Rachel A.
Morrison
a,
Sarah
Brookes
a,
Theodore J.
Puls
b,
Abigail
Cox
c,
Hongyu
Gao
d,
Yunlong
Liu
d and
Sherry L.
Voytik-Harbin
*ae
aWeldon School of Biomedical Engineering, Purdue University, West Lafayette, IN 47907, USA. E-mail: harbins@purdue.edu
bGeniPhys Inc., Zionsville, IN 46077, USA
cDepartment of Comparative Pathobiology, Purdue University, West Lafayette, IN 47907, USA
dDepartment of Medical & Molecular Genetics, Indiana University School of Medicine, Indianapolis, IN 46202, USA
eDepartment of Basic Medical Sciences, Purdue University, West Lafayette, IN 47907, USA
First published on 21st March 2023
The efficacy and longevity of medical implants and devices is largely determined by the host immune response, which extends along a continuum from pro-inflammatory/pro-fibrotic to anti-inflammatory/pro-regenerative. Using a rat subcutaneous implantation model, along with histological and transcriptomics analyses, we characterized the tissue response to a collagen polymeric scaffold fabricated from polymerizable type I oligomeric collagen (Oligomer) in comparison to commercial synthetic and collagen-based products. In contrast to commercial biomaterials, no evidence of an immune-mediated foreign body reaction, fibrosis, or bioresorption was observed with Oligomer scaffolds for beyond 60 days. Oligomer scaffolds were noninflammatory, eliciting minimal innate inflammation and immune cell accumulation similar to sham surgical controls. Genes associated with Th2 and regulatory T cells were instead upregulated, implying a novel pathway to immune tolerance and regenerative remodeling for biomaterials.
It is well established that nature-derived or synthetic materials not inherent to the host are recognized as foreign objects by the immune system. Also recognized as foreign are remnant cellular debris and modified biomolecules (e.g., denatured, chemically modified), which are often associated with tissue-derived or other biological materials. Implantation of these foreign materials yields well-characterized FBRs, which are initiated by the immune system at the tissue-implant interface.10–14 Mechanistically, FBRs are often described as having multiple sequential but overlapping phases, including protein adsorption, acute inflammation, chronic inflammation, foreign body giant cell (FBGC) formation, and fibrosis/fibrous capsule formation.2,15 This reaction initiates immediately upon tissue contact, with blood-derived and interstitial protein adsorption to the material surface and associated complement activation (the first step of innate immunity). In turn, chemokines and other chemoattractants guide cells of the innate or non-specific immune system, notably neutrophils, mast cells, dendritic cells, monocytes, and macrophages, to accumulate at the implant site. Continued activation of macrophages, as well as their fusion to form multi-nucleated FBGCs, leads to a state of chronic inflammation, with engagement of adaptive immune cell populations (e.g., lymphocytes including cytotoxic T cells, T helper cells, and B cells), recruitment of additional cell types (e.g., fibroblasts, myofibroblasts), and ultimately deposition of dense fibrous connective tissue (fibrotic capsule) that walls off the foreign material. This process is further exacerbated by phagocytosis, where neutrophils, macrophages, and FBGCs work to actively engulf and break down foreign materials, releasing reactive oxygen species and matrix degradative enzymes.
While FBGCs, fibrosis, and fibrous capsule formation are commonly reported outcomes following biomaterial implantation, immune cell phenotypes and cellular signaling pathways, material degradation or bioresorption, thickness of the fibrous capsule, and consequences of the inflammatory response vary with biomaterial type and implantation microenvironment.16 Early mechanistic studies focusing on the innate immune system showed that macrophage polarization and phenotype play a major role in determining biomaterial FBRs and outcomes.17 However, growing evidence now suggests that biomaterial tissue reaction initiation and resolution are driven by crosstalk between both innate and adaptive immune cell players. More specifically, macrophages, depending upon their phenotype and cytokine/chemokine profile, recruit and differentially engage CD8+ cytotoxic T cells and various CD4+ T helper cells.18–20 Although T helper cells are activated through antigen presentation by macrophages and dendritic cells, the biomaterial microenvironment greatly influences subtype, including differentiation along pro-inflammatory Th1/Th17 pathways or Th2/regulatory T (Treg) pathways that suppress inflammation and promote immune homeostasis.21 Given that both innate and adaptive immune cells operate along phenotypic continuums extending from pro-inflammatory/pro-fibrotic to anti-inflammatory/pro-regenerative, it is this delicate balance that ultimately determines biomaterial tissue response and healing outcomes.21 To influence this balance, many current biomaterial design strategies focus on modulating the tissue response by manipulating specific physicochemical properties, including size and shape,22 pore size,23 surface properties (e.g., hydrophilicity,24 surface charge,25 chemical functionalization26,27), mechanical properties,28,29 and degradability.30 To further augment biomaterials and promote favorable host reactions, researchers have also commonly explored controlled release of immunomodulatory agents (e.g., nitric oxide,31 IL-4,32 IFNγ32) and surface functionalization with bioactive coatings (e.g., IL-4,33 ECM components17). Overall, these contemporary approaches highlight the importance of tissue and immune response considerations in the design and translation of next-generation biomaterials.
Given that type I collagen represents the major structural and mechanical framework of tissues, numerous collagen-based biomaterials exist within the market today. Scaffolds fashioned from decellularized animal and human tissues represent a prominent type of collagen-based product. To create these products, tissues are processed to eliminate cellular components (via physical, chemical, and/or enzymatic methods) while maintaining the complex molecular composition, architecture, mechanical properties, and bioactivity inherent to the ECM.4 Other collagen-based products, including freeze-dried collagen sponges, are fabricated from more refined starting materials, namely tissue particulate consisting of microfibrillar collagen or collagen hydrolysates, with hydrolysates representing denatured and/or enzymatically treated collagen molecules cleaved into shorter chains of amino acids.4,34 All commercial collagen-based materials, regardless of source material and method of manufacture, engage innate and adaptive immune systems, resulting in a distinct FBR, commonly referenced as constructive remodeling, that features immune-mediated material bioresorption (phagocytosis and proteolytic breakdown) and fibrous tissue formation and remodeling.35–38 Depending on a number of factors (e.g., source material, extent of decellularization, processing and exogenous crosslinking, material configuration, implant location), tissue responses can vary between collagen-based products, affecting macrophage polarization, number and distribution of FBGCs, degradation rates, and tissue response outcomes.39 Overall, the complexity of present-day collagen-based biomaterials and their associated design poses ongoing challenges to the elucidation of their mechanism of action and predictive modulation of their immune response.6
Our design strategy for next-generation engineered collagen materials employs a unique starting material consisting of a polymerizable type I collagen protein known as oligomeric collagen (Oligomer; also known as Collymer™). Oligomer is readily extracted and purified from porcine dermis, as well as other collagen-containing tissues from animal and human sources, thereby eliminating cellular and other immunogenic components.40,41 When acidic solutions of Oligomer are brought to physiologic conditions (e.g., pH, ionic strength, and electrolyte composition) by mixing with a buffer, polymerization is initiated, giving rise to highly-interconnected and physically-stable fibrillar scaffolds, without use of exogenous additives or crosslinking agents.40–42 To date, engineering design tools (e.g., computational modeling) and scalable fabrication methods (e.g., compression densification, extrusion) have been applied for customization of wide variety of collagen polymeric material formats, including in situ scaffold-forming formulations, high-strength thin materials (e.g., sheets, cylinders), compression-resistant constructs, and materials with gradient or aligned fibrillar microstructures.42–50In vivo, these biomaterials exhibit a unique noninflammatory, regenerative healing response with no FBR or immune-mediated bioresorption, which has been documented by implanting various material formats in a number of different anatomical locations, including dermis, subcutaneous tissue, skeletal muscle, intraperitoneal cavity, larynx, and breast.44,46–50
To extend this work and continue to foster the bench-to-bedside translation of this biopolymer technology, we conducted a rat subcutaneous implantation study (Fig. 1) to systematically and mechanistically define the immune response and tissue reaction of a high-density Oligomer scaffold compared to commercial biomaterials with well characterized and distinguishable tissue responses, ranging from rapidly bioresorbable to nonresorbable. HeliCote® was selected as a representative bioresorbable collagen sponge material that is routinely used for wound and dental applications. HeliCote is bioengineered from bovine tendon microfibrillar collagen particulate that has been lyophilized and subjected to dehydrothermal crosslinking, yielding a material that bioresorbs within 10 to 14 days according to manufacturer specifications. Prolene® mesh, on the other hand, was selected as an exemplar nondegradable biomaterial synthesized from the thermoplastic polymer polypropylene. Because of superior burst and tensile strength, Prolene meshes are routinely used to reinforce tissues, including those associated with ventral and inguinal hernias, pelvic organ prolapses, and stress urinary incontinence. Study outcomes, which were measured at 3-, 7-, and 14-days following implantation, included gross and histological assessments of tissue reactions, immune cell identification via immunostaining, and RNA-sequencing (RNA-seq) transcriptomics analysis, which provided expression levels of roughly 12,000 gene products. Gross and histological assessments were also performed 60 days following Oligomer scaffold implantation to define more long-term tissue responses. Collectively, results from this study further demonstrate that engineered collagen polymeric materials fashioned from Oligomer elicit a unique regenerative mechanism of action that benefits from host immune tolerance. Additionally, results suggest new paradigms where inflammation and immune-mediated bioresorption of collagen materials are not required for desirable endogenous tissue regeneration outcomes.
A detailed histological analysis of H&E- (Fig. 2B and Fig. S1B, C†) and MTC- (Fig. S2†) stained sections provided additional insights into the immunophenotypic response of Oligomer scaffolds compared to conventional implantable materials. Histologically, the Oligomer scaffold appeared as a homogenous and bright pink (eosinophilic) material that stained similarly to subcutaneous collagen (Fig. 2B and Fig. S1B, C†). At day 3, a relatively small number of mononuclear leukocytes and fibroblasts (1–2 cells thick) were seen accumulating around the Oligomer scaffold surface; however, by day 7, this cell density declined to match densities and distributions observed within the sham surgical site (Fig. 2B). Importantly, there were no FBGCs observed at any time point, demonstrating no evidence of chronic inflammation. Additionally, cell infiltration of the Oligomer scaffold was minimal, and no scaffold degradation or abnormal histopathology of the surrounding subcutaneous tissue was noted. Similar findings were observed for 60-day Oligomer scaffold implants, with no evidence of a long-term, deleterious tissue reaction grossly or histologically (Fig. S1†).
By contrast, commercial collagen and mesh showed an increase in cellularity over time, with significant infiltration around and within the materials (Fig. 2B and Fig. S2†). At 3 days, the commercial collagen appeared as an integrated meshwork of variably-sized, pink-staining particulate, which was well circumscribed and encapsulated with a robust mixture of inflammatory cells, fibroblasts, and collagen. Gradual resorption of the commercial collagen over the 14-day study period was apparent, which was accompanied by an increase in fibrous connective tissue deposition (Fig. 2B and Fig. S2†). Histological analysis also revealed progressive cell infiltration and active proteolytic degradation and phagocytosis by macrophages and FBGCs (Fig. 2B, arrows and Fig. S2†). Additionally, a moderate inflammatory reaction, characterized by infiltrating lymphocytes, plasma cells, and macrophages, was observed within the nearby subcutaneous tissue. An inflammatory-mediated FBR was also observed with the mesh. Histologically, the mesh appeared as multiple unstained spaces, which became surrounded by a mix of inflammatory cells over time and increasing amounts of fibrous connective tissue (Fig. 2B and Fig. S2†). Circumscribing the implant was a mix of neutrophils, macrophages, lymphocytes, plasma cells, and abundant plump fibroblasts. By 7 days post implantation, FBGCs were evident, with increasing numbers noted at day 14 (Fig. 2B, arrows). At all time points, a significant inflammatory reaction was noted adjacent to the mesh. This reaction was seen extending into the overlying panniculus carnosus muscle on day 14. Additionally, by 14 days, a dense capsule of fibrous connective tissue surrounded the mesh implant, indicative of a classic FBR (Fig. 2A and Fig. S2†).
Because macrophages are considered major players in the FBR to biomaterials,32,33 immunostaining for the pan-macrophage surface marker CD68 (Fig. 3) was performed to support further spatial and temporal cellular characterization. As shown in Fig. 3, modest increases in CD68+ cells were noted within sham and Oligomer sites at day 3, with levels decreasing to normal subcutaneous levels by 14 days. On the other hand, increased numbers of CD68+ macrophages were found associated with commercial collagen and mesh implants, with the apparent density and depth of material infiltration increasing over the 14-day period (Fig. 3). For commercial collagen, high numbers of CD68+ macrophages were seen accumulating at the material-tissue interface at day 3, with significant but somewhat lesser numbers distributed throughout the adjacent subcutaneous tissue and penetrating into the material. CD68+ staining remained high at days 7 and 14 but showed more uniform distribution, as macrophages progressively infiltrated the material over time (Fig. 3). On the other hand, CD68+ macrophages formed a thin layer along the mesh surface on day 3, with even greater numbers distributed within the surrounding stroma (Fig. 3). These macrophages appeared to increase in density over the duration of the study. Overall, gross, histological, and immunostaining results suggested that the three materials elicited strikingly different tissue reactions and cellular responses, with the Oligomer scaffold showing an uncommon tissue response that was not driven by macrophages and showed no significant foreign body response or bioresorption.
Principal component analysis (PCA) at each time point revealed different group profiles, with distinct clusters converging by day 14 (Fig. S3†). Given that day 14 yielded the most stable, distinct responses, this time point was the primary focus for further volcano plot and KEGG analyses. Volcano plots, which visualize differentially expressed genes (DEGs), revealed little difference between sham and normal tissue at day 14, with only 80 differentially expressed genes (DEGs) identified under the condition of |log2 fold changes| > 1 and FDR < 0.01 (Fig. 4A). For Oligomer scaffold, 856 DEGs were identified, which was substantially less than commercial collagen (2162 DEGs) and mesh (1732 DEGs) (Fig. 4B–D). KEGG overenrichment analyses further informed the transcriptomic analysis through quantification of the pathways most significantly associated with DEGs for each material, and the statistical relevance of the relation to that pathway was also evaluated (adjusted p-values <0.05). As shown in Fig. 4E–G, all three materials showed engagement of several immune-related pathways, such as cytokine–cytokine receptor interaction, hematopoietic cell lineage, cell adhesion molecules, complement and coagulation cascades, and chemokine signaling pathways. However, the number of DEGs within these pathways was lowest for Oligomer scaffold. Interestingly, KEGG analysis also suggested differential involvement of T cell mediated adaptive immune pathways for Oligomer scaffold and commercial collagen, while the mesh showed primarily innate immunity pathway activation (Fig. 4E–G). Mesh implantation also led to overenrichment of the ECM-receptor pathway, which is consistent with fibrosis (Fig. 4G).58
Informed by KEGG analysis outcomes, more detailed evaluations of gene subsets were conducted, beginning with those related to innate immune cell (e.g., neutrophil and macrophage) phenotype and activation. As shown in Fig. 5A, the profiles for Oligomer scaffold and sham were most similar and paralleled histological observations. Both showed transient upregulation of several innate inflammation genes, including the macrophage associated genes Ccl9, Csf2, Spp1, and Il1rn, with levels decreasing over the 14-day study period. A notable exception was Ccl17 and Ccl22, which for Oligomer scaffold maintained increased expression at 7 and 14 days (FDR < 0.01; Ccl17: log2FC = 5.3 at 7 days, log2FC = 4.5 at 14 days; Ccl22: log2FC = 5.4 at 7 days, log2FC = 4.9 at 14 days). Interestingly, these genes encode chemokines known to recruit Tregs.7 Additionally, high and sustained expression of Cxcl6, a chemoattractant of neutrophils and potent angiogenic chemokine,59 was observed for Oligomer scaffold but not sham (Fig. 5A). By contrast, commercial collagen and mesh showed broad upregulation of innate inflammation genes with no discernible temporal trends (Fig. 5A). At 14 days, commercial collagen showed high expression of genes associated with neutrophil chemotaxis, angiogenesis, and macrophage lineage, activation, and chemotaxis (FDR < 0.01: Cxcl6 log2FC = 10.0, Itgam log2FC = 3.4, Cd68 log2FC = 2.7, Cd86 log2FC = 1.7, Ccl2 log2FC = 1.5, Ccl3 log2FC = 2.2, Ccl7 log2FC = 2.0, Ccl9 log2FC = 7.5, Csf2 log2FC = 6.6, Spp1 = 9.7). Upregulation of macrophage phenotype and function genes was also observed with commercial collagen, including Ccl17, Ccl22, Ccl20, Il1rn, Arg1, Il10, Cxcl13, and Ifng, many of which foster engagement of the adaptive immune system.18,60 The heatmaps for mesh and commercial collagen were most similar with the exceptions that Itgam, Cd86, Ccl3, Ccl17, Ccl22, Ccl20, Il1rn, Arg1, and Cxcl1 showed somewhat lesser or no upregulation, and Ifng was downregulated for the mesh group (Fig. 5A).
Since FBGCs, which are a hallmark of FBRs, appeared to play a significant role in the tissue responses to commercial collagen and mesh but not Oligomer scaffold, signaling pathways associated with macrophage fusion and FBGC formation were also evaluated in detail. As shown in Fig. 5B, FBGC heatmaps for Oligomer scaffold and sham appeared largely similar, suggesting a transient response to the surgical procedure and confirming a lack of FBGC involvement. Commercial collagen, on the other hand, showed upregulation of all genes within the FBGC subset. In fact, factors known to stimulate macrophage fusion, namely Oscar (log2FC = 4.4 at 14 days), Csf2 (log2FC = 6.6 at 14 days), Mmp9 (log2FC = 6.4 at 14 days), and Spp1 (log2FC = 9.7 at 14 days), were amongst the most highly upregulated factors throughout the 14-day study period (Fig. 5B).13,58,61 Fusion process genes (Trem2, Tyrobrp, and Syk) and markers associated with FBGC and phagosomes (Cd44 and Itgb2) also showed moderate upregulation across all time points.10,13 Again, mesh exhibited a similar profile to commercial collagen, with the exceptions that lesser upregulation of Mmp9 (log2FC = 3.7 at 14 days) and Itgb2 (log2FC = 2.2 at 14 days) and no upregulation of Syk were observed. These differences are likely associated with divergent pathways and outcomes for FBRs, with commercial collagen and mesh yielding bioresorption and fibrotic encapsulation, respectively.
Additional mechanistic insight into the differential material responses was gleaned by analyzing gene subsets associated with ECM synthesis, assembly, and degradation. Once again, Oligomer scaffold and sham profiles were largely similar, with various genes showing modest and transient upregulation, likely owing to the surgical procedure (Fig. 5C). Transcriptomics showed that Oligomer scaffold, like sham, maintained baseline expression of matrix metalloproteinases, Mmp13 and Mmp9, and the tissue inhibitor of matrix metalloproteinases (Timp1), further confirming the lack of observed bioresorption, material degradation, and fibrotic response (Fig. 5E). As shown in Fig. 5C and 5E, only commercial collagen featured a robust and consistent upregulation of Mmp9 (log2FC = 6.5 at 14 days), Mmp13 (log2FC = 12.6 at 14 days), and Itgb2 (log2FC = 3.5 at 14 days). Upregulation of this gene combination corroborated the active phagocytosis and rapid ECM proteolysis observed histologically.13,58,62 Modest upregulation of genes favoring ECM deposition and maintenance was also observed, including Tgfb1 (log2FC = 1.1 at 14 days), P4ha1 (log2FC = 1.3 at 14 days), Plod2 (log2FC = 4.9 at 14 days), and Timp1 (log2FC = 2.6 at 14 days) (Fig. 5C). On the other hand, the mesh showed engagement of pathways consistent with fibrosis and fibrotic capsule formation,63,64 with strong upregulation of Plod2 (log2FC = 3.9 at 14 days), Thbs4 (log2FC = 5.0 at 14 days) as well as Timp1 (log2FC = 3.2 at 14 days) (Fig. 5C). This was complemented by more moderate upregulation of additional ECM synthesis and assembly genes,65 including Col3a1 (log2FC = 1.9 at 14 days), Col5a3 (log2FC = 2.3 at 14 days), Fn1 (log2FC = 1.4 at 14 days), P4ha1 (log2FC = 1.1 at 14 days), Adamts14 (log2FC = 2.0 at 14 days), and Lox (log2FC = 1.8 at 14 days) (Fig. 5C).
Given the known crosstalk between innate and adaptive immune components and the KEGG analyses implicating certain T lymphocyte pathways, a more in-depth analysis of adaptive immunity genes was performed. As shown in Fig. 5D, the groups showed distinct profiles, with Oligomer scaffold and commercial collagen exhibiting the most similarities. Both groups showed upregulation of a broad array of genes associated with T cell surface markers (CD3e), T cell activation and proliferation (Sit1, Icos, Cd27, Cd28, Pdcd1, Ctla4), and inflammation (Il18, Il21), especially at later time points (Fig. 5D). Additionally, both Oligomer and commercial collagen exhibited similar expression patterns for genes associated with Treg (Foxp3, Il2ra, Ccr4) and Th2 (Gata3, Il13) pathways (Olig: Foxp3 log2FC = 5.3, Il2ra log2FC = 4.7, Ccr4 log2FC = 6.2, Gata3 log2FC = 2.3, Il13 log2FC = 5.8 at 14 days; C-Col: Foxp3 log2FC = 6.6, Il2ra log2FC = 5.8, Ccr4 log2FC = 7.4, Gata3 log2FC = 3.0, Il13 log2FC = 7.1 at 14 days). Interestingly, heatmaps also revealed a number Th1/Th17 pathway genes that were only upregulated for commercial collagen (Fig. 5D). These factors included the characteristic cytokine of Th1 and Th17 cells,21Il17a (log2FC = 8.3 at 14 days); Irf4 (log2FC = 1.1 at 14 days), a transcription factor implicated in induction, amplification, and stabilization of the Th17 phenotype;66Ccr6 (log2FC = 2.5 at 14 days), a receptor upregulated on Th17 cells; Ccl20 (log2FC = 3.8 at 14 days), a known Th17 and Treg chemokine;67 and Ifng (log2FC = 2.0 at 14 days) and Tgfb1 (log2FC = 1.1 at 14 days). The heatmap for mesh (Fig. 5D), on the other hand, suggested only modest T cell engagement, which is consistent with previous reports for nonabsorbable materials.63,68,69 Similarly, sham implantation showed minimal engagement of the adaptive immune system (Fig. 5D). Collectively, transcriptomics data corroborated gross and histological findings and further documented that Oligomer-based materials have an uncommon tissue response and mode of action, where it engages cells to yield immune tolerance.
Given that transcriptomics data suggested a differentiating adaptive immune response for Oligomer scaffold, immunostaining for T cell markers CD3, CD4, CD8, Foxp3, and IL-17 was conducted to corroborate gene expression profiles. As shown in Fig. 6A, the 14-day tissue reaction associated with commercial collagen comprised the greatest number of CD3+ lymphocytes, with more moderate numbers identified around mesh and only a few associated with Oligomer scaffold and sham. Further analysis of subtypes showed that commercial collagen featured a mixed population of CD4+ T helper and CD8+ cytotoxic T cells (Fig. 6B and C), including Foxp3+ and IL-17+ T helper cell subtypes (Fig. 6D and E). Mesh also showed a mix of CD4+ and CD8+ cells (Fig. 6B and C), with lesser IL-17+ (Fig. 6E) and few Foxp3+ (Fig. 6D) cells. Of the limited lymphocytes associated with Oligomer scaffolds and sham, Oligomer showed a modest bias towards increased Foxp3+ Treg cells (Fig. 6). Overall, immunostaining results corroborated transcriptomic findings, confirming that Oligomer scaffold favored a subtle T cell response involving Tregs while commercial collagen and mesh induced robust and mixed adaptive immune cell engagement.
Neutrophils and macrophages are dominate innate immune cell players during both early and late phases of FBRs to implantable materials, including metals (e.g., titanium), synthetic polymers (e.g., silicone, polypropylene, polycaprolactone), biological materials (e.g., collagen sponges, decellularized ECM scaffolds), and nature-derived materials (e.g., alginate, silk).17,19,22,24,68 Targeted cell knockout or depletion studies show that macrophages are primary contributors to the fibrotic response,69 and both neutrophils and macrophages participate in degradation of collagen-based materials.71 Consistent with this notion as well as other published preclinical and human clinical studies,72–77 we show that the commercial mesh, when implanted subcutaneously, induced fibrotic capsule formation driven by large numbers of macrophages and FBGCs. The filamentous mesh structure supported rapid cell accumulation around and throughout the mesh as well as an extended inflammatory reaction that progressed outward into the surrounding normal tissue over time. Commercial collagen featured a similar temporal accumulation of macrophages and FBGCs; however, its material composition and porous microstructure led to a different cellular infiltration pattern. In this case, cells penetrated deeper into the material from the tissue interface over time, with progressive material degradation. A distinct response was observed with Oligomer scaffold, where macrophage activation appeared transient, similar to the sham control. Because macrophages did not appear to drive the Oligomer scaffold response, definition of macrophage polarization according to traditional M1/M2 types was considered outside the scope of this work. Broadly, transcriptomics showed common and persistent upregulation of Cxcl6, Cd68, Ccl9, Csf2, and Spp1 for both commercial materials, and increased expression of Cd86, Ccl2, Il1rn, Arg1, Ccl20, Ccl22, Il10, Cxcl13, and Ifng for commercial collagen. Such inflammatory cytokine and cell marker profiles are consistent with mixed or hybrid macrophage phenotypes as suggested by other published work.78,79 For both materials, FBGCs were readily identifiable at 7- and 14-day time points, with corroborating upregulation of macrophage fusion genes (Oscar, Csf2, Spp1, Trem2, and Tyrobp) and FBGC surface markers (Cd44). Mmp9, Mmp13, and Itgb2 expression was enhanced only for commercial collagen, which is consistent with the active phagocytosis and proteolysis of collagen materials.58,62
By contrast, Oligomer scaffold showed minimal innate immune cell accumulation and activation, with associated histological findings and heatmaps that largely mirrored the extent and temporal response of the sham surgical group. A notable exception was the early and persistent upregulation of Ccl22 and Ccl17, which was observed with Oligomer scaffold but not sham. These macrophage and dendritic cell derived chemokines are potent recruiters of Th2 and Treg cells, which are required for maintenance of self-tolerance, controlling inflammation, and preventing autoimmune diseases.80–83 It is reasonable to suppose that the apparent recognition of Oligomer scaffold as “self” by the host innate immune system is owing to its composition and structural features. Immune reactions elicited by conventional collagen materials, whether fabricated from decellularized tissues or processed tissue particulate (e.g., microfibrillar collagen, hydrolysates, gelatin) are largely driven by immunogenic cell remnants (e.g., membrane and intracellular), non-collagenous molecules, as well as leachables (e.g., detergents) and deleterious effects caused by material processing and formatting (e.g., denaturation, exogenous crosslinking, sterilization).39,84 By contrast, Oligomer materials are fabricated from highly-purified type I collagen protein building blocks, comprising full-length collagen molecules held together by the natural, non-reducible intermolecular crosslink chemistry known as histidinohydroxylysinonorleucine (HHL).41,85 The component collagen polypeptide chains (e.g., α1 and α2) and intermolecular crosslink chemistry are highly conserved across species,86 with reported values for porcine and human sequence homologies being 97% and 94% for component α1 and α2 chains, respectively.87 Relevant to the present study, sequence homology between porcine and rat collagen is slightly less but still high, with values of 91% for the α1 chain and 90% for α2 chain. Further, the structure of Oligomer scaffolds closely resembles native collagen found within tissue ECMs,42 which likely contributes to its noninflammatory response. The molecular self-assembly (polymerization) of Oligomer occurs at physiologically-relevant pH and ionic strength and is driven by hydrophobic and electrostatic interactions, yielding highly interconnected fibrillar collagen materials.41 The component D-banded fibrils, with roughly 67 nm periodicity, are similar to those found in connective tissues,42 facilitating cellular recognition, macromolecular interactions, and mechanochemical signaling. In summary, it is reasonable to suppose that tissue response differences between Oligomer scaffold and commercial collagen are largely driven by differences in both molecular composition and multi-scale structural features.
While the innate immune system plays a critical role in biomaterial FBRs, more recent investigations have focused on adaptive immune system participation (see Adusei et al., 202121 for recent review). In short, stimulation of lymphocyte migration, proliferation, and downstream maturation occurs following innate immunity engagement through antigen-presenting dendritic cells and macrophages, along with secreted cytokines. Integrative signaling between innate and adaptive immune components, in turn, drives biomaterial responses, ranging from tissue regeneration to tissue repair (fibrosis) to chronic inflammation.2,16 In the present study, transcriptomics data indicated adaptive immune system engagement within the first week following Oligomer scaffold implantation, with upregulation of Th2 and Treg pathway genes dominating by 14 days. Specifically, Oligomer induced high levels of Ccr4, which encodes a Treg receptor that binds the recruitment chemokines CCL17 and CCL22.83 Treg participation was further confirmed by the upregulation of Forkhead box P3 (Foxp3), a transcription factor essential for Treg maturation and function, as well as Treg surface receptors Ctla4 (cytotoxic T lymphocyte antigen), Icos (inducible T cell costimulatory), and Il2ra (interleukin 2 receptor α-chain).80,88,89 Previous studies have documented that recruited and resident Treg populations control inflammation after tissue injury and restrain immune responses to self and foreign antigens by modulating neutrophil and macrophage function and suppressing CD4+ and CD8+ T cell-mediated inflammation.90,91 Moreover, it has been suggested that Tregs directly facilitate tissue regeneration via activating local stem and progenitor cell populations.92–94 In addition to promoting Treg activity, Oligomer scaffold implantation led to increased Th2 pathway gene expression, including the key transcription factor GATA3 and cytokine IL-13.95 Th2 cells, in addition to Tregs, are known suppressors of Th1- and Th17-driven inflammation, with many studies suggesting a critical role of IL-13 and IL-4 activated macrophages in the resolution of inflammation and the restoration of immune homeostasis.96,97 Th2 cells have also been identified as a required cell type for the constructive remodeling response observed with decellularized ECM scaffolds.78,98,99 Overall, the promotion of Th2/Treg signaling by Oligomer scaffolds suggests that adaptive immune engagement is likely a key attribute of its unique noninflammatory mechanism of action.
Th2/Treg pathways were also upregulated by commercial collagen; however, this activity was counterbalanced by engagement of pro-inflammatory Th1 and pro-fibrotic Th17 pathways. Specifically, Il17a, Ccl20, Irf4, and Ccr6, which are Th17 pathway gene products, as well as Ifng, the signature gene for the Th1 pathway, were upregulated exclusively with commercial collagen, especially at later time points. Given the high numbers of macrophages with mixed phenotypical expression, the observed mixed T cell population is not unexpected. Still, the balance between Th1/Th17 and Th2/Treg pathways is an important determinant of fibrosis, as demonstrated by biomaterial outcomes and disease pathogenesis.21,80 Interestingly, Th17 and Treg subtypes have interconnected differentiation pathways and antagonist functions.100,101 Th17 cells are associated with autoimmunity, inflammation, and fibrosis whereas Tregs inhibit these processes and maintain immune homeostasis. More specifically, the extent of capsular fibrosis observed clinically for failed silicone breast and polypropylene mesh implants was shown to be dependent on a shift in the balance towards Th1/Th17 activity and inversely proportional to the number of Tregs.74,102 Additionally, IL-13 has been identified as a driver of myofibroblast differentiation and collagen deposition,95,103,104 further supporting the significance of the T helper cell balance during tissue repair. In a distinct mechanism from the commercial collagen, the lymphocytes observed histologically surrounding the commercial mesh showed CD4+ and CD8+ T cell populations, with more moderate upregulation of lymphoid-specific pathways evident from transcriptomics data across the 14-day study period. For mesh, there appeared to be greater macrophage and FBGC accumulation compared to T cell numbers that was observed histologically. In support, such results are consistent with previous studies that show that monocyte-derived macrophages dominate the chronic inflammation and FBR to polypropylene mesh at both early and late stages following implantation in mice and humans.72 Overall, findings from the present study further demonstrate the importance of the Th2/Treg and Th1/Th17 balance in determining material responses, with Oligomer materials exhibiting prominent Th2/Treg signaling and no evidence of Th1/Th17 activation, thereby shifting the balance towards immune tolerance and away from inflammation, material bioresorption, and fibrosis.
Present-day collagen materials, regardless of their method of fabrication, undergo immune-mediated bioresorption with fibrous connective tissue formation and remodeling.34,105 To control degradation rate and enhance mechanical stability, materials are routinely subjected to various types and levels of exogenous crosslinking, which increases covalent bonding between chemical moieties. Common methods employed both commercially and non-commercially include those involving glutaraldehyde, isocyanates, dehydrothermal, and carbodiimide.4 Not only is the specificity of these crosslinking reactions difficult to control within complex biological materials, but the extent of crosslinking is also positively correlated with enhanced FBR, fibrous capsule formation, cytotoxicity, and calcification (see Delgado et al., 201539 for review). In the present study, commercial collagen showed rapid bioresorption, which is consistent with modest crosslinking and the 10- to 14-day resorption rating of this product. As with other collagen materials, cell-mediated material phagocytosis and proteolysis was facilitated by neutrophils, macrophages, and FBGCs, as well as associated matrix metalloproteinases MMP-13 and MMP-9, which are known to degrade intact collagen fibrils and collagen fragments, respectively.10,105,106 This degradation process is accompanied by the influx of fibroblasts and other stromal cell populations that deposit ECM components, including collagen, and progressively remodel the fibrous connective tissue. This well-known collagen biomaterial response is referred to as “constructive remodeling”, with tissue response outcomes “culminating in the formation of a tissue that is usually site appropriate, at least partially functional, and devoid of any persistent or chronic inflammatory reaction”.71,107 This remodeling process does not represent tissue morphogenesis or regeneration in the true sense, but rather is best described as a departure from the default wound contraction and scar tissue formation observed with tissue injury.34,107 Interestingly, in most situations when conventional collagen materials are exogenously crosslinked, improved persistence is achieved, but chronic inflammation and a more robust FBR are also observed.39 There is an increased number and prolonged presence of macrophages and FBGCs, enhanced type 1 immune response, and enhanced fibrosis and/or fibrotic encapsulation, ultimately yielding undesirable outcomes.39,84,107 This has led to suppositions within the biomaterials field that inflammation and biological scaffold degradation are requisite processes to achieve favorable constructive remodeling outcomes.107,108
The immune response and regenerative remodeling observed following Oligomer material implantation is distinct from conventional collagen materials, thereby challenging current paradigms. Unlike commercial collagen scaffolds, Oligomer scaffolds do not undergo degradation (bioresorption) following implantation in the body. Oligomer scaffolds undergo cellular remodeling via mechanisms that largely involve cell-collagen mechanobiological signaling and collagen metabolism (turnover). It is notable that these mechanisms occur in absence of a material-induced inflammatory response, which is important since inflammatory signals and processes may modulate or even override these normal tissue homeostatic mechanisms.109,110 This and previously published work show that Oligomer materials persist within both non-wound and wound environments, where they maintain their structural and physical integrity, and exhibit physiological collagen turnover and regenerative remodeling, all of which is dependent upon the material format and implant microenvironment.44,47,49,50 The slow turnover observed with Oligomer materials is not surprising, since fibrillar collagen in normal connective tissues is naturally metabolized (turned over) at a very slow and controlled rate via a process that is distinct from immune-mediated bioresorption.111,112 In native tissues, molecules of fibrillar collagens are held together by mature intermolecular crosslink chemistries that influence tissue mechanical properties and collagen metabolism during tissue turnover and remodeling.113,114 As an inherent component of polymerizable oligomeric collagen, these mature, trivalent crosslink chemistries contribute to the distinct suprafibrillar self-assembly observed with these collagen building blocks, which in turn yields scaffolds with improved mechanical integrity and proteolytic resistance compared to those formed by other polymerizable collagens.40,41
The rate of cellularization and regenerative remodeling observed following implantation of Oligomer materials is dependent upon material geometry (e.g., volume and shape), fibrillar density and microstructure, implantation microenvironment, and whether there is an associated tissue injury or void. In general, the remodeling rate increases with decreased scaffold volume, increased scaffold surface area, increased scaffold fibrillar density, and presence of tissue void/injury. The present study shows that Oligomer materials fashioned as relatively thin, high-density scaffolds show limited cellular infiltration and slow remodeling over a 60-day period when implanted subcutaneously, where there is limited tissue injury and no tissue void and limited disruption of the tissue mechanobiological continuum. Similar slow remodeling, along with immunoprotective properties, have been reported previously following subcutaneous implantation of islets (syngeneic, allogeneic, or xenogeneic) encapsulated within polymerized oligomeric collagen.48,49 More specifically, euglycemia was induced for beyond 90 days in streptozotocin-induced type 1 diabetic mice following injection/implantation of 250 islets encapsulated in 500 μL of oligomeric collagen (4.2 mg/cm3 in density) in two separate locations along their dorsum. In this case, the noninflammatory and persistent oligomeric scaffold supported and protected the foreign islets, preventing detection by the host immune system. In a separate study, thin, high-density Oligomer scaffolds, similar to those evaluated in the present study, were applied for reconstruction of a full-thickness mucosal layer defect (∼3–4 cm2) within a porcine hemilaryngectomy model.47 In that study, the thin, high-strength sheet was reported to provide immediate and suitable barrier function, with progressive regenerative remodeling occurring over an 8-week period. The newly formed mucosa was histologically similar to normal mucosa, featuring a stratified squamous epithelium supported by a vascularized and cellularized stromal layer with evidence of gland formation. Similar endogenous regeneration was reported when an Oligomer soft tissue filler formulation was evaluated in a porcine lumpectomy (breast conserving surgery) model.50 In this study, the in situ scaffold-forming collagen (∼8 mg/cm3 in density; 4–8 mL volume) conformed to and filled the complex surgical void, where it prevented contraction and supported breast tissue neogenesis, including adipose tissue and mammary glands and ducts over a 16-week period. Collectively, the regenerative remodeling observed with Oligomer materials to date has many similarities to processes associated with tissue development and morphogenesis, highlighting the importance of minimizing inflammation and maintaining the mechanobiological continuum between the tissue and material.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3bm00091e |
This journal is © The Royal Society of Chemistry 2023 |