POS0456 AUTOPHAGY-RELATED RISK LOCI IN SLE AND THEIR ROLE IN NEUTROPHILS

Background

Systemic lupus erythematous (SLE) is an autoimmune disease with ~150 established susceptibility risk loci. Genome wide-association (GWA) studies in SLE cases and controls of Korean ancestry identified the SLE risk locus ATG16L2-P2RY2, and rs11235604 as a SLE-associated missense variant (R220W) of Autophagy Related 16 Like 2 (ATG16L2) [1]. PRDM1-ATG5 is also an SLE risk locus in European populations that is implicated in autophagy. Autophagy plays a crucial role in neutrophil extracellular trap (NET) formation, degranulation, and limiting autoantigens in blood. Dysregulated autophagy has been implicated in SLE pathology and poor disease outcomes. The function of ATG16L2 is unknown, but evidence suggests it may function as a negative regulator of autophagosome formation [2].

Objectives

To identify autophagy-related SLE risk variants shared across different ancestry populations and define the role of ATG16L2 in SLE and autophagy.

Methods

SLE case-control GWA scans from European (7568 cases; 1082 controls), African American (4336 cases; 935 controls), Hispanic (3752 cases; 1840 controls), and Korean (1173 cases; 4213 controls) populations were imputed and SNP associations tested. Meta-analysis was performed, then Bayesian statistics were used to define a credible SNP set. Bioinformatic analyses (RegulomeDB, promoter capture Hi-C, eQTLs, etc.) further prioritized SNPs based on predicted functionality. The functional significance of autophagy SLE risk genes, ATG16L1, ATG16L,2 and ATG5, were tested by CRISPR knockout (KO) in PLB-985 cell line. CRISPR-targeted single cell clones were screened for ATG16L1, ATG16L2 or ATG5 deletion using qPCR, NanoPore sequencing, and Western blotting. Changes in autophagy were assessed by Western blotting and confocal microscopy.

Results

Transracial fine-mapping of PRDM1-ATG5 locus identified two SNP associations shared across the credible sets in all populations: rs56886418 (p=1.38x10-5) located in the intron of PRDM1 and rs77791277 (p=1.38x10-5) that tagged a group of SNPs in strong linkage disequilibrium. Cross comparison of the credible SNP sets and bioinformatic analyses of shared SNPs identified rs533733 and rs9373843 as additional likely functional variants. Bioinformatic analyses prioritized rs56886418, an eQTL for ATG5 (p=0.05) and PRDM1 (p=4.75x10-7) in blood cells positioned in a topologically associated domain (TAD) that may interact with ATG5 and PRDM1 promoters in EBV-transformed B cells. SNP rs533733 is an eQTL for ATG5 in neutrophils (p=0.006) and is in a TAD 6.4kb 3’ of PRDM1 that interacts with the ATG5 promoter region where rs9373843 (eQTL of ATG5 in neutrophils (p=0.04)) is positioned. These data suggest that risk SNPs on the PRDM1-ATG5 locus may modulate ATG5 expression and autophagy in specific cell types by modulating the local chromatin regulatory network.

To assess the roles of ATG5, ATG16L1 and ATG16L2 in autophagy, PMA/I-induced hallmarks of autophagy, LC3-I and LC3-II conversion and p62 protein aggregation, were assessed in homozygous and heterozygous ATG5, ATG16L1, or ATG16L2 CRISPR KO PLB-985 cells by Western blotting and confocal microscopy. Loss of ATG5 or ATG16L1 impaired PMA/I-induced autophagosome formation in myeloid-like and differentiated neutrophil-like PLB-985 cells. In contrast, loss of ATG16L2 elevated the conversion of LC3-I to LC3-II and p62 protein aggregation in both cell types, suggesting that ATG16L2 may inhibit autophagy.

Conclusion

Functional characterization of SNPs on the PRDM1-ATG5 and ATG16L2-P2RY2 loci, and the functional characterization of ATG16L2 in myeloid and neutrophil cell lines, provide new insights into the mechanisms that regulate autophagy in health and disease. Ongoing studies will focus on in vitro validation of predicted functional SNPs and will introduce ATG16L2 rs11235604 risk variant in PLB-985 cells to assess its importance in autophagy.

References

[1]Lessard CJ, et al. Arthritis Rheumatol. 2016; 68(5):1197-1209.

[2]Wible DJ, et al. Cell Discov. 2019; 5:42.

Disclosure of Interests

None declared.

OP0110 CELL-TYPE-SPECIFIC TRANSCRIPTOME ARCHITECTURE UNDERLYING THE ESTABLISHMENT AND EXACERBATION OF SYSTEMIC LUPUS ERYTHEMATOSUS

Background

Systemic lupus erythematosus (SLE) is a complex autoimmune disease with unknown etiology involving multiple immune cells and has diverse clinical phenotypes. This heterogeneous nature has hampered a better understanding of SLE pathogenesis and the development of effective therapeutic agents. While recent single-cell RNA sequencing studies of SLE identified several important cell subpopulations, they were limited by sparse expression information at single-cell level and small sample sizes.

Objectives

This study aimed to elucidate the dysregulated gene expression pattern linked to multiple clinical statuses of SLE with a fine cellular resolution and higher sensitivity. We also attempted to resolve a complex interaction between risk variants and the transcriptome dysregulation seen in SLE patients.

Methods

We conducted a large-scale bulk transcriptome study of 6,386 RNA-sequencing data including 27 purified immune cell types in peripheral blood from 136 SLE and 89 healthy donors in the Immune Cell Gene Expression Atlas from the University of Tokyo (ImmuNexUT) cohort1. At enrollment, SLE patients had diverse clinical manifestations (disease activity, organ involvement and treatment profiles) and 22 patients were re-evaluated after belimumab treatment.

Results

We first profiled two distinct cell-type-specific transcriptomic signatures: disease-state and disease-activity signatures, reflecting disease establishment and exacerbation, respectively.

After confirming the high replicability of both signatures in independent cohorts, we identified candidates of biological processes unique to each signature: e.g., upregulated E2F transcriptional activity in Th1, CD8+ memory T-lineage and NK cells, and dynamic increase of IL21 and CXCL13 in Th1 cells in an active phase of SLE. Pathway analysis highlighted the importance of immunometabolic process for SLE (e.g., oxidative phosphorylation) in cell-type-specific resolution.

Moreover, we demonstrated cell-type-specific contributions to diverse organ involvement, e.g., Th1 for mucocutaneous, monocyte-lineage cells for musculoskeletal, neutrophil-lineage cells for renal activity, respectively.

We also observed the strong associations of disease-activity signatures with treatment effect: (i) belimumab suppressed activity signatures from B-lineage cells, especially in good responders and (ii) mycophenolate mofetil substantially suppressed activity signatures from plasmablast, Th1, and central memory CD8 cells.

However, through stratified LD score regression using large-scale SLE-GWASs, we revealed that disease-activity signatures were less enriched around SLE risk variants than disease-state signatures. Consistent with this result, the directions of SLE risk alleles’ expression quantitative trait locus (eQTL) effects were significantly concordant with the directions of disease-state signatures, but not with those of activity signatures. These findings suggested that the current genetic case-control studies may not well capture clinically vital biology linked to drug target discovery for SLE. Meanwhile, we also detected some examples of activity signatures that might contribute to the disease risk by modulating risk allele’s eQTL effects.

Figure 1.

Conclusion

We identified comprehensive gene signatures reflecting the establishment and exacerbation of SLE, which provide essential foundations for future genomic, genetic, and clinical studies.

References

[1]Ota, M. et al. Dynamic landscape of immune cell-specific gene regulation in immune-mediated diseases. Cell 2021;184:3006-21.e17.

Acknowledgements

This study was supported by Chugai Pharmaceutical Co., Ltd., Tokyo, Japan; the Ministry of Education, Culture, Sports; and the Japan Agency for Medical Research and Development (AMED) (JP21tm0424221 and JP21zf0127004).

Disclosure of Interests

Masahiro Nakano: None declared, Mineto Ota Grant/research support from: Mineto Ota belongs to the Social Cooperation Program, Department of functional genomics and immunological diseases, supported by Chugai Pharmaceutical., Yusuke Takeshima Grant/research support from: Yusuke Takeshima belonged to the Social Cooperation Program, Department of functional genomics and immunological diseases, supported by Chugai Pharmaceutical., Yukiko Iwasaki: None declared, Hiroaki Hatano: None declared, Yasuo Nagafuchi Grant/research support from: Yasuo Nagafuchi belongs to the Social Cooperation Program, Department of functional genomics and immunological diseases, supported by Chugai Pharmaceutical., Kwangwoo Kim: None declared, So-Young Bang: None declared, Hye Soon Lee: None declared, Hirofumi Shoda: None declared, Xuejun Zhang: None declared, Sang-Cheol Bae: None declared, Chikashi Terao: None declared, Kazuhiko Yamamoto: None declared, Tomohisa Okamura Grant/research support from: Tomohisa Okamura belongs to the Social Cooperation Program, Department of functional genomics and immunological diseases, supported by Chugai Pharmaceutical., Kazuyoshi Ishigaki: None declared, Keishi Fujio Speakers bureau: Keishi Fujio receives speaker fees from Chugai Pharmaceutical., Consultant of: Keishi Fujio receives consulting honoraria from Chugai Pharmaceutical., Grant/research support from: Keishi Fujio receives research support from Chugai Pharmaceutical.

Development of a high-resolution typing method for SLA-3, swine MHC class I antigen 3

We developed a high-resolution and comprehensive typing method for swine leukocyte antigen 3 (SLA-3), an MHC class I gene, employing locus-specific genomic PCR followed by subsequent direct sequencing. A total of 292 individuals from nine pure, one cross-breed and six cell lines were successfully typed. A total of 21 SLA-3 alleles were identified, of which four were found to be novel alleles. However, the allelic diversity of SLA-3 was lower than that of previously reported class I genes, SLA-1 and -2. More SLA-3 alleles were observed in the Landrace and Yorkshire breeds than the other breeds. SLA-3*04:01 was identified in seven out of nine breeds and was the most widely distributed allele across all breeds. Therefore, the typing method reported in this study completes our efforts to develop high-resolution typing methods for major SLA molecules, facilitating the combined analysis of major SLA genes from field samples, which is important to understand the relationship between the adaptive immune responses against pathogens and the immunogenetic makeup of an individual.

An Open Medical Platform to Share Source Code and Various Pre-Trained Weights for Models to Use in Deep Learning Research

Deep learning-based applications have great potential to enhance the quality of medical services. The power of deep learning depends on open databases and innovation. Radiologists can act as important mediators between deep learning and medicine by simultaneously playing pioneering and gatekeeping roles. The application of deep learning technology in medicine is sometimes restricted by ethical or legal issues, including patient privacy and confidentiality, data ownership, and limitations in patient agreement. In this paper, we present an open platform, MI2RLNet, for sharing source code and various pre-trained weights for models to use in downstream tasks, including education, application, and transfer learning, to encourage deep learning research in radiology. In addition, we describe how to use this open platform in the GitHub environment. Our source code and models may contribute to further deep learning research in radiology, which may facilitate applications in medicine and healthcare, especially in medical imaging, in the near future. All code is available at https://github.com/mi2rl/MI2RLNet.

A population-based study for difference in contribution of type 2 diabetes mellitus to cardiovascular outcomes in adult cancer survivors and general population without history of cancer

Background/Introduction

Type 2 diabetes mellitus (T2DM) is a well-established risk factor for cardiovascular disease (CVD). However, evidence is scarce on the difference in burden of T2DM as a cardiovascular risk factor in adult cancer survivors compared to the general population without history of cancer. This evidence gap needs to be addressed as increased risk for CVD in adult cancer survivors compared to the general population has been consistently reported from observational studies.

Purpose

To investigate the association of T2DM with CVD in adult cancer survivors and general population without history of cancer and compare the magnitude of associations for each population using propensity score-matched analyses

Methods

For this population-based cohort study, we used the National Health Insurance Service-National Sample Cohort (NHIS-NSC) established in the Republic of Korea (2002–2015) to identity adult cancer survivors who were aged 18 years or older and survived more than 12 months after the first-ever cancer diagnosis and general population without history of cancer matched for age and sex in a 1:1 ratio. In both cohorts, those with history of CVD prior to the index date were excluded. We used Cox proportional hazards model to compute hazard ratios (HR) and 95% confidence intervals (95% CI) for incident CVD in those with T2DM compared to those without T2DM in both adult cancer survivors and the general population after adjusting for shared confounding factors (income status, cigarette smoking, alcohol consumption, physical inactivity, body mass index, blood pressure, lipid abnormalities, family history of CVD, and aspirin use). Furthermore, we used Q statistic to compare the magnitude of association between T2DM and CVD in this matched cohort.

Results

The overall age and sex matched cohort in the NHIS-NSC comprised of 5,163 adult cancer survivors and the equivalent number of participants categorized as general population without history of cancer. The adjusted HR for CVD among adult cancer survivors with T2DM, as compared to without T2DM was 2.27 (95% CI: 1.25–4.11). Similarly, T2DM was associated with an increased risk of CVD in the general population without cancer (HR=1.88; 95% CI: 1.03–3.42). However, no strong statistical evidence was found for difference in contribution of T2DM to CVD risk between adult cancer survivors and general population without history of cancer (Pheterogeneity = 0.662).

Conclusions

In this population-based cohort study, T2DM was associated with an increased risk for CVD in both adult cancer survivors and those without history of cancer. Difference in magnitude of T2DM-CVD associations between adult cancer survivors and the general population without history of cancer was not substantial from a statistical standpoint. However, adult cancer survivor with T2DM may still need additional clinical attention for CVD prevention compared to those without history of cancer.

Funding Acknowledgement

Type of funding sources: Public grant(s) – National budget only. Main funding source(s): National Cancer Center (Republic of Korea)

Treatment potential of LPCN 1144 on liver health and metabolic regulation in a non-genomic, high fat diet induced NASH rabbit model

PURPOSE

Low free testosterone (T) level in men is independently associated with presence and severity of Non-Alcoholic Steatohepatitis (NASH). The histological and molecular effects of oral testosterone prodrug LPCN 1144 treatment on hepatic fibrosis and NASH features are unknown. A metabolic syndrome-induced NASH model in rabbits consuming high fat diet (HFD) has been previously used to assess treatment effects of injectable T on hepatic fibrosis and NASH features. Here we present results on LPCN 1144 in this HFD-induced, NASH preclinical model.

METHODS

Male rabbits were randomly assigned to five groups: regular diet (RD), HFD, HFD + 1144 vehicle (HFD + Veh), HFD + 1144 (1144), and HFD + 1144 + α-tocopherol (1144 + ALPHA). Rabbits were sacrificed after 12 weeks for liver histological, biochemical and genetic analyses. Histological scores were obtained through Giemsa (inflammation), Masson's trichrome (steatosis and ballooning), and Picrosirius Red (fibrosis) staining.

RESULTS

Compared to RD, HFD and HFD + Veh significantly worsened NASH features and hepatic fibrosis. Considering HFD and HFD + Veh arms, histological and biomarker features were not significantly different. Both 1144 and 1144 + ALPHA arms improved mean histological scores of NASH as compared to HFD arm. Importantly, percentage of fibrosis was improved in both 1144 (p < 0.05) and 1144 + ALPHA (p = 0.05) treatment arms vs. HFD. Both treatment arms also reduced HFD-induced inflammation and fibrosis mRNA markers. Furthermore, 1144 treatments significantly improved HFD-induced metabolic dysfunctions.

CONCLUSIONS

Histological and biomarker analyses demonstrate that LPCN 1144 improved HFD-induced hepatic fibrosis and NASH biochemical, biomolecular and histochemical features. These preclinical findings support a therapeutic potential of LPCN 1144 in the treatment of NASH and of hepatic fibrosis.