction signaling pathway were activated
   (Figures [109]S19–S22, Supporting Information). Gene enrichment
   analysis presented that integrin binding‐related genes associated with
   M2 polarization were enriched in M‐type self‐assembling matrixes
   (Figure [110]3B), which was consistent with the results in the above
   gene expression profile analysis. The volcano map showed that compared
   with the control group, the expression of ARG1 in P‐type, R‐type, and
   M‐type groups was gradually up‐regulated. The Venn map showed M‐type,
   P‐type, and R‐type were all different compared with the control group,
   among that the M‐ type had the largest number of differential genes
   (Figure [111]S23 Supporting Information). These results suggest that
   the macrophages in the M‐type self‐assembling matrixes had a stronger
   immune activation effect than those in the R‐type and P‐type
   self‐assembling matrixes.

Figure 3.

   Figure 3
   [112]Open in a new tab

   A) A heatmap representing major altered genes in three datasets of
   macrophages treated with P‐type, R‐type, and M‐type matrixes.
   Macrophages with LPS+IFN‐γ treatment were used as a negative control.
   The color bar indicates normalized z‐score intensity‐based absolute
   quantification. B) Gene enrichment analysis showed that M‐type matrixes
   were enriched in integrin binding‐related genes related to M2
   polarization. C) Schematic diagram of specific molecular signal
   transduction in M2 macrophages induced by chiral matrixes. D) Western
   blotting analysis revealed the expression of mechanoinductive and
   transducing proteins (P‐PI3K, P‐FAK, P‐AKT) in M‐type, P‐type and
   R‐type matrixes. E) Immunofluorescence analysis showed the expression
   of CD44 and ITGB in M‐type, P‐type and R‐type matrixes, scale :5 µm. F)
   Quantitative RT‐qPCR analysis revealed the expression of typical
   M2‐related genes and mechanism pathway genes in macrophages in M‐type,
   P‐type and R‐type matrixes.

   Recent research has reported ECM assembling and remodeling,
   cytoskeletal remodeling, and tissue repair were closely related to the
   polarization of M2 macrophages,^[ [113]^5 , [114]^21 ^] which inspired
   us to explore the underlying molecular mechanism of macrophage
   polarization induced by supramolecular chirality. We referred to the M2
   polarization signaling pathway of macrophages and found most likely
   signaling pathway involved in adhesion‐induced polarization might be
   the PI3K/AKT/mTOR signal axis.^[ [115]^22 ^] Western blotting results
   showed the phosphorylation of PI3K, FAK, and Akt1 was found to be
   significantly higher in the M‐type self‐assembling matrixes than in
   P‐type and R‐type self‐assembling matrixes (Figure [116]3D).
   Immunofluorescence staining exhibited that the expression of integrin
   β1 (Itgβ1) and CD44 were up‐regulated in the P‐type, M‐type, and R‐type
   self‐assembling matrixes, respectively. While the higher expression of
   Itgβ1 and CD44 in M‐type matrixes than in P‐type and R‐type matrixes
   (Figure [117]3E) indicated that more focal adhesions of cellular were
   formed in the M‐type matrixes.^[ [118]^23 ^] Therefore, we proposed 3D
   chiral self‐assembling matrixes microenvironment potentially affects
   FAK/Rho/ROCK signaling pathways activated by integrin and Rho/ROCK
   signaling pathways activated by CD44, which are involved in upstream
   signal transduction of the PI3K/AKT/mTOR signal axis
   (Figure [119]3C).^[ [120]^24 ^]

   To test the hypothesis that the expression pattern of polarized genes
   in macrophages mediated by the chirality of self‐assembling matrixes
   may be related to adhesion and cytoskeletal remodeling,^[ [121]^30 ^]
   Subsequently, we focused on the expression of genes that related to
   adhesion and cytoskeletal remodeling. RT‐qPCR assays were performed to
   quantify the expression of mechano‐related genes in macrophages
   cultured with P‐type, M‐type, and R‐type self‐assembling matrixes,
   respectively. The expression of typical M2‐related genes and the
   mechanism pathway genes were significantly up‐regulated in the M‐type
   self‐assembling matrixes compared to the P‐type and R‐type
   self‐assembling matrixes (Figure [122]3F). Y27632 was used to
   antagonize the downstream signal of ROCK and BEZ235, as well as
   specifically block the downstream signal of PI3K and mTOR.^[ [123]^25
   ^] There were no significant differences in the expression of
   M2‐related genes (ITGB1, CD44, RAC1, Vinculin, SOCS1, RhoA, mTOR, and
   IRF4) in P‐type, R‐type, and M‐type self‐assembling matrixes
   (Figure [124]S24, Supporting Information), respectively. Moreover, the
   results that Y27632 inhibits M2 polarization of macrophages in chiral
   self‐assembling matrixes, indicate FAK/Rho/ROCK signal axe plays a key
   role in supramolecular chirality‐dependent macrophage polarization.
   Above them, our findings suggest that chirality‐dependent macrophage
   polarization is initiated by divergent mechanical sensations. The
   cell‐matrix interactions do not occur directly but through
   ligand‐integrin receptor binding.^[ [125]^30 ^] We characterized the
   affinity of matrix chirality to fibronectin (FN), the absorption assay
   showed significantly increased immunofluorescence staining of FN in the
   M‐type matrix, which indicated that the M‐type matrix facilitated an
   increased amount of ligand adsorption than the P‐type and R‐type
   matrixes (Figure [126]S25, Supporting Information).

2.5. Classic Molecular Dynamic Simulation

   Classic molecular dynamic (MD) simulations were explored to gain more
   insight into the initial enantioselectivity between different chiral
   molecules (L‐BA/D‐BA) and FnIII7‐10 (as the key cell‐binding domains of
   fibronectin (FN), which plays a vital character in mediating
   cell‐material interaction^[ [127]^7 , [128]^26 ^]). The snapshots
   demonstrated that the binding of D‐BA and FnIII7‐10 could reach a
   stable stage within 50 ns, while the interaction between L‐BA and
   FnIII7‐10 could not reach an equilibrium stage until 50 ns (Figure
   [129]4A). This result suggested that FnIII7‐10 had a higher
   stereo‐affinity for D‐BA than L‐BA. The representative 3D and 2D
   structures of the adhesive domain complex indicate that greater
   coordination is achieved between D‐BA and FnIII7‐10 than between L‐BA
   and FNIII7‐10 (Figure [130]4B).^[ [131]^7c ^] The smaller
   root‐mean‐square deviation (RMSD) value of FnIII7‐10 in the D‐BA/L‐BA
   and FnIII7‐10 complex (0.82 nm for the L‐BA and FnIII7‐10 complex vs.
   0.65 nm for the D‐BA and FnIII7‐10 complex; Figure [132]4C) and the
   smaller root‐mean‐square fluctuation (RMSF) value of FnIII7‐10 in the
   D‐BA/L‐BA and FnIII7‐10 complex (0.60 for D‐BA and FnIII7‐10 complex
   vs. 0.42 nm for the L‐BA and FnIII7‐10 complex; Figure [133]4D)
   indicated that the high interdomain elasticity and flexibility of the
   RGD configuration in the D‐BA and FnIII7‐10 complex could facilitate
   integrins binding to the initial downstream cell response.^[ [134]^7c
   ^] The lower interaction energy between FnIII7 and D‐BA (−105 kJ mol^−1
   for L‐BA and FnIII7‐10 complex, −181 kJ mol^−1 for D‐BA and FnIII7‐10
   complex, Figure [135]4E) further confirmed stronger binding of
   FnIII7‐10 on D‐BA than on L‐BA. Taken together, the MD simulation data
   confirmed a greater stereo‐affinity interaction of FnIII7‐10 to D‐BA
   than to L‐BA (Figure [136]S26, Supporting Information), ultimately
   leading to the effects of chirality regulating the polarization of
   macrophage.

Figure 4.

   Figure 4
   [137]Open in a new tab

   MD simulations proved the stereoselective interaction between FnIII7‐10
   and L‐BA/D‐BA molecule. A) Snapshots. B) The binding structure in
   equilibrium between FnIII7‐10 and L‐BA/D‐BA molecule. C) The
   root‐mean‐square deviation (RMSD) D) and root‐mean‐square fluctuation
   (RMSF) values of the RGD residues in the FnIII7‐10 and L‐BA/D‐BA
   molecule. E) Computed average interaction energies between FnIII7‐10
   and L‐BA/D‐BA molecule.

2.6. 3D Chiral Self‐Assembling Matrixes on Enhance Repair of MI

   Macrophages are the primary responder cells of the immune system which
   infiltrate the infarcted myocardium for cardiac repair. After MI
   occurs, M1 macrophages mainly remove debris, necrotic cardiomyocytes
   (CMs), and apoptotic neutrophils, thereby causing inflammation. In the
   subsequent inflammation resolution stage, M2 macrophages dominantly
   promote tissue reconstruction and regeneration. During this process,
   the regulation of macrophage phenotype and inflammatory responses is
   critical for cardiac recovery.^[ [138]^19 , [139]^27 ^] Encourage by
   the promising in vitro results, we carried out the in vivo experiments
   employing the MI model of mice. A control group with PBS and
   self‐assembling matrixes groups were implanted into the MI region of
   mice, respectively, to study the regulation of macrophage phenotype and
   inflammatory responses in vivo (Figure [140]5A). The proportion of
   CD206‐positive and CD86‐positive macrophages in the heart muscle was
   then assessed on days 1 and 3. Tissue immunofluorescence co‐staining
   and quantitative analysis showed that the proportion of CD206‐positive
   macrophages in M‐type self‐assembled matrices was significantly
   increased compared with P‐type and R‐type self‐assembled matrices
   (Figure [141]5B,C), while the proportion of CD86‐positive macrophages
   in M‐type self‐assembled matrices was significantly decreased compared
   with P‐type and R‐type self‐assembled matrices (Figure [142]5B,D). In
   addition, by evaluating the inflammatory factor iNOS and CD163 surface
   protein in the myocardial, the results showed that the proportion of
   CD163‐positive macrophages in the M‐type self‐assembly matrix was
   significantly increased compared with the P‐type and R‐type
   self‐assembly matrix, while the proportion of iNOS positive expression
   in the M‐type self‐assembly matrix was significantly decreased compared
   with the P‐type and R‐type self‐assembly matrix (Figure [143]S27,
   Supporting Information). The comparison of biostabilities of L‐BA and
   D‐BA showed that the degradation rate of assemblies dramatically
   depended on their supramolecular chirality, namely, M‐type matrixes
   seemed to be more resistant to hydrolyzed than the P‐type matrix, which
   has been reported previously by our group.^[ [144]^28 ^] Additionally,
   quantitative IR783B‐NHS‐doped three self‐assembling matrixes were
   injected into the MI region of mice, respectively, and effective
   accumulation and degradation of the substances in the infarcted heart
   were observed (Figure [145]S28, Supporting Information), which M‐type
   matrixes seemed to be more resistant to be hydrolyzed than the P‐type
   and R‐type matrixes. Moreover, the results of the detection of serum
   liver and kidney function in mice showed that the injection of chiral
   self‐assembling matrixes has no significant effect on the liver and
   kidney function of mice (Figure [146]S29, Supporting Information).
   These data suggested the chiral self‐assembling matrixes have good
   biosafety. MI can cause severe pathological reactions, mainly
   ventricular remodeling thus changes in left ventricular (LV) thickness
   and mass.^[ [147]^29 ^] Subsequently, cardiac functions were evaluated
   28 days after MI to verify the effect of chiral self‐assembling
   matrixes on the MI model of mice. The left ventricular function of
   different types of mice was analyzed by echocardiography
   (Figure [148]5E). Compared with the sham operation group without MI,
   the left ventricular diameter of all MI mice was increased, and the
   left ventricular systolic function was decreased. The M‐type
   self‐assembling matrixes showed the smallest increase in left
   ventricular internal diameter at the end‐diastole (LVIDd) and left
   ventricular internal diameter at the end‐systole (LVIDs) in all
   experiment groups. Compared with Sham, MI, P‐type, and R‐type
   self‐assembling matrixes, the infarcted hearts treated with the M‐type
   self‐assembling matrixes exhibited a significantly increased left
   ventricular ejection fraction (LVEF) and left ventricular fraction
   shortening (LVFS) (Figure [149]5F–I).

Figure 5.

   Figure 5
   [150]Open in a new tab

   A) Time axis of the myocardial repair process detected in the MI region
   with implanting chiral matrixes. B) Immunofluorescence staining showed
   that the M1 macrophage‐related surface marker CD86 and M2
   macrophage‐related surface marker CD206 were co‐stained in MI region
   after implantation of M‐type, P‐type and R‐type matrixes (scale
   :100 µm). C,D) Quantitative analysis showed the expression of CD206 and
   CD86 in M‐type, P‐type and R‐type matrixes. E) Echocardiography images
   of MI model of mice in each group 28 days after implanting chiral
   matrixes. F–I) Left ventricular internal diameter at the end‐systole
   (LVIDs), left ventricular internal diameter at the end‐diastole
   (LVIDd), left ventricular ejection fraction (LVEF), and left
   ventricular fraction shortening (LVFS) at 28 days after implanting
   chiral matrixes.

   In the pathologic process of MI, excessive degradation of ECM adversely
   affects cardiac function.^[ [151]^30 ^] Herein, 3D chiral
   self‐assembling matrixes provide structural support for the infarcted
   tissue microenvironment and further, promote the repairing of the MI
   region. Masson is trichromatic and Sirius red staining was used to
   analyze the cardiac fibrosis after 28 days of treatment (Figure
   [152]6A). Compared with the MI group, the injection of chiral
   self‐assembling matrixes groups inhibited the thinning of the left
   ventricular (LV) wall, as well as the M‐type self‐assembling matrixes
   has the most significant inhibitory effects (Figure [153]6B).
   Quantitative analysis correspondingly showed the M‐type self‐assembling
   matrixes had the least thinning and fibrosis of the LV wall compared
   with MI, R‐type, and P‐type matrixes (Figure [154]6C,D). The
   regeneration of blood vessels is crucial in the process of repairing
   MI. Finally, an immunohistochemical examination was performed 4 weeks
   later, and tissue sections of the site of MI were taken to evaluate the
   degree of fibrosis, which CD31 and α‐SMA could as the index of cardiac
   regeneration. The expression of CD31 and α‐SMA was significantly higher
   in the M‐type self‐assembling matrixes than in the MI group, P‐type,
   and R‐type self‐assembling matrixes (Figure [155]6E). These results
   were consistent with the above echocardiographic data. Pathological
   diagnosis of hematoxylin‐eosin (HE) staining showed there was no
   significant toxicity to the heart of injected mice with different
   chiral self‐assembling matrixes treatments compared to control groups
   (Figure [156]S30, Supporting Information). In addition, collagen is
   mainly distributed in IZ and BZ. On day 28, the levels of P‐type,
   R‐type, and M‐type group type III collagen and type I collagen were
   higher than those in the MI group. We further detected collagen
   subtypes by immunostaining. The ratio of type III/type I collagen in
   M‐type was significantly higher than in the other groups
   (Figure [157]6F), IZ (Figure [158]6G), and BZ (Figure [159]6H). The
   implanted bionic 3D chiral self‐assembling matrixes developed in this
   work could achieve improved cardiac function without the assistance of
   stem cells, drugs, or genes. The significantly improved restoration of
   cardiac function further confirms the advantages and application
   prospects of chiral self‐assembling matrixes in cardiac repair.

Figure 6.

   Figure 6
   [160]Open in a new tab

   A) Panorama of Masson's trichromatic horizontal heart images at 28 days
   postoperatively. B) Masson trichromatic and Sirius red staining were
   used to evaluate the anti‐fibrosis effect of chiral matrixes, Scale
   bar: 50 µm. C) Percentage of the fibrotic area in each left ventricle
   after coronary artery ligation. D) Quantitative analysis of myocardial
   remodeling by ventricular wall thickness in infarct area. E) α‐SMA and
   CD31 immunofluorescence co‐staining were performed 28 days after
   implanting chiral matrixes, Scale bar: 50 µm. F). On day 28, the
   distribution of type III collagen and type I collagen levels in IZ
   region in MI, M‐type, P‐type and R‐type matrixes groups. G‐H)
   Quantitative analysis of the ratio of type III collagen to type I
   collagen in IZ and BZ region was measured in the MI, M‐type, P‐type and
   R‐type matrixes groups, Scale bar: 3 mm and 400 µm.

3. Conclusion

   In summary, bionic 3D ECM based on chiral supramolecular matrixes is
   employed to manipulate the regulation of macrophages’ immune response
   as well as enhance the repair of myocardial infarction. Results imply
   that M‐type self‐assembling matrixes significantly inhibit inflammation
   and promote damaged myocardial repair by upregulating M2 macrophage
   polarization and downstream immune signaling compared with P‐type and
   R‐type self‐assembling matrixes. The corresponding infarct size was
   reduced, and the thickness of the left ventricular wall was increased
   substantially in the MI model with implanting chiral self‐assembling
   matrixes. The significantly improved recovery of cardiac function
   further verified the excellent therapeutic efficacy of M‐type matrixes
   in cardiac repair. MD data demonstrate that M‐type self‐assembling
   matrixes display higher stereo‐affinity to cellular binding compared to
   P‐type and R‐type self‐assembling matrixes, which enhance the
   clustering of mechanosensitive Itgβ1, activate FAK and Rho‐ROCK
   signaling pathway, as well as downstream PI3K/AKT/mTOR signaling axes
   to promote M2 polarization. This work revealed that the 3D
   supramolecular chirality microenvironment of biomaterials could serve
   as a promising regulator to affect macrophage behavioral responses and
   provide insights for the design of immunoregulatory biomaterials for
   the intended use.

Conflict of Interest

   The authors declare no conflict of interest.

Supporting information

   Supporting Information
   [161]Click here for additional data file.^ (2.3MB, pdf)

Acknowledgements