Ethics statement
All animal care and procedures performed in this study were reviewed and approved by the George Washington University (GWU) Institutional Animal Care and Use Committee (IACUC), protocol #A456, which complies with the guidelines stated in the National Institutes of Health’s (NIH) Guide for the Care and Use of Laboratory Animals. This manuscript was written in accordance with the ARRIVE guidelines (Animal Research: Reporting of In Vivo Experiments).
Mice
A total of 120 mice were included in this study. Equal numbers of 10-week-old male and female C57BL/6 J mice were obtained from The Jackson Laboratory (https://www.jax.org/). Mice were housed in groups of five in micro-isolator cages in the ABSL-3 laboratory at GWU at a constant room temperature (68–75°F) and humidity (30–70%) and with a 12-h light/dark cycle. Cages were lined with 1/8 inch corncob bedding, and each cage contained nesting materials and paperboard tunnels. Mice were given free access to food (PicoLab Verified 75 IF Irradiated) and water.
Mice were anesthetized with isoflurane, then injected with 25 µl (2.75 × 105 plaque forming units (PFU)) of CHIKV subcutaneously in the caudoventral aspect of the left hind foot near the tarsal joint. Mice were monitored at least once daily for signs of illness, such as lethargy, joint swelling, lameness, and limb gnawing. At 19 dpi, all mice were euthanized via CO2 asphyxiation, followed by a cardiac stick.
Virus
The strain of CHIKV used for this study was from the East, Central, and South African (ECSA) lineage that was initially isolated from a human case in South Africa (SAH-2123) and was supplied by the World Reference Center for Emerging Viruses and Arboviruses (Galveston, TX, USA). This strain was used in a previous study performed by our lab to determine a mouse model for studying chronic CHIKV arthritis3.
All studies performed with viable CHIKV were performed in Biosafety Cabinets in a certified BSL-3 laboratory and were conducted with the approval of the GWU Institutional Biosafety Committee, protocol #IBC-19–026.
CHIKV was passaged once in Vero cells (CCL-81, ATCC) in DMEM containing antibiotics and antimycotics, then quantified by viral plaque assay and quantitative real-time – polymerase chain reaction (qRT-PCR). The calculated titer of the stock virus was 1.1E7 PFU/ml, yielding a predicted limit of detection for the qRT-PCR assay of 0.01 PFU/ml.
Viremia detection
At 48 h post-infection, 200 µl of blood was collected from the submandibular vein for serum analysis to confirm mice were CHIKV positive by qRT-PCR. The Zymo Research Quick DNA/RNA Kit was used to isolate viral RNA, and the viral load was quantified using the Invitrogen Superscript III Platinum One-Step qRT-PCR Kit and CHIKV primer/probe set22. RT-PCR was analyzed using a LightCycler 96 Instrument (Roche Diagnostics) with thermal cycling conditions as follows: one cycle at 50 °C for 900 s, one cycle at 95 °C for 120 s, and 45 cycles at 95 °C for 3 s, followed by 55 °C for 30 s and a cooling cycle at 37 °C for 30 s.
Treatment with rIL2 and monoclonal antibody
At 16 dpi (the point at which all mice were CHIKV-negative), mice were treated once per day for three days by intraperitoneal injection with phosphate buffered saline (PBS) as a control, 1.5 µg/day recombinant mouse IL2, 5 µg/day anti-IL2 monoclonal antibody (mAb; clone JES6-1), or an rIL2/anti-IL2 complex containing 1.5 µg of rIL2 and 5 µg IL2 mAb. Thirty mice were included in each treatment group (15 males and 15 females).
Serum IL2 analysis
Peripheral blood was collected from the submandibular vein at 2, 14, 16, and 18 dpi and via cardiac puncture at 19 dpi. Blood was allowed to clot at room temperature for 30 min, then centrifuged at 3000 × g for 10 min. Serum was aliquoted and stored at − 80 °C until analysis. The IL2 Mouse ProQuantum Immunoassay Kit (Invitrogen) was used to analyze serum IL2 levels.
Flow cytometry
Spleens were dissected from mice following euthanasia, and splenocytes were collected by mashing the spleens through a 70-µm cell strainer. Red blood cells were lysed using ACK lysis buffer (Thermo Fisher Scientific). Cells were then stained for flow cytometry using the Biolegend antibodies FITC CD4 (clone RM4-5), PE-Cy7 CD25 (clone PC61.5), PE FoxP3 (clone MF-14), APC Helios (clone 22F6), and PerCP-Cy5.5 FR4 (clone 12A5), and Fixable Live/Dead Aqua stain from Invitrogen, as well as the FoxP3 Fixation/Permeabilization kit from eBiosciences. All antibodies were titrated prior to use in the study.
A BD LSRII flow cytometer was used to process samples, and analysis was performed using FlowJo Software. Compensation was performed for each fluorophore using CompOne beads (Thermo Fisher Scientific) and the ArC Amine Reactive Compensation Bead Kit (Thermo Fisher Scientific) for the Live/Dead stain. Treg cells were defined as CD4 + CD25hiFoxP3 + , and Teff cells were defined as CD4 + CD25 + FoxP3-. Further, FR4 (folate receptor 4) was used in combination with CD25 staining to define activated Tregs (FR4hiCD25hi), activated effector T cells (FR4intCD25int-hi), and naïve T cells (FR4loCD25lo)14.
Histology staining
After euthanasia, hindlimbs were dissected and fixed in 10% formaldehyde at 4 °C for a minimum of 48 h. Fixed tissues were rinsed with distilled water, and any remaining skin was removed. Decalcification of bones was achieved using a solution containing EDTA, sodium hydroxide, and polyvinylpyrrolidone (PVP) with a pH of 7.0–7.4. Tissues were submerged in the decalcification solution at 4 °C with agitation for 28 days, and the solution was changed every other day to maintain a pH of 7.0–7.4.
After decalcification, tissues were rinsed with distilled water and processed in PBS three times for 5 min each, in distilled water three times for 5 min each, in 50% ethanol for 5 min, and finally, in 70% ethanol for 5 min. After the final wash, the tissues were stored in 70% ethanol at 4 °C until processed.
To process for histopathological evaluation, tissue samples were dehydrated in increasing ethanol concentrations, xylene, and embedded in paraffin wax, from which five µm-thick sections were adhered to positively charged glass slides. Samples were stained with hematoxylin and eosin (H&E) for evaluation, and Masson’s trichrome stains were performed on select slides. All histologic evaluations were performed by a board-certified pathologist, who was blinded to the treatment group. Two blinded reads were performed one month apart, and the histologic score was averaged. All slides were evaluated using an Olympus BK46 microscope, and digital images were obtained using a Spot Imaging digital camera and Spot Pathsuite 2.0.
Histologic inflammation scoring system
The generation of the semi-quantitative histologic inflammation scoring system used in this study has been previously described3. In short, a microscopic evaluation of bone and joint tissue assessed inflammatory reaction and injury in (1) the synovium, (2) articular cartilage, (3) skeletal muscle and soft tissue, (4) periosteum, and (5) cortical bone. The scoring ranged from zero (no injury/inflammation) to two (significant injury/inflammation) for each of the five components, with a total possible composite score of 0 to 10. The scoring system was based on the maximal degree of inflammation and injury seen on day seven in our pilot study3.
Tarsal joint measurement
Tarsal joint width and breadth were measured daily using calipers and recorded in mm. Each joint measurement was performed three consecutive times and averaged to account for variability. The width was measured directly posterior to the tarsal joint foot pad projection, with only enough pressure to cause slight flaring of the toes. The breadth was measured at the same spot as the width, but the calipers were rotated around the foot 90 degrees. Footpad size was calculated as width x breadth, and the degree of inflammation was expressed as the increase in size relative to the pre-infection measurement (day -1), obtained by the following formula: [(x – day -1)/day -1)], where x is the footpad size measurement for a given day post-infection.
Statistical analysis
SASS and GraphPad Prism Software (version 9.4.1) were used to analyze and graph data, with a p < 0.05 considered statistically significant. The change in joint inflammation was examined by treatment group, from baseline to the treatment period following CHIKV infection (baseline vs. days 16–19), using a random effects mixed model. This nests observations within subjects to account for correlated measurements on the same subjects over time. The ‘treatment’ time point was averaged across days 16–19. The treatment group-by-time interaction was also assessed to determine whether the slopes over time differed by group. The group and time main effects were tested to ascertain whether joint size differed across groups and time. Additionally, all analyses comparing groups were performed using a one-way ANOVA followed by Dunnett’s post-test comparing each treatment group to the control (PBS). Treg/Teff ratios were calculated, and mean ratios were analyzed by treatment group using a general linear model adjusted for sex. Histology composite scores were analyzed using a two-way ANOVA (mixed methods). Two independent ratings were done for each subject for the histology score, and test–retest reliability was assessed using the Pearson r, Bland–Altman plot and weighted kappa.
To determine sample size, raw data was used from Miner et al.’s seven-day change in foot-pad swelling in mice inoculated with CHIKV, comparing controls vs. mice treated with abatacept, to establish the expected effect size for IL2 treatment on ankle swelling23. Their effect size was 0.97. In order to have a power > 0.95 to detect this effect in a two-tailed (between-groups) t test with an alpha of 0.05, our study needed at least 29 mice per treatment group.