Neonatal human dermal fibroblasts (nHDFs) were sourced from ATCC (PCS-201-010). Cells were cultured in IMDM (GIBCO), 10% FBS (Hyclone), 2 mM L-glutamine (GIBCO), and gentamicin (GIBCO). Cells were maintained at 37°C 20% O₂/5% CO₂ in complete medium with medium exchanges every 3–4 days as needed. Cells were grown until near confluent and passaged using TrypLE (GIBCO).

Neonatal human dermal fibroblasts were plated in T25 flasks and transduced with TDO2 activation lentiviral particles (Santa Cruz Biotech) at MOI:20 in complete medium. After 24 h of transduction, the virus was removed, and fresh complete medium was added for cell recovery for a further 24 h. Cells were then subjected to selection by 5.0 μg/mL puromycin for approximately 3–4 days. Following selection, complete medium was replaced and cells were grown and passaged.

Extracellular vesicles were harvested from primary nHDFs at passage 5-7, from normal and transduced cells using a 15-day serum starvation method previously described (Walravens et al., 2021). Briefly, cells were grown to near confluence (∼90%) at 20% O₂/5% CO₂ at 37°C. Cell bed was washed 2x with warmed phosphate-buffered saline (PBS) and then incubated in IMDM without serum supplementation for 15 days in the same environment. Conditioned medium was collected, centrifuged at 3,000 × g for 10 min to remove dead cells and debris, then filtered through a 0.45-μm PES filter to remove apoptotic bodies and protein aggregates, and frozen for later use at −80°C. EVs were purified using centrifugal ultrafiltration with a 100-kDa molecular weight cutoff filter (Sigma-Millipore). EV preparations, before and after concentration were analyzed by NTA using the Malvern Nanosight NS300 Instrument (Malvern Instruments) with the following acquisition parameters: camera levels of 15, detection level less than or equal to 5, number of videos taken = 5, and video length of 30 s.

Extracellular vesicles were collected and prepared as described above. After 100 kDa ultrafiltration, EVs were further purified using size exclusion chromatography (SEC) columns (SBI). Briefly, 1.0 mL of concentrated EVs was added to each chromatographic column and incubated at room temperature with rotation for 30–35 min. EVs were eluted from the column by centrifugation at 500 × g. EV size and concentration were analyzed by NTA as described above. Protein content of EV preparations was quantified using a BCA assay (Pierce).

Bone marrow-derived progenitor cells were collected from 3-month-old female Wistar Kyoto rats and differentiated into bone marrow-derived macrophages (BMDMs) by culturing with 20 ng/mL recombinant M-CSF (Life Technologies). Briefly, whole bone marrow cells were collected via aspiration with ice-cold PBS. Cells were filtered using a 70-μm cell strainer and centrifuged at 400 × g for 10 min at 4°C to pellet. The cell pellet was resuspended in 10 mL ACK buffer (GIBCO) for 30 s. ACK was quenched with IMDM + 10% FBS, and cells were centrifuged as described above. Cells were resuspended in complete medium; IMDM + 10% FBS + 20 ng/mL M-CSF and counted. Cells were seeded into six-well plates at 8.0e10⁴ cells/well, or equivalent. Cells were incubated at 37°C with 20% O₂ and 5% CO₂. Fresh complete medium was exchanged on day 3 and cells monitored for confluence. Test compounds were administered once BMDM cultures reached ∼75% confluence. Serum concentration was reduced to 1% during assays to facilitate EV uptake.

Primary bone marrow macrophages were collected as previously described and plated in 96-well plates at a density of 4.0e⁴ cells/well in complete medium (IMDM + 10% FBS + human recombinant M-CSF 20 ng/mL). After attachment and maturation (∼3 days), complete medium was removed and replaced with IMDM w/1% FBS for all test conditions. 4.0e⁶ EVs were added ∼1 h after medium change along with lipopolysaccharide (LPS) (10 ng/mL, Sigma). Control conditions were cultured in IMDM w/10% and 1% FBS. Cells were grown overnight after EV & LPS addition and proliferation quantified using a bromodeoxyuridine (BrdU) Cell Proliferation ELISA (Abcam, ab126556).

Bone marrow-derived monocytes were seeded onto 100 mm cell culture dishes and differentiated into mature macrophages using IMDM + 10% FBS + 20 ng/mL human recombinant M-CSF (Life Technologies). Upon reaching 75% confluence, BMDMs were lifted using ice-cold PBS + 2 mM EDTA. Cells were quantified and re-seeded onto 8.0-μM pore size transwells (Costar). Cells were recovered overnight at 37°C, 5% CO₂ with complete medium (IMDM + 10% FBS). One hour before EV administration, complete medium was removed, cells were gently washed 1x with serum-free IMDM, and medium was replaced with IMDM + 1% FBS. Cells were incubated with EVs overnight and then fixed with 4.0% PFA. Cells were gently removed from the upper side of the transwell using a cotton swab. The underside of the transwell was stained for 20 min at RT using Crystal Violet. After staining, cells were gently washed several times with PBS until the wash ran clear. Ten images were captured of each transwell at 10x magnification (three per condition). Quantification of cell migration was done using ImageJ.

Total cell RNA was isolated using the RNeasy Plus Mini Kit (Qiagen) according to the manufacturer’s protocol. Total EV RNA was isolated using the miRNeasy Advanced Serum Plasma Kit (Qiagen). Total cell RNA was quantified using NanoDrop and diluted using diH₂O. Total EV RNA was quantified by Qubit (Thermo Fisher Scientific). Cellular RNA Reverse Transcription was performed using the High-Capacity RNA-to-cDNA kit (Life Technologies) with 1 μg RNA per reaction. PCR reactions were performed on the QuantStudio 7 Flex Real-Time PCR System (Applied Biosystems) using TaqMan Fast Advanced Master Mix (Life Technologies, 4444556) and TaqMan primers. Each reaction was performed in triplicate. The gene expression assays used for this study are summarized in Supplementary Table 1.

Cell lysates were collected for ELISA and western blot from six-well plates. Cells were washed 1x with ice-cold PBS. Cells were lysed in-well with 75 μL 1 × lysis buffer with phospo/protease inhibitors (Thermo Fisher Scientific). The cell lysate was incubated on ice for 15 min, sonicated twice for 10 s each, and centrifuged at 15,000 × g for 15 min at 4°C. The supernatant was collected and frozen for later use at −80°C. Protein lysates were quantified using a Pierce BCA Protein Assay kit (Thermo Fisher Scientific).

Electrophoresis was conducted using NuPage 4–12% Bis-Tris protein gels (Life Technologies) using 25 μg protein per well. HPVD Membrane transfer was performed using the Turbo Transfer System (BIO-RAD) after gel electrophoresis. Blocking was performed using 5% non-fat milk in TBS + 20% Teen, 1 h at RT. Primary antibody staining was done overnight at 4°C. Secondary HRP antibody staining was done for 90 min at RT and then detected by SuperSignal West Pico PLUS Chemiluminescent Substrate (Thermo Fisher Scientific). Antibodies used in this study are summarized in Supplementary Table 2.

Interleukin-6 ELISA (R&D Systems, Quantikine ELISA) was performed according to the manufacturer’s protocol. Samples concentration for testing was 1.5 mg mL^–1.

Cell and EV RNA samples were sequenced at the Cedars-Sinai Genomics Core. Total RNA and Small RNA were analyzed using an Illumina NextSeq 500 platform for cell and EV samples, respectively.

Proteomics of nHDF-EV and nHDF^TDO2-EV was conducted by Creative Proteomics (Shirley, NY, United States) using 200 μg protein per sample. Data analysis and processing were done using FunRich (Pathan et al., 2015).

All animal studies were conducted under approved protocols from the Institutional Animal Care and Use Committee protocols.

Acute myocardial infarction was induced in 3-month-old male C57/B-L6 mice as described previously (Ibrahim et al., 2014). Within 10 min of left anterior descending artery ligation, a total of 1 × 10⁵ cells (or vehicle) were administered via 3 × 8μL injections intramyocardially.

Echocardiography was performed in the mouse model of acute myocardial infarction at 1 day (baseline) and 21 days after surgery using Vevo 3100 Imaging System (Visual Sonics) as described (Ibrahim et al., 2019). The average of the left ventricular ejection fraction was analyzed from multiple left ventricular end-diastolic and left ventricular end-systolic measurements.

GraphPad Prism 9.0 (GraphPad Software) was used to analyze the data. A comparison of three or more groups was performed using two-way or one-way ANOVA followed by Sidek’s post hoc multiple comparison test for paired groups. Two-group comparisons were analyzed using two-tailed unpaired t-tests with a confidence interval of 95%. RNA sequencing data were analyzed for differential expression, fold change, and unsupervised PCA using DESeq2 (Anders and Huber, 2010; Love et al., 2014).

Normal human dermal fibroblasts (HDFs) treated with the beta-catenin activator 6-bromoindirubin-3’-oxime (BIO) lead to upregulation of TDO2 more than any other gene (Supplementary Figure 1A). Therefore, to examine the role of TDO2, we transduced neonatal human fibroblasts (nHDFs) with lentivirus containing the TDO2 transgene under the control of a constitutive promoter. Upon transduction and subsequent selection by puromycin, expression of TDO2 was 50-fold higher than non-transduced cells (Supplementary Figure 1B). TDO2 activation had a significant impact on the nHDF transcriptome. Transcriptomic sequencing identified 3,000 differentially expressed genes and over 400 unique genes in TDO2 transduced cells (Supplementary Figures 1D–E). Increased expression of TDO2 was further confirmed by the sequencing data as well (Supplementary Figure 1F).

To explore the effect of TDO2 activation on the immunomodulatory capacity of nHDFs, rat BMDMs were co-cultured with TDO2-transduced nHDFs (nHDF^TDO2), un-transduced nHDFs (nHDF^UNT), or LPS as an activation control. Examination of macrophage polarization and inflammatory markers showed no changes in arginase 1, IL-6, and IL-1B expression following co-culture (Figures 1A–C). Nos2 expression was significantly decreased compared to both LPS- and nHDF-co-cultured BMDMs (Figure 1D). Decreased expression of the protein, iNOS, was further confirmed by western blot (Figures 1E,F). Interestingly, while no changes were observed in IL-6 transcription (Figure 1B), secreted IL-6 in the conditioned culture medium was lower as observed by ELISA which suggests potential post-translational regulation (Figure 1G). To investigate the potential mediator of the immunoregulatory effect of TDO2-activated nHDFs, we assessed the levels of secreted kynurenine. TDO2 is a rate-limiting enzyme in the conversion of L-tryptophan to N-formyl-L-kynurenine. Kynurenine and its downstream metabolites play roles in anti-inflammation and vascular relaxation (Nguyen et al., 2010; Wang et al., 2010). However, subsequent analysis of conditioned medium by ELISA revealed no increase in kynurenine (Supplementary Figure 1C). Therefore, TDO2 activation endows nHDFs with immunomodulatory capacities as shown by their ability to blunt IL-6 and Nos2 expression in co-cultured macrophages. Furthermore, this effect is not driven by the synthesis of kynurenine.

Having ruled out the role of kynurenine, we investigated changes in EV payload post TDO2 activation in nHDFs. nHDF^TDO2 (nHDF^TDO2-EVs) and control nHDF EVs (nHDF-EVs) were conditioned using a 15-day serum-starvation protocol described by us previously (Ibrahim et al., 2014). EVs were isolated using ultrafiltration with 100 kDa molecular weight cut-off followed by buffer exchange with PBS. The size and concentration of EVs were similar between the nHDF and nHDF^TDO2 groups (Figures 2A–C). Further characterization was completed demonstrating the presence of conserved EV markers such as HSP70, CD81, and CD63 (Figure 2D). The absence of the endoplasmic reticulum (ER) protein Calnexin in both preparations demonstrates equivalent purification during EV concentration (Figure 2D). Total RNA and protein content were comparable in both groups (Figures 2E,F). Subsequent RNA sequencing of the vesicles revealed a slight enrichment of total small RNAs in the TDO2 group (Figure 2G). This increase appeared to be primarily driven by increases in piRNA and tRNA (Figure 2H).

To investigate the function of EVs from nHDF^TDO2, BMDMs were treated directly with EVs from each group and assayed for the same markers of macrophage and inflammatory markers described earlier. Significant reductions were observed in the expression of Arg-1, Nos2, IL-1B, and Nos2, and ADAM17 was observed after overnight treatment with EVs (Figures 3A–D and Supplementary Figure 4A). Additional inflammation markers were investigated but were not observed to have significantly changed expression levels (Supplementary Figure 4B). Reduced iNOS levels in BMDM were confirmed by western blot (Figures 3E,F). Reductions in IL-6 secretion were observed by ELISA (Figure 3G). Both EV groups also stimulated BMDM proliferation equally as shown by BrdU incorporation (Figure 3H). The effect of EVs on migration was also evaluated using a modified Bowden’s Chamber assay. Interestingly, nHDF^TDO2-EVs enhanced macrophage migration compared to other groups (Figure 3I). Taken together, these data demonstrate a potent immunomodulatory effect of nHDF^TDO2 in macrophages. This effect was reflected by inherent changes in the inflammatory profile of the cells rather than impairing proliferation or infiltration. To rule out bioactivity from extra-vesicle proteins, EV preparations were further purified using SEC. SEC purification yielded preparations with nearly 10-fold less protein compared to the ultrafiltration only preparation (Supplementary Figure 2A). Particle numbers were equivalent between both groups (Supplementary Figures 2B,C). Equivalently to the ultrafiltration product, SEC-purified EVs attenuated Nos2 and IL6 (Supplementary Figures 2E,F) which suggests that the immunomodulatory effects on macrophages are mediated by the EVs. Additionally, very few differences were observed when comparing the proteomic composition of nHDF and nHDF^TDO2 derived EVs (Supplementary Figure 3). These data suggest that EV-associated proteins are not responsible for the immunomodulatory effects observed.

To investigate the mechanism by which EVs from nHDF^TDO2 regulate macrophage activation, we performed transcriptomic analysis on macrophages treated with nHDF^TDO2-EVs and nHDF-EVs. Macrophages treated with nHDF-EVs had significant transcriptomic changes including the activation of several pro-inflammatory genes (Figure 4A). In contrast, macrophages treated with nHDF^TDO2-EVs had a much more muted effect (Figure 4B). This is further reflected using principle component analysis where nHDF^TDO2-EV-treated groups are more proximal to the untreated group than the nHDF-EV-treated group (Figure 4C). Indeed, nHDF^TDO2-EVs nearly reverse the inflammatory effects of nHDF-EV treatment (Figure 4D). Given this major reversal of inflammatory phenotype, we interrogated nuclear factor-kappa b (NFκB) genes as it is a central regulator of inflammation (Taniguchi and Karin, 2018). Analysis of the NFκB inflammatory complex shows potent abatement compared to the nHDF-EV-treated group and equivalent to the untreated group (Supplementary Figure 4G). Direct comparison of canonical inflammatory markers and pathways shows universal upregulation in macrophages treated with nHDF-EVs versus those treated with nHDF^TDO2-EVs (Figures 4E–J) including decreased expression of all constituents of the NFκB complex (Figure 4I). This led us to suspect whether the nHDF^TDO2-EVs might be inert and incapable of signaling to macrophages. To test this hypothesis, we performed a sequential exposure experiment whereby macrophages were treated first with nHDF^TDO2-EVs followed by exposure to nHDF-EVs. If the nHDF^TDO2-EVs are truly inert, then the nHDF-EVs should still induce inflammatory activation. If the nHDF^TDO2-EVs induce anergic modulation in macrophages, then macrophage activation by nHDF-EVs will be attenuated. Indeed, pre-treating macrophages with nHDF^TDO2-EVs blunted their ability toward inflammatory activation by nHDF-EVs. This is shown by decreased levels of Arg-1, Nos2, and IL-6, and a significant increase in IL-10 expression after treatment with nHDF-EVs (Supplementary Figures 4C–F). Taken together, these data demonstrate a unique ability of nHDF^TDO2-EVs to induce immunomodulation through inducing anergia in macrophages. This effect is mediated in part through blunting NFκB activation.

Having observed these profound effects in macrophages, we sought to investigate the effect of nHDF^TDO2 in a cardiac injury model where macrophages play an active role in injury and resolution. We investigated the therapeutic capacity of these cells in a well-established mouse model of acute myocardial infarction used by our group to establish therapeutic potency of cells and EVs (Figure 5A). Results showed significant improvement of cardiac function at 21-day post-injury in nHDF^TDO2-treated hearts (compared to nHDF-treated hearts; Figure 5B). Observation of both B-mode M-mode images reveals an observable difference in left ventricular wall contractility (Figure 5C). Changes in end-diastolic and systolic volumes were decreased in the nHDF-TDO2-EV treated hearts, indicating preservation of end-systolic volumes and maintenance of ejection fraction (Figures 5D,E).