Camilla Recordati

Background: Orthopedic implant-associated infections are difficult to eradicate because of biofilm formation and impaired healing. FK-16 and GF-17 are short synthetic antimicrobial fragments of the human cathelicidin LL-37; FK-16 consists of 16 amino acids and differs from the 17-amino-acid GF-17 by the absence of its N-terminal glycine. This study aimed to assess their acute tolerability following single-dose intravenous administration in mice and to evaluate the local antibacterial efficacy of FK-16 delivered via peptide-enriched silk fibroin (PSF) scaffolds in a rat model of methicillin-resistant Staphylococcus epidermidis (MRSE)-associated femoral nonunion. Methods: Thirty-six BALB/c mice (18 males and 18 females) received vehicle, FK-16, or GF-17 (0.1 or 1 mg/kg for each peptide) and underwent clinical and functional assessments, hemolysis testing, biochemical analyses, organ-weight measurements, and histopathological examination. Thirty-two male Wistar rats were assigned to a noninfected control group or infected groups treated with unloaded PSF, vancomycin-loaded PSF, or FK-16-loaded PSF. On day 56, antibacterial activity and bone healing were evaluated using microbiological culture, micro-computed tomography, and histopathology. Results: No mortality, persistent neurobehavioral abnormalities, or treatment-related organ lesions occurred in mice. Transient hind-limb paralysis occurred in one mouse treated with FK-16 and one treated with GF-17 at 1 mg/kg. Biochemical and hemolysis-related changes were isolated and did not follow a consistent dose–response pattern; no corresponding treatment-related organ lesions were observed. In rats, viable MRSE was recovered from 3/4 untreated specimens and none from those treated with vancomycin or FK-16. FK-16 was associated with partial normalization of the circulating ratio of osteoprotegerin to receptor activator of nuclear factor-κB ligand without significantly reducing early systemic inflammation. Bone mineral density, tissue mineral density, and bone volume fraction increased numerically, but no quantitative micro-computed tomography healing parameter improved significantly; the radiographic osteomyelitis score decreased significantly. Untreated PSF showed absent bridging, abundant fibrovascular tissue, chronic inflammation, rare Gram-positive cocci, and higher histological scores than noninfected controls; both treated groups showed minimal inflammation but incomplete bridging. Conclusions: FK-16 and GF-17 showed generally acceptable acute tolerability despite transient high-dose neurological effects. Locally delivered FK-16 demonstrated antibacterial activity, reduced osteomyelitis severity, and was associated with minimal histological evidence of inflammation, but did not significantly improve quantitative bone healing or restore complete bridging by day 56.

Ketamine, a dissociative anesthetic commonly used in pediatric medicine, may induce neurotoxic effects when administered during early neonatal life. Although ketamine-induced neuronal apoptosis has been previously documented, less is known about glial cell responses or temporal evolution of injury. This study investigated the neurotoxic effects of ketamine (20 mg/kg, 5 intraperitoneal injections administered 90 minutes apart) in postnatal day 7 Sprague-Dawley rats of both sexes, compared with age-matched controls. Brains were collected 2 and 16 hours after the last injection and evaluated by histology, immunohistochemistry, and immunofluorescence using markers of apoptosis, DNA damage, and cell populations (Cleaved Caspase-3, γH2AX, NeuN, Iba-1, GFAP, and Olig2). Quantification of apoptotic and immunostained cells in the cortex, hippocampus, and corpus callosum was performed using standardized digital image analysis. Ketamine induced a significant increase in apoptosis and DNA damage at 2 hours, concomitant with an elevation in Olig2-positive cells, suggesting a potential compensatory response. Neuronal, astrocytic, and microglial cells were unaffected. Intrinsic apoptosis primarily involved neurons and oligodendrocytes, with microglia and astrocytes engaged in phagocytosis of apoptotic cells. At 16 hours, a return to baseline levels was observed. These findings offer a temporal and cellular characterization of ketamine-induced neurotoxicity in the developing rat brain.

Figure S7. A) Frequency of CD4+ Foxp3+ regulatory T cells in colon LP of DSS-treated Csf3r+/+ and Csf3r-/-. Unpaired Student’s t-Test. Data are mean ± SEM. ns: not significant.

Figure S2. Neutrophil deficiency is associated with intestinal dysbiosis. Related to Figure 2 A) Representative histological images of H&E-stained colon section from Csf3r+/+(left) and Csf3r-/- (right) mice, after one cycle of DSS, showing the increased inflammatory infiltrate and the presence of bacterial accretions (indicated by black arrow) in Csf3r-/-. B-C) Shannon Index representing the evenness in the overall number of bacterial species between wild-type and neutrophil deficient mice untreated (B) and DSS-treated mice (C). D-E) Taxonomic analysis of the phyla that compose the microbiota of feces of untreated (D) and DSS-treated (E) Csf3r+/+ (untreated n=4, DSS-treated n=3) and Csf3r-/- (untreated n=4, DSS-treated n=4) mice. F) Microbiota depletion efficiency measured by qPCR for 16S gene: complete depletion is achieved after 20 days of antibiotic treatment. G) Body weight loss during DSS-induced acute colitis in Csf3r+/+(n=11) and Csf3r-/- (n=9) mice with and without cohousing (Csf3r+/+Cohoused n=5 and Csf3r-/-Cohoused n=5). H) Body weight loss of AOM/DSS treated Csf3r+/+(n=7) and Csf3r-/- (n=5) mice with and without cohousing (Csf3r+/+Cohoused n=5 and Csf3r-/-Cohoused n=5); I) Macroscopic polyp count at the experimental endpoint. A-I) One experiment. B-C) Unpaired Student’s t-Test. F, I) Multiple Student’s t-Test. G-H) Wilcoxon matched-pairs signed rank test. Data are mean ± SEM. *** p < 0.001 ** p < 0.01 * p < 0.05.

Supplementary Table 2. List of critical reagents and antibodies used.

Figure S1. Neutrophil deficiency is associated with increased lymphoid aggregates. Related to Figure 1.A) Number of total lymphoid structures and B) number of lymphoid structures with follicular dendritic cells detected by RNAscope staining for the follicular dendritic cell marker Mfge8 in colon tissue sections from Csf3r+/+(n=8) and Csf3r-/- (n=5) mice after AOM/DSS-induced CRC. C) Representative images of RNAscope analysis for Mfge8 in colon tissue sections from Csf3r-/- mice after AOM/DSS-induced CRC: aggregates with follicular structures are visible as Mfge8+ (red spot): on the top is visible a lymphoid aggregate at two different magnifications, with diffuse Mfge8+ cells; on the bottom right lymphoid microaggregate with Mfge8+ cells; on the bottom left aggregate without follicular structures (Mfge8+). D) Representative dot-plot of neutrophils frequency in blood (top panels) and colon LP (bottom panels) of Csf3r-/- mice and Csf3r-/- mice 4hrs and 18hrs after adoptive transfer of neutrophils. E-F) Frequency of neutrophils in blood (D) and colon LP (E) of DSS-treated Csf3r-/- mice (n=5) and Csf3r-/- mice 4hrs (n=5) and 18hrs (n=5) after adoptive transfer of neutrophils. A-B) Representative data of three independent experiments. C-E) One experiment. A-B, D-E) Unpaired Student’s t-Test. Data are mean ± SEM. *** p < 0.001 ** p < 0.01 * p < 0.05.

Figure S4. Neutrophil deficiency is associated with decreased expression of IL-22 by γδ T cells. Related to Figure 4-5. A-C) Representative gating strategy used to gate myeloid population (A), lymphoid population (B) and innate lymphoid cells (ILC) (C) in colon LP. UTC: unconventional T cells (TCRβ+, CD8-, CD4-). D) IL-22 expression in myeloid cells derived from DSS-treated Csf3r+/+(n=3), Csf3r-/-(n=3) stimulated 4 hours with IL-23 analyzed by FACS. E-F) Expression of IL-22 by lymphoid cells derived from DSS-treated Csf3r+/+(n=4), Csf3r-/- (n=4) stimulated 4 hours with IL-23 plus IL-1β (E) and PMA plus ionomycin (F) analyzed by FACS. A-F) Representative data of three independent experiments. D-F) Multiple t-test. Data are mean ± SEM. *** p < 0.001 ** p < 0.01 * p < 0.05.

Figure S5. γδ T cell subsets activation in DSS-treated mice. Related to Figure 4-5 A) Representative gating strategy for γδ T cell subsets in colon LP. B) Frequency of γδ T cell subsets from colon tissue of DSS-treated mice in colon LP. C) AhR expression in γδ T cells subset from colon LP of DSS-treated mice. D-E) Expression of IL-22 by γδ T cell subsets stimulated 4 hours with IL-23 plus IL-1β (D) and PMA plus ionomycin (E) analyzed by flow cytometry. F-G) Expression of IL-17 by γδ T cell subsets stimulated 4 hours with IL-23 plus IL-1β (F) and PMA plus ionomycin (G) analyzed by flow cytometry. H) RorγT expression in γδ T cell subsets from colon LP of DSS-treated mice. I) Body weight loss during DSS-induced acute colitis in Csf3r+/+(n=4), Csf3r-/- (n=4) and Csf3r-/- mice upon adoptive transfer of neutrophils (NΦ) (n=4) with and without antibiotic (ABX) oral treatment (Csf3r+/++ABX n=4 and Csf3r-/-+ABX n=4, Csf3r-/-+NΦ+ABX n=4). Red arrows indicated days of neutrophils transfer. A-H) Representative data of two independent experiments, Csf3r+/+(n=4), Csf3r-/- (n=4). I) One experiments. B-H) Multiple t-test. I) Wilcoxon matched-pairs signed rank test. Data are mean ± SEM. *** p < 0.001 ** p < 0.01 * p < 0.05.

Figure S3. Neutrophils drive the expression of IL-23 by BMDMs. Related to Figure 3. A) Total number of ulcers in the colon of DSS-treated Csfr3+/+ (n=6) and Csf3r-/- (n=7) mice. B) IL-23 levels detected by ELISA in supernatants of BMDMs, neutrophils (NΦ) and BMDM-NΦ co-cultures after stimulation with GM-CSF+CpG. In indicated conditions neutrophils were pre-treated with diphenyleneiodonium (DPI) (10µM) or BMDM-NΦ were cultured in transwells (for all conditions n=6). C) Percentage of viable neutrophils (AnnexinV-/PI-) cultured alone or after co-culture with BMDMs. D) Schematic representation of the treatment schedule for anti-IL-22 treatment and neutrophil adoptive cell transfer during acute colitis. Days of treatment are indicated by red arrows. E) Percentage of body weight loss during DSS-induced acute colitis in Csf3r+/+(n=4), Csf3r-/-(n=4) and Csf3r-/- mice upon adoptive transfer of neutrophils treated with anti-IL-22 or isotype control (50µg/mouse) via i.p. injection (Csf3r-/- mice+Isotype n=3; Csf3r-/-+anti-IL-22 n=3). A) Representative data of five independent experiments. B) Representative data of three independent experiments. C) Representative of two independent experiments. E) One experiment. A) Unpaired Student’s t-Test. B-C) Multiple Student’s t-Test. E) Wilcoxon matched-pairs signed rank test. Data are mean ± SEM. *** p < 0.001 ** p < 0.01 * p < 0.05.

Figure S6. Neutrophil infiltration is associated with tissue repair gene signatures in patients with ulcerative colitis. Related to Figure 6. A-E) Enrichment plots for gene ontology term epithelial cell (GO:Epithelial Cell Proliferation (A, D); GO:Epithelial Cell Development (B, E); GO:Maintanance of Gastrointestinal Epithilium (C)) in CSF3Rhigh versus CSF3Rlow UC patients GSE109142 (A-C) GSE87473 (D-E). F-K) Enrichment plots for gene ontology term antimicrobial response (GO:Response to molecule of bacterial origin (F, I); GO:Defense response to bacterium; GO:Antimicrobial humoral immune response mediated by antimicrobial peptides) in CSF3Rhigh versus CSF3Rlow UC patients GSE109142 (F-H) and GSE87473 (I-K). CSF3R gene expression values were stratified by quartiles and patients belonging to the upper (GSE109142 n=51; GSE87473 n=21) and lower GSE109142 n=51; GSE97473 n=21) quartiles were considered for differential expression analysis. FDR, false discovery rate.