Cul5^fl/fl LysM-Cre mice were generated by crossing Cul5^fl/fl mice with LysM^Cre mice (Jackson Laboratory, Bar Harbor, ME, strain #004781) to obtain LysM^Cre Cul5^fl/fl mice. Male and female mice with conditional deletion of Cul5 are referred to as Cul5^fl/fl LysM-Cre and these mice are on a C57BL/6 background. To the extent possible, LysM-Cre littermates were used as controls. All mice were bred in-house under specific pathogen-free conditions in the animal facility at the Children’s Hospital of Philadelphia (CHOP). The mice were housed at 18–23°C with 40–60% humidity, with 12-h light/12-h dark cycles. All mice, unless stated otherwise, were 8–12 weeks of age, and both sexes were used without randomization or blinding. Animal housing, care, and experimental procedures were performed in compliance with the CHOP Institutional Animal Care and Use Committee.

Cul5^fl/fl LysM-Cre and WT mice were injected s.c. with 200 μg of MOG_35–55 (Genemed Synthesis) emulsified in 5 mg/ml CFA (BD Biosciences). On the same day and again 48 hours later, mice received an intraperitoneal injection of 200 ng of pertussis toxin (Sigma-Aldrich). As negative controls, mice were left unimmunized. Mice were then examined daily for the clinical signs of EAE in a blinded fashion. The disease severity of EAE was scored using the standard clinical score as previously described: 0, no disease; 0.5, distal limp tail; 1, compete limp tail; 1.5, limp tail and hindlimb weakness; 2, unilateral partial hindlimb paralysis; 2.5, bilateral partial hindlimb paralysis; 3, complete bilateral hindlimb paralysis; 3.5 complete bilateral hindlimb paralysis and partial forelimb paralysis; 4, moribund (completely paralyzed); 5, death.

Mice were euthanized according to IACUC guidelines and perfused with PBS through the left ventricle of the heart using a 25G butterfly needle attached to a 10-mL syringe. Spinal cord, brain, and brain stem were removed by dissection. CNS tissues was processed through mechanical dissociation in 5 ml of digestion medium (collagenase I and Ia (Sigma-Aldrich, catalog nos. C0130 and C9891, respectively), and 20 μg/ml DNAse I (Roche, catalog no. 10104159001) and digested at 37°C for 30 min. Enzymatic digestion was stopped by adding EDTA to a final concentration of 5 mM. To homogenize the tissue, the suspension was repeatedly sucked up and ejected using a 5-ml syringe with an 18G needle until a uniform milky homogenate was formed, avoiding excessive foaming. The homogenate was filtered through a 70-μm nylon cell strainer into a 50-ml Falcon tube, which was then filled with media (RPMI 1640 with 1% FCS) and centrifuged at 1300 rpm for 8 min at 4°C to pellet the cells and myelin. The supernatant was discarded by pouring it off gently, and the pellet was resuspended in a final volume of 5 ml of 30% Percoll and mixed. The 30% Percoll-containing CNS mixture was poured gently into a 15-ml tube containing 5 ml of 70% Percoll and then centrifuged at 1800 rpm for 30 min at 4°C, without brakes. After centrifugation, a PBS–Percoll gradient formed, with each layer containing a specific fraction of the homogenate. The myelin was carefully sucked off using a suction pump or pipette. The large middle layer containing the leukocytes was transferred without the bottom layer of RBCs into a Falcon tube. The supernatant was discarded, and the cells were resuspended in media (RPMI 1640 with 1% FCS).

Bone marrow-derived macrophages were generated from the femurs and tibias of mice as previously described (35). Mice were euthanized according to institutional animal care guidelines, and femurs and tibias were harvested under sterile conditions. Bone marrow cells were flushed from the bones using a 25G needle and DMEM+GlutaMAX ^™(Gibco), medium supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin. The cell suspension was filtered through a 70 µm cell strainer to remove debris and centrifuged at 400 × g for 7 minutes at room temperature. The pellet was resuspended in red blood cell lysis buffer for 2 minutes at room temperature, washed with PBS, and plated in non-tissue culture-treated Petri dishes at a density of 5 × 10⁶ cells/mL in differentiation media.

Differentiation media consisted of complete DMEM+GlutaMAX ^™ supplemented with 10 ng/mL macrophage colony-stimulating factor (M-CSF; PeproTech). Cells were incubated at 37°C, 5% CO₂, with fresh media containing 10 ng/mL M-CSF replenished on day 3. By day 6, adherent cells were considered fully differentiated BMDMs. Macrophage identity was confirmed by flow cytometry for CD11b and F4/80 expression and morphological assessment under a light microscope.

For experimental assays, BMDMs were detached using ice-cold PBS containing 2 mM EDTA and collected for downstream applications. Cells were either stimulated with IL-4 (PeproTech) 10ng/mL, IL-13 10ng/mL (PeproTech), LPS 100ng/mL (Sigma Aldrich) or IFN-γ (PeproTech) 10ng/mL for the time points indicated.

Single-cell suspensions were stained with a fixable viability dye (LIVE/DEAD fixable blue stain), then pretreated with unlabeled anti-CD16/CD32 (Fc Block, BD Pharmingen). Cells were then stained in FACS buffer (PBS containing 2.5% FCS and 0.1% sodium azide) with mixtures of directly conjugated Abs.

For intracellular cytokine staining, a single-cell suspension was incubated with MOG_35–55 (Ge (30 ng/ml; Genemed Synthesis) and ionomycin (1 μM; Abcam) in the presence of GolgiPlug (1 μg/ml; BD Biosciences, catalog no. 555029) and GolgiStop (1 μg/ml; BD Biosciences, catalog no. 554724) for 4 h at 37°C and stained using the Cytofix/Cytoperm kit (BD Biosciences). For Foxp3 staining, a Foxp3 staining kit (eBioscience) was used according to the manufacturer’s instruction. Samples were analyzed using an LSRFortessa flow cytometer (BD Biosciences) at the CHOP Flow Cytometry Core Facility. Data were analyzed using FlowJo software v10 (Tree Star, Ashland, OR). Results are expressed as the percentage of positive cells or median fluorescence intensity (MFI).

Bone marrow-derived macrophages (BMDMs) were generated as described previously and collected on day 6 post-differentiation. Cells were washed twice with ice-cold PBS and lysed in RIPA buffer (50 mM Tris-HCl pH 7.4, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS) supplemented with protease and phosphatase inhibitors. Lysates were incubated on ice for 30 minutes, followed by centrifugation at 14,000 × g for 10 minutes at 4°C. Supernatants were collected, and protein concentrations were determined using the BCA protein assay kit.

SDS-PAGE and protein transfer: Equal amounts of protein (20–40 µg) were mixed with 2× Laemmli sample buffer containing β-mercaptoethanol and boiled at 95°C for 5 minutes. Samples were loaded onto a 4–12% Bis-Tris polyacrylamide gel and separated by SDS-PAGE at 120V for ~120 minutes in 1× running buffer. Proteins were transferred onto a membrane using a wet transfer system at 40V for 120 minutes.

Immunoblotting: Membranes were blocked with TBS-Tween (TBS-T, 0.1% Tween-20) for 1 hour at room temperature. Membranes were incubated overnight at 4°C with primary antibodies diluted in blocking buffer. After washing three times with TBS-T (5 minutes each), membranes were incubated with secondary antibodies (1:5,000) for 1 hour at room temperature.

Cul5^fl/fl LysM-Cre and WT mice were induced with EAE (as previously described). CNS was collected at day 12 post-induction and cells was sorted for CD45^hi CD11b⁺ macrophages. Data were processed using R for statistical analysis. Results were quantile-normalized and analyzed for differential abundance. Differentially abundant proteins were identified using a Limma t-test, and protein identifiers were mapped to their corresponding gene names. Gene-level data were subsequently analyzed for enrichment using the MSigDB Hallmark Gene Dataset. The heatmap was generated using the heatmap package, displaying the top 20 upregulated and downregulated genes. The volcano plot was generated with a significance threshold of p < 0.05.

Sorted cells underwent lysis and solubilization in 2% SDS-based lysis buffer. Lysate was reduced and alkylated before undergoing protein aggregation capture (PAC) cleanup and in-solution digestion with Trypsin/LysC in an automated KingFisher APEX system (68). The resulting peptides were desalted using C18 stage tips, dried via vacuum centrifugation, and reconstituted in 0.1% TFA containing iRT peptides.

Peptides were analyzed on a TimsTOF Pro2 mass spectrometer coupled with a NanoElute 2 nanoLC system using data independent acquisition (DIA). Subsequently, raw data was searched using Spectronaut (69), and bioinformatics analysis was conducted in R.

CUL5 is expressed across various tissues and cell types, playing a crucial role in protein degradation and cellular homeostasis (36). Its expression is particularly enriched in myeloid cells, suggesting a specialized function in regulating immune responses (37). CUL5 protein levels, however, vary depending on cell type and differentiation status (38), and its precise roles in myeloid cell biology, development, differentiation and neuroinflammation remain unexplored. Given that CUL5 works with SOCS-box containing proteins that are known to target signaling mediators (36, 39), particularly those downstream of cytokine receptors, we sought to test whether loss of CUL5- in myeloid cells would impact macrophage development. We generated mice in which Cul5 is selectively deleted in macrophages, monocytes, and neutrophils by crossing Cul5^fl/fl to LysM-Cre mice to produce Cul5^fl/fl LysM-Cre mice. To confirm that Cul5 was deleted, we generated BMDMs from WT and Cul5^fl/fl LysM-Cre mice and analyzed CUL5 protein levels by western blot ( Figure 1A ). We found that Cul5 was effectively deleted in BMDMs from Cul5^fl/fl LysM-Cre mice. We next assessed whether loss of Cul5 impacted macrophage differentiation capacity in vitro using bone marrow cells cultured in the presence or M-CSF ( Figures 1B, C ). BMDM cultures from WT and Cul5^fl/f LysM-Cre mice showed similar frequencies of cells expressing F4/80 and CD11b on days 2, 4, and 6 post-seeding ( Figures 1B ), showing no significant differences between the genotypes ( Figure 1C ).

To determine whether loss of Cul5 impacts myeloid cell differentiation in vivo, we assessed WT and Cul5^fl/fl LysM-Cre mice for frequencies and numbers of myeloid and lymphoid subsets using flow cytometry. We focused on organs known to be important for myeloid cell differentiation and sites of neuroinflammation. We found that the overall numbers of cells in these tissues were similar between WT and Cul5^fl/fl LysM-Cre mice ( Figure 1D ). Analysis of specific types of myeloid and lymphoid cells in bone marrow ( Figure 1E ), spleen ( Figure 1F ) and brain ( Figure 1G ) showed no substantial differences in myeloid and lymphoid cell abundance, nor frequency ( Supplementary Figure S1 ). These results suggested that the loss of Cul5 in myeloid cells did not significantly impact the development or differentiation of myeloid lineages under homeostatic conditions.

EAE is the most widely used mouse model of MS recapitulating key pathological features including immune-mediated demyelination and neuroinflammation (40). It is initiated when myelin specific CD4⁺ T cells become activated (40, 41) and the blood-brain barrier (BBB) is disrupted, allowing immune cell infiltration and myelin destruction. While EAE is considered to be T cell-mediated, macrophages play a critical role in disease progression by promoting the reactivation of T cells within the CNS (42). To test the role of Cul5 in myeloid cell-mediated neuroinflammation, EAE was induced in 8-10-week WT and Cul5^fl/fl LysM-Cre mice as illustrated ( Figure 2A ). Mice were monitored for weight loss and disease development (clinical score). Surprisingly, while both WT and Cul5^fl/fl LysM-Cre mice started to show increased clinical scores by day 10 post-induction, disease in Cul5^fl/fl LysM-Cre mice did not continue to progress ( Figure 2B ). Analysis at day 14 post-induction revealed that Cul5^fl/fl LysM-Cre mice exhibited significantly lower clinical scores compared to their WT counterparts ( Figure 2B ). This time point corresponds to the peak of the disease in the EAE model. Consistent with this, while WT mice showed a reduction of weight as EAE developed, Cul5^fl/fl LysM-Cre mice did not lose weight following EAE induction ( Figure 2C ).

Immune cell recruitment into the CNS is a key initiator of neuroinflammation, as it drives tissue damage and disease progression. To assess the types of cells in the CNS of WT and Cul5^fl/fl LysM-Cre mice, we isolated cells from the CNS and analyzed the landscape of cell types present in the CNS using flow cytometry. As expected, WT mice showed an increase in CD4⁺ T cells over control mice, as expected following EAE induction ( Figure 2D ). In contrast, Cul5^fl/fl LysM-Cre mice showed a significant reduction of CD4⁺ T cells in the CNS compared to EAE challenged WT mice ( Figure 2E ). Cul5^fl/fl LysM-Cre mice also had decreased numbers of neutrophils ( Figure 2F ) and CD45^hi microglia/macrophages ( Figure 2G ), but not CD45^int microglia/macrophages ( Figure 2H ) when compared to WT mice. Based on published data, these CD45^int and CD45^hi cells are considered to be a broad representation of microglia/macrophages across reactive states. Furthermore, when we analyzed CD45^hiMCHII⁺CD80⁺ microglia/macrophages, which represent activated macrophages with antigen-presenting characteristics, we again found fewer in Cul5^fl/fl LysM-Cre mice compared to WT ( Figure 2I ). These findings suggest that Cul5 deletion ameliorates immune cell accumulation in the CNS, thereby dampening inflammation and disease severity​.

CD4⁺ T cells are primary drivers of EAE pathogenesis. Peripheral activation of CD4⁺ T cells, and their differentiation into pro-inflammatory Th1 and Th17 subsets, occurs prior to their migration into the CNS, where they help to recruit neutrophils. Since a loss of Cul5 in macrophages might impact their ability to act as APCs, we next aimed to evaluate the peripheral activation of CD4⁺ T cells during the initiation of disease.

To assess whether Cul5 deletion in myeloid cells influences CD4⁺ T cell activation and cytokine production, we analyzed cells in the draining lymph nodes (dLNs) and spleens as well as CNS on day 8 after induction of EAE. We isolated cells from the dLNs, spleens, and CNS from LysM-Cre and WT mice at day 8 post-immunization and quantified total numbers of CD4⁺ T cells as well as their cytokine-production. To assess cytokine production, we stimulated T cells with MOG_35–55 in vitro and analyzed intracellular cytokine production by flow cytometry. While numbers of CD44+ ‘activated’ CD4+ T cells were significantly increased in both EAE cohorts over controls, we found that T cell numbers in the dLNs were comparable between Cul5^fl/fl LysM-Cre and WT mice, as evidenced by similar absolute numbers of total CD4⁺ T and CD44+ ‘activated’ CD4⁺ T cells ( Figures 3A, B ). Furthermore, we found no significant differences in the numbers of MOG_35–55 reactive IL-17⁺ ( Figure 3C ), TNF-α⁺ ( Figure 3D ), or IFN-γ⁺ ( Figure 3E ) producing T cells in dLNs between Cul5^fl/fl LysM-Cre and WT mice. Furthermore, the overall numbers of T cells in the spleens were comparable between Cul5^fl/fl LysM-Cre and WT mice ( Figure 3F ), as were the numbers of MOG_35–55 reactive IL-17⁺ producing CD4⁺ T cells ( Figure 3G ).

Following peripheral activation, MOG reactive CD4⁺ T cells migrate into the CNS. Thus, we next sought to assess whether Cul5-deletion in myeloid cells impacted T cell migration into the CNS. On day 8 post EAE induction, cells from the CNS of WT and Cul5^fl/fl LysM-Cre mice were analyzed using flow cytometry. We found that the number of cells in the CD45^hi gate was comparable between WT and Cul5^fl/fl LysM-Cre mice ( Figure 3H ). Furthermore, the total number of CD4⁺ T cells was similar between WT and Cul5^fl/fl LysM-Cre mice ( Figures 3I ). These data suggest that the loss of Cul5 in myeloid cells had no significant impact in T cell priming or migration into the CNS following EAE induction. This is consistent with WT and Cul5^fl/fl LysM-Cre mice showing a similar increase in disease score on days 9–11 post EAE induction.

Once in the CNS, T cells accumulate in the perivascular space before fully infiltrating into the parenchyma (43). Local APCs, such as microglia and macrophages, present CNS-derived antigens to CD4⁺ T cells, thereby reactivating them (44). This local reactivation is essential for T cell persistence in the CNS and it enhances the release of pro-inflammatory cytokines, which further disrupts the BBB and recruits additional immune cells (45). Among these APCs, infiltrating macrophages play a crucial role by presenting CNS-derived antigens via MHCII and providing key costimulatory signals, such as CD80/CD86, to enhance T cell activation (46). This interaction is necessary to amplify the immune response, leading to further immune cell recruitment. Given that Cul5^fl/fl LysM-Cre mice showed similar numbers of infiltrating T cells and macrophages on day 8 post EAE induction compared to WT mice, but showed decreased numbers of both types of cells by day 14 post-induction, we decided to further investigate macrophage numbers and phenotype on day 12, when Cul5^fl/fl LysM-Cre mice are just starting to show a decrease in EAE disease severity.

Specifically, WT and Cul5^fl/fl LysM-Cre mice were induced with EAE and on day 12 cells were isolated from the CNS and analyzed by flow cytometry. We did not observe a significant difference in the overall numbers of CD45^hi microglia/macrophages ( Figure 4A ), or neutrophils ( Figure 4B ) in the CNS at this time point. However, when we analyzed the phenotype of CD45^hi microglia/macrophages, we found a significant increase in the number of Arginase 1 expressing cells in Cul5^fl/fl LysM-Cre mice compared to WT ( Figure 4C ), suggesting that Cul5-deficient macrophages were biased toward an anti-inflammatory or reparative macrophage phenotype. Arginase 1 was predominantly expressed by CD45^hi cells, and not by CD45^int cells, consistent with a study by Greenhalgh et al. (47), where lineage tracing further identified CD45^hi cells as infiltrating monocytes. ( Figure 4D ). In our study, the increase in CD45^hi/Arg1 cells correlated with a change in regulatory T cell frequencies. Specifically, while the overall number of CD4⁺ T cells was not different between Cul5^fl/fl LysM-Cre mice compared to WT ( Figure 4E ), the frequency of FoxP3⁺ regulatory T cells (Tregs) was significantly higher in Cul5^fl/fl LysM-Cre mice ( Figure 4F ) compared to WT controls. This increase in FoxP3⁺ Tregs, coupled with the decrease in conventional CD4⁺ T cells ( Figure 4G ) and enhanced presence of Arginase 1⁺ macrophages, supported that Cul5 deficiency in myeloid cells may actively promote an anti-inflammatory state within the CNS.

The production of Arginase 1 by myeloid subsets, particularly macrophages, creates a tissue environment with insufficient essential amino acid concentration which restricts T cell activation and proliferation (48, 49). Given the increase in Arginase 1 positive microglia/macrophages in Cul5^fl/fl LysM-Cre mice, we next sought to test whether loss of Cul5 might bias macrophages toward an M2 phenotype in vitro. We examined Arginase 1 expression in WT and Cul5-deficient macrophages cultured in IL-4 or IL-13. BMDMs generated from WT and Cul5^fl/fl LysM-Cre mice were cultured for 24 hours in either IL-4 or IL-13 and cells were analyzed by flow cytometry. Both IL-4 and IL-13 were able to drive expression of Arginase 1 in WT cells ( Figure 5A ). Interestingly, Cul5-deficient BMDMs showed a significant increase in the percentage of Arginase 1⁺ cells following IL-4 stimulation compared to WT cells ( Figure 5A ). In contrast, the percentages of Arginase 1⁺ macrophages were similar between the two genotypes when cells were cultured in IL-13 ( Figure 5A ). To determine whether CUL5 might also play a role in macrophage pro-inflammatory responses, we cultured WT and Cul5^fl/fl LysM-Cre BMDMs with LPS and IFN-γ. Percentages of both CD80 and CD86 remained unchanged between groups, suggesting that Cul5 loss does not prevent a pro-inflammatory APC phenotype induction in macrophages ( Figure 5B ).

To gain a deeper understanding of how Cul5 loss impacts CNS myeloid subsets, we conducted whole-cell proteomic analysis on WT and Cul5-deficient CD45^hi microglia/macrophages directly isolated from the CNS of EAE-induced mice at day 12 post-induction. To do this we employed high-resolution mass spectrometry ( Figure 6A ). Our CNS isolation and fluorescence-activated cell sorting (FACS) strategies yielded approximately 40,000 cells per sample, enabling the identification of an average of 4,000 proteins per condition ( Supplementary Figure S3 ).

We applied a statistical threshold of p < 0.05 and a greater than 2-fold change to evaluate proteins significantly differentially expressed in Cul5^fl/fl LysM-Cre macrophages compared to their WT counterparts ( Figures 6B, C ). Among the top 10 differentially expressed proteins, we observed a notable reduction in key mediators of pro-inflammatory responses, including Clec4e ( Figure 6D ), a C-type lectin receptor known to modulate macrophage activation and pathogen recognition (50), Ssr1 ( Figure 6E ), a component of the translocon-associated protein complex that may contribute to antigen processing and MHC presentation (51); and Lyz2, a hallmark myeloid antimicrobial enzyme associated with phagocytosis and inflammatory responses (52, 53) ( Figure 6F ).

Interestingly, we also identified a significant downregulation of Atm (ataxia telangiectasia mutated kinase) in Cul5-deficient macrophages ( Figure 6G ). Atm plays an important role in DNA damage responses, cellular stress adaptation, response to metabolic changes and immune regulation (54–56). Moreover, ATM is crucial for shaping macrophage pro-inflammatory phenotype transition, by controlling IRF5 and its post-translational modifications (57). Together, these observations support the hypothesis that CUL5 has an important function in microglia/macrophages genomic stability and stress adaptation mechanisms in the CNS microenvironment.

Among the top upregulated proteins, we found Serpinb6 ( Figure 6H ), which plays a role in cellular damage responses (58, 59). Additionally, Cul5-deficient cells showed upregulation of Aldh5a1 ( Figure 6I ), a key enzyme in GABA metabolism and oxidative stress responses (4, 60), which may suggest a metabolic reprogramming or foamy macrophage-like features. These findings highlight that CUL5 plays an important role in the mechanisms by which CNS microglia/macrophage subsets function during neuroinflammation.