Macrophage depletion protects against cisplatin-induced ototoxicity and nephrotoxicity

Cisplatin is a widely used anticancer drug with notable side effects including ototoxicity and nephrotoxicity. Macrophages, the major resident immune cells in the cochlea and kidney, are important drivers of both inflammatory and tissue repair responses. To investigate the roles of macrophages in cisplatin-induced toxicities, we used PLX3397, a U.S. Food and Drug Administration–approved inhibitor of the colony-stimulating factor 1 receptor, to eliminate tissue-resident macrophages. Mice treated with cisplatin alone had considerable hearing loss (ototoxicity) and kidney injury (nephrotoxicity). Macrophage ablation resulted in significantly reduced hearing loss and had greater outer hair cell survival. Macrophage ablation also protected against cisplatin-induced nephrotoxicity, as evidenced by markedly reduced tubular injury and fibrosis. Mechanistically, our data suggest that the protective effect of macrophage ablation against cisplatin-induced ototoxicity and nephrotoxicity is mediated by reduced platinum accumulation in both the inner ear and the kidney. Together, our data indicate that ablation of tissue-resident macrophages represents an important strategy for mitigating cisplatin-induced ototoxicity and nephrotoxicity.


Fig. S2. Macrophage ablation followed by partial repopulation resulted in comparable depletion of macrophages in various compartments within the cochlea (Experiment 1).
After completing three cycles of the cisplatin administration protocol with initial PLX3397 treatment through once-in three days oral gavage, followed by subsequent endpoint auditory testing, cochleae were harvested.CX3CR1 GFP -positive macrophages were quantified in various compartments within the cochlea, including (A) modiolus, (B) Rosenthal's canal, and (C) organ of Corti.(A-C) Statistical analysis was performed using one-way ANOVA with Tukey's multiple comparisons test.(D) Representative images of the middle turn of the cochlea showing macrophage density and distribution in different cochlear compartments in the four treatment groups: Saline/Vehicle, Saline/PLX3397, Cisplatin/Vehicle, and Cisplatin/PLX3397.Nuclei were stained with Hoechst 33342 (blue).Scale bar, 100 μm.

Fig. S3. Macrophage ablation followed by partial repopulation resulted in protection against
OHC dysfunction in mice of both sexes (Experiment 1).DPOAEs were recorded to assess OHC function in female and male mice.An emission at 2f1-f2 was considered present when its amplitude exceeded -5dB (dotted line).Sample groups are represented by color-coded line: saline/vehicle (blue line), saline/PLX3397 (purple line), cisplatin/vehicle (red line), and cisplatin/PLX3397 (green line).(A) female and (B) male mice.Grey line indicates the biological noise floor.Statistical analyses were performed using two way-ANOVA with Tukey's multiple comparisons test (main column effect).Mean±SEM, n=10-16 mice (5-7F and 5-9M) per experimental group (total 49 mice; 24F and 25M).Statistical analyses to identify the difference in protection between males and females were performed using Mixed Effect Modeling with the R package "lmerTest" as described in the Methods.

Fig. S4. Co-administration of cisplatin and PLX3397 protects against cisplatin-induced impairment of wave I latency (Experiment 1).
(A-D) Auditory sensitivity was measured by ABRs at baseline and endpoint and wave I amplitude and latency were assessed at low frequencies at (A,C) 8kHz and (B,D) 11.2kHz.Statistical analyses were performed using two way-ANOVA with Tukey's multiple comparisons test (main column effect).Mean±SEM, n=10-15 mice (4-6F and 5-9M) per experimental group (total 46 mice; 21F and 25M).

Fig. S5
. Macrophage ablation followed by partial repopulation using PLX3397 protects against cisplatin-induced SGN loss (Experiment 1).Inner ear tissues were harvested following endpoint auditory testing and stained for Tuj-1 (red) to visualize and quantify SGNs.Nuclei were stained with Hoechst 33342 (blue).(A) Representative images and (B) quantification demonstrate that cisplatin-induced SGN loss occurred primarily in the apex, with no significant loss in the middle or basal regions.PLX3397 protected against cisplatin-induced loss of SGNs.(A) Scale bar, 100 μm.(B) Mean±SD, n=6 cochleae (3F and 3M) per experimental group (total 24 cochleae; 12F and 12M).Statistical analysis was performed using one way-ANOVA with Tukey's multiple comparisons test.

Fig. S6. Sustained macrophage ablation resulted in comparable depletion of macrophages in various compartments within the cochlea (Experiment 2).
Following three cycles of cisplatin administration protocol with daily oral gavage of PLX3397, and subsequent auditory testing at the endpoint, cochleae were collected.CX3CR1 GFP -positive macrophages were visualized and quantified in various compartments within the cochlea including (A) modiolus, (B) Rosenthal's canal, and (C) organ of Corti.(A-C) Statistical analysis was performed using one-way ANOVA with Tukey's multiple comparisons test.(D) Representative images of the middle turn of the cochlea showing macrophage density and distribution in different cochlear compartments in the four treatment groups: Saline/Vehicle, Saline/PLX3397, Cisplatin/Vehicle, and Cisplatin/PLX3397.Nuclei were stained with Hoechst 33342 (blue).Scale bar, 100 μm.

Fig. S7. Sustained macrophage ablation using PLX3397 ablates macrophages in the osseous spiral lamina and the stria vascularis (Experiment 2). (A) Representative images and
(B) quantitative analysis indicate that daily administration of PLX3397 in Experiment 2 resulted in ablation of macrophages in the osseous spiral lamina overall with the greatest ablation efficiency (92%) in the apex and 80.5% ablation in the basal region of the cochlea, when compared to mice treated with saline/vehicle.In addition, quantitative analysis indicated a significant reduction in macrophages following cisplatin treatment (cisplatin/vehicle-treated mice) compared to mice treated with saline/vehicle, with the middle regions exhibiting the greatest loss (Apex: 27.3%; Mid-Apex: 45.9%; Mid-Base: 48.9%; Base: 28.3%; Base-Hook: 18.9% reduction).Scale bar, 400 μm.Mean±SD, n=6 cochleae (3F and 3M) per experimental group (total 24 cochleae; 12F and 12M).Statistical analysis was performed using one-way ANOVA with Tukey's multiple comparisons test.Statistical comparisons (asterisks or n.s.) are color-coded as described in Methods.(C-D) Stria vascularis wholemounts were dissected from basal and middle regions of the cochlear lateral wall.(C) Representative images and (D) quantitative analysis of CX3CR1 GFP -positive PVMs in the stria vascularis.Scale bar, 100 μm.(B,D) Mean±SD, n=3-6 cochleae (2-3F and 2-3M) per experimental group (total 19 cochleae; 9F and 10M).Statistical analysis was performed using one-way ANOVA with Tukey's multiple comparisons test.

Fig. S8. CSF1R inhibition via PLX3397 treatment does not affect peripheral immune cells in the spleen (Experiment 2).
Spleens were isolated and single cell suspensions were used for flow cytometric analyses.(A) A representative gating strategy was used to analyze immune cell profiling in the spleen.The absolute numbers of immune cells in the spleen were analyzed: (B) CD45+ Leukocytes, (C) T cells, (D) B cells, (E) NK cells, (F) CD11b+ macrophages, (G) CD11b+ CX3CR1+ macrophages, (H) CD11b+ CX3CR1-macrophages, and (I) neutrophils.Cisplatin reduced the numbers of all analyzed leukocytes.Sustained macrophage ablation via PLX3397 administration did not significantly alter the numbers of leukocytes in the spleen.Mean±SD, n=7-9 spleens (3-4F and 4-6M) per experimental group (total 34 spleens; 14F and 20M).Statistical analysis was performed using Kruskal-Wallis test with Dunn's multiple comparisons test.

Fig. S9. Cisplatin does not induce macrophage activation and immune cell infiltration into the cochlea (Experiment 2).
Inner ear tissues were harvested following endpoint auditory testing and stained for CX3CR1 GFP (green), Iba-1 (red), and CD45 (magenta).Iba-1 expression, known to increase with macrophage activation, was visualized to assess the activation state of cochlear macrophages.Representative image of (A) the whole cochlear section (Scale bar, 300 μm) and (B) magnified inset images (from white box in panel A) of the cochlear modiolus (Scale bar, 50 μm).Nuclei were stained with Hoechst 33342 (blue).(C-E) Quantification of CX3CR1 GFP and CD45 labeling was performed to assess (C) the number of [CX3CR1 GFP +CD45+] macrophages, (D) the percentage of [CX3CR1 GFP +CD45+] macrophages to total number of CD45+ cells ((CX3CR1 GFP +CD45+ / total CD45+))x100), and (E) the number of infiltrating immune cells into the cochlea by determining the number of cells that are [CX3CR1 GFP -CD45+].Mean±SD, n=6 cochleae (2F and 3-4M) per experimental group (total 11 cochleae; 4F and 7M).Statistical analysis was performed using a Student's t-test.

Fig. S10. Sustained macrophage ablation using PLX3397 resulted in complete protection against cisplatin-induced OHC dysfunction in both male and female mice (Experiment 2).
OHC function was evaluated by DPOAE in (A) female and (B) male mice.DPOAEs were considered present at 2f1-f2 when the DPOAE amplitude surpassed the -5dB threshold (dotted line).The grey line represents the biological noise floor.Data are shown as Mean±SEM, n=8-9 mice (4-5F and 4-5M) per experimental group (total 34 mice; 16F and 16M).Statistical analysis was performed using two way-ANOVA with Tukey's multiple comparisons test (main column effect).

Fig. S11. Sustained macrophage ablation protected against cisplatin-induced impairment of wave I latency (Experiment 2).
(A-D) Auditory sensitivity was measured by ABRs at baseline and endpoint and wave I amplitude and latency were assessed at low frequencies at (A,C) 8kHz and (B,D) 11.2kHz.Statistical analyses were performed using two way-ANOVA with Tukey's multiple comparisons test (main column effect).Mean±SEM, n=7-9 mice (3-4F and 4-5M) per experimental group (total 32 mice; 14F and 18M).

Fig. S14
. Sustained depletion of macrophages via PLX3397 treatment protects against cisplatin-induced nephrotoxicity in both female and male mice (Experiment 2).(A) Plasma blood urea nitrogen (BUN) levels, (B) neutrophil gelatinase-associated lipocalin (NGAL) levels, (C) tubular injury scores assessed using Periodic Acid-Schiff staining, (D) fibrotic area assessed using Masson-Trichrome staining, (E) cell-associated fibrillar collagen-positive areas assessed by Sirius red staining, and (F) fibronectin-positive areas were evaluated and presented separately for female and male mice.(A-F) Data are expressed as mean±SD, n=8-9 blood and kidney samples (4F and 4-5M) per experimental group (total 34 mice; 16F and 18M).P-values were calculated using oneway ANOVA with Tukey's multiple comparisons test.

Fig. S15. Sustained macrophage ablation via PLX3397 treatment prevented cisplatininduced elevation of cytokines and toll-like receptors (Experiment 2).
Total RNA was extracted from the kidney, and the gene expression levels of cytokines (Tnf and Tgfb1) and toll-like receptors (Tlr2 and Tlr4) were measured using RT-PCR from the four treatment groups: Saline/Vehicle, Saline/PLX3397, Cisplatin/Vehicle, and Cisplatin/PLX3397.The expression levels were normalized to Gapdh (internal control) and are expressed relative to Saline/Vehicle-treated group.(A-F) Data are expressed as mean±SD, n=5-6 kidney samples (2-3F and 2-3M) per experimental group (total 22 mice; 11F and 11M).P-values were calculated using one-way ANOVA with Tukey's multiple comparisons test.

Fig. S16
. Incomplete macrophage ablation via PLX3397 treatment protects against cisplatininduced nephrotoxicity in both female and male mice (Experiment 1).Blood was collected in heparin-containing syringe from the inferior vena cava and spun down.Plasma was used to measure (A) plasma blood urea nitrogen (BUN) levels and (B) neutrophil gelatinase-associated lipocalin (NGAL) levels to assess kidney injury in (A,D) all, (B,E) female, and (C,F) male mice.(A-F) Data are expressed as mean±SD, n=8-11 blood and kidney samples per experimental group (total 39 mice; 17F and 22M).P-values were calculated using one-way ANOVA with Tukey's multiple comparisons test.