TNFAIP3 in circulating and parenchymal myeloid lineage critically controls monocytes, monocytes-derived cells and microglia


 Background. The intracellular ubiquitin-ending enzyme TNFAIP3 is one of the most potent inhibitor of the pro-inflammatory NF-kB pathway. Single nucleotide polymorphisms in the TNFAIP3 locus have been associated to autoimmune inflammatory disorders, including Multiple Sclerosis (MS). Previously, we reported a TNFAIP3 down-regulated gene expression level in blood and specifically in CD14+ monocytes obtained from treatment naïve MS patients in comparison to healthy controls (HC). Notably, myeloid cells which include monocytes, exert a key role in the neuro-inflammatory pathogenic process of MS.Methods. Here, we evaluated the effect of the specific TNFAIP3 deficiency in myeloid cells including monocytes, monocyte-derived cells (M-MDC) and microglia analyzing the lymphoid organs and central nervous system (CNS) of experimental mice. The TNFAIP3 deletion is induced using conditional knock-out mice for the myeloid lineage. Flow cytometry and histological procedures were applied to evaluate the immune cell population of spleen, lymph nodes and bone marrow and the microglial cell density in CNS, respectively.Results. Here, we found that the deletion of TNFAIP3 in myeloid cells induces a reduced body weight, a decrease in the percentage number of M-MDC and of common monocyte and granulocyte precursor cells. We also reported that the deletion of TNFAIP3 in myeloid cells reports an increased microglial cell density in brain.Conclusions. Collectively, the results suggest that the presence of TNFAIP3 in myeloid cells critically controls the development of M-MDC in lymphoid organ and of microglia in brain.


Background
Inflammation is an adaptive physiological response to cell injury characterized by the production of pro-and anti-inflammatory mediators that establish both innate and acquired immune response (1). During physiological conditions, inflammation restore tissue integrity and function. However, in case of prolonged or non-resolving inflammatory events, inflammation can evolve into detrimental process contributing to the pathogenesis of chronic inflammatory disorders, such as autoimmune diseases. Many of the signaling molecules produced by inflammatory events are under the control of the transcription factor nuclear factor kappa-light-chainenhancer of activated B cells (NF-kB), able to induce inflammatory responses through the activation and the production of pro-inflammatory cytokines, cellular mediators and response genes. In this contest, the intracellular ubiquitin-ending enzyme TNFAIP3 is one of the most potent inhibitor of the NF-kB pathway (2).
TNFAIP3 has an ubiquitin ligase C-terminal and a de-ubiquitinase N-terminal domain (3) and both cooperate to inhibit the NF-kB signaling through the ubiquitination of different NF-kB-signaling molecules.
Single nucleotide polymorphisms (SNPs) in the TNFAIP3 locus have been associated to autoimmune disorders such as systemic lupus erythematous (SLE) (4), rheumatoid arthritis (RA) (4,5), psoriasis (6), type one diabetes (7), coeliac disease (7) and multiple sclerosis (MS) (8). Previously, we reported a TNFAIP3 downregulated gene expression level in the whole blood and in peripheral blood mononuclear cells (PBMCs) obtained from treatment naïve MS patients in comparison to healthy controls (HC) (9)(10)(11)(12). Notably, we observed the abnormal expression levels of TNFAIP3 being reverted in patients with MS during pregnancy, which represents a transitory state of immune tolerance associated with reduced disease activity (9). We also demonstrated a negative correlation between the TNFAIP3 gene expression level and the clinical parameters such as the relapse rate and Expanded Disability Status Scale (EDSS) score, indicating that the most aggressive form of MS are characterized by lower level of TNFAIP3 expression (10).
Furthermore, we found that the peripheral blood cell population most affected by the reduced expression of TNFAIP3 in MS patients are CD14+ monocytes (13).
Consistently, different studies highlighted an important role of myeloid cells which include monocytes, monocytes-derived cells (M-MDC) such as macrophages and dendritic cells and microglia, in the pathogenic autoimmune process of MS. Indeed, activated macrophages and microglia are hallmarks of active lesions with ongoing demyelination and axonal injury in MS able to induce neuro-inflammation in the brain parenchyma (14).
To date the specific evaluation of the effects of the TNFAIP3 deficiency in monocytes derives from few studies with murine models. Interestingly, mice deficient in TNFAIP3 either completely (15) or in hematopoietic cells (16) developed severe inflammation, hypersensitive to both lipopolysaccharide and TNF-alpha, cachexia and premature death. In contrast, the circulating myeloid-TNFAIP3 deficient mice did not develop similar signs. However, these mice showed spontaneous polyarthritis and high levels of inflammatory cytokines in their serum, consistent with a sustained NF-kB activation (17). Recently, Voet and colleagues investigated the role of TNFAIP3 in microglia using a targeting strategy based on the longevity and capacity of self-renewal of myeloid cells (18). Notably, mice

Statistical analysis
Statistical analysis was performed using GraphPad Prism(GraphPad Software, version 5; San Diego, CA,US). Continuous data were presented as medians and ranges and categorical data were given as counts and percentages. Normality of distribution was assessed by the Shapiro-Wilk test. The differences between groups were calculated with Mann-Whitney U test. p values < 0.05 were considered significant.

Results
To characterize the phenotype of adult TNFAIP3 cx3cr1-KO mice, their body weight was recorded at 3 months of age. Notably, TNFAIP3 cx3cr1-KO mice (n=7) showed a lower body weight compared to their WT littermates (n=7) (Figure 1 These results indicate that the deletion of TNFAIP3 in myeloid cells leads to an altered immunological phenotype not only in myeloid linage, as expected but also in lymphocytes, such as B cells. In order to corroborate the results obtained on spleen, the immunological phenotype analysis of lymph nodes obtained from the same micewas performed (Figure 3; WT n=8, TNFAIP3 cx3cr1-KO n=4). Notably, TNFAIP3 cx3cr1-KO showed a reduction of the percentage number of macrophages ( Figure 3A Specifically, we investigated the consequences of TNFAIP3 ablation in the common myeloid precursors (CMPs, Figure 4A) and in the common monocyte and granulocyte precursor cells (CMGPs, Figure 4B). No differences were highlighted in the percentage number of the CMPs between WT and TNFAIP3 cx3cr1-KO mice ( Figure 4A, Mann-Whitney test p>0.05). Conversely, the analysis revealed a decrease of CMGPs in TNFAIP3 cx3cr1-KO mice compared to their WT littermates ( Figure 4B, Mann-Whitney test p=0.006).
Considering that the TNFAIP3 cx3cr1-KO mice lead the absence of TNFAIP3 gene also in microglia and that the microglial phenotype has been already studied in the inducible TNFAIP3 cx3cr1-KO mice by Voet and colleagues, we confirmed their results on microglial cell density in the CNS parenchyma. Specifically, the immunohistochemistry analysis was performed in the corpus callosum and in the spinal cord of 3 months-old WT (n=6) and TNFAIP3 cx3cr1-KO (n=3) littermates.
Notably, the evaluated regions represent the most affected areas by demyelination in the human MS and in the EAE murine model. Representative images of coronal section of corpus callosum and spinal cord immunostained with Iba1 antibody were reported in Figure 5. The quantification analysis of the microglial cell density showed that TNFAIP3 cx3cr1-KO mice have an increased microglial cell density in the corpus callosum compared to their WT littermates ( Figure 6A, Mann-Whitney test p=0.024). No differences were detected in the microglial cell density of spinal cord between WT and TNFAIP3 cx3cr1-KO mice ( Figure 6B-D, Mann-Whitney test p>0.05).
These data indicate that the lack of TNFAIP3 influence also the myeloid glial compartment.

Discussion
TNFAIP3 is a central gatekeeper in inflammation and able to inhibit the NF-kB pathway. TNFAIP3 has been genetically associated with different inflammatory autoimmune diseases (4-7), among which MS (8). We previously demonstrated a significant down-regulation of the TNFAIP3 transcript level in peripheral blood obtained from treatment-naïve MS patients in comparison to HC (9)(10)(11)(12)22). Further researches revealed that the unbalanced expression of TNFAIP3 in MS was mainly due to the monocyte cell population (13). Interestingly, the analysis of the role of TNFAIP3 in myeloid cells deriving from studies in murine models is still incomplete.
In fact, previous work induced the deficiency of TNFAIP3 in M-MDC (17) or in microglia (18), but never in the entire myeloid compartment. Here, we evaluated the effects of the TNFAIP3 lack in myeloid lineage, including M-MDC and microglia in a conditional KO murine model. TNFAIP3 cx3cr1-KO mice were vital and showed a macroscopic normal appearance. Notably, we did not observe in the TNFAIP3 cx3cr1-KO mice signs of paws inflammation, as reported in the conditional KO model in which the TNFAIP3 gene was deleted only in M-MDC and not in microglia (17). On the other hands, our results are in agreement with the data obtained from Voet and colleagues in which the deficiency of TNFAIP3 in microglia is not related to alteration in gross anatomy of mice. However, we observed a body weight reduction in TNFAIP3 cx3cr1-KO mice compared to their WT littermates. To date, no results related to the body weight in TNFAIP3 KO mice are still reported. In order to detect the specific role of TNFAIP3 in myeloid cells, we performed cytofluorimetric analysis on the spleen of these conditional KO mice. Notably, we revealed, in agreement with previous studies, a strong decreased number of macrophages, monocytes and DCs in TNFAIP3 cx3cr1-KO mice in comparison to WT. Accordingly, in lymph nodes of TNFAIP3 cx3cr1-KO mice we observed a decreased amount in macrophages in comparison to WT mice. In addition, we also observed an increased number in DCs, B and CD8+ T cells in lymph nodes of TNFAIP3 cx3cr1-KO mice, in agreement with the studies performed by (17).
These observations suggest that TNFAIP3 is involved in the development of myeloid lineage. In addition, these data also suggest that a de-regulated production of M-MDC due to the lack of TNFAIP3 could be related to general altered immune phenotype involving not only myeloid cells but also lymphocytes, such as B cells and NK.
Thanks to the analysis performed on the bone marrow of the conditional TNFAIP3 cx3cr1-KO mice, we highlighted for the first time that the immunological alteration did not occur only during the development of immune cells but it was already presents in the precursor cells.
Related to the CNS parenchyma, we observed, in agreement with the Voet's study, an increased microglia cell number in the corpus callosum of TNFAIP3 cx3cr1-KO .
However, considering the great analysis performed by Voet and colleagues we decide to not deep insight in this aspect that it has been already fully dissected.

Conclusions
In conclusion, the obtained results support the suggested role of TNFAIP3 in myeloid cell. In addition to its role in M-MDC, the defects of TNFAIP3 is confirmed to be