Reply

ogenesis of SLE has not been fully elucidated. Therefore, searching for the mechanisms involved in the development of SLE and targeting potential mechanisms for treatment remain topics of investigation. Absent in Melanoma 2 (AIM2) is a member of the pyrin and hematopoietic interferon (IFN)–inducible nuclear domain proteins. This protein has wide-ranging, propyroptotic properties. Regarding innate immunity, AIM2 serves as a cytoplasmic double-stranded DNA (dsDNA) sensor, regulating the initiation of innate immune responses (2). The recognition of dsDNA by AIM2 results in the assembly of a large multiprotein oligomeric complex known as the inflammasome, which regulates interleukin-1β (IL-1β) and IL-18 generation and induces cell death. In the cytosol, sensing of dsDNA by AIM2 is important for protection against invading pathogens, such as bacteria, viruses, and fungi. Conversely, the response of AIM2 to dsDNA released by damaged host cells may lead to the production of different cytokines that are involved in the pathogenesis of sterile inflammatory diseases, such as skin and kidney diseases. AIM2 contributes to lung tumorigenesis through the inflammasomedependent release of IL-1β and the regulation of mitochondrial dynamics. The AIM2 inflammasome becomes activated in the presence of atherosclerotic plaque, abdominal wall aortic aneurysm, and injured myocardium. To date, the findings of studies on AIM2 in the setting of lupus have been inconsistent. In a recent study by Dr. Lu and colleagues (1), it was found that Aim2 gene–deficient mice developed lupus, demonstrated by high histologic scores and high serum levels of dsDNA, myeloperoxidase, proteinase 3, albumin, and urea nitrogen, after pristane treatment. Increased infiltration of dendritic cells, macrophages, neutrophils, B cells, and T cells as well as high type I IFN signatures were also found in the kidneys of Aim2 gene–deficient mice treated with pristane. Aim2 deficiency resulted in elevated expression of type I IFN–induced genes in the kidneys, suggesting that AIM2 inhibits the development of lupus (1). Similarly, Panchanathan et al predicted that Aim2 deficiency contributed to increased susceptibility to lupus (2). In their study, Aim2 gene–deficient mice had high expression of p202 protein, which correlated with increased expression of IFNβ, STAT1, and IFN-inducible genes (2). In contrast, Huang et al evaluated Aim2 expression in lupus nephritis patients and found that Aim2 was highly expressed in the glomerular cells of patients with lupus nephritis class II (3). Yang et al found that in SLE patients, Aim2 expression was increased in germinal center B cells and plasma cells from peripheral blood, and in tonsil memory and skin lesions (4). Proportions of CD19+ cells were down-regulated in the lymph nodes and spleens of mice with lupus and Aim2 gene-deficient B cells. Aim2 deficiency in B cells attenuated the development of lupus, as shown by reduced glomerulonephritis, lower proportions of germinal center B cells, T follicular helper cells, and plasma cells, up-regulated secretion of B lymphocyte–induced maturation protein 1 (BLIMP-1), and reduced expression of Bcl-6 (4). Interestingly, knockout of the genes for BLIMP-1 and Bcl-6 did not affect Aim2 expression. These findings suggest that AIM2 promotes lupus pathogenesis by regulating BLIMP-1/Bcl-6 axis–mediated B cell differentiation. Additionally, Zhang et al found that Aim2 expression was related to severity of disease in SLE patients and mice with lupus (5). Aim2 expression was elevated in apoptotic DNA–induced macrophages and was related to macrophage activation. Knockout of Aim2 blunted apoptotic DNA–induced macrophage activation and suppressed lupus development by impeding macrophage activation and reducing inflammatory responses (5). Based on the findings reviewed above, it is uncertain whether AIM2 is able to inhibit lupus development or promote SLE pathogenesis. Therefore, more studies with Aim2 gene deficiency or conditional knockout are needed to further clarify the role of AIM2 in lupus. Supported by grants from the National Natural Science Foundation of China (81701606). Author disclosures are available at https:// onlinelibrary.wiley.com/action/downloadSupplement?doi=10.1002%2Fart. 42084&file=art42084-sup-0001-Disclosureform.pdf.

manifestations, such as lupus nephritis (1). To date, a clear pathogenesis of SLE has not been fully elucidated. Therefore, searching for the mechanisms involved in the development of SLE and targeting potential mechanisms for treatment remain topics of investigation.
Absent in Melanoma 2 (AIM2) is a member of the pyrin and hematopoietic interferon (IFN)-inducible nuclear domain proteins. This protein has wide-ranging, propyroptotic properties. Regarding innate immunity, AIM2 serves as a cytoplasmic double-stranded DNA (dsDNA) sensor, regulating the initiation of innate immune responses (2). The recognition of dsDNA by AIM2 results in the assembly of a large multiprotein oligomeric complex known as the inflammasome, which regulates interleukin-1β (IL-1β) and IL-18 generation and induces cell death. In the cytosol, sensing of dsDNA by AIM2 is important for protection against invading pathogens, such as bacteria, viruses, and fungi. Conversely, the response of AIM2 to dsDNA released by damaged host cells may lead to the production of different cytokines that are involved in the pathogenesis of sterile inflammatory diseases, such as skin and kidney diseases. AIM2 contributes to lung tumorigenesis through the inflammasomedependent release of IL-1β and the regulation of mitochondrial dynamics. The AIM2 inflammasome becomes activated in the presence of atherosclerotic plaque, abdominal wall aortic aneurysm, and injured myocardium.
To date, the findings of studies on AIM2 in the setting of lupus have been inconsistent. In a recent study by Dr. Lu and colleagues (1), it was found that Aim2 gene-deficient mice developed lupus, demonstrated by high histologic scores and high serum levels of dsDNA, myeloperoxidase, proteinase 3, albumin, and urea nitrogen, after pristane treatment. Increased infiltration of dendritic cells, macrophages, neutrophils, B cells, and T cells as well as high type I IFN signatures were also found in the kidneys of Aim2 gene-deficient mice treated with pristane. Aim2 deficiency resulted in elevated expression of type I IFN-induced genes in the kidneys, suggesting that AIM2 inhibits the development of lupus (1).
Similarly, Panchanathan et al predicted that Aim2 deficiency contributed to increased susceptibility to lupus (2). In their study, Aim2 gene-deficient mice had high expression of p202 protein, which correlated with increased expression of IFNβ, STAT1, and IFN-inducible genes (2). In contrast, Huang et al evaluated Aim2 expression in lupus nephritis patients and found that Aim2 was highly expressed in the glomerular cells of patients with lupus nephritis class II (3).
Yang et al found that in SLE patients, Aim2 expression was increased in germinal center B cells and plasma cells from peripheral blood, and in tonsil memory and skin lesions (4). Proportions of CD19+ cells were down-regulated in the lymph nodes and spleens of mice with lupus and Aim2 gene-deficient B cells. Aim2 deficiency in B cells attenuated the development of lupus, as shown by reduced glomerulonephritis, lower proportions of germinal center B cells, T follicular helper cells, and plasma cells, up-regulated secretion of B lymphocyte-induced maturation protein 1 (BLIMP-1), and reduced expression of Bcl-6 (4). Interestingly, knockout of the genes for BLIMP-1 and Bcl-6 did not affect Aim2 expression. These findings suggest that AIM2 promotes lupus pathogenesis by regulating BLIMP-1/Bcl-6 axis-mediated B cell differentiation.
Additionally, Zhang et al found that Aim2 expression was related to severity of disease in SLE patients and mice with lupus (5). Aim2 expression was elevated in apoptotic DNA-induced macrophages and was related to macrophage activation. Knockout of Aim2 blunted apoptotic DNA-induced macrophage activation and suppressed lupus development by impeding macrophage activation and reducing inflammatory responses (5).
Based on the findings reviewed above, it is uncertain whether AIM2 is able to inhibit lupus development or promote SLE pathogenesis. Therefore, more studies with Aim2 gene deficiency or conditional knockout are needed to further clarify the role of AIM2 in lupus.

Reply
To the Editor: We thank Drs. Xu and Huang for their interest in our recently published article. As we mentioned in our report, Aim2 deficiency leads to elevated type I IFN generation and SLE-like nephritis, which has been implied in previous publications (1,2). The finding of elevated type I IFN signatures in both resting and pristanetreated Aim2 −/− mice provides direct evidence that AIM2 inhibits type I IFN production, not only upon DNA transfection in cultured cells (3,4), but also in the SLE mouse model in vivo. Of note, Aim2 deficiency results in not only elevated type I IFN production, but also an enhanced response to IFNβ stimulation (2). Importantly, our work revealed the mechanism through which AIM2 represses type I IFN: sumoylation-mediated inhibition of type I IFN transcription requires ubiquitin-conjugating enzyme 2i (Ube2i) (5), and Aim2 partners with Ube2i for its optimal function. In the absence of AIM2, the activity of Ube2i is compromised, which leads to decreased cellular sumoylation; thus, a stronger type I IFN signal appears and more severe lupus nephritis develops.
With regard to another autoimmune disease model (multiple sclerosis), 2 recent studies demonstrated that Aim2 deficiency in Treg cells or microglia exacerbates the development of experimental autoimmune encephalomyelitis (EAE) (6,7). In both Treg cells and microglia, Aim2 deficiency resulted in stronger expression of type I IFN-stimulated genes (ISGs) including Irf7 and Ddx58, as well as Stat1 phosphorylation (6,7). Therefore, although the Aim2 −/− mice we used were on a mixed (129 × B6) genetic background, Aim2 −/− mice on the C57BL/6 background also exhibited elevated type I IFN signatures and enhanced autoimmune disease activity, demonstrating that, at least in macrophages, Treg cells, and microglia, AIM2 inhibits type I IFN signaling (6,7). Conditional knockout of Aim2 in macrophages has not been achieved, but Aim2 is mainly expressed in innate immune cells. Therefore, the findings with universal Aim2 deletion should be largely attributed to innate immune cells including macrophages.
Additionally, in our study Aim2 −/− Rag1 −/− mice clearly exhibited improvement in disease activity, implying that adaptive immune cells are involved in the nephritis of Aim2 −/− mice, despite deficiency of Aim2 in the B cells and T cells of these mice. In light of findings reported by Yang et al (8), it would be interesting to compare Aim2 deficiency in different types of cells among animals in the same facility with SLE development upon pristane injection, because lupus induction is a long process and microbiota may affect disease development.
With regard to the elevated Aim2 expression in SLE patients and apoptotic DNA-treated BALB/c mice (8,9), it should be noted that the Aim2 gene itself is an ISG whose expression can be induced by IFN. Moreover, as demonstrated in the study by Yang et al, IL-10 can also induce the expression of Aim2 (8). Therefore, the elevated expression of Aim2 is likely a result of SLE development rather than a driving factor of the disease. It would be interesting to generate a line of Aim2-overexpressing mice to investigate whether enforced expression of this gene leads to enhancement or alleviation of autoimmune diseases, including SLE and EAE. We read with great interest the report by Dr. Prisco et al (1), describing their study which demonstrated much greater odds of both restrictive and obstructive pulmonary patterns among rheumatoid arthritis (RA) cases, compared with controls, in an analysis of data from the UK Biobank. However, some concerns are worth addressing.
In discussions about pulmonary function abnormalities, a crucial variable is the presence of chronic respiratory disease.