Six-week old female and male BALB/cJ WT mice were purchased from Janvier Labs, Saint-Berthevin, France. BALB/c ΔdblGata1, BALB/c IL-4r/IL-5^−/−, BALB/c μMT, and C57BL/6 Rag2/IL-2rγ^−/− were bred at the animal facility “Haus für Experimentelle Therapie” of the University Hospital Bonn. ΔdblGata1 and μMT mice were originally obtained from Jackson Laboratory (Bar Harbor, United States), IL-4r/IL-5^−/− from Prof. Dr. Klaus Matthaei (Matthaei, Stem Cell & Gene Targeting Laboratory, ANU College of Medicine, Biology and Environment, Canberra, Australia) and Rag2/IL-2rγ^−/− from Taconic Biosciences Inc. (Cologne, Germany).

For the experiments, mice were housed in individually ventilated cages with unlimited access to food and water and a 12-h day/night cycle within the animal facility at the Institute for Medical Microbiology, Immunology and Parasitology (IMMIP), University Hospital Bonn. All animal experiments were performed according to EU Directive 2010/63/EU and approved by the appropriate state authorities Landesamt für Natur-, Umwelt- und Verbraucherschutz, Recklinghausen, Germany (AZ: 84–02.04.2015.A507, 81–02.04.2020.A244, 81–02.05.40.18.057).

Six to twenty week old male and female mice were naturally infected with L. sigmodontis via exposure to the tropical rat mite, Ornithonyssus bacoti, carrying the infective L3 larvae as described elsewhere (Reichwald et al., 2022). In short, mice were placed in cages with bedding material containing the mites. After 24 h, the bedding material containing the mites was removed, and the cages were placed on top of a plastic tub with disinfectant below the cages and no direct contact with the mice. After an additional 24 h, mice were moved into standard cages, and the bedding material was exchanged daily for 5 days to remove any remaining mites.

For the OXF treatment, a commercially available formulation of OXF (Dolthene) was used and dissolved in corn oil (Sigma-Aldrich, St. Louis, United States). Vehicle controls received only corn oil. For all experiments, treatment with OXF was performed orally, twice per day, 8 h apart starting 35 days post infection (dpi) after the development of adult worms (28–35 dpi) but before the onset of microfilaremia (50–56 dpi) (Hübner et al., 2009). For the experiments with immunodeficient strains, mice received either 5 or 12.5 mg/kg OXF in a total volume of 5 mL/kg per treatment for 5 days (see Supplementary Table S1 for a detailed breakdown of treatments in immunodeficient mice) (Hübner et al., 2020). For the 3-day combination therapy, mice received 12.5 mg/kg OXF for 3 or 5 days, and for the 2-day combination therapy, mice received either 12.5 or 25 mg/kg OXF per treatment for 2 or 5 days. For the 3-day combination therapy, 2 μg of interleukin-4 (IL-4), IL-5 or IL-33 were given intranasally in a volume of 10 μL under short anesthesia induced with 2% isoflurane (AbbVie, Wiesbaden, Germany) once per day for 3 days (Johansson et al., 2018; Beckert et al., 2020). For the 2-day combination therapy, 2 μg of IL-5 was given intranasally in a volume of 10 μL under short anesthesia induced with 2% isoflurane (AbbVie, Wiesbaden, Germany) once per day for 2 days.

For the FBZ treatment, FBZ (Sigma-Aldrich, St. Louis, United States) was dissolved in distilled water with 0.1% v/v tween80 (Sigma-Aldrich, St. Louis, United States) and 0.5% w/v hydroxyethyl cellulose (Sigma-Aldrich, St. Louis, United States). Vehicle controls received only distilled water with 0.1% v/v tween80 and 0.5% w/v hydroxyethyl cellulose. Treatment was performed 35 dpi for 2 or 5 days once per day via subcutaneous injections (Hübner et al., 2019). All mice received 2 mg/kg FBZ in a volume of 5 mL/kg per treatment (see Supplementary Table S1 for a detailed breakdown of treatments in immunodeficient mice).

Necropsies were performed to quantify the adult worm burden and the number of microfilariae in the peripheral blood at 70 dpi. Mice were euthanized with an overdose of isoflurane, and adult worms were isolated via lavage of the thoracic cavity with 8–10 mL PBS (Thermo Fisher Scientific, Waltham, United States). Isolated worms were counted and the gender was determined. To quantify the microfilariae burden at 56 and 70 dpi, 50 μL of peripheral blood was drawn from the facial vein into EDTA tubes (Sarstedt AG & Co. KG, Nümbrecht, Germany) and incubated with 950 μL red blood cell lysis buffer (Thermo Fisher Scientific, Waltham, United States) for 10 min at room temperature (RT). The blood was centrifuged at 400 × g for 5 min at RT, the supernatant was discarded, and microfilariae in the pellet were counted with a bright-field microscope (Zeiss, Göttingen, Germany) as previously described (Reichwald et al., 2022).

To quantify the embryonal stages, intact, female adult worms were fixed with 4% formaldehyde (Sigma-Aldrich, St. Louis, United States) in PBS (Thermo Fisher Scientific, Waltham, United States) for 24 h and then stored in 60% v/v ethanol in distilled water at RT until further analysis. Worms were then transferred into 1.5 mL Eppendorf tubes (Eppendorf SE, Hamburg, Germany) containing 100 μL Hinkelmann solution (0.5% eosin yellow, 0.5% phenol, 0.185% formaldehyde in distilled water) and crushed with a mortar. Embryonal stages [oocyte, morula, pretzel, stretched microfilaria, degenerated early stage (altered oocyte/morula) and degenerated late stage (altered pretzel/microfilaria)] were then quantified with a bright-field microscope (Zeiss, Göttingen, Germany) as previously described (Ziewer et al., 2012).

Lungs from L. sigmodontis-infected WT BALB/c mice (n = 6–7 per group) were inflated and fixed twice in 4% formalin for 24 h each. Lungs were then dehydrated in ethanol baths of increasing concentrations from 70 to 100% and placed in toluene before paraffin embedding. All sections were cut deep enough to see the main bronchi and perivascular adventitial spaces (PVS). Seven-micron-thick serial sections were prepared and various stainings were performed. (i) Hematoxylin and eosin staining was used to visualize lung structure and performed as follows: Sections were incubated with Mayer’s hematoxylin solution for 5 min, rinsed with tap water for 20 s and then incubated with 1% eosin solution for 1 min. (ii) Alcian Blue—Periodic Acid Schiff (AB-PAS) staining to visualize mucus-producing cells was performed as follows: Sections were incubated with 1% Alcian blue in 3% acetic acid for 20 min, rinsed with tap water and distilled water for 2 min each, incubated with 0.5% periodic acid for 5 min and again rinsed with distilled water. Next, sections were counterstained with Schiff’s reagent for 10 min, rinsed with tap water for 5 min and stained with hematoxylin for 1 min. Finally, sections were rinsed again with tap water for 2 min and then differentiated with acid alcohol. (iii) Luxol Fast Blue staining to visualize eosinophils was performed as follows: Sections were incubated with Luxol blue for 20 min, rinsed with running tap water for 2 min and then counterstained with Mayer’s hematoxylin for 10 s. The number of Luxol blue positive eosinophils was determined in 15 fields of view with an ×100 objective using Olympus BH2. The other sections were scanned at the MNHN light microscopy facility (CeMIM, Centre de Microscopie et d’IMagerie numerique, MNHN Paris) with a NanoZoomer S60 digital slide scanner (Hamamatsu) and images were analyzed with QuPath 0.3 software (Bankhead et al., 2017). For cell infiltration in PVS, 3–5 PVS areas per mouse were segmented manually, Hematoxylin positive nucleus detection was done using the “Cell detection” tool in QuPath and results were expressed as “number of nuclei/mm² of PVS.” Minimal thickness of bronchial arteries (in micrometer) was measured from 3 to 6 arteries of similar diameter per mouse.

The thoracic cavity lavage was performed with 8–10 mL PBS. The lavage was centrifuged at 400 × g for 5 min at 4°C and the supernatant was discarded. The pellet was resuspended in 1 mL red blood cell lysis buffer (Thermo Fisher Scientific, Waltham, United States). Cells were then washed with PBS containing 1% v/v FCS (PAN Biotech, Aidenbach, Germany) and 2 mM EDTA (Carl Rohe, Karlsruhe, Germany), resuspended in PBS containing 1% v/v FCS and 2 mM EDTA and counted with a CasyR TT Cell Counter (Schärfe Systems, Reutlingen, Germany). 1 × 10⁶ cells were used for flow cytometric analysis.

To isolate splenocytes, spleens were perfused with 3 mL of a 0.5 mg/mL collagenase VIII solution (Roche, Basel, Switzerland), cut into small pieces and incubated at 37°C for 30 min on a shaker with 200 rpm. Five milliliter PBS containing 1% v/v FCS and 2 mM EDTA were then added and the minced spleens were pushed through a 70 μm metal sieve to generate single-cell suspensions. Cells were then centrifuged at 400 × g for 5 min at 4°C and the supernatant was discarded. Red blood cell lysis was performed by incubating the pellet in 1 mL red blood cell lysis buffer (Thermo Fisher Scientific, Waltham, United States). Cells were then washed with PBS containing 1% v/v FCS (PAN Biotech, Aidenbach, Germany) and 2 mM EDTA (Carl Rohe, Karlsruhe, Germany), resuspended in PBS containing 1% v/v FCS and 2 mM EDTA and counted with a CasyR TT Cell Counter (Schärfe Systems, Reutlingen, Germany). 1 × 10⁶ cells were used for flow cytometric analysis.

Cells from the thoracic cavity and spleen were analyzed with one surface staining to identify lymphoid cells and two intracellular stainings to identify T helper cell subsets and myeloid cells. For the surface staining, cells were incubated for 20 min on ice in a mastermix in PBS with 1% v/v FCS, 0.1% v/v rat IgG (Sigma-Aldrich, St. Louis, United States) containing the following antibodies: anti-CD3 (Al700, clone GK1.5, BioLegend), anti-CD4 (BV605, clone RM4-5, BioLegend), anti-CD5 (PerCP Cy5.5, 53–7.3, BioLegend), anti-CD8 (PE, 53–6.7, BioLegend), and anti-CD19 (APC, eBio1D3, Thermo Fisher Scientific). Cells were then washed twice with PBS containing 1% v/v FCS, resuspended in 200 μL PBS containing 1% v/v FCS and 2 mM EDTA and filtered through 70 μm gauze (Labomedic, Bonn, Germany) before measurement with a CytoFLEX S (Beckmann Coulter, Brea, United States) and further analysis with FlowJo V10 (FlowJo, Ashland, United States).

For the intracellular staining, cells were incubated in a fixation/permeabilization buffer (Thermo Fisher Scientific, Waltham, United States) for 20 min at RT. Cells were then centrifuged at 400 × g for 5 min at 4°C and incubated overnight in PBS containing 1% w/v bovine serum albumin fraction V (PAA Laboratories, Cölbe, Germany) and 0.1% v/v rat IgG (Sigma-Aldrich, St. Louis, United States) at 4°C. The following day, cells were centrifuged at 400 × g for 5 min at 4°C and incubated in a permeabilization buffer (Thermo Fisher Scientific, Waltham, United States) for 20 min at RT. Cells were again centrifuged at 400 × g for 5 min at 4°C and stained with master mixes in PBS containing 1% w/v bovine serum albumin fraction V (PAA Laboratories, Cölbe, Germany) and the following antibodies for 45 min at 4°C in the dark: anti-CD3 (BV510, clone 145-2C11, BioLegend), anti-CD4 (Al700, clone 17A2, BioLegend), anti-CD8 (PerCP Cy5.5, clone 53–6.7, BioLegend), anti-CD11b (Al700, clone M1/70, BioLegend), anti-CD11c (BV605, clone N418, BioLegend), anti-CD25 (BV421, PC61, BioLegend), anti-GATA-3 (Al488, clone 16E10A23, BioLegend), anti-FOXP3 (PE-Cy7, FJK-16S, Thermo Fisher Scientific), anti-I-ab (BV421, M5/114.15.2, BioLegend), anti-Ly6C (APC-Cy7, clone HK1.4, BioLegend), anti-Ly6G (PE-Cy7, clone 1A8, BioLegend), anti-RELM-α (APC, clone DS8RELM, Thermo Fisher Scientific), anti-RORγT (PE, clone AFKJS-9, Thermo Fisher Scientific), Siglec-F (PE, clone E50-2440, BD Biosciences) and anti-T-bet (APC, clone 4B10, BioLegend). Cells were then washed twice with PBS containing 1% v/v FCS, resuspended in 200 μL PBS containing 1% v/v FCS and 2 mM EDTA and filtered through 70 μm gauze (Labomedic, Bonn, Germany) before measurement with a CytoFLEX S (Beckmann Coulter, Brea, United States) and further analysis with FlowJo V10 (FlowJo, Ashland, United States).

Statistical analysis was performed with GraphPad Prism software version 8/9 (GraphPad Software, San Diego, United States). The choice of statistical analysis was based on sample size and distribution of samples. Normality was assessed via the Shapiro–Wilk test. For normally distributed data, a One-Way ANOVA with Dunnet’s multiple comparison test was performed. For non-parametric data, the Kruskal-Wallis test followed by Dunn’s multiple comparisons test was used to assess significant differences between 3 or more groups or the Mann–Whitney-U test for differences between two groups. Data for adult worms, microfilariae, immune cells and histological analysis are shown as median with interquartile range. Data for embryonal stages in stacked bar graphs are shown as mean with standard error of mean. p < 0.05 were considered to be significant.

To characterize the role of the immune system during anti-filarial treatment, we naturally infected WT BALB/c mice and several immunodeficient strains, i.e., ΔdblGata1, IL-4r/IL-5^−/−, μMT, and Rag2/IL-2rγ^−/−, with Litomosoides sigmodontis and treated them orally with 5 or 12.5 mg/kg OXF twice per day for 5 days after the development of adult worms (35 dpi) but before the appearance of microfilariae at 50–56 dpi. Necropsies were performed 70 dpi to quantify worm burden and immunological changes.

Treatment with both 5 and 12.5 mg/kg OXF led to a statistically significant reduction of the median adult worm burden only in BALB/c WT (88 and 94% respectively) and ΔdblGata1 mice (80 and 90%) (Figure 1A). The reduction of the adult worm burden in the other strains was either not statistically significant with a median reduction of up to 50% or only statistically significant for one treatment condition, i.e., a reduction of 64% in μMT mice after treatment with 5 mg/kg OXF (Figure 1A). Interestingly, the strain with the most severe immunodeficiency (Rag2/IL-2rγ^−/−) displayed no reduction of the adult worm burden after treatment at all (Figure 1A). The effect of the treatment on the numbers of microfilariae in the peripheral blood presented a different picture (Figure 1B). Here we could observe a median reduction of 100% in BALB/c WT mice as well as the mature B cell-and antibody-deficient μMT mice whereas all other strains displayed no (ΔdblGata1) or only impaired effects on microfilariae numbers (IL-4r/IL-5^−/−, Rag2/IL-2rγ^−/−) (Figure 1B). Accordingly, treatment in immunodeficient mice led to fewer adult worm or microfilariae-negative animals (with the exception of μMT mice) compared to the treatment in BALB/c WT animals (Table 1). Importantly, the treatment was performed 35–39 dpi while microfilariae are only detectable in the peripheral blood after 49–56 days indicating that the effect of OXF was based on changes in the embryogenesis of adult worms rather than a direct effect against microfilariae.

To further assess the fertility of female worms, we analyzed drug effects on embryogenesis. Therefore, female worms were homogenized and stained with Hinkelmann solution and different embryonal stages, i.e., oocyte, morulae, pretzel, stretched microfilariae as well as degenerated early (oocyte, morulae) and degenerated late stages (pretzel, microfilariae), were counted with a bright field microscope (Figure 1C). Here, IL-4r/IL-5^−/− and Rag2/IL-2rγ^−/− mice had significantly more embryonal stages in the untreated mice than the other strains and treatment with OXF led to a substantial reduction of embryonal stages in all tested strains (Figures 1D–H). In line with the microfilariae numbers (Figure 1B), a closer examination of the treatment effect revealed a nearly complete elimination of late stages (<10 pretzel or stretched microfilariae per worm) only in BALB/c WT and μMT mice but not the other immunodeficient strains (Supplementary Table S2). Taken together, the macrofilaricidal efficacy of OXF appears to be influenced by the innate and adaptive immune system with a more substantial decrease in effectiveness in more severe immunodeficient strains. In contrast, the effect on microfilaremia appears to be influenced by components of innate and adaptive immunity but independent of mature B cells or antibodies and mediated via a reduction in fertility of adult female worms.

Next, we analyzed the efficacy of FBZ in immunodeficient mice by treating mice that were naturally infected with L. sigmodontis with 2 mg/kg FBZ subcutaneously for 2 or 5 days once per day starting treatment at 35 dpi. Necropsies were again performed 70 dpi to investigate the worm burden and immunological changes.

Similar to OXF, the macrofilaricidal efficacy of FBZ was reduced in all immunodeficient strains (Figure 2A). Treatment with FBZ in BALB/c WT mice led to a median reduction of 100% with 56 or 93% of animals cleared of all adult worms after 2 and 5 days of treatment, respectively (Table 2). By contrast, reductions in the immunodeficient strains ranged from 0 to 92% with significant reductions only in μMT mice after 2 days of FBZ and IL-4r/IL-5^−/− mice after 5 days of FBZ treatment (Figure 2A). In addition, FBZ treatment only achieved a partial clearance of adult worms in the immunodeficient strains with the most significant effect in ΔdblGata1 mice with 27 and 30% of animals being adult worm free, respectively (Table 2). Interestingly, the effect of FBZ on microfilariae was drastically different compared to the effect of OXF (Figure 2B). All mice with the exception of one IL-4r/IL-5^−/− mouse had no microfilariae detectable in the peripheral blood after FBZ treatment (Table 2). Results from the analysis of the embryonal development supported the microfilariae data showing significant reductions in all embryonal stages after treatment (Figures 2C–G). Furthermore, late stages were almost completely eliminated in all strains, and only early or degenerated stages could be detected (Supplementary Table S3). Hence we conclude that the macrofilaricidal efficacy of FBZ is dependent on the immune system. However, the prevention of microfilaremia is immune system-independent.

Next, we aimed to further characterize the impact of the host’s immune status by analyzing changes in immune cell populations in the thoracic cavity, the site of adult worm residency, and the spleen, which plays a crucial role in the elimination of microfilariae, after treatment (see Supplementary Figures S1, S2 for gating strategies). OXF and FBZ treatment led to distinct, strain-and compound-specific changes in both spleen (Figures 3A–D) and thoracic cavity (Figures 3E–H) compared to corresponding vehicle controls.

In BALB/c WT mice (shown in green in Figure 3), treatment with both OXF and FBZ induced only limited changes in cell frequencies in the spleen, with the strongest reductions observed for eosinophils, Ly6C(hi) monocytes and cDC2s after OXF treatment (Figure 3A) and Th1 cells, cDCs2 and pDCs after FBZ treatment (Figure 3C) while the total cell count was reduced for almost all cell types (Figures 3B,D). By contrast, treatment with OXF led to an overall increase in cell frequencies and total cell counts in ΔdblGata1 mice (blue, Figures 3A,B). Treatment with FBZ in ΔdblGata1 mice, on the other hand, led to broadly similar changes as in BALB/c WT mice with decreases in Th1, cDC2, and pDC frequencies and overall decreased total cell counts (Figures 3C,D). For IL-4r/IL-5^−/− mice, treatment with OXF led to no significant changes in cell frequencies or total counts (red, Figures 3A,B), whereas FBZ treatment reduced frequencies and total cell counts of lymphocytes (except Tregs) while myeloid cells except for cDC2s and pDCs were increased (Figures 3C,D). Rag2/IL-2ry^−/− had overall decreased frequencies and total cell counts after both OXF and FBZ treatment (purple, Figures 3A–D). μMT mice had decreased lymphocyte frequencies after OXF treatment, but unchanged or slightly elevated cell counts (black, Figures 3A,B) and overall increased lymphocyte numbers after FBZ treatment (Figures 3C,D). Interestingly, myeloid cell numbers were mostly increased after OXF treatment, whereas FBZ treatment reduced neutrophils, eosinophils and cDC numbers, while other myeloid cells were increased in μMT mice (Figures 3C,D).

In the thoracic cavity of WT mice, treatment with both OXF and FBZ led to an increase in most lymphocyte numbers except for Th1 cells (green, Figures 3E–H). Myeloid cells, on the other hand, were mostly decreased after OXF treatment or essentially unchanged, except for strongly decreased eosinophils after FBZ treatment (Figures 3E–H). For both ΔdblGata1 and IL-4r/IL-5^−/− mice, OXF treatment had only minor effects on cell numbers in the thoracic cavity, whereas treatment with FBZ led to a significant decrease in CD4⁺ (especially Th1) and CD8⁺ T cell numbers (blue and red, Figures 3E–H). For Rag2/IL-2rγ^−/−, OXF treatment induced only limited changes in thoracic cavity cell numbers while treatment with FBZ strongly reduced total cell numbers (purple, Figures 3E–H). By contrast, treatment in μMT mice revealed more extensive changes after OXF rather than FBZ treatment (black, Figures 3E–H). Overall, OXF and FBZ induced marked differences in both spleen and thoracic cavity cell compositions with distinct differential patterns in each strain.

Based on the reduced treatment efficacy of OXF and FBZ observed in immunodeficient mice, we hypothesized that stimulating the immune system during drug treatment might improve the efficacy or enable an equally effective treatment using a shorter treatment regimen. To investigate this, we naturally infected BALB/c WT mice with L. sigmodontis and treated them with 12.5 mg/kg OXF for 3 days (suboptimal time) with or without additional intranasal application of IL-4, IL-5, or IL-33 starting 35 dpi (Figure 4). The cytokines were given intranasally to induce a local immune response at the site of infection, i.e., the thoracic cavity (Jackson-Jones et al., 2016).

Treatment with OXF for 3 days alone led to no statistically significant difference of the adult worm burden, with a 33% median reduction compared to the vehicle controls (Figure 4A). Addition of IL-4 and IL-33 improved the median adult worm burden reduction to 58%. The combination with IL-5 boosted the efficacy to 91% achieving a similar reduction as a 5-day OXF treatment even though all three combination therapies did not lead to statistically significant differences in comparison to the vehicle control (Dunn’s post-hoc test). However, the combination with IL-5 did lead to significant differences compared to both the vehicle control (p = 0.025) as well as the 3-day OXF treatment (p = 0.004) via direct comparison (Mann–Whitney U test). Treatment with the cytokines alone did not lead to a reduction of the adult worm burden (Figure 4A). Besides the improved macrofilaricidal efficacy, the combination of OXF with IL-5 reduced the number of microfilariae-positive animals from 4/13 to 1/13 compared to OXF alone (Figure 4B). Analysis of the embryonal development showed a substantial reduction of all stages after OXF treatment (Figure 4C). Importantly, significant numbers of late stages remained after 3-day OXF treatment (137.1 ± 61.8 pretzel stages, 36.7 ± 24.9 stretched microfilariae, mean ± SEM) which were almost completely eliminated in the OXF + IL-5 group (5.8 ± 4.3 pretzel stages, 0.0 ± 0.0 stretched microfilariae) (Supplementary Table S4). Based on these results, we also tested a treatment regimen of only 2 days. However, a combination therapy of 12.5 or 25 mg/kg OXF with IL-5 for only 2 days failed to significantly reduce the adult worm burden or microfilariae numbers (Supplementary Figure S3).

Interestingly, flow cytometric analysis of the 3-day OXF combination therapy revealed a reduction in eosinophil frequencies in the spleen in mice treated with OXF, cytokines (IL-4, IL-5, IL-33) or a combination thereof compared to the vehicle controls (Figure 4D). The composition of immune cells in the thoracic cavity was mostly unchanged after treatment with an intriguing increase in monocyte numbers after 3-day OXF treatment (Figures 4F,G). To assess any pathological changes in adjacent organs caused by the combination therapy, we analyzed histological changes in the lungs of treated mice (Figure 5). Here, we observed that the treatment with IL-33 appeared to cause vascular inflammation with increased cell infiltration into the pulmonary vascular stroma (Figures 5A,B) and increased thickness of bronchial arteries (Figures 5C,D). Apart from these pathological changes, treatment with OXF, the combination of OXF with cytokines and the cytokines by themselves led to a reduction in mucus covering the surfaces of bronchioles (Figures 5E,F). Lastly, as expected, treatment with IL-5 led to an infiltration of eosinophils into the lung (Figures 5G,H).

Overall, the combination of OXF with IL-5 improved the macrofilaricidal efficacy, reduced the number of microfilariae-positive animals, and no pathological changes were observed in the lungs of treated mice.