Engineered macroporous gelatin scaffolds enhance lymph node fibroblastic reticular cell identity and enable diabetogenic T cell immunomodulation.

Engineered Macroporous Gelatin Scaffolds Enhance Lymph Node Fibroblastic Reticular Cell Identity and Enable Diabetogenic T Cell Immunomodulation - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Search Log in Primary site navigation Search Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Biomaterials . Author manuscript; available in PMC: 2026 May 31. Published in final edited form as: Biomaterials. 2025 May 31;324:123460. doi: 10.1016/j.biomaterials.2025.123460 Engineered Macroporous Gelatin Scaffolds Enhance Lymph Node Fibroblastic Reticular Cell Identity and Enable Diabetogenic T Cell Immunomodulation Leonor N Teles †, ‡ , Logan A Beatty ‡ , Ana V Hernandez † , Marvin A Mendoza Jr † , Zachary M Wilkes †, ‡ , Vivien P Dominick ‡ , Michelle T Argy Telias ‡ , Benjamin Miller # , Chun-Yuh Huang ‡ , Camillo Bechi Genzano $, % , Fotios M Andreopoulos ‡ , Edward A Dauer ‡ , Remi J Creusot $, % , Alice A Tomei †, ‡, ⭙, * Author information Article notes Copyright and License information † Diabetes Research Institute, University of Miami Miller School of Medicine, Miami, FL. 1450 NW 10th Ave, Miami, FL 33136, USA ‡ Department of Biomedical Engineering, University of Miami, 1251 Memorial Dr, Coral Gables, FL 33146, USA # Department of Biomedical Engineering, Johns Hopkins University, 3400 N. Charles Street Wyman Park Building, Suite 400 West, Baltimore, MD 21218, USA $ Columbia Center for Translational Immunology, Department of Medicine, Columbia University Irving Medical Center, 650 West 168th St. BB15-1501 New York, NY 10032, USA % Naomi Berrie Diabetes Center, Columbia University Medical Center, Russ Berrie Medical Science Pavilion, 1150 St Nicholas Ave, New York, NY 10032, USA ⭙ Department of Microbiology and Immunology, University of Miami Miller School of Medicine, Miami, FL. 1450 NW 10th Ave, Miami, FL 33136, USA AUTHOR DECLARATION The work submitted has not been published and is not under consideration for publication elsewhere. L.N.T. and A.A.T. conceived the project. L.N.T., Z.M.W., A.A.T., C. Y. H., C.B.G., F.M.A. E.A.D. and R.J.C. interpreted results. L.N.T., L.A.B., A.V.H., M.A.M., A.A.T., C.B.G., Z.M.W., C.Y.H., and R.J.C. designed experiments. L.N.T., A.V.H., L.A.B., M.A.M., M.T.A.T., V.D., and B.M. performed experiments. L.N.T. analyzed data. L.N.T. wrote the manuscript with input from all authors. A.A.T. supervised the project. All authors have approved the final version of the manuscript. * Corresponding Author atomei@med.miami.edu Issue date 2026 Jan. PMC Copyright notice PMCID: PMC12823137  NIHMSID: NIHMS2132223  PMID: 40472409 The publisher's version of this article is available at Biomaterials Abstract Current treatments for autoimmune diseases like Type 1 Diabetes (T1D) carry significant risks because they lack tissue specificity. A promising strategy is to achieve persistent presentation of relevant antigens (Ags) in non-inflamed sites by tolerogenic Ag-presenting cells (APCs) like fibroblastic reticular cells (FRCs). FRCs build lymph node (LN) reticula and act as immunomodulatory non-professional APCs. However, their therapeutic potential for Ag-specific immunomodulation for T1D remains unexplored. We engineered 3D FRC-based reticula using freeze-dried macroporous gelatin scaffolds with customizable pore diameters (small: <50μm, medium: <200 μm, large: <300 μm) to evaluate FRC phenotype and FRC-T cell interactions, leveraging FRCs’ ability to build dynamic LN reticula that expand and contract during inflammation. Our scaffolds promoted FRC viability, reticular formation, FRC phenotypic marker expression, and extracellular matrix secretion compared to 2D culture. GFP-Luciferase fusion (GLF)-expressing FRCs subcutaneously implanted in our scaffolds survived for at least 21 days regardless of pore size. Implantation in the vascularized fat pad led to graft rejection by day 14 in pre-diabetic NOD mice but not in immunodeficient NODscid. Our scaffolds outperformed clinically-used biologic gels, demonstrated the limitations of the NOD mouse model in longitudinal imaging of GLF + cell survival, and validated our scaffold-based FRC delivery approach for future therapeutic applications. Co-culture of T cells in scaffolds with FRCs presenting T1D Ags enabled Ag-specific T cell engagement with reduced cytotoxic and increased anergic and regulatory phenotypes. Our results validate the use of macroporous gelatin scaffolds to enhance FRC phenotypic markers, in vivo survival, and diabetogenic cell immunomodulation in vitro . Keywords: Immune engineering, type 1 diabetes, antigen-specific therapies, stromal cells, 3D culture, lymphoid organs, tolerance Introduction Autoimmune diseases (ADs) like Type 1 Diabetes (T1D) are characterized by a breakdown in tolerance to self-antigens (Ag), wherein autoreactive T cells destroy self-tissues [ 1 ]. Current treatments rely on systemic approaches with undesirable side effects and limited duration [ 2 , 3 ]. Therefore, there is a need for Ag-specific therapies that specifically target autoreactive T cells and modulate autoimmune responses more precisely and durably. Ag-specific therapies present self-Ags at alternative sites using various types of Ag-presenting cells (APCs) to elicit tolerogenic Ag-specific T cell responses. These therapies are safer than systemic, non-specific approaches but face challenges such as limited Ag exposure duration and the complexity of APC-T cell interactions [ 4 – 6 ]. Professional APCs can transition from tolerogenic to immunogenic states under inflammation, exacerbating autoimmunity [ 7 ]. Conversely, non-professional APCs remain poorly immunogenic even during inflammation [ 8 ], yet current therapies do not utilize these cells due to their limited Ag capture capacity. To address these limitations, we devised a novel strategy involving a tissue-engineered three-dimensional (3D) platform with Ag-presenting fibroblastic reticular cells (FRCs). Fibroblastic reticular cells (FRCs), a subpopulation of lymph node (LN) stromal cells, function as non-professional APCs with immunomodulatory properties [ 7 , 9 – 12 ]. In the LN paracortex, the podoplanin (gp38)-expressing CD45 − CD31 − FRCs form tensile collagen fiber networks, reinforcing the LN and supporting recruitment (via CCL19/CCL21 secretion), survival (via IL-7 secretion), and interaction of T cells with FRCs [ 13 , 14 ]. FRCs modulate LN structure by adjusting contractility and secreting extracellular matrix (ECM) components like laminin and collagen [ 15 , 16 ]. During acute immune responses, FRCs relax, resulting in cellular elongation and enlarged reticular pores, facilitating LN expansion, immune cell trafficking, and T cell proliferation [ 15 – 18 ]. FRCs exhibit immunomodulatory effects through nitric oxide, prostaglandin E2, TGF-β, and other factors [ 11 , 19 ]. They maintain Ag-specific peripheral T cell tolerance via AIRE-independent peripheral tissue Ag expression [ 20 – 23 ]. Upon inflammation, FRCs upregulate MHC class I/II, immune checkpoints (PD-L1), while expressing limited co-stimulatory molecules (low CD80, no CD86) [ 11 , 20 , 24 ]. In transgenic mice with FRC-restricted Ag expression, this non-immunogenic phenotype deleted Ag-specific naïve T cells after engagement [ 20 ]. Overall, FRCs have shown therapeutic potential in preclinical models of ischemic injury [ 25 ], sepsis [ 26 ], and cardiac transplant [ 27 ]. These findings highlight FRCs’ potential for Ag-specific therapies in ADs. However, FRC-based platforms for AD treatment remain to be established. Traditional two-dimensional (2D) FRC cultures fail to replicate the reticular, porous LN paracortex and the complex FRC-T cell interactions found in vivo . In contrast, 3D culture better preserves fibroblast phenotypes [ 28 , 29 ] and supports physiological cell-ECM interactions [ 28 , 30 ], justifying their utilization to maintain native LN FRC identity. We have previously demonstrated that genetically engineered FRCs overexpressing T1D-specific Ags seeded in commercially available 3D collagen scaffolds with 400 μm pores generate reticula and engage diabetogenic T cells more efficiently than 2D cultures [ 12 ]. However, these pore sizes exceed native LN FRC reticular pores, ranging from 20–200 μm [ 18 ]. To identify the optimal macroporous scaffold platform for FRC-based therapies, we designed freeze-dried gelatin scaffolds with varying pore sizes (small pores: ~50 μm, medium pores: <200 μm, and large pores: <300 μm), given that LN size and FRC reticula can expand up to 10-fold during adaptive immune responses [ 11 , 27 , 31 , 32 ]. While previous studies modeled LN microenvironments using hydrogels [ 33 ], computational models [ 34 , 35 ], and decellularized LN ECM [ 36 , 37 ], the impact of 3D FRC culture in macroporous gelatin scaffolds and the effects, if any, of reticular pore size on FRC phenotypical marker expression, FRC survival, and FRC-T cell interactions remains underexplored. Using freeze-dried scaffolds that allow easy tuning of pore size during fabrication, we developed 3D macroporous gelatin scaffolds to investigate effects of culture substrate on FRC phenotype, survival, and FRC-T cell interactions and test whether scaffold pore size in the range of normal to inflamed LN reticula would affects these measured outcomes. Here, we report the fabrication and characterization of freeze-dried gelatin macroporous scaffolds with customizable pore sizes and assessed the effects of 3D culture with specific pore sizes on FRC phenotypic marker expression and ECM secretion compared to 2D culture in vitro . Scaffold biocompatibility and FRC localization and survival in macroporous scaffolds compared to clinically applicable biologic gels were evaluated in vivo in T1D mouse models. Using pulse treatment with T1D Ag peptides to achieve presentation of these Ags in FRCs, we studied FRC interactions with diabetogenic T cells in a reticula-like environment, exploring their in vitro immunomodulatory properties. This work validates the use of macroporous gelatin scaffolds for FRC delivery, and it establishes a foundation for developing FRC-based platforms to induce Ag-specific tolerance in T1D. Materials and Methods Scaffold Fabrication and Characterization Scaffolds were fabricated by adapting previously established methods [ 38 – 40 ]. 2% glutaraldehyde (GTA, Sigma, Saint Louis, MO) solution was added to 3 or 5% sterilized gelatin type B (bovine skin powder, Lot G9391, Sigma) solutions in a 1:14 ratio. The mixed solution was allowed to crosslink for 2, 4, 6, or 24 hours (gelatin-glutaraldehyde gels). The cross-linked scaffolds were placed in 100 mM glycine solution (Sigma) at room temperature (RT) for 1 hour to quench unreacted aldehyde groups. The gels were then placed in distilled water for at least 2 hours. Scaffolds were frozen at −20°C, −80°C for at least 6 hours or via rapid freezing using liquid nitrogen (−200 °C). Frozen gels were then lyophilized overnight, and the resulting freeze-dried scaffolds were shaped into 1-mm height by 4-mm diameter discs to fit a 96-well plate or 8-mm to fit a 24-well plate using a biopsy puncher (Sklar Instruments). A schematic of the fabrication method is depicted in Fig. 1A . Figure 1. Open in a new tab Macroporous scaffold fabrication and characterization. (A) Freeze-dried macroporous scaffold fabrication schematic. (B) Swelling ratio before freeze-drying (n=5). (C-E) Effect of crosslinking time (C), freezing temperature prior to freeze-drying (D), and gelatin B concentration (E) on macroporous scaffold pore size measured as Feret’s diameter of scaffold cross-sections via SEM imaging (average of 15 pores per n=3 scaffolds). (F) Swelling ratio of freeze-dried small, medium and large pore size range scaffolds (n=5). (G) Scaffold porosity by liquid displacement method (n=5). (H) Scaffold degradation by % weight reduction after 21 days in culture (n=3). (I) Young’s Moduli by 10% unconfined compression testing (n=5). (J) Representative SEM images of scaffolds with small, medium and large pore size ranges (scale bar = 200 μm). Data represented as mean ± standard deviation. P-values were determined by student’s t-test or ordinary one-way ANOVA followed by Tukey’s multiple comparison test. (*p<0.01; **p<0.001; ***p < 0.0001; ****p < 0.00001; ns, not significant, p > 0.05). Swelling Ratio To assess the extent of crosslinking, gelatin-glutaraldehyde gels were left to dry at RT overnight and then placed in distilled water overnight. Alternatively, to evaluate post-fabrication swelling ratio, scaffolds were frozen and freeze-dried, weighed (W: weight) and then placed in distilled water overnight. Swelling ratios were assessed after at least 16-hour equilibration in water as: Swelling Ratio = W wet − W dry / W dry Free Amine Quantification 2,4,6-Trinitrobenzene Sulfonic Acid (TNBSA) assay was performed to quantify unreacted, free amino groups, as previously reported [ 41 ]. Briefly, 3 mg of dry scaffolds crosslinked for 0 (non-crosslinked control), 2, 4, 6, or 24 hours were hydrated for 30 minutes in 0.5 mL of 4% (w/v) sodium bicarbonate solution. 0.5 mL of 0.05% (w/v) TNBSA solution was added and heated at 40°C for 2 hours. Then, 1.5 mL of 6 M hydrochloric acid was added, and samples were hydrolyzed at 60°C for 90 minutes. The reaction mixture was diluted with 2.5 mL of distilled water, cooled to room temperature, and absorbance was measured at 320 nm to quantify free amino groups as an alternative way to quantify gelatin crosslinking with GTA. Scaffold Pore Size A JSM-6010PLUS/LA analytical Scanning Electron Microscope (SEM, JEOL) was used to image the dry scaffolds and determine the pore morphology after gold sputtering (Desk V, Denton Vacuum), and pore sizes were quantified by image analysis. An accelerating voltage of 2 kV, working distance of 12 mm, magnification between 45X and 65X, and spot size of 50 were used. Feret diameter was used to quantify the pore size of at least 15 pores per scaffold, which were then averaged and measured across n=3 fabrication batches and conditions using ImageJ on SEM scaffold images. Scaffold Porosity, Density, Surface Area and Volume, Degradation and Compressive Moduli Scaffold porosity and density were quantified via liquid displacement, as previously reported [ 42 ]. Scaffold surface area to volume ratio was measured using IMARIS (Oxford Instruments). Scaffold degradation was assessed by weighing the scaffolds on day 0, placing them in murine FRC media [ 43 ] for 21 days and then weighing the scaffolds 21 days later after lyophilizing the scaffolds to remove the water. Unconfined compression testing (Electroforce 5500 DHR-2, TA Instruments) with 10% deformation was performed to measure the scaffolds’ Young’s Moduli to estimate scaffold stiffness. Cylindrical, 6-mm diameter, cylindrical-shaped scaffolds submerged in 1X PBS (Gibco, Massachusetts) were used for mechanical testing at a rate of 0.048 mm/s. Scaffold height, around 6 mm, was measured before and after compression. FRC Isolation and Culture Female non-obese diabetic (NOD) mice purchased from The Jackson Laboratory (Bar Harbor, Maine) at 12 weeks of age were used for FRC enzymatic isolation from skin-draining LNs as previously reported [ 43 ]. After purification by cell sorting (Beckman Coulter MoFlo Astrios EQ cell sorter) as gp38 + CD45 − CD31 − live cells and expansion in 2D culture (to passage 20) and GFP-Luciferase fusion (GLF + FRCs, passage 25). 2nd generation lentiviral constructs were produced using the pHR2-GLF lentiviral vector plasmid, which encodes a GFP-firefly luciferase fusion gene, packaging plasmid, and pVSV-G-expressing plasmid (3:2:1 ratio) as previously described [ 44 ]. Untransduced cells were used for most in vitro cultures, whereas GLF + FRCs were used for in vivo studies and corresponding in vitro controls. For in vitro studies, scaffolds were sterilized under ultraviolet light for 2 hours, and 200,000 FRCs in 10 μL of media were seeded onto scaffolds with different pore sizes and allowed to attach for an hour before supplementing the wells with 200 μL of cell culture media. Cell number and viability were assessed via RealTime-MT and CellTiter-Glo Luminescent assays (Promega) per manufacturer’s protocol. Seeding efficiency was calculated as the percentage of FRCs remaining 24 hours after relative to the initial 200,000 FRCs seeded. FRC Luminescence-based In vitro and In vivo Viability For luminescence-based studies, 800,000 GLF + FRCs were seeded in 8-mm-diameter scaffolds and either kept in in vitro culture or implanted into a non-vascularized, dorsal subcutaneous site of pre-diabetic 8-week-old female and male NOD mice (control: empty scaffolds). Alternatively, GLF + FRCs were seeded onto macroporous gelatin scaffolds or within biologic gels formed by combining NOD mice plasma and recombinant human thrombin (Recothrom), as previously reported [ 45 ], and implanted into the gonadal fat pads of male and female NOD mice or immunodeficient NODscid mice. Survival of GLF + FRCs in the implant site was assessed in vitro and in vivo longitudinally using whole mouse optical imaging (IVIS Spectrum). Imaging was done 15 minutes after intraperitoneal injection of Luciferin (IVISbrite D-luciferin prepared according to manufacturer protocols, PerkinElmer, Waltham, MA) to determine FRC viability over time. Vascularization of FRC-scaffold implant site was assessed in NODscid mice by injecting IVISense Vascular 750 Fluorescent Probe AngioSense (Revvity, Waltham, MA), followed by whole mouse optical imaging (IVIS Spectrum). Quantification was performed using the Living Image software (PerkinElmer, Waltham, MA) after placing the region of interest over the scaffold implantation sites. Average radiance (photons/second/cm 2 /steradian) or Radiant Efficiency ([p/]/[μW/cm 2 ]) were obtained and log 2 -transformed data was analyzed and reported. Control scaffold radiance or radiant efficiency were averaged and subtracted for each range per time point. Control scaffolds were harvested from mice 21 days after implantation for histological analysis in formalin-fixed and paraffin-embedded 5μm-thick sections. FRC Immunofluorescence Reticular formation during extended culture times in scaffolds, as well as fibroblast structural components and ECM marker expression after 21 days in culture were assessed by immunofluorescence staining and imaging using a confocal microscope (Stellaris, Leica Instruments). The antibody list used for immunofluorescence is shown in Supplementary Table 1 . Reticular formation was quantified by measuring the F-actin-positive Feret diameter of cross-sections of scaffolds with different pore sizes seeded with FRCs and stained with Phalloidin (Invitrogen). For fluorescence intensity quantification, IMARIS was used to obtain Fluorescent Intensity Sum for each channel and then normalized to DAPI/Nuclei signal for each respective scaffold. Since the scaffolds are made of gelatin, a denatured product of collagen, the scaffolds alone stained positive for Polyclonal Collagen. To account for this, Polyclonal Collagen Fluorescence Intensity values from IMARIS were normalized to the fluorescence values of acellular scaffolds for each corresponding pore size. FRC Flow cytometry FRC marker expression (gp38 and PDGFRα) was measured via flow cytometry (CytoFLEX, Beckman Coulter) after FRCs were cultured in the scaffolds for 7, 14 and 21 days. The list of antibodies used for flow cytometry is shown in Supplementary Table 2 . FRC proliferation was measured by CellTrace dilution after culture of CellTrace (Invitrogen)-labeled FRCs in the scaffolds for 14 days. Cell proliferation was calculated as the percent of CellTrace Violet dilution after the highest peak to account for an initial dilution of the dye due to protein turnover and time, unrelated to cell proliferation. CellTrace Violet-labeled (Invitrogen) FRCs were treated with 20 μg/mL of Mitomycin C (Sigma) for an unproliferated, negative control. RNA Isolation and Quantitative RT-PCR Total mRNA of FRCs cultured in scaffolds with small, medium or large pore size ranges for 24 hours was extracted according to the manufacturer’s protocol (Vazyme), followed by RNA clean-up (New England Biolabs). For cells grown in 3D, the scaffolds were lysed for 5 minutes in RL Buffer (Vazyme) with a tissue homogenizer (Qiagen) before RNA extraction. The extracted RNA yield was quantified using a NanoDrop 2000 Spectrophotometer (ThermoFisher) and RNA samples that did not meet the desired purity levels (260/280 ratio below 1.8 and 230/260 ratio below 2.0) were discarded. Up to 5 μg of RNA in 7 μL of PCR-grade water (ThermoFisher) was subjected to reverse transcription using the SuperScript ™ IV First-Strand Synthesis System (ThermoFisher), and negative reverse transcriptase controls were used. qPCR was run in a StepOne apparatus (ThermoFisher) using TaqMan Advanced MasterMix for qPCR (ThermoFisher) and FAM TaqMan ™ 20X Gene Expression Assay Probes (ThermoFisher). Probes used for RT-qPCR studies are shown in Supplementary Table 3 . 40 ng of cDNA was used per reaction. A no-template control was run as a negative control for each probe, which had no detectable Ct. Three passage-matched, biological replicates were used per culture conditions and two technical replicates per biological replicate were plated for RT-qPCR. Relative gene expression was quantified using 2 −ΔΔCt methods and normalized to TBP (TATA Box Binding Protein) and RNA18s (18s ribosomal RNA) expression as reference genes. For CCL21, gene expression was exclusively detected on FRCs cultured in 3D scaffolds but not in 2D cultures, and, as we were unable to apply the 2 −ΔΔCt method, we instead employed the ΔCT method, comparing the log 2 -transformed CCL21 expression values among the various 3D scaffolds. TGF-β2 ELISA FRCs in 2D flasks, FRCs in scaffolds, and empty scaffold controls were cultured in 1% FBS serum medium for 48 hours to deprive the culture of residual TGF-β2 in FBS. After 24 hours in reduced serum medium, cell supernatant and control medium were collected and stored at −80°C. Active TGF-β2 levels were measured using a mouse TGF-β2 duo-set ELISA kit (DY7346–05) and supplementary ancillary reagent kit (DY008B) per manufacturer’s instructions (R&D Systems, Minneapolis, MN). Results were reported by constructing a standard curve and calculating the fold change between 3D scaffolds and 2D cell culture. The fold change was calculated by subtracting the average raw absorbance readings from the control medium from the raw absorbance readings recorded in experimental samples. Fourier Transform Traction Force Microscopy on FRCs Fourier transform traction force microscopy (TFM) was employed using soft, compliant 1.2 kPa polyacrylamide (PA) hydrogels as previously reported [ 46 – 50 ]. Briefly, 6-well 20mm glass diameter plates (MatTek) were treated with a bind silane solution for 1 hour and air-dried. The PA-gel solution was prepared using ultra-pure, 40% acrylamide (Bio-Rad) and 2% bis-acrylamide (Bio-Rad), and 0.5 μm diameter red fluorescent carboxylate modified microsphere beads (Invitrogen). The gel solution was degassed for 30 minutes. Then, after the addition of polymerizing agents, ammonium persulfate (BIO-RAD), and TEMED (N,N,N’,N’-tetramethylethane-1,2-diamine) (BIO-RAD), the solution was added to the treated dishes for polymerization and sealed under 18mm circular coverslips. PA gels were functionalized with Sulfo-SANPAH (Proteochem) by diluting Sulfo-SANPAH in 0.1M HEPES (Fischer-Scientific) to achieve a final concentration of 10 mM. The Sulfo-SANPAH solution was placed on the PA gels and placed under a UV lamp for 20 minutes for photoactivation. Following SANPAH UV-burning and subsequent HEPES and PBS washes, gels were incubated overnight at 4°C with 0.1 mg/mL collagen I (Advanced Biomatrix). Excess collagen was removed the following day, and FRCs were seeded at a concentration of 40,000 cells/mL after culture in 2D flasks or scaffolds with small, medium or large pore diameters. FRCs were harvested from the scaffolds as reported [ 43 ] and allowed to adhere overnight. 2 hours before the experiment, FRCs were treated with Calcein AM (Invitrogen). After live-image acquisition, FRCs were exposed to 0.2% Triton-X 100 for 10 minutes to detach the cells from the PA gel to obtain a stress-free image, which was utilized for calculating the tractions. Cell-induced gel deformations were computed using a custom-written particle image velocimetry (PIV) routine in MATLAB [ 51 , 52 ]. Splenocyte Culture and Biocompatibility Studies Splenocytes were isolated from 12-week-old female NOD or C57BL/6 mice, and 200,000 cells were seeded in each scaffold and cultured in 300 μL of 5% FBS media supplemented with interleukin-2 to ensure cell survival in the absence of Ag-driven stimulation. Alternatively, 200,000 RAW 264.7 macrophages were seeded in each scaffold and cultured in 200 μL of GM-CSF-supplemented media. Controls included splenocytes or macrophages cultured in 2D with media (negative control for macrophages) or LPS (100 ng/mL, positive control for both macrophage and splenocyte activation). Following 72 hours of culture, 200 μL of conditioned media from each experimental group and control was collected for assessment of IL-6, IL-10, and TNF-α secretion by ELISA (R&D Systems, Minneapolis, MN) per manufacturer’s protocol. The remaining 100 μL of media was retained for MTS assay to compare splenocyte proliferation in each condition. In vivo Immunofluorescence and Collagen Deposition Scaffold integration within host tissue was analyzed by immunofluorescence and histological staining. Staining with antibodies to label blood (CD31 + α-SMA + ) and lymphatic (CD31 + gp38 + ) vasculature markers, and with antibodies to assess host cell infiltration (macrophages: CD68, Arg1, and iNOS; T cells: CD3; B cells: CD19) was performed on formalin-fixed/paraffin-embedded 5μm-thick sections and sections were imaged by confocal microscopy. Fibrosis as collagen deposition around the scaffolds was assessed via Masson’s trichrome staining (Abcam) per manufacturer protocol. Cell infiltration in each scaffold was quantified using the QuPath Software (Queen’s University, Belfast, Northern Ireland) [ 53 ]. Antibodies for immunofluorescence are listed in Supplementary Table 4 . Animal Studies All animal procedures were approved by the Institutional Animal Care and Use Committee of the University of Miami (IACUC protocol #IPROTO202400000648). NOD/ShiLtJ (NOD), C57BL/6 (B6), NOD.Cg-Prkdc scid /J (NODscid), NOD.Cg-Tg(TcraTcrbNY8.3)1Pesa/DvsJ (NY8.3), NOD.Cg-Tg(TcraBDC2.5,TcrbBDC2.5)1Doi/DoiJ (BDC2.5), NOD.B6-Ptprc b /6908MrkTacJ (CD45.2), NOD.CgTcra tm1Mjo Tg (TcrbG9C8)2Fsw/J, and NOD.CgTcra tm1Mjo Tg(TcraG9C8)22Fsw/J mice were purchased from The Jackson Laboratory (Bar Harbor, Maine). G9C8 mice were generated by crossing TcrbG9C mice with TcraG9C8 mice. In vitro Co-Culture of FRCs and T cells 200,000 FRCs were seeded in scaffolds and pre-conditioned with 10 ng/mL of interferon-gamma (IFN-γ) for 72 hours to upregulate MHC-I, MHC-II and PD-L1 ( Fig. S1 ), as previously reported [ 54 ], and with exogenous T1D peptides (0.01 μg/mL INS-Valine [ 55 ] peptide for G9C8 Ag-specific CD8 + T cells [ 55 ], 0.011 μg/mL IGRP peptide [ 56 ] for NY8.3 Ag-specific CD8 + T cells, and 0.013 μg/mL p79 peptide [ 57 , 58 ] for BDC2.5 Ag-specific CD4 + T cells) for 24 hours for presentation by FRCs to Ag-specific T cells. 300,000 T cells were added to the tissue-engineered reticula and cultured for 3 days. Naïve Ag-specific T cells were isolated from splenocytes of mice with transgenic T cell receptors (TCRs) specific for selected T1D Ags (BDC2.5, NY8.3, G9C8) through magnetic activated cell sorting (STEMCELL Technologies) for CD4 + or CD8 + T cells following manufacturer protocol. T cells were also labeled with CellTrace Violet (Invitrogen) to assess proliferation prior to addition to the tissue-engineered FRC scaffolds. IGRP 206–214 -reactive CD8 + T cells were obtained from NY8.3 mice, InsB 15–23 -reactive CD8 + T cells from G9C8 mice, and p79/HIP2.5-reactive CD4 + T cells from BDC2.5 mice bred with CD45.2 mice for congenic labeling. Ag-specific p79/HIP2.5-reactive CD4 + T cells (BDC2.5 + CD45.1 + CD45.2 − ) were diluted 1:10 or 1:2 with polyclonal, non-Ag-specific T cells (BDC2.5 − CD45.1 − CD45.2 + ) to better simulate physiological conditions, where fewer Ag-specific T cells than polyclonal T cells are present. Negative controls included no-Ag pulsed FRCs (no exogenous Ag addition), and αCD3/CD28 Dynabeads (Gibco) cultured with 200,000 T cells, as per manufacturer protocol and previously reported [ 12 ], were used as positive control for T cell activation and proliferation. The following co-culture conditions were used: Condition 1: 90% Polyclonal NOD (CD45.2 + Vβ4 − ) + 10% NOD BDC2.5 Ag-specific (CD45.2 − Vβ4 + ) CD4 + T cells + p79 peptide (0.013 μg/mL); Condition 2: 50% Polyclonal NOD (CD45.2 + Vβ4 − ) + 50% NOD BDC2.5 Ag-specific (CD45.2 − Vβ4 + ) CD4 + T cells + p79 peptide (0.013 μg/mL); Condition 3: 50% NOD NY8.3 Ag-specific CD8 + T cells + 50% NOD BDC2.5 Ag-specific CD4 + T cells + IGRP 206–214 peptide (0.011 μg/mL) + p79 peptide (0.013 μg/mL); Condition 4: 100% NOD G9C8 Ag-specific CD8 + T cells + INSB 15–23 peptide (0.01 μg/mL). T cell proliferation was assessed by CellTrace dilution and phenotype after co-culture with FRCs was measured using flow cytometry (Cytek Aurora) and Kaluza analysis as previously reported [ 59 ]. Activated, Ag-specific T cells were considered as % proliferated CD44 + CD25 + out of Ag-specific CD4 + or CD8 + T cells; anergic as % CD73 + FR4 + among Ag-specific CD4 + CD44 + CD25 + Foxp3 − T cells; regulatory as % Foxp3 + CD25 + among Ag-specific CD4 + CD44 + CD25 + T cells; cytotoxic as % proliferated granzyme B + (GrB + ) Ag-specific CD8 + T cells. The antibody list used for flow cytometry is shown in Supplementary Table 5 . Data Analysis Data is represented as mean ± standard deviation or standard error of the mean. Data was graphed using GraphPad Prism version 10.0.0 for Windows (GraphPad Software, Boston, Massachusetts USA) and p-values were determined by student’s t-test, Mann-Whitney U test, ordinary one-way ANOVA followed by Tukey’s multiple comparison test with Šidák correction where appropriate, Kruskal-Wallis test for data not normally distributed, and Benjamini-Hochberg False Discovery Rate Correction for significances RT-qPCR log2 value. (*p<0.01; **p<0.001; ***p < 0.0001; ****p< 0.00001; ns, not significant, p > 0.05). Results The pore size of macroporous gelatin scaffolds fabricated by freeze-drying depends on the freezing temperature, gelatin concentration, and crosslinking time Macroporous gelatin-GTA scaffolds with different pore sizes were fabricated via freeze-drying ( Fig. 1A ) and adapted from previously described methods [ 38 , 40 , 60 ]. The swelling ratios of the crosslinked gels were measured before scaffolds were frozen to assess the extent of crosslinking as a function of the GTA crosslinking time. After 4 hours of crosslinking, the gels were likely fully crosslinked, as indicated by the swelling ratio ( Fig. 1B ), pore size ( Fig. 1C ), and the remaining, unreacted amino groups ( Fig. S2A ), which did not further decrease after 4–24 hours. In contrast, gels crosslinked for 2 hours showed significant differences in swelling ratio, average pore size and unreacted amino groups yielding higher absorbance by TNBSA assay, compared to those crosslinked for 4, 6, or 24 hours. Scaffold pore size, measured as average Feret diameter on SEM images through ImageJ, depended on the freezing temperature before freeze-drying and the gelatin B concentration. We found that lower freezing temperatures ( Fig. 1D ) and higher gelatin B concentrations ( Fig. 1E ) resulted in smaller scaffold pore sizes ( Table 1 ). Overall, the freezing temperature showed the biggest impact on scaffold pore sizes. Table 1. Scaffold Pore size (μm) Summary Gelatin Concentration (%) Crosslinking Time (hours) Freezing Temperature −200°C (Liquid Nitrogen) −80°C −20°C 5 2 86.44 ± 7.43 242.03 ± 13.94 5 4 43.01 ± 0,36 181.24 ± 10.74 5 6 44.20 ± 1.55 171.61 ± 8.82 5 24 44.65 ± 1.36 171.39 ± 15.59 247.48 ± 14.49 3 24 248.17 ± 9.14 Open in a new tab Based on these results, we chose three distinct scaffold pore size ranges (small, medium, and large) to study FRCs 3D culture in scaffolds that replicate FRC reticular organization in normal vs. inflamed (up to 10-fold expansion) LNs and test whether scaffold pore size affects FRC phenotypic marker expression and FRC-mediated T cell immunomodulation. Scaffolds with small (<50 μm, 44.65 ± 1.36 μm), medium (<200 μm, 171.39 ± 15.59 μm), and large (<300 μm, 247.48 ± 14.49 μm) pore size ranges were fabricated by freezing the crosslinked 5% gelatin-GTA solution at −200°C (rapid freezing in liquid nitrogen), −80°C overnight, and −20°C overnight, respectively. The swelling ratio, porosity, degradation rates, and mechanical properties of scaffolds with small, medium, and large pore size ranges were then measured, as these factors impact molecular diffusion and cell seeding and viability. We found that small pore scaffolds had the highest swelling ratio (6.42 ± 0.58) compared to medium (5.61 ± 0.35, p=0.03) and large pores (5.44 ± 0.26, p=0.01) ( Fig. 1F ). Scaffolds with small pore sizes were also slightly less porous (87.22 ± 3.08 %) ( Fig. 1G ), denser (0.142 ± 0.03 mg/mL) ( Fig. S2B ) and with a higher Surface Area/Volume ratio ( Fig. S2C ) than medium (93.84 ± 0.95 %, p=0.004; 0.11 ± 0.004 mg/mL, p=0.02) and large pore (90.87 ± 0.72 %, p=0.03; 0.083 ± 0.006 mg/mL, p=0.003) scaffolds. All scaffolds were at least 87% porous ( Fig. 1G ). Additionally, there was no difference in scaffold degradation after 21-day incubation in cell culture media across all pore size ranges ( Fig. 1H ) (p>0.92). Lastly, scaffolds with small pore sizes had a significantly higher Young’s moduli (49.45 ± 6.97 kPa) compared to scaffolds with large pore sizes (39.72 ± 2.93 kPa, p= 0.04) ( Fig. 1I ). SEM images of scaffolds with small, medium, and large pore size ranges selected for the following studies are shown in Figure 1J . Macroporous gelatin scaffolds do not trigger in vitro macrophage or splenocyte activation and integrate within in vivo NOD mouse subcutaneous tissue Small, medium, and large pore size range scaffolds did not promote NOD ( Fig. 2 A – C ) or B6 splenocyte ( Fig. S3A – C ) or macrophage ( Fig. S4 ) activation, as there was no significant increase in secretion of pro-inflammatory cytokines TNF-α ( Fig. 2A , S3A ), IL-6 ( Fig. 2B , S3B ) or anti-inflammatory cytokine IL-10 ( Fig. 2C , S3C ) compared to controls. Cytokine secretion levels were comparable to untreated cells (- Ctl, p>0.99 for small and medium, and p=0.65 for large scaffolds) and lower than LPS-treated cells. Moreover, none of the scaffolds promoted splenocyte proliferation ( Fig. 2D , S3D ), measured via MTS assay absorbance, which was comparable to untreated cells (- Ctl, p>0.99). These results suggest that acellular scaffolds did not promote an immune response in vitro in NOD or B6 mouse models. Figure 2. Open in a new tab Macroporous scaffold biocompatibility. (A-D) Pro (A,B) and (C) anti-inflammatory (C) cytokine concentrations measured by ELISA and proliferation by MTS assay (D) after splenocyte culture in macroporous scaffolds with small (S), medium (M) and large (L) pore size ranges (n=3–5). (E) Representative images of Masson’s trichrome staining and immunofluorescence evaluation of macroporous gelatin scaffolds after transplantation for 21 days in the dorsal subcutaneous space of 8-week-old NOD mice (scale bar: light microscopy: 200 μm; confocal microscopy: 50 μm). The four immunofluorescence panels refer to lymphatic and blood vessels (“Vasculature”), B and T cells (“Adaptive”), M2-like macrophages (“Innate 1”) and M1-like macrophages (“Innate 2”). (F) Quantification of collagen fibrotic capsule thickness around implanted scaffolds (n=4). (G-H) Quantification of CD45 + immune cells inside (G) and around (H) the implanted scaffolds (n=3). (I) Macrophage M2 (Arg-1 + ) and M1 (iNOS + ) ratio inside implanted scaffolds (n=3). Data represented as mean ± standard deviation. P-values were determined by Kruskal-Wallis test for ELISA studies or ordinary one-way ANOVA followed by Tukey’s multiple comparison test. (ns, not significant, p>0.05). In addition, acellular scaffolds implanted in the dorsal subcutaneous site of 8-week-old NOD mice integrated well with the host tissue, as evidenced by scaffold vascularization and immune cell infiltration ( Fig. 2E ). As shown in the vasculature panel of Fig. 2E , the colocalization of CD31 and gp38 indicates the presence of lymphatic vessels in and around the scaffold region. Immature blood vessels were identified by CD31 + gp38 − regions, while mature blood vessels, typically larger, were identified by CD31 + regions surrounded by α-SMA + pericytes. The adaptive immune cell panel demonstrates the presence of CD3 + T cells and CD19 + B cells within and around the scaffold region of the implants. Further, the innate immune cell panel 1 identified Arg1 + CD68 + anti-inflammatory M2-like macrophages around the scaffold region, which act to reduce localized inflammation at the scaffold implant site, in contrast to iNOS + CD45 + pro-inflammatory M1-like macrophages shown in the innate immune cell panel 2. Moreover, moderate fibrosis was present around all scaffolds, but there was no difference in thickness of fibrosis-mediated collagen deposition around scaffold pore size ranges, quantified by ImageJ on Trichrome stained sections (p>0.1780) ( Fig. 2F ). Then, we calculated the density of immune cells (% CD45 + cells) in the scaffold ( Fig. 2G ) or boundary regions ( Fig. 2H ) of implanted, acellular scaffolds, relative to the total number of cells (DAPI + ) by QuPath. We found no significant difference in immune cell infiltration between scaffold pore size, both inside (p=0.64) and around (p=0.37) the scaffolds. Lastly, there was no significant difference in the ratio of anti-inflammatory to pro-inflammatory-like macrophages (M2/M1 macrophage ratios) inside the acellular scaffolds of varying pore sizes ( Fig. 2I ) (p=0.62), assessed by quantifying the relative number of M2-like (CD68 + Arg-1 + cells) and M1-like (CD45 + iNOS + cells) macrophages via immunofluorescence, confocal imaging, and QuPath image analysis. Macroporous Gelatin Scaffolds sustain Fibroblastic Reticular Cell survival in vitro and in vivo To test the feasibility of macroporous gelatin scaffolds with different pore sizes as an immunoengineered biomaterial platform for FRC delivery, we first measured FRC seeding efficiency, viability and proliferation in small, medium, and large pore size range scaffolds. There was no significant difference in the seeding efficiency (% of cells remaining in scaffold after 24 hours) across scaffold pore size ranges ( Fig. 3A ) (p=0.18), with values between 40 and 60%. We then assessed FRC survival relative to cell number on day 1 number in scaffolds with different pore ranges via cell reduction ( Fig. 3B ) and ATP-based ( Fig. 3C ) assays. We found that FRCs survived in the scaffolds, regardless of pore size, for at least 21 days in vitro . However, results from the ATP-based assay suggested that FRCs in larger pores had increased survival (70.19 ± 2.31 %, p=0.02) on day 21, which could indicate increased cellular division and/or remodeling activity. These results were confirmed via flow cytometry after labeling FRCs with CellTrace dye ( Fig. 3D , gating strategy in Fig. S5 ). FRCs in scaffolds with larger pores divided slightly more, compared to FRCs in small and medium pore sizes (80.5 ± 1.40 %, p<0.001) after 14 days in the scaffolds. CellTrace dye was no longer detectable after 21 days (data not shown) due to cell protein turnover and proliferation, and the dye fading over time. We further confirmed FRC survival in vitro using GLF + FRCs seeded in scaffolds by IVIS longitudinal imaging ( Fig. 3E ) and found no significant differences in FRC survival between scaffold pore sizes. Figure 3. Open in a new tab Seeding efficiency, viability, and proliferation of FRCs seeded in scaffolds with small (S), medium (M), or large (L) pore size ranges. (A) % FRC seeding efficiency 24 hours after seeding 200,000 FRCs in the scaffolds (n=3). (B-C) % FRC survival relative to day 1 after in vitro culture in scaffolds for 3, 5, 7, 14 and 21 days by RealTime MT assay (B) and CellTiter Glo ATP assay (C) (n=3–5). (D) % FRC proliferation after 14 days in the scaffolds by CellTrace dilution via flow cytometry (n=3). (E) Survival of GFP-Luciferase fusion (GLF)-expressing FRCs in vitro (n=6). Data represented as mean ± standard deviation. P-values were determined by ordinary two ANOVA followed by Tukey and Šidák correction. (***p < 0.0001; ns, not significant, p > 0.05). To investigate GLF + FRC survival in scaffolds in vivo , we implanted GLF + cell-seeded scaffolds in the subcutaneous site of female pre-diabetic NOD mice and found no difference in GLF + FRC survival between scaffold pore size ( Fig. 4A ) or decrease in GLF signal over 21 days. However, when we tested GLF + FRC survival in the NOD gonadal fat pad, a more vascularized site than the subcutaneous site, we observed a complete loss of luciferase signal by day 14 ( Fig. 4B ). To determine if the loss was caused by unfavorable in vivo FRC survival in gelatin scaffolds, we implanted GLF + FRCs in biologic gels in validated preclinical and clinical T1D models [ 45 , 61 ].Since no significant differences in GLF + FRCs survival were observed between different scaffold pore sizes, we proceeded with the medium pore size scaffolds. GLF + FRCs in gelatin macroporous scaffolds showed increased survival compared to the biologic gels at day 7 ( Fig. 4C ). However, on day 14, the luciferase signal was again lost in all groups. These results suggested that the higher vascularization of the fat pad site compared to the subcutaneous site may have facilitated adaptive immune cell infiltration, leading to graft rejection. Given previous reports of NOD mice adaptive immunity against fluorescent protein antigens in tumor [ 62 – 64 ] and autoimmunity models [ 65 ], we hypothesized that the loss of signal might be due to xenoreactivity against GLF protein expressed by GLF + FRCs. To address this, we repeated experiments using NODscid mice, which lack adaptive immunity. We observed sustained luminescence signal from GLF + FRCs, with gelatin scaffolds once again outperforming the biologic scaffold ( Fig. 4D ). Notably, FRCs remained localized in the implant site and within the scaffolds, as confirmed by ex vivo fat pad implant site imaging at day 30 ( Fig. S6A ), and bioluminescence indicative of FRC survival was higher in gelatin scaffolds compared to biologic gels. Additionally, since FRCs have been reported to secrete VEGF and promote vascularization, we assessed vascularization of the graft site using a fluorescent vascular probe ( Fig. S6B ). Mice implanted with FRCs exhibited increased vascularization compared to sham mice, and no difference was found between GLF + FRCs in biologic gels and gelatin scaffolds. This high vascularization may have contributed to accelerated immune cell infiltration and graft rejection in the fat pad compared to the less vascularized subcutaneous site NOD model. Comparable vascularization between the biologic gels and the macroporous gelatin scaffolds indicates that higher FRC survival in macroporous gelatin scaffolds is not due to enhanced scaffold revascularization. Figure 4. Open in a new tab In vivo performance of GFP-Luciferase fusion (GLF)-expressing FRCs in mice. (A-B) Survival of FRCs (800k cells/scaffold) in 3D scaffolds after transplantation in the dorsal subcutaneous (A, S: small, M: medium, L: large microporous gelatin scaffold) and vascularized gonadal fat pad (B. M: medium pore only) sites of pre-diabetic 8 wk-old NOD mice by IVIS longitudinal imaging. N=3 scaffolds/mouse. (C-D) Survival of FRCs (400k cells/scaffold) in the vascularized gonadal fat pad of 8 wk-old NOD (C) and of immunodeficient NODscid mice (D) comparing macroporous gelatin scaffolds (Gelatin) to biologic scaffold hydrogels (Biologic) (N=4 scaffold/mouse). Data are represented as mean ± standard deviation. P-values were determined by ordinary two ANOVA followed by Tukey and Šidák correction. (**p < 0.001; *p < 0.01 ns, not significant, p > 0.05). Scaffolds promote FRC phenotypic identity, TGF-β2 secretion, lymph node-like reticula formation, subsequently altering cellular contractility We then analyzed transcriptional and translational expression of FRC phenotypic markers known for their structural, immunological, and regulatory roles in 3D culture in different pore-size scaffolds compared to 2D culture controls, allowing us to assess how scaffold architecture influences FRC function. FRCs cultured in 3D scaffolds with small, medium, or large pore sizes showed a significant increase in FRC phenotypical markers gp38 and PDGFRα, measured via flow cytometry after 21-day culture compared to passage-matched FRCs cultured in 2D conditions ( Fig. 5A , gating strategy in Fig. S5 on live, FRC singlets) (p<0.001). More importantly, FRCs cultured in small pore size scaffolds showed the highest upregulation of FRC phenotypical markers compared to medium and large pore size scaffolds (54.57 ± 0.43 %, p<0.001). Levels of gp38 canonical FRC marker varied depending on the scaffold pore size ( Fig. 5B ) and time ( Fig. S7 ), with small pore scaffolds yielding the highest expression of gp38 hi FRCs (58.48 ± 1.22 %, p<0.001) after 21-day culture in scaffolds. Figure 5. Open in a new tab Characterization of FRC engineered reticula fabricated by culturing murine FRCs in macroporous gelatin scaffolds with small, medium, and large pores for up to 21 days. (A) FRC marker expression in scaffolds compared to 2D culture measured by flow cytometry (n=3) after 21 days. (B) Representative parts of whole charts of FRC gp38 expression (low, medium or high) cultured in scaffolds for 21 days (average of n=3). (C-G) Normalized gene expression of gp38, collagen I, CCL21 (gene expression exclusively detected on FRCs in 3D but not 2D), α-SMA, TGF-β2 and laminin α−5 relative to FRCs cultured in 2D (n=3). (D) Laminin α−4 and laminin α−5 relative gene expression ratio (n=3). (E) TGF-β2 secretion by ELISA after 48 hours in scaffolds in FBS-deprived media relative to FRCs cultured in 2D(n=3). (F) FRC reticular pore quantification after 1, 7, 14 and 21 days in scaffolds (average Feret’s diameter of at least 10 reticular pores per n=3 scaffolds per time point). (G) Representative immunofluorescence and confocal microscopy images (F-actin: green, nuclei: blue, scale bar = 100 μm) after 1 day, 7 days, 14 days and 21 days of FRCs in scaffolds. (H) Generated forces from FRCs on a soft, compliant 1.2 kPa hydrogel quantified by traction force microscopy after culturing in 2D and 3D scaffolds with small, medium, and large pore sizes. Data represented as mean ± standard deviation. P-values were determined by ordinary one-way ANOVA followed by Tukey’s multiple comparison test. (*p<0.01; **p<0.001; ****p < 0.00001; ns, not significant, p > 0.05). Via RT-qPCR, we confirmed that Gp38 (Pdpn) gene expression was increased in small and medium pore scaffolds, compared to large pore scaffolds ( Fig. 5C ). All 3D conditions showed a significant increase in Gp38 gene expression compared to passage-matched FRCs cultured in 2D ( Fig. S8A ), with FRCs cultured in small and medium pore sizes showing higher gp38 gene expression levels compared to large pore scaffolds. There was also a significant increase in collagen I (Col1a1) gene expression compared to 2D ( Fig. 5C , S8B ) for small and medium pore size scaffolds, although no difference was observed in gene expression between scaffold pore sizes. Additionally, Ccl21b gene expression was the most upregulated in medium pore size scaffolds, compared to those with small and large pores ( Fig. 5C , S8C ), and was not detected in 2D cultures. Notably, previous studies reported that CCL21 expression is only upregulated when stromal cells undergo laminar flow conditions [ 66 ]. Thus, the observed upregulation of Ccl21b in our 3D scaffolds, without flow, suggests a unique aspect of our 3D macroporous scaffolds and partial recovery of the FRC phenotypic identity. Ccl19, however, was not detected in either condition (data not shown). Conversely, α-SMA (Acta2) and Laminin α−5 (Lama5) gene expressions decreased, while Laminin α−4 (Lama4) expression increased across all scaffold conditions compared to 2D culture ( Fig. 5C , S8D – F ). No significant differences were observed between scaffold pore sizes, except for a reduction in Laminin α−5 expression in large pore scaffolds compared to medium pores. Additionally, the pro-tolerogenic Laminin α−4 to Laminin α−5 gene expression ratio, as previously reported [ 31 ], whereby higher Laminin α−4 and lower Laminin α−5 levels promoted T cell suppression and a tolerogenic environment, along with Tgfb2 gene expression, was higher in small and medium pore scaffolds compared to large pores ( Fig. 5D ). While Tgfb1 was not detected (data not shown), FRCs cultured in 3D scaffolds showed a 15-fold increase in active TGF-β2 levels compared to 2D conditions (p<0.001), with no significant differences observed among the three pore sizes (p>0.05) ( Fig. 5E , S8G ). These results were further validated via immunofluorescence staining of fibroblast and ECM markers ( Fig. S9 – 10 ). FRCs in 3D scaffolds secreted ECM proteins (Pan-laminin, Polyclonal Collagen I and Collagen IV) and maintained fibroblast marker expression (α-SMA and vimentin). Similar to RT-qPCR results, there was no significant difference in collagen, α-SMA, and laminin fluorescence intensity between scaffold pore sizes ( Fig. S10 ). We then quantified the pore size of the filamentous actin (F-actin) networks formed by FRCs around the scaffold pores over time to evaluate whether scaffolds could promote FRC organization into LN-like reticular structure. While F-actin itself is an intracellular structure, its organization reflects the cells’ ability to spread, align, and form interconnected networks that contribute to the extracellular architecture ( Fig. S11 ). We observed a decrease in the average pore size of these FRC networks across all scaffold conditions between days 1 and 21 ( Fig. 5F , G , Table 2 ). This progressive remodeling suggests the formation of tissue-engineered FRC reticula. To determine whether the observed decrease in pore size was due to FRC-mediated remodeling rather than scaffold swelling, we incubated scaffolds without FRCs in media for 21 days. Afterward, we seeded FRCs onto scaffolds for 24 hours and compared their F-actin-defined pore sizes to those of FRCs cultured in scaffolds for the full 21 days ( Fig. S8H ). FRCs cultured for 21 days exhibited significantly smaller pores across all scaffolds compared to 24-hour FRC-seeded controls. To further assess cell infiltration in scaffolds, integration, and 3D reticular organization IMARIS software was used to generate oblique slicers in the XY, XZ, and YZ planes, confirming FRC 3D distribution throughout the scaffold as evidenced by their organization in multiple planes ( Fig. S11 ). Table 2. FRC Filamentous Actin Structure Pore size (μm) Summary Time (Day) Scaffold Pore Size Range Small Medium Large 1 34.64 ± 3.06 103.20 ± 8.80 221.79 ± 14.27 7 33.77 ± 2.33 91.75 ± 4.96 217.24 ± 8.26 14 21.57 ± 1.54 86.36 ± 4.10 182.50 ± 22.22 21 19.68 ± 1.88 82.66 ± 5.96 166.74 ± 8.94 Open in a new tab Lastly, traction force microscopy was used to assess whether the differences in FRC phenotypic identity observed between FRCs cultured in small, medium, and large pore size scaffolds, or 2D conditions corresponded to significant functional changes in cell behavior, particularly contractility via cell traction forces ( Fig. 5H , S12 ). Our results revealed a significant difference in FRC contractility between 2D-cultured FRCs and FRCs cultured in large pore scaffolds (p<0.01), as well as between FRCs in small and large pore scaffolds (p<0.01). FRC scaffold pore size and polyclonal to Ag-specific T cell ratio shape Ag-specific T cell immunomodulation It has been previously shown that 3D FRC culture increases FRC-T cell interactions and engagement of Ag-specific T cells [ 12 ]. However, these studies used commercially available collagen scaffolds with pore sizes larger than the FRC reticula in LNs. Therefore, here, we used our tissue-engineered FRC reticula of small, medium, and large LN pore sizes which better mimic LN FRC reticula in normal and inflamed LNs to evaluate the immunomodulatory effects of 3D-cultured FRCs to decrease effector function of diabetogenic T cells. We also used FRCs that present selected T1D-relevant Ags in the form of pulsed peptides instead of previously-used genetically engineered FRCs that express these Ags [ 12 ]. Specifically, we aimed to study whether scaffold pore size affected FRC-T cell interactions, T cell engagement and regulation in different FRC-T cell co-culture conditions mimicking different abundances of diabetogenic T cells ( Table 3 ). We hypothesized that tighter FRC networks in small pore scaffolds would increase T cell engagement in cultures with more Ag-specific T cells. We then performed co-cultures with polyclonal T cells to better mimic physiological conditions where diabetogenic T cells represent the minority of T cells. In the presence of more Ag-specific T cells, we hypothesized that the tighter networks in small pore scaffolds would likely increase interactions due to reduced space. However, when polyclonal T cells are present, the limited space in these tighter networks would restrict T cell motility, potentially reducing the ability of Ag-specific T cells to interact with antigen-expressing FRCs. Table 3. FRC and T cell co-culture experimental conditions Co-Culture Ag-specific T cell Polyclonal T cell Ratio (Polyclonal: Ag-specific) Exogenous T1D peptide 1 p79/HIP2.5-reactive CD4 + T cells (MHC-II) CD45.2 + CD3 + BDC2.5 – T cells 90/10 p79 2 p79/HIP2.5-reactive CD4 + T cells (MHC-II) CD45.2 + CD3 + BDC2.5 – T cells 50/50 p79 3 p79/HIP2.5-reactive CD4 + T cells (MHC-II); and IGRP 206–214 - reactive CD8 + T cells (MHC-I) None 0/100 (50/50 Ag-specific CD4 + : Ag-specificCD8 + ) p79 + IGRP 206–214 4 INS B15–23 CD8 + T cells (MHC-I) from G9C8 mice None 0/100 INS B15–23 Open in a new tab FRCs were pre-seeded in scaffolds with varying pore size ranges (small: S, medium: M and large: L pore sizes) and conditioned with IFN-γ ( Fig. S1 ), to increase their Ag presentation capability, and selected T1D Ag peptides (S, M, L) or no Ag (S-, M-, L-), and co-cultured with Ag-specific CD4 + T cells (from BDC2.5 mice) and/or CD8 + T cells (from either NY8.3 or G9C8 mice) and/or polyclonal (Poly) T cells for 3 days, as indicated in Table 3 . As controls for professional APC engagement, we activated T cells with αCD3/CD28 Dynabeads (+ Ctl). T cell engagement by FRCs was characterized as proliferation (CellTrace dilution) ( Fig. 6A ), TCR engagement (as % CD44 + CD25 + ) ( Fig. 6B ), cytotoxicity (as % proliferated GrB + ) ( Fig. 6C ), regulatory (as % CD25 + FoxP3 + ) ( Fig. 6D ) and anergic (as % CD73 + FR4 + Foxp3 − ) ( Fig. 6E ) phenotype among Ag-specific T cells, and relative abundance of regulatory and anergic cells among other T cells. Gating strategy is shown in Fig. S13 . Figure 6. Open in a new tab Engagement and immunomodulation of diabetogenic, Ag-specific T cells in 72-hour 3D scaffold co-culture with FRCs. (A) Representative flow cytometry plots of Ag-specific T cell (CD4 + or CD8 + ) proliferation measured by CellTrace dilution after interaction with FRCs presenting T1D Ags in co-culture with (1) more polyclonal T cells (90% CD45.2 + BDC2.5 − CD3 + T cells + 10% CD45.2 − BDC2.5 + CD4 + T cells), (2) more Ag-specific T cells (50% CD45.2 + BDC2.5 − CD3 + T cells + 50% CD45.2 − BDC2.5 + CD4 + T cells), (3) 50% BDC2.5 + CD4 + T cells + 50% NY8.3 + CD8 + T cells, and (4) 100% G9C8 + CD8 + T cells (n=3–4). (B) Representative flow cytometry plots and Ag-specific T cell engagement in co-culture with FRCs quantified by CD44 and CD25 upregulation on Ag-specific (BDC2.5 or G9C8) T cells (n=3–4). (C) Representative flow cytometry plots and Ag-specific (NY8.3 or G9C8) CD8 + T cell cytotoxic phenotype based on Granzyme B intracellular expression (n=3–4). (D) Representative flow cytometry plots and Ag-specific, CD4 + T reg phenotype based on Foxp3 and CD25 expression, gated on engaged (CD44 + CD25 + ), Ag-specific CD4 + T cells (n=3). (E) Representative flow cytometry plots and CD4 + anergic T cell phenotype of engaged, Ag-specific (BDC2.5) CD4 + T cells based on CD73 + FR4 + expression (gating out regulatory (Foxp3 + ) T cells) (n=3). (F) Representative parts of whole charts of Ag-specific, engaged, CD4 + T cell phenotype as regulatory, anergic, or other/effector T cells (average of n=3). Co-culture flow cytometry controls include co-cultures without addition of soluble Ag (negative control) and αCD3/CD28 Dynabeads (positive control). P-values were determined by ordinary one-way ANOVA followed by Tukey’s multiple comparison test or Kruskal-Wallis test for data not normally distributed. (*p<0.05; **p<0.01; ***p<0.001; ****p<0.0001; ns, not significant, p > 0.05). As expected, we found proliferation ( Fig. 6A ) and stimulation ( Fig. 6B ) of specific CD4 + and CD8 + T cells by FRCs presenting T1D Ags (S, M, L) but not by FRCs without Ags (S-, M-, L-). As hypothesized, we confirmed that larger FRC reticular pores decrease the ability of Ag-expressing FRCs to expand ( Fig. 6A ) and engage ( Fig. 6B ) specific T cells in vitro in conditions with more Ag-specific T cells (conditions 2, 3, 4) after 3-day co-culture, indicating that smaller pore sizes are preferable for T cell immunomodulation in those conditions. However, smaller pores likely reduce T cell motility due to limited space. In conditions with more polyclonal than Ag-specific T cells (co-culture condition 1), larger pores facilitated greater T cell proliferation. Overall, we found that Ag-specific T cells engaged by Ag-presenting FRCs (S, M, L) have negligible cytotoxicity ( Fig. 6C ), and increased regulatory ( Fig. 6D , S14A ) and anergic ( Fig. 6E , S14B ) phenotype compared to αCD3/CD28-treated T cells (+ Ctl). We also did not find an effect of scaffold pore size or T cell ratio (conditions 1–3) on T cell phenotype (relative ratio of T regs and anergic T cells among other T cells) after engagement and proliferation ( Fig. 6F ). Importantly, T cell engagement was Ag-dependent, since effects were seen with FRC presenting Ags (S, M, L) and in Ag-specific CD4 + CD45.2 − T cells ( Fig. 6 ), but not in CD4 + CD45.2 + polyclonal T cells ( Fig. S15 ). Discussion To address the need for more targeted immunomodulation of Ag-specific T cells in T1D, we developed tissue-engineered FRC reticula with customizable pore sizes to mimic the dynamic LN FRC reticula, ranging from 20–200 μm, and examined whether 3D culture in scaffolds with varying pore sizes affects FRC phenotype and FRC-T cell interactions. We first demonstrated the fabrication of macroporous gelatin scaffolds with pore sizes adjusted by modifying freezing temperature, gelatin concentration, and crosslinking time. Despite having used the same lot of gelatin, this natural material contains collagen fragments, cytokines, chemokines, growth factors, and small metabolic molecules that vary between manufacturers’ lot, which may have impacted our studies. Corroborating previous work [ 38 – 40 , 67 – 70 ], lower freezing temperatures and higher gelatin concentrations produced smaller pores. The mechanical properties of the scaffolds also varied with pore size, as smaller pores resulted in higher stiffness (Young’s Modulus) than larger pore size scaffolds. This is likely due to the increased density and reduced porosity of these scaffolds [ 39 ], confirmed via liquid displacement experiments. Similar findings by Zhao et al. showed that larger pores reduce mechanical strength [ 71 ]. Overall, the high porosity across all scaffold sizes supports enhanced cell infiltration and tissue integration, as previously highlighted [ 72 ]. Our macroporous scaffolds also exhibited desirable biocompatibility, integrating into subcutaneous tissue in NOD mice without significant immune activation. Minimal cytokine secretion (IL-6, TNF-α, IL-10) and splenocyte proliferation in NOD and B6 mouse models indicated a lack of pro-inflammatory or anti-inflammatory responses, supporting their suitability for in vivo applications. Moderate fibrosis observed around the scaffolds aligns with expected biomaterial responses essential for host integration [ 73 ]. While smaller pores can increase fibrous encapsulation, and larger pores improve tissue integration and vascularization [ 73 – 77 ], the pore size range in our scaffolds (~250 μm) and the 21-day evaluation period were likely insufficient to observe such differences, compared to longer reported studies (21 days, compared to 12 weeks [ 78 ]). The presence of immature blood vessels within and surrounding the implanted scaffolds further suggests successful biocompatible integration with native host tissue. Mature blood vessels, indicated by α-SMA coverage [ 79 ], were observed at the scaffold border. Moreover, the presence of a higher ratio of M2-like macrophages compared to M1-like macrophages 21 days post-implantation, coupled with the absence of excessive fibrotic capsule formation, indicates minimal foreign body response [ 80 ]. Our studies suggest efficient incorporation and remodeling of our scaffolds by the host tissue. Our findings also highlight the limitations of using NOD mice for in vivo longitudinal monitoring of GLF + cells by IVIS imaging when cells are implanted in vascularized sites. These results align with previous studies in cancer [ 62 – 64 ] and T1D models [ 65 ], thus emphasizing the importance of carefully selecting appropriate animal models for future research. We hypothesized that a porous, 3D platform for FRCs would better preserve FRC identity, while assessing how varying pore sizes influenced FRC marker expression. Our findings confirm that our 3D culture system supports FRC viability and enhances their phenotypic identity, particularly their immunomodulatory properties (via TGF-β2) and ECM production as initially hypothesized. The upregulation of FRC phenotypical markers gp38 and PDGFRα in 3D scaffolds compared to 2D culture highlights the importance of a 3D environment for maintaining FRC phenotype. Small pore scaffolds showed the highest gp38 expression after 21 days, suggesting a closer mimicry of FRCs in their native LN environment. The observed differences in gp38 expression levels may reflect post-transcriptional regulation. Given the role of gp38 in lymphatic development, immune cell interaction, and tissue architecture [ 81 ], variations in its expression could influence FRC activation, tissue organization, or immune cell migration. As gp38 supports interactions with platelets and dendritic cells via CLEC-2 [ 15 , 64 – 69 ], future tri-culture studies with dendritic cells, FRCs, and T cells will assess its role in LN expansion upon immune activation using our scaffolds. CCL21 expression, essential for lymphocyte recruitment [ 66 , 82 , 83 ], was highest in medium-pore 3D scaffolds, supporting the use of this pore size for in vivo studies requiring lymphocyte homing. Unlike 2D cultures, where CCL21 was lost, 3D scaffolds maintained its expression, demonstrating their potential to preserve FRC cytokine expression. Laminin α−4/Laminin α−5 ratios in FRCs also varied by pore size, with small and medium pores favoring higher ratios, which have been associated with a tolerogenic niche that enables allograft survival and promotes regulatory T cells [ 31 ]. Additionally, ECM components, crucial for structural support [ 84 ], showed increased expression in FRCs cultured in scaffolds compared to 2D culture. Lastly, our scaffolds also supported FRC network formation, essential for LN structural support and immune cell movement. Reduced α-SMA expression, a myofibroblast marker associated with fibrosis and inflammation in LNs, in our 3D scaffolds highlights the role of mechanical properties in regulating FRC phenotype. Nonetheless, our scaffolds exhibited Young’s Moduli at least twice that of reported lymph node (LN) stiffnesses [ 85 ]. Yuen et al. showed that in vivo Young’s Modulus of porcine cervical LNs is 26 kPa, comparable in magnitude to our engineered macroporous scaffolds. Further studies on inter-species and LN zone variability [ 86 ], could provide deeper insights into LN mechanical properties. Additionally, increased TGF-β2 secretion in 3D cultures indicates FRC-mediated immune modulation, particularly Treg expansion [ 87 ], supporting the promise of FRC-based immune therapies. Similar findings by Velankar et al. using self-assembling peptides showed FRC clustering, enhanced T cell infiltration, and an immunosuppressive environment, highlighting the advantages of 3D FRC culture [ 88 ]. The most pronounced phenotypic differences between FRCs cultured in scaffolds, particularly between small and large pore scaffolds, were observed in gp38 expression levels. Moreover, since large pore scaffolds had a significantly lower Young’s modulus, this may have contributed to the reduced contractility of FRCs due to the lower stiffness of these scaffolds. Although α-SMA gene expression did not show significant changes, this difference in scaffold stiffness likely translated into diminished contractile function in FRCs cultured in large pore scaffolds. Additionally, TGF-β also plays a crucial role in cellular contractility, with higher levels resulting in increased fibroblast contractility [ 89 ]. While not significantly seen at the gene expression level, likely due to the greater gene fold-change differences from comparing 2D to 3D FRC culture, small scaffolds had slightly higher Tgfb2 gene expression than large scaffolds, which may have contributed to the observed differences in contractility. Notably, while α-SMA is commonly associated with fibroblast contractility [ 89 ], the variability observed in gp38 levels suggests that these phenotypic markers may also synergistically impact contractile behavior. Our traction force microscopy studies provided valuable insights into how phenotypic alterations in FRCs can influence their contractility, a critical aspect of their role in lymph node mechanics during inflammation [ 15 ]. Nonetheless, the overnight culture of FRCs in soft gels for traction force microscopy may have mitigated more subtle differences in contractility. Despite this limitation, the results underscore the importance of phenotypic changes in dictating FRC functionality within the lymph node microenvironment. Our 3D platform provides a pre-clinical foundation for implantable tissue-engineered therapies for ADs and represents a significant advancement in the development of biomaterials for tissue engineering and immunomodulation. Given FRCs’ promising immunomodulatory effects [ 90 , 91 ], we co-cultured our tissue-engineered FRC reticula of varying pore sizes with diabetogenic T cells. These effects occur through Ag-dependent and Ag-independent mechanisms [ 13 ], mainly via the expression of PD-L1 by FRCs [ 20 , 92 , 93 ], which can induce T cell tolerance [ 24 , 94 , 95 ] and anergy, a state where T cells acquire reduced responsiveness to their cognate Ag upon re-exposure. Moreover, APC culture in 3D substrates increases surface area for T cell engagement, aiding ex vivo activation and expansion of tumor-specific T cells [ 96 , 97 ]. Due to the extended timeframe required for diabetes to develop in NOD mice (starting at 12 weeks of age) and the potential complications arising from repopulation with other stromal and immune cell types upon FRC scaffold implantation, we focused on in vitro studies. We assessed whether the immunomodulatory effects of FRCs would reduce diabetogenic, autoreactive T cell phenotype in T1D models using our tissue-engineered reticula. The surface area to volume ratio (SA/V) of 3D scaffolds is crucial for enhancing cell-cell interactions, which promote processes like proliferation and immune modulation. We hypothesized that increasing the SA/V ratio (highest in small pore scaffolds) would lead to higher T cell proliferation. Our findings confirmed this hypothesis and showed that smaller pore sizes were more effective at engaging Ag-specific T cells, likely due to tighter FRC networks that increase T cell-FRC interactions. However, when the percentage of Ag-specific T cells added to the co-culture was lower, larger pore scaffolds favored interactions between FRCs and the smaller number of Ag-specific T cells. In smaller pore size scaffolds, these Ag-specific T cells might have been hindered by polyclonal cells that overcrowded the more limited space available in these conditions. Larger pores may have also enhanced T cell motility, while also providing additional space for T cell proliferation. Conversely, in smaller pores, with decreased porosity, the few Ag-specific T cells in the culture had less opportunity to encounter Ag-presenting FRCs. Our results showed that FRCs in 3D scaffolds exhibit Ag-specific immunomodulatory capabilities, as polyclonal T cells did not demonstrate significant proliferation compared to Ag-specific conditions. Despite this lack of significant interaction, we found a slight upregulation of CD44 and CD25 in a few polyclonal T cells compared to negative polyclonal controls without Ag. This may be due to the influence of cytokines secreted by activated Ag-specific T cells, acting on a small population of likely Ag-experienced polyclonal T cells poised to respond to these stimuli. This suggests that while direct Ag recognition is crucial for targeted T cell proliferation, the local cytokine milieu [ 98 , 99 ], influenced by Ag-specific interactions, can also modulate the activation status of surrounding T cells in an Ag-independent manner. Additionally, the T cell pool may have contained a minor subset of Ag-specific T cells contributing to this effect. This phenomenon highlights the complex interplay between Ag-specific and bystander T cell activation within the LN microenvironment, emphasizing the role of FRCs in both Ag presentation to T cells and shaping the overall immune response through cytokine-mediated signaling pathways. Moreover, the negligible cytotoxicity and increased regulatory and anergic T cell phenotypes observed in CD4 + T cells engaged by Ag-presenting FRCs further emphasize the potential of our platform for immunomodulatory applications. Peripheral induction of T regs occurs through Ag presentation with TGF-β [ 100 ], which we confirmed is secreted by our tissue-engineered reticula. The minimal differences in T cell phenotype across scaffold pore sizes suggest that while pore size affects T cell engagement, it does not significantly impact the overall regulatory capacity of FRC-T cell interactions. Future studies will explore FRC-based immunomodulation in in vivo T1D mouse models, focusing on memory T cells, as FRCs play a crucial role in their maintenance within LNs [ 101 ]. Conclusion Herein, we demonstrate that tissue-engineered FRC reticula with different pore sizes can be fabricated using freeze-dried macroporous gelatin scaffolds. Our scaffolds are biocompatible and support the survival and function of FRCs in vitro and in vivo , modulate Ag-specific T cell phenotype and tolerogenic engagement, and provide a promising platform for developing Ag-specific immunotherapies for T1D and other ADs. We also provide evidence for the limitations of the NOD model for longitudinal imaging monitoring of GLF + cells in highly vascularized implant sites. The upregulation of gp38 and CCL21 highlights the critical nature of a 3D culture environment in maintaining the integrity of LN networks and promoting immune cell trafficking, which is significantly lost in 2D cultures. Our results also highlight the complex relationship between scaffold architecture, FRC phenotypic identity, and consequent changes functional behavior in FRCs. Future studies will explore the in vivo therapeutic effect of our tissue-engineered reticula in AD models. Supplementary Material 1 NIHMS2132223-supplement-1.docx (5.3MB, docx) Fabricated freeze-dried macroporous gelatin scaffolds with customizable pore diameters to mimic naive and inflamed lymph node paracortex reticula. Seeded fibroblastic reticular cells (FRCs) from non-obese diabetic (NOD) mice lymph nodes onto scaffolds to create tissue-engineered reticula in Type 1 Diabetes models. Demonstrated that scaffold pore size influences in vitro FRC phenotypic identity, extracellular matrix secretion, and diabetogenic T cell regulation. Showed that scaffolds support in vivo FRC viability, local retention, and host integration in NOD mice. ACKNOWLEDGMENTS The authors would like to thank the staff at the Diabetes Research Institute for their expertise, research services, and support. Namely, the authors would like to acknowledge Dr. Oliver Umland from the Flow Cytometry Core facility, Dr. Maria Boulina from the Analytical Imaging Core facility, Kevin Johnson from the Histology Core, Dr. Joel Szust, Greycy Vega, Yelena Gadea, Shirley Vega and Roy Morales from the Animal Core facility, and Dr. Christopher Fraker, Dr. Armando Mendez and Dr. Angela Szeto for their generosity in lending the plate reader and lyophilizer. The authors would also like to acknowledge Dr. Courtney Dumont in Biomedical Engineering at the University of Miami for her generosity in lending the lyophilizer. Schematics were created using Biorender ( biorender.com ). Research reported in this publication was performed in part at the Cancer Modeling Shared Resource (CMSR) of the Sylvester Comprehensive Cancer Center at the University of Miami, RRID: SCR_022891, which is supported by the National Cancer Institute (NCI) of the National Institutes of Health (NIH) under award number P30CA240139. FUNDING SOURCES Funding was provided by the National Institute of Health (NIH) grants R56DK131087, R56AI151217-01 and R01DK141150. This research was also supported by the Diabetes Research Initiative Foundation at the University of Miami. The authors acknowledge the critical support and resources provided by these institutions, which have significantly contributed to the advancement of this work. The content is solely the responsibility of the authors and does not necessarily represent the official views of the University of Miami or the Diabetes Research Institute Foundation. ABBREVIATIONS ATP Adenosine triphosphate α-SMA Alpha smooth muscle actin Ag Antigen APC Antigen-presenting cell Arg1 Arginase 1 AD Autoimmune diseases CCL21 C-C motif chemokine 21 Ct Cycle threshold ELISA Enzyme-linked immunosorbent assay ECM Extracellular matrix FBS Fetal bovine serum FRC Fibroblastic reticular cell GTA Gelatin-glutaraldehyde GLF GFP-Luciferase fusion iNOS Inducible Nitric Oxide Synthase IL-10 Interleukin-10 IL-6 Interleukin-6 LAMA4 Laminin Subunit Alpha 4 LAMA5 Laminin Subunit Alpha 5 L Large LPS Lipopolysaccharide LN Lymph node MHC Major Histocompatibility Complex M Medium NOD Non-obese diabetic PDGFRα Platelet-derived growth factor receptor Alpha PCR Polymerase Chain Reaction T reg Regulatory T cell RT-PCR Reverse transcription polymerase chain reaction RT-qPCR Reverse transcription quantitative real-time PCR SEM Scanning Electron Microscopy S Small SA Surface area 3D Three-dimensional TGF-β2 Transforming growth factor-beta 2 Tg Transgenic TNF-α Tumor necrosis factor alpha 2D Two-dimensional T1D Type 1 diabetes V Volume Footnotes Declaration of interests The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Alice A. Tomei reports financial support was provided by National Institutes of Health. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. REFERENCES [1]. Pugliese A, Autoreactive T cells in type 1 diabetes, J Clin Invest 127(8) (2017) 2881–2891. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [2]. Herold KC, Gitelman SE, Gottlieb PA, Knecht LA, Raymond R, Ramos EL, Teplizumab: A Disease-Modifying Therapy for Type 1 Diabetes That Preserves beta-Cell Function, Diabetes Care 46(10) (2023) 1848–1856. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [3]. Felton JL, Griffin KJ, Oram RA, Speake C, Long SA, Onengut-Gumuscu S, Rich SS, Monaco GSF, Evans-Molina C, DiMeglio LA, Ismail HM, Steck AK, Dabelea D, Johnson RK, Urazbayeva M, Gitelman S, Wentworth JM, Redondo MJ, Sims EK, Ada P /Easd, Disease-modifying therapies and features linked to treatment response in type 1 diabetes prevention: a systematic review, Commun Med (Lond) 3(1) (2023) 130. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [4]. Bone RN, Evans-Molina C, Combination Immunotherapy for Type 1 Diabetes, Curr Diab Rep 17(7) (2017) 50. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [5]. Firdessa Fite R, Bechi Genzano C, Mallone R, Creusot RJ, Epitope-based precision immunotherapy of Type 1 diabetes, Hum Vaccin Immunother 19(1) (2023) 2154098. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [6]. Kenison JE, Stevens NA, Quintana FJ, Therapeutic induction of antigen-specific immune tolerance, Nat Rev Immunol 24(5) (2024) 338–357. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [7]. Castenmiller C, Keumatio-Doungtsop BC, van Ree R, de Jong EC, van Kooyk Y, Tolerogenic Immunotherapy: Targeting DC Surface Receptors to Induce Antigen-Specific Tolerance, Front Immunol 12 (2021) 643240. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [8]. Kambayashi T, Laufer TM, Atypical MHC class II-expressing antigen-presenting cells: can anything replace a dendritic cell?, Nat Rev Immunol 14(11) (2014) 719–30. [ DOI ] [ PubMed ] [ Google Scholar ] [9]. Ferreira BO, Gamarra LF, Nucci MP, Oliveira FA, Rego GNA, Marti L, LN-Derived Fibroblastic Reticular Cells and Their Impact on T Cell Response-A Systematic Review, Cells 10(5) (2021). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [10]. Hirosue S, Dubrot J, Modes of Antigen Presentation by Lymph Node Stromal Cells and Their Immunological Implications, Front Immunol 6 (2015) 446. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [11]. Li L, Wu J, Abdi R, Jewell CM, Bromberg JS, Lymph node fibroblastic reticular cells steer immune responses, Trends Immunol 42(8) (2021) 723–734. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [12]. Gonzalez Badillo F, Zisi Tegou F, Masina R, Wright S, Scully M, Harwell L, Lupp M, Postigo-Fernandez J, Creusot RJ, Tomei AA, Tissue-Engineered Stromal Reticula to Study Lymph Node Fibroblastic Reticular Cells in Type I Diabetes, Cell Mol Bioeng 13(5) (2020) 419–434. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [13]. Fletcher AL, Acton SE, Knoblich K, Lymph node fibroblastic reticular cells in health and disease, Nat Rev Immunol 15(6) (2015) 350–61. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [14]. Bajenoff M, Egen JG, Koo LY, Laugier JP, Brau F, Glaichenhaus N, Germain RN, Stromal cell networks regulate lymphocyte entry, migration, and territoriality in lymph nodes, Immunity 25(6) (2006) 989–1001. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [15]. Astarita JL, Cremasco V, Fu J, Darnell MC, Peck JR, Nieves-Bonilla JM, Song K, Kondo Y, Woodruff MC, Gogineni A, Onder L, Ludewig B, Weimer RM, Carroll MC, Mooney DJ, Xia L, Turley SJ, The CLEC-2-podoplanin axis controls the contractility of fibroblastic reticular cells and lymph node microarchitecture, Nat Immunol 16(1) (2015) 75–84. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [16]. Martinez VG, Pankova V, Krasny L, Singh T, Makris S, White IJ, Benjamin AC, Dertschnig S, Horsnell HL, Kriston-Vizi J, Burden JJ, Huang PH, Tape CJ, Acton SE, Fibroblastic Reticular Cells Control Conduit Matrix Deposition during Lymph Node Expansion, Cell Rep 29(9) (2019) 2810–2822 e5. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [17]. Dasoveanu DC, Shipman WD, Chia JJ, Chyou S, Lu TT, Regulation of Lymph Node Vascular-Stromal Compartment by Dendritic Cells, Trends Immunol 37(11) (2016) 764–777. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [18]. Assen FP, Abe J, Hons M, Hauschild R, Shamipour S, Kaufmann WA, Costanzo T, Krens G, Brown M, Ludewig B, Hippenmeyer S, Heisenberg CP, Weninger W, Hannezo E, Luther SA, Stein JV, Sixt M, Multitier mechanics control stromal adaptations in the swelling lymph node, Nat Immunol 23(8) (2022) 1246–1255. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [19]. Knoblich K, Cruz Migoni S, Siew SM, Jinks E, Kaul B, Jeffery HC, Baker AT, Suliman M, Vrzalikova K, Mehenna H, Murray PG, Barone F, Oo YH, Newsome PN, Hirschfield G, Kelly D, Lee SP, Parekkadan B, Turley SJ, Fletcher AL, The human lymph node microenvironment unilaterally regulates T-cell activation and differentiation, PLoS Biol 16(9) (2018) e2005046. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [20]. Fletcher AL, Lukacs-Kornek V, Reynoso ED, Pinner SE, Bellemare-Pelletier A, Curry MS, Collier AR, Boyd RL, Turley SJ, Lymph node fibroblastic reticular cells directly present peripheral tissue antigen under steady-state and inflammatory conditions, J Exp Med 207(4) (2010) 689–97. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [21]. Yip L, Su L, Sheng D, Chang P, Atkinson M, Czesak M, Albert PR, Collier AR, Turley SJ, Fathman CG, Creusot RJ, Deaf1 isoforms control the expression of genes encoding peripheral tissue antigens in the pancreatic lymph nodes during type 1 diabetes, Nat Immunol 10(9) (2009) 1026–33. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [22]. Yip L, Creusot RJ, Pager CT, Sarnow P, Fathman CG, Reduced DEAF1 function during type 1 diabetes inhibits translation in lymph node stromal cells by suppressing Eif4g3, J Mol Cell Biol 5(2) (2013) 99–110. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [23]. Dertschnig S, Evans P, Santos ESP, Manzo T, Ferrer IR, Stauss HJ, Bennett CL, Chakraverty R, Graft-versus-host disease reduces lymph node display of tissue-restricted self-antigens and promotes autoimmunity, J Clin Invest 130(4) (2020) 1896–1911. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [24]. Keir ME, Freeman GJ, Sharpe AH, PD-1 regulates self-reactive CD8+ T cell responses to antigen in lymph nodes and tissues, J Immunol 179(8) (2007) 5064–70. [ DOI ] [ PubMed ] [ Google Scholar ] [25]. Maarouf OH, Uehara M, Kasinath V, Solhjou Z, Banouni N, Bahmani B, Jiang L, Yilmam OA, Guleria I, Lovitch SB, Grogan JL, Fiorina P, Sage PT, Bromberg JS, McGrath MM, Abdi R, Repetitive ischemic injuries to the kidneys result in lymph node fibrosis and impaired healing, JCI Insight 3(13) (2018). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [26]. Fletcher AL, Elman JS, Astarita J, Murray R, Saeidi N, D’Rozario J, Knoblich K, Brown FD, Schildberg FA, Nieves JM, Heng TS, Boyd RL, Turley SJ, Parekkadan B, Lymph node fibroblastic reticular cell transplants show robust therapeutic efficacy in high-mortality murine sepsis, Sci Transl Med 6(249) (2014) 249ra109. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [27]. Li X, Zhao J, Kasinath V, Uehara M, Jiang L, Banouni N, McGrath MM, Ichimura T, Fiorina P, Lemos DR, Shin SR, Ware CF, Bromberg JS, Abdi R, Lymph node fibroblastic reticular cells deposit fibrosis-associated collagen following organ transplantation, J Clin Invest 130(8) (2020) 4182–4194. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [28]. Birgersdotter A, Sandberg R, Ernberg I, Gene expression perturbation in vitro--a growing case for three-dimensional (3D) culture systems, Semin Cancer Biol 15(5) (2005) 405–12. [ DOI ] [ PubMed ] [ Google Scholar ] [29]. Smithmyer ME, Cassel SE, Kloxin AM, Bridging 2D and 3D culture: probing impact of extracellular environment on fibroblast activation in layered hydrogels, AIChE J 65(12) (2019). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [30]. Clement V, Roy V, Pare B, Goulet CR, Deschenes LT, Berthod F, Bolduc S, Gros-Louis F, Tridimensional cell culture of dermal fibroblasts promotes exosome-mediated secretion of extracellular matrix proteins, Sci Rep 12(1) (2022) 19786. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [31]. Li L, Shirkey MW, Zhang T, Xiong Y, Piao W, Saxena V, Paluskievicz C, Lee Y, Toney N, Cerel BM, Li Q, Simon T, Smith KD, Hippen KL, Blazar BR, Abdi R, Bromberg JS, The lymph node stromal laminin alpha5 shapes alloimmunity, J Clin Invest 130(5) (2020) 2602–2619. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [32]. German SV, Abalymov AA, Kurochkin MA, Kan Y, Gorin DA, Novoselova MV, Plug-and-Play Lymph Node-on-Chip: Secondary Tumor Modeling by the Combination of Cell Spheroid, Collagen Sponge and T-Cells, Int J Mol Sci 24(4) (2023). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [33]. Kim J, Wu B, Niedzielski SM, Hill MT, Coleman RM, Ono A, Shikanov A, Characterizing natural hydrogel for reconstruction of three-dimensional lymphoid stromal network to model T-cell interactions, J Biomed Mater Res A 103(8) (2015) 2701–10. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [34]. Shou Y, Johnson SC, Quek YJ, Li X, Tay A, Integrative lymph node-mimicking models created with biomaterials and computational tools to study the immune system, Mater Today Bio 14 (2022) 100269. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [35]. Shanti A, Samara B, Abdullah A, Hallfors N, Accoto D, Sapudom J, Alatoom A, Teo J, Danti S, Stefanini C, Multi-Compartment 3D-Cultured Organ-on-a-Chip: Towards a Biomimetic Lymph Node for Drug Development, Pharmaceutics 12(5) (2020). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [36]. Lin HJ, Wang W, Huang YY, Liao WT, Lin TY, Lin SY, Liu DZ, Decellularized Lymph Node Scaffolding as a Carrier for Dendritic Cells to Induce Anti-Tumor Immunity, Pharmaceutics 11(11) (2019). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [37]. Choi YS, Jeong E, Lee JS, Kim SK, Jo SH, Kim YG, Sung HJ, Cho SW, Jin Y, Immunomodulatory Scaffolds Derived from Lymph Node Extracellular Matrices, ACS Appl Mater Interfaces 13(12) (2021) 14037–14049. [ DOI ] [ PubMed ] [ Google Scholar ] [38]. Jiang S, Lyu C, Zhao P, Li W, Kong W, Huang C, Genin GM, Du Y, Cryoprotectant enables structural control of porous scaffolds for exploration of cellular mechano-responsiveness in 3D, Nat Commun 10(1) (2019) 3491. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [39]. Dathathri E, Thakur G, Koteshwara KB, Kumar NVA, Rodrigues FC, Investigating the effect of freezing temperature and cross-linking on modulating drug release from chitosan scaffolds, Chem Pap 74(6) (2020) 1759–1768. [ Google Scholar ] [40]. Yang G, Xiao Z, Long H, Ma K, Zhang J, Ren X, Zhang J, Assessment of the characteristics and biocompatibility of gelatin sponge scaffolds prepared by various crosslinking methods, Sci Rep 8(1) (2018) 1616. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [41]. Gallo N, Natali ML, Curci C, Picerno A, Gallone A, Vulpi M, Vitarelli A, Ditonno P, Cascione M, Sallustio F, Rinaldi R, Sannino A, Salvatore L, Analysis of the Physico-Chemical, Mechanical and Biological Properties of Crosslinked Type-I Collagen from Horse Tendon: Towards the Development of Ideal Scaffolding Material for Urethral Regeneration, Materials (Basel) 14(24) (2021). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [42]. Guan J, Fujimoto KL, Sacks MS, Wagner WR, Preparation and characterization of highly porous, biodegradable polyurethane scaffolds for soft tissue applications, Biomaterials 26(18) (2005) 3961–71. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [43]. Fletcher AL, Malhotra D, Acton SE, Lukacs-Kornek V, Bellemare-Pelletier A, Curry M, Armant M, Turley SJ, Reproducible isolation of lymph node stromal cells reveals site-dependent differences in fibroblastic reticular cells, Front Immunol 2 (2011) 35. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [44]. Creusot RJ, Yaghoubi SS, Kodama K, Dang DN, Dang VH, Breckpot K, Thielemans K, Gambhir SS, Fathman CG, Tissue-targeted therapy of autoimmune diabetes using dendritic cells transduced to express IL-4 in NOD mice, Clin Immunol 127(2) (2008) 176–87. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [45]. Berman DM, Molano RD, Fotino C, Ulissi U, Gimeno J, Mendez AJ, Kenyon NM, Kenyon NS, Andrews DM, Ricordi C, Pileggi A, Bioengineering the Endocrine Pancreas: Intraomental Islet Transplantation Within a Biologic Resorbable Scaffold, Diabetes 65(5) (2016) 1350–61. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [46]. Islam MM, Steward RL Jr., Probing Endothelial Cell Mechanics Through Connexin 43 Disruption, Exp Mech 59(3) (2019) 327–336. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [47]. Steward R Jr., Tambe D, Hardin CC, Krishnan R, Fredberg JJ, Fluid shear, intercellular stress, and endothelial cell alignment, Am J Physiol Cell Physiol 308(8) (2015) C657–64. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [48]. Tambe DT, Hardin CC, Angelini TE, Rajendran K, Park CY, Serra-Picamal X, Zhou EH, Zaman MH, Butler JP, Weitz DA, Fredberg JJ, Trepat X, Collective cell guidance by cooperative intercellular forces, Nat Mater 10(6) (2011) 469–75. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [49]. Butler JP, Tolic-Norrelykke IM, Fabry B, Fredberg JJ, Traction fields, moments, and strain energy that cells exert on their surroundings, Am J Physiol Cell Physiol 282(3) (2002) C595–605. [ DOI ] [ PubMed ] [ Google Scholar ] [50]. Xavier Trepat MRW, Thomas E Angelini, Emil Millet, David A. Weitz, James P. Butler, Jeffrey J. Fredberg Physical forces during collective cell migration, Nature Physics 5 (2009) 426–430. [ Google Scholar ] [51]. Wu J, Steward RL Jr., Disturbed fluid flow reinforces endothelial tractions and intercellular stresses, J Biomech 169 (2024) 112156. [ DOI ] [ PubMed ] [ Google Scholar ] [52]. Subramanian Balachandar VA, Steward RL Jr., Extracellular matrix composition alters endothelial force transmission, Am J Physiol Cell Physiol 325(1) (2023) C314–C323. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [53]. Bankhead P, Loughrey MB, Fernandez JA, Dombrowski Y, McArt DG, Dunne PD, McQuaid S, Gray RT, Murray LJ, Coleman HG, James JA, Salto-Tellez M, Hamilton PW, QuPath: Open source software for digital pathology image analysis, Sci Rep 7(1) (2017) 16878. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [54]. Perez-Shibayama C, Islander U, Lutge M, Cheng HW, Onder L, Ring SS, De Martin A, Novkovic M, Colston J, Gil-Cruz C, Ludewig B, Type I interferon signaling in fibroblastic reticular cells prevents exhaustive activation of antiviral CD8(+) T cells, Sci Immunol 5(51) (2020). [ DOI ] [ PubMed ] [ Google Scholar ] [55]. Wong FS, Moustakas AK, Wen L, Papadopoulos GK, Janeway CA Jr., Analysis of structure and function relationships of an autoantigenic peptide of insulin bound to H-2K(d) that stimulates CD8 T cells in insulin-dependent diabetes mellitus, Proc Natl Acad Sci U S A 99(8) (2002) 5551–6. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [56]. Krishnamurthy B, Mariana L, Gellert SA, Colman PG, Harrison LC, Lew AM, Santamaria P, Thomas HE, Kay TW, Autoimmunity to both proinsulin and IGRP is required for diabetes in nonobese diabetic 8.3 TCR transgenic mice, J Immunol 180(7) (2008) 4458–64. [ DOI ] [ PubMed ] [ Google Scholar ] [57]. You S, Chen C, Lee WH, Wu CH, Judkowski V, Pinilla C, Wilson DB, Liu CP, Detection and characterization of T cells specific for BDC2.5 T cell-stimulating peptides, J Immunol 170(8) (2003) 4011–20. [ DOI ] [ PubMed ] [ Google Scholar ] [58]. Judkowski V, Pinilla C, Schroder K, Tucker L, Sarvetnick N, Wilson DB, Identification of MHC class II-restricted peptide ligands, including a glutamic acid decarboxylase 65 sequence, that stimulate diabetogenic T cells from transgenic BDC2.5 nonobese diabetic mice, J Immunol 166(2) (2001) 908–17. [ DOI ] [ PubMed ] [ Google Scholar ] [59]. Zisi Tegou F, Velluto D, Stock AA, Fitzgerald SN, Stealey S, Zustiak SP, Bayer AL, Tomei AA, CCL21 and beta-cell antigen releasing hydrogels as tolerance-inducing therapy in Type I diabetes, J Control Release 348 (2022) 499–517. [ DOI ] [ PubMed ] [ Google Scholar ] [60]. Takemoto S, Morimoto N, Kimura Y, Taira T, Kitagawa T, Tomihata K, Tabata Y, Suzuki S, Preparation of collagen/gelatin sponge scaffold for sustained release of bFGF, Tissue Eng Part A 14(10) (2008) 1629–38. [ DOI ] [ PubMed ] [ Google Scholar ] [61]. Baidal DA, Ricordi C, Berman DM, Alvarez A, Padilla N, Ciancio G, Linetsky E, Pileggi A, Alejandro R, Bioengineering of an Intraabdominal Endocrine Pancreas, N Engl J Med 376(19) (2017) 1887–1889. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [62]. Schultheiss C, Binder M, Overcoming unintended immunogenicity in immunocompetent mouse models of metastasis: the case of GFP, Signal Transduct Target Ther 7(1) (2022) 68. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [63]. Ansari AM, Ahmed AK, Matsangos AE, Lay F, Born LJ, Marti G, Harmon JW, Sun Z, Cellular GFP Toxicity and Immunogenicity: Potential Confounders in in Vivo Cell Tracking Experiments, Stem Cell Rev Rep 12(5) (2016) 553–559. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [64]. Grzelak CA, Goddard ET, Lederer EE, Rajaram K, Dai J, Shor RE, Lim AR, Kim J, Beronja S, Funnell APW, Ghajar CM, Elimination of fluorescent protein immunogenicity permits modeling of metastasis in immune-competent settings, Cancer Cell 40(1) (2022) 1–2. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [65]. Aldrich VR, Hernandez-Rovira BB, Chandwani A, Abdulreda MH, NOD Mice-Good Model for T1D but Not Without Limitations, Cell Transplant 29 (2020) 963689720939127. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [66]. Tomei AA, Siegert S, Britschgi MR, Luther SA, Swartz MA, Fluid flow regulates stromal cell organization and CCL21 expression in a tissue-engineered lymph node microenvironment, J Immunol 183(7) (2009) 4273–83. [ DOI ] [ PubMed ] [ Google Scholar ] [67]. Haugh MG, Murphy CM, O’Brien FJ, Novel freeze-drying methods to produce a range of collagen-glycosaminoglycan scaffolds with tailored mean pore sizes, Tissue Eng Part C Methods 16(5) (2010) 887–94. [ DOI ] [ PubMed ] [ Google Scholar ] [68]. Ko YG, Kawazoe N, Tateishi T, Chen GP, Preparation of Novel Collagen Sponges Using an Ice Particulate Template, J Bioact Compat Pol 25(4) (2010) 360–373. [ Google Scholar ] [69]. Yang L, Tanabe K, Miura T, Yoshinari M, Takemoto S, Shintani S, Kasahara M, Influence of lyophilization factors and gelatin concentration on pore structures of atelocollagen/gelatin sponge biomaterial, Dent Mater J 36(4) (2017) 429–437. [ DOI ] [ PubMed ] [ Google Scholar ] [70]. O’Brien FJ, Harley BA, Yannas IV, Gibson LJ, The effect of pore size on cell adhesion in collagen-GAG scaffolds, Biomaterials 26(4) (2005) 433–41. [ DOI ] [ PubMed ] [ Google Scholar ] [71]. Zhao HX, Li LH, Ding S, Liu CX, Ai JY, Effect of porous structure and pore size on mechanical strength of 3D-printed comby scaffolds, Mater Lett 223 (2018) 21–24. [ Google Scholar ] [72]. Mukasheva F, Adilova L, Dyussenbinov A, Yernaimanova B, Abilev M, Akilbekova D, Optimizing scaffold pore size for tissue engineering: insights across various tissue types, Front Bioeng Biotechnol 12 (2024) 1444986. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [73]. Afewerki S, Sheikhi A, Kannan S, Ahadian S, Khademhosseini A, Gelatin-polysaccharide composite scaffolds for 3D cell culture and tissue engineering: Towards natural therapeutics, Bioeng Transl Med 4(1) (2019) 96–115. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [74]. Feng B, Jinkang Z, Zhen W, Jianxi L, Jiang C, Jian L, Guolin M, Xin D, The effect of pore size on tissue ingrowth and neovascularization in porous bioceramics of controlled architecture in vivo, Biomed Mater 6(1) (2011) 015007. [ DOI ] [ PubMed ] [ Google Scholar ] [75]. Loh QL, Choong C, Three-dimensional scaffolds for tissue engineering applications: role of porosity and pore size, Tissue Eng Part B Rev 19(6) (2013) 485–502. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [76]. Lukin I, Erezuma I, Maeso L, Zarate J, Desimone MF, Al-Tel TH, Dolatshahi-Pirouz A, Orive G, Progress in Gelatin as Biomaterial for Tissue Engineering, Pharmaceutics 14(6) (2022). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [77]. Sussman EM, Halpin MC, Muster J, Moon RT, Ratner BD, Porous implants modulate healing and induce shifts in local macrophage polarization in the foreign body reaction, Ann Biomed Eng 42(7) (2014) 1508–16. [ DOI ] [ PubMed ] [ Google Scholar ] [78]. Wang CY, Hong PD, Wang DH, Cherng JH, Chang SJ, Liu CC, Fang TJ, Wang YW, Polymeric Gelatin Scaffolds Affect Mesenchymal Stem Cell Differentiation and Its Diverse Applications in Tissue Engineering, Int J Mol Sci 21(22) (2020). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [79]. Brown EL, Lefebvre TL, Sweeney PW, Stolz BJ, Grohl J, Hacker L, Huang Z, Couturier DL, Harrington HA, Byrne HM, Bohndiek SE, Quantification of vascular networks in photoacoustic mesoscopy, Photoacoustics 26 (2022) 100357. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [80]. Sadtler K, Allen BW, Estrellas K, Housseau F, Pardoll DM, Elisseeff JH, The Scaffold Immune Microenvironment: Biomaterial-Mediated Immune Polarization in Traumatic and Nontraumatic Applications<sup/>, Tissue Eng Part A 23(19–20) (2017) 1044–1053. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [81]. Astarita JL, Acton SE, Turley SJ, Podoplanin: emerging functions in development, the immune system, and cancer, Front Immunol 3 (2012) 283. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [82]. Acton SE, Farrugia AJ, Astarita JL, Mourao-Sa D, Jenkins RP, Nye E, Hooper S, van Blijswijk J, Rogers NC, Snelgrove KJ, Rosewell I, Moita LF, Stamp G, Turley SJ, Sahai E, Reis e Sousa C, Dendritic cells control fibroblastic reticular network tension and lymph node expansion, Nature 514(7523) (2014) 498–502. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [83]. Horsnell HL, Tetley RJ, De Belly H, Makris S, Millward LJ, Benjamin AC, Heeringa LA, de Winde CM, Paluch EK, Mao Y, Acton SE, Lymph node homeostasis and adaptation to immune challenge resolved by fibroblast network mechanics, Nat Immunol 23(8) (2022) 1169–1182. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [84]. Yamada KM, Doyle AD, Lu J, Cell-3D matrix interactions: recent advances and opportunities, Trends Cell Biol 32(10) (2022) 883–895. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [85]. Wing-Han Yuen Q, Zheng YP, Huang YP, He JF, Chung-Wai Cheung J, Ying M, In-vitro Strain and Modulus Measurements in Porcine Cervical Lymph Nodes, Open Biomed Eng J 5 (2011) 39–46. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [86]. Ramirez A, Sriram V, Abbouchi Y, Patolia R, Passaro E, Kaluzienski M, Maisel K, Inflammation modulates lymph node biomechanics in a sex-dependent manner, bioRxiv (2025). [ Google Scholar ] [87]. Horwitz DA, Zheng SG, Gray JD, The role of the combination of IL-2 and TGF-beta or IL-10 in the generation and function of CD4+ CD25+ and CD8+ regulatory T cell subsets, J Leukoc Biol 74(4) (2003) 471–8. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [88]. Velankar KY, Liu W, Hartmeier PR, Veleke SR, Reddy GA, Clegg B, Gawalt ES, Fan Y, Meng WS, Fibril-Guided Three-Dimensional Assembly of Human Fibroblastic Reticular Cells, ACS Appl Bio Mater 7(6) (2024) 3953–3963. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [89]. Link PA, Choi KM, Diaz Espinosa AM, Jones DL, Gao AY, Haak AJ, Tschumperlin DJ, Combined control of the fibroblast contractile program by YAP and TAZ, Am J Physiol Lung Cell Mol Physiol 322(1) (2022) L23–L32. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [90]. Perez-Shibayama C, Gil-Cruz C, Ludewig B, Fibroblastic reticular cells at the nexus of innate and adaptive immune responses, Immunol Rev 289(1) (2019) 31–41. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [91]. Valencia J, Jimenez E, Martinez VG, Del Amo BG, Hidalgo L, Entrena A, Fernandez-Sevilla LM, Del Rio F, Varas A, Vicente A, Sacedon R, Characterization of human fibroblastic reticular cells as potential immunotherapeutic tools, Cytotherapy 19(5) (2017) 640–653. [ DOI ] [ PubMed ] [ Google Scholar ] [92]. Kawasaki K, Noma K, Kato T, Ohara T, Tanabe S, Takeda Y, Matsumoto H, Nishimura S, Kunitomo T, Akai M, Kobayashi T, Nishiwaki N, Kashima H, Maeda N, Kikuchi S, Tazawa H, Shirakawa Y, Fujiwara T, PD-L1-expressing cancer-associated fibroblasts induce tumor immunosuppression and contribute to poor clinical outcome in esophageal cancer, Cancer Immunol Immunother 72(11) (2023) 3787–3802. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [93]. Mueller SN, Matloubian M, Clemens DM, Sharpe AH, Freeman GJ, Gangappa S, Larsen CP, Ahmed R, Viral targeting of fibroblastic reticular cells contributes to immunosuppression and persistence during chronic infection, Proc Natl Acad Sci U S A 104(39) (2007) 15430–5. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [94]. Martin-Orozco N, Wang YH, Yagita H, Dong C, Cutting Edge: Programmed death (PD) ligand-1/PD-1 interaction is required for CD8+ T cell tolerance to tissue antigens, J Immunol 177(12) (2006) 8291–5. [ DOI ] [ PubMed ] [ Google Scholar ] [95]. Carter L, Fouser LA, Jussif J, Fitz L, Deng B, Wood CR, Collins M, Honjo T, Freeman GJ, Carreno BM, PD-1:PD-L inhibitory pathway affects both CD4(+) and CD8(+) T cells and is overcome by IL-2, Eur J Immunol 32(3) (2002) 634–43. [ DOI ] [ PubMed ] [ Google Scholar ] [96]. Cheung AS, Zhang DKY, Koshy ST, Mooney DJ, Scaffolds that mimic antigen-presenting cells enable ex vivo expansion of primary T cells, Nat Biotechnol 36(2) (2018) 160–169. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [97]. Dang A, De Leo S, Bogdanowicz DR, Yuan DJ, Fernandes SM, Brown JR, Lu HH, Kam LC, Enhanced activation and expansion of T cells using mechanically soft elastomer fibers, Adv Biosyst 2(2) (2018). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [98]. Roncarolo MG, Gregori S, Battaglia M, Bacchetta R, Fleischhauer K, Levings MK, Interleukin-10-secreting type 1 regulatory T cells in rodents and humans, Immunol Rev 212 (2006) 28–50. [ DOI ] [ PubMed ] [ Google Scholar ] [99]. Jankovic D, Kugler DG, Sher A, IL-10 production by CD4+ effector T cells: a mechanism for self-regulation, Mucosal Immunol 3(3) (2010) 239–46. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [100]. Chen W, Jin W, Hardegen N, Lei KJ, Li L, Marinos N, McGrady G, Wahl SM, Conversion of peripheral CD4+CD25− naive T cells to CD4+CD25+ regulatory T cells by TGF-beta induction of transcription factor Foxp3, J Exp Med 198(12) (2003) 1875–86. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [101]. Knop L, Deiser K, Bank U, Witte A, Mohr J, Philipsen L, Fehling HJ, Muller AJ, Kalinke U, Schuler T, IL-7 derived from lymph node fibroblastic reticular cells is dispensable for naive T cell homeostasis but crucial for central memory T cell survival, Eur J Immunol 50(6) (2020) 846–857. [ DOI ] [ PubMed ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials 1 NIHMS2132223-supplement-1.docx (5.3MB, docx) ACTIONS View on publisher site PDF (2.5 MB) Cite Collections Permalink RESOURCES Similar articles Cited by other articles Links to NCBI Databases Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top

Related documents

• Safety and Efficacy Evaluation of 4th Generation Safety-engineered CAR T Cells Targeting Sarcomas#158216
• Detection of pancreatic beta cell mass in vivo in humans: studies in individuals with long-standing type 1 diabetes and in individuals with obesity#18841
• A dynamic vascular-inflammatory-fibrotic loop in systemic sclerosis: an integrative immuno-vascular framework.#199155
• The Efficacy and Safety of Tislelizumab Combined With Anlotinib and S1 Plus Oxaliplatin as Neoadjuvant Therapy for the Locally Advanced Adenocarcinoma of Esophagogastric Junction#272356
• Cell Culture Media Optimization#982912