IL-17A stimulates erythropoiesis by tuning the Erythropoietin-feedback circuit

Feedback control of erythropoiesis exemplifies conflicting goals in tissue homeostasis: maintaining fast reactivity to stress while minimizing proliferative burden on progenitors in the steady state. Here we show that these conflicting goals are tuned through the combinatorial action of cytokines. We find that IL-17A, a pro-inflammatory cytokine, mediates striking synergism with the negative feedback signal erythropoietin (Epo) in vivo, accelerating the erythropoietic response to hypoxia. A model of erythropoietic control shows increased reactivity may occur through two cell circuit designs, with one having far lower constitutive progenitor burden in normoxia. IL-17A acts through this optimal design by sensitizing progenitors to Epo, a model supported by multiple experimental observations. We suggest that IL-17A signals impending hypoxia during infections, tuning erythropoiesis in favor of a faster stress response. Our study highlights IL-17A as a potential erythropoietic therapeutic agent and serves as a model of homeostatic tuning in stem and progenitor cell circuits. Overall design: This study examines hematopoietic responses through single-cell RNA sequencing analysis of bone marrow and spleen cells from a subset of mice treated with IL-17A, Erythropoietin (Epo), or their combination over a 72-hour period. The experiment comprised four treatment groups. The first group received IL-17A administered subcutaneously at 200 ng/g body weight every 12 hours for 72 hours, totaling six doses. The second group received Erythropoietin (Procrit) administered subcutaneously at 0.25 IU/g every 24 hours for 72 hours, totaling three doses. The third group received a combination of both treatments following the same dosing schedules. The control group received vehicle (PBS) injections at a volume of 4 µL/g body weight, with timing matched to the treatment groups. For single-cell RNA sequencing analysis, eight mice were selected, with two mice per treatment group. From each mouse, both bone marrow and spleen tissues were collected, resulting in sixteen total samples. All tissues were harvested at 72 hours after the initial treatment and were collected in SB5 buffer containing PBS supplemented with 0.2% BSA, 0.08% glucose, and 5mM EDTA. Bone marrow was processed by flushing bones with SB5 using a 26-gauge needle, while spleen samples were mechanically dissociated. All samples were filtered through 100 µM cell strainers and centrifuged at 2000rpm for 10 minutes. The cells were enriched for CD117-expressing cells using the EasySepTM Mouse CD117 Positive Selection Kit at 4°C, followed by live cell enrichment using OptiPrep density gradient centrifugation with 40%, 18%, 12%, and 5% gradient layers, centrifuged at 800g for 15 minutes. Five independent single-cell RNA sequencing libraries were generated using the 10X Genomics Chromium platform. The first library contained all vehicle-treated samples (two bone marrow and two spleen samples from two mice). The second library contained all IL-17A-treated samples (two bone marrow and two spleen samples from two mice). The third library contained all Epo-treated samples (two bone marrow and two spleen samples from two mice). The fourth library contained all IL-17A plus Epo-treated samples (two bone marrow and two spleen samples from two mice). The fifth library contained all samples pooled together (eight bone marrow and eight spleen samples from all eight mice), providing technical validation through sample multiplexing. Sample preparation for single-cell RNA sequencing involved MULTIseq labeling using MULTI-Seq Lipid-Modified Oligos with unique sample barcode oligos for multiplexing. Cells were prepared at a concentration of 1000 cells/µL in PBS with 0.04% BSA. Sequencing was performed using the 10X Genomics Chromium X instrument with Chromium Single Cell 3' v3.1 chemistry. Libraries were sequenced on the Illumina NovaSeq 6000 using S4 200 cycle and SP 100 cycle flow cells, targeting 25,000 reads per cell for gene expression and 5,000 reads per cell for sample barcodes. Quality control was performed using Agilent 2100 Bioanalyzer, Agilent Tape station, and KAPA Library Quantification.