Robust monosodium urate crystal-induced transcriptomic changes in multiple human immune cell systems
Nicholas A Sumpter, Nils Asmann, Brenda Kischkel, Pariyaphon Lertprachakwong, Paloma M Guzzardo, Mayank Tanson, Stephane E Castel, Heather A Arnett, Jeffrey Edberg, Tony R Merriman, Leo A B Joosten, Richard Reynolds
Affiliation(s):
Department Of Internal Medicine, Radboud University Medical Center, Nijmegen, Netherlands
Objectives
There is limited and conflicting data available for the transcriptomic response of human immune cells to MSU crystals. Using three new datasets and one published dataset, we aimed to identify robust MSU-induced transcriptomic changes in three different human immune cell systems.
Methodology
Two new cohorts were included in this study, with all experiments performed independently between the two cohorts. The first included 6 gout cases and 3 controls recruited at the University of Alabama at Birmingham, Alabama, USA (UAB). The second included 10 controls recruited at Radboud University Medical Center, Nijmegen, Netherlands (RUMC). Blood samples were taken from all 19 individuals, with peripheral blood mononuclear cells (PBMCs) isolated using standard methodology. Fresh PBMCs were stimulated for either 8h or 24h with RPMI (control), lipopolysaccharide (LPS), or MSUc.
An independent study was also performed in THP-1 cells. First, THP-1 cells were differentiated over 24h with PMA, then rested for a further 48h. The cells were then stimulated for 18h with DMSO (control), MSUc, or MSUc + MCC950 inflammasome inhibitor.
After stimulation, RNA was isolated from the cells, then sent for RNA sequencing. Reads were aligned to the genome to produce gene counts. RNA count matrices were analysed in R using either a linear mixed model approach (for both PBMC studies) or using DESeq2 (for the THP-1 study). In all cases, significance was determined by an FDR-corrected P-value < 0.05.
Summary statistics were also obtained for a published dataset of human monocyte-derived macrophages (hMDM) from 5 donors stimulated for 5h with MSUc compared to PBS control, with genes selected based on the additional criteria of an absolute log2 foldchange > 1 (Cobo et al., 2022).
Findings: We validated a clear transcriptomic signature associated with MSUc stimulation, with each dataset exhibiting between 1,974 and 4,779 differentially expressed genes in the MSUc condition compared to control samples. In total, 1,039 genes were significant and directionally consistent in both PBMC datasets. Of these, 55 were also significant in the THP-1 and hMDM datasets with consistent directionality, with 41 of these showing an opposite effect with inflammasome inhibition in THP-1 cells, and 14 of these showing either no effect or an opposite effect with LPS in PBMCs (Figure 1 & Table 1). Of these 14 genes, CSF1 was the only gene also implicated by the gout GWAS (Major et al., 2024).
Significance: This study highlights genes that are extremely likely to be involved in the inflammatory mechanism of a gout flare, with CSF1 being strongly implicated. CSF1 encodes the M-CSF protein, which is involved in the differentiation of monocytes to macrophages. The gout GWAS risk allele (rs2938616-T) results in decreased expression of CSF1 in primary fibroblasts, while MSUc stimulation consistently increases the expression of this gene in human immune cells. This suggests that those individuals with the gout-risk allele would mount an insufficient CSF1 response to MSUc deposition in the joint, leading to an increased risk of gout. In line with this, the six individuals with gout in the UAB cohort showed a reduced CSF1 response to MSUc compared to the three controls in this cohort, though this needs validation in a larger cohort (Figure 2). This finding suggests that elevated levels of CSF1 are protective of gout, perhaps driving appropriate monocyte differentiation that controls the response to MSUc.
References:
1. Cobo, I., Cheng, A., Murillo-Saich, J., Coras, R., Torres, A., Abe, Y., ... & Guma, M. (2022). Monosodium urate crystals regulate a unique JNK-dependent macrophage metabolic and inflammatory response. Cell reports, 38(10).
2. Major, T. J., Takei, R., Matsuo, H., Leask, M. P., Sumpter, N. A., Topless, R. K., ... & Merriman, T. R. (2024). A genome-wide association analysis reveals new pathogenic pathways in gout. Nature genetics, 56(11), 2392-2406.
