Causal Effects of AL Amyloidosis on Autoimmune Diseases and the Role of Inflammatory Cytokines: Insights from Mendelian Randomization

doi:10.21203/rs.3.rs-7503974/v1

Download PDF

Research Article

Causal Effects of AL Amyloidosis on Autoimmune Diseases and the Role of Inflammatory Cytokines: Insights from Mendelian Randomization

https://doi.org/10.21203/rs.3.rs-7503974/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

This study aims to investigate the causal effect of amyloid light-chain (AL) amyloidosis on autoimmune diseases (ADs), with a focus on identifying inflammatory cytokines as potential mediators from a genetic perspective.

Methods

Instrumental variables strongly associated with AL amyloidosis were identified from the largest available genetic cohort. Two-sample Mendelian randomization (MR) analyses were conducted to assess causal associations, followed by rigorous sensitivity analyses and validation in independent cohorts. Protein-protein interaction (PPI) analyses were applied to explore cytokine networks mediating the observed associations.

Results

AL amyloidosis was associated with an increased risk of celiac disease and overall AD risk but exhibited a protective effect against inflammatory bowel disease (IBD) and ulcerative colitis (UC). Protein quantitative trait locus (pQTL) analysis quantified inflammatory cytokine changes across these conditions, and the results were visualized.

Conclusions

Genetically predicted AL amyloidosis increases the risk of celiac disease and overall autoimmune disease susceptibility while reducing the risk of IBD and UC. These associations may involve inflammatory cytokines such as IL-6, CCL3, and BAFF, suggesting a potential mechanistic pathway worthy of further investigation.

AL amyloidosis

autoimmune diseases

Mendelian randomization study

circulating inflammatory cytokines

protein quantitative trait loci

celiac disease

inflammatory bowel disease

Immunoglobulin light chain (AL) amyloidosis is a plasma cell dyscrasia in which aberrant proliferation leads to the accumulation of monoclonal light chains and subsequent multiorgan damage[1, 2]. Beyond structural tissue injury, the potential for these circulating free light chains to mediate pathological effects through antigen-binding activity should not be neglected[3]. Notably, serum-free light chain levels often increase years before the onset of clinical symptoms and even before the development of AL amyloidosis[4]. Recent clinical observations worldwide have reported that AL amyloidosis may coexist with several autoimmune diseases (ADs), with AL amyloidosis often preceding the onset of these conditions[5–7]. Autoimmune diseases, driven by an overactive immune response targeting the body’s own components, encompass more than 80 chronic inflammatory and destructive conditions, such as rheumatoid arthritis, Sjögren’s syndrome, inflammatory bowel disease (IBD), and type 1 diabetes mellitus[8, 9]. These diseases share common pathogenic mechanisms that largely involve proinflammatory factors, B cells, plasma cells, and autoantigen-targeting antibodies[8]. The interplay between immature plasma cells and dysfunctional immunoglobulins in AL amyloidosis, overactive proinflammatory factors, and a disturbed immune system in ADs remains unclear. Despite these clinical observations, the causal effects of AL amyloidosis on the risk of developing ADs are not yet well understood.

Given the complexity of these conditions, it is crucial to explore the genetic underpinnings linking AL amyloidosis and ADs. Mendelian randomization (MR) provides a powerful framework for evaluating causal associations by leveraging genetic variants as instrumental variables (IVs). This approach enables the assessment of whether genetic susceptibility to AL amyloidosis contributes to the risk of autoimmune diseases, while minimizing confounding effects and reverse causality that often limit traditional observational study designs[10]. Owing to the increasing availability of genome-wide association study (GWAS) data, the efficiency of screening high-throughput data and the power of statistical evaluation have significantly improved[11]. Furthermore, protein-protein interaction (PPI) analysis offers additional insight into cytokine networks identified through text-mining or experimental data and simultaneously supports the MR results, thereby enhancing the robustness of our findings[12, 13].

Prior studies have mainly described the co-occurrence of autoimmune diseases in patients with AL amyloidosis, while only a few have examined the odds ratios (ORs) for developing these conditions within this population[6]. Furthermore, no published studies have evaluated the causal relationship between AL amyloidosis and ADs or examined the mediating role of inflammatory factors. This study applied a two-sample MR approach to evaluate the causal impact of AL amyloidosis on multiple ADs, with special attention to the potential mediating role of inflammatory cytokines. Additionally, PPI analysis was conducted to explore the cytokine networks potentially involved in this process. Understanding how these cytokines influence the development of ADs in the context of AL amyloidosis could not only elucidate the connection between the two diseases but also offer novel perspectives into potential therapeutic targets.

Study design

This study aimed to determine whether patients with AL amyloidosis are more susceptible to ADs and to identify potential risk factors. The investigation was conducted in three stages, as illustrated in Fig. 1. First, preliminary MR analyses were applied to examine the potential causal influence of AL amyloidosis on 21 ADs. Next, the stability of these associations was evaluated using multiple statistical approaches, including tests for heterogeneity, pleiotropy, and leave-one-out procedures, to minimize bias and assess sensitivity. To corroborate the credibility of these findings, the observed effects of AL amyloidosis on related autoimmune diseases were further examined in independent validation cohorts. Lastly, MR results were complemented with protein-protein interaction analyses to explore mechanistic links, focusing on inflammatory mediators such as chemokines, interleukins, and other cytokines.

Data sources

To minimize race-related confounding factors, this study focused on a population predominantly of European ancestry. The exposure and outcome data were sourced from two independent cohorts to prevent overlap-related bias[14].

Genetic instruments indicating the diagnosis of AL amyloidosis were extracted from summary statistics of the largest published dataset, comprising 1,351 AL amyloidosis patients (595 from Germany, 474 from the UK, and 282 from Italy) and 7,589 controls of European ancestry[15]. The diagnosis of amyloidosis has traditionally relied on histological confirmation as described[16]. A total of 21 ADs were defined as outcomes, with summary-level data obtained from publicly available GWAS datasets in the IEU Open GWAS project[17] or GWAS Catalog[18]. The GWAS IDs and detailed descriptions of these datasets are provided in Table 1. Independent cohorts were used for secondary validation: summary statistics for IBD were drawn from a GWAS meta-analysis involving 12,824 cases and 21,770 controls[19]; UC data were derived from a GWAS meta-analysis compiled by the IBD Genetics Consortium, comprising 6,687 cases and 19,718 controls across six research centers[20]; and celiac disease replication data were obtained from a large prospective cohort in the UK Biobank, including 1,855 patients and 334,783 white British controls[21].

Plasma protein quantitative trait loci (pQTLs) offer insights into the genetic variation underlying interindividual differences in plasma protein levels, facilitating precision medicine and clinical decision-making[22, 23]. pQTLs related to inflammatory factors were retrieved from a proteome-level genetic mapping study, which linked genetic factors to circulating protein levels in a cohort of 3,301 healthy participants from 25 centers across England[23].

Clinical trial number: not applicable.

Table 1
Accession numbers of ADs from public databases and detailed information
Autoimmune diseases	Online database	NCase	Sample size	Population	SNPs	DOI
Multiple sclerosis	Open GWAS	47,429	115,803	European	6,304,359	10.1126/science.aav7188
Type 1 diabetes	Open GWAS	18,942	520,580	European	59,999,551	10.1038/s41586-021-03552-w
Autoimmune hepatitis	Open GWAS	821	485,234	European, East Asian	24,198,482	10.1038/s41588-021-00931-x
Primary biliary cirrhosis	GWAS Catalog	8,021	24,510	European	5,054,572	10.1016/j.jhep.2021.04.055
Graves' disease	Open GWAS	1,678	458,620	European, East Asian	24,189,816	10.1038/s41588-021-00931-x
Hypothyroidism	Open GWAS	22,997	198,472	European	16,380,353	10.1038/s41586-022-05473-8
Inflammatory bowel disease	Open GWAS	25,042	59,957	European	9,619,016	10.1038/ng.3760
Crohn's disease	Open GWAS	12,194	40,266	European	9,457,998	10.1038/ng.3760
Ulcerative colitis	Open GWAS	12,366	45,975	European	9,474,559	10.1038/ng.3760
Coeliac disease	Open GWAS	1,973	16,380,438	European	16,380,438	10.1038/s41586-022-05473-8
Addison's disease	Open GWAS	1,223	5,320	European	7,068,382	10.1038/s41467-021-21015-8
Sjogren's syndrome	Open GWAS	NA	407,746	European	11,039,117	10.1038/s41588-021-00870-7
Systemic lupus erythematosus	Open GWAS	5,201	14,267	European	7,071,163	10.1038/ng.3434
Rheumatoid arthritis	Open GWAS	8,255	417,256	European, East Asian	24,175,266	10.1038/s41588-021-00931-x
Ankylosing spondylitis	Open GWAS	9,069	22,647	European	99,962	10.1038/ng.2667
Plaque psoriasis	Open GWAS	1,684	399,883	European	28,000,000	10.1038/s41588-018-0184-y
Psoriatic arthritis	Open GWAS	5,065	26,351	European	8,558,403	10.1002/art.42154
IGA Glomerulonephritis	Open GWAS	15,587	477,784	European, East Asian	24,182,646	10.1038/s41588-021-00931-x
Idiopathic Thrombocytopenic Purpura	GWAS Catalog	843	668,150	European, East Asian	25,845,163	10.1038/s41588-021-00931-x
autonomic nervous system disease	GWAS Catalog	266	395,475	European	28,000,000	10.1038/s41588-018-0184-y
Autoimmune diseases ^a	Open GWAS	42,202	218,792	European	16,380,466	10.1038/s41586-022-05473-8

^a Autoimmune diseases encompass a variety of conditions, including 45 subtypes. These subtypes include rheumatoid arthritis, relapsing polychondritis, systemic lupus erythematosus, Sjogren syndrome, systemic scleroderma, dermatomyositis, inflammatory bowel disease, Churg–Strauss syndrome, microscopic colitis, Wegener granulomatosis, acute disseminated encephalomyelitis, type I diabetes mellitus, multiple sclerosis, hypothyroidism, rheumatic fever, demyelinating diseases of the central nervous system, autoimmune hemolytic anemia, drug-induced autoimmune hemolytic anemia, autoimmune hepatitis, autoimmune polyglandular syndrome, autoimmune thyroiditis, autoimmune hyperthyroidism, narcolepsy and cataplexy, myasthenia gravis, psoriasis, systemic sclerosis, Behçet disease, other demyelinating diseases of the central nervous system, disorders of the myoneural junction and muscle classified elsewhere, celiac disease, primary biliary cholangitis, Guillain–Barré syndrome, vitiligo, alopecia areata, idiopathic thrombocytopenic purpura, adrenocortical insufficiency, hypersensitivity angiitis, IgA nephropathy, vitamin B12 deficiency anemia, anterior iridocyclitis, Graves ophthalmopathy, pemphigoid, dermatitis herpetiformis, and mixed connective tissue disease.

Selection of instrumental variables

Using single nucleotide polymorphisms (SNPs) as instrumental variables (IVs) inherently mitigates confounding bias that often affects traditional observational studies. Consequently, the validity of the MR approach depends heavily on the selection of appropriate IVs. To ensure validity, all IVs used in this study were required to meet 3 fundamental assumptions: relevance, independence, and exclusion restriction. Candidate SNPs were screened according to strict criteria, with those associated with AL amyloidosis selected at a genome-wide significance threshold of P < 5×10^− 8. For inflammatory cytokines, a more lenient threshold of 5×10^− 6 was used to include a larger pool of IVs. The strength of each genetic instrument was gauged using the F statistic, with values above 10 indicating sufficient instrument robustness for reliable MR estimation. To mitigate confounding effects from linkage disequilibrium (LD), SNPs were clustered within a ± 10,000 kb range, applying an LD threshold of 𝑟² < 0.001, on the basis of the 1000 Genomes European reference panel. Only SNPs with a minor allele frequency (MAF) > 0.01 were included. For exposure-associated SNPs not present in the outcome dataset, proxy SNPs (in high linkage disequilibrium, 𝑟² >0.8) were identified and used as substitutes.

Two-sample MR analysis

We implemented five MR methods based on the statistical properties of the exposure and outcome variables: inverse variance weighted (IVW), MR-Egger, weighted median, simple mode, and weighted mode. To evaluate horizontal pleiotropy, we employed the MR-Egger intercept test and the MR-PRESSO global test. Heterogeneity IVs were subsequently assessed via Cochran’s Q test. Finally, the robustness of the causal inference was further verified through leave-one-out sensitivity analysis, which confirmed that no single SNP was solely responsible for the observed effects.

To determine the appropriate analytical approach, the presence of heterogeneity was evaluated using Cochran’s Q statistic. The choice of MR estimator depended on the results of pleiotropy and heterogeneity assessments. In the absence of significant directional pleiotropy (pleiotropy test P > 0.05), the IVW method was chosen as the primary estimator due to its superior statistical power under the assumption of balanced pleiotropy. Conversely, if pleiotropy was detected (P < 0.05), MR Egger was employed, as it accounts for directional pleiotropy via an intercept term[24, 25].

If heterogeneity was not significant (P ≥ 0.05), IVW was retained as the primary method. For significant heterogeneity (P < 0.05), the IV selection threshold was tightened to P < 5×10^− 8, reducing the number of IVs to a more stringent subset to ensure validity. If the number of instruments remained sufficient after this filtering, MR-PRESSO was implemented to identify and remove outliers. The outlier-corrected IVW estimate was then used once outliers were detected. In cases where heterogeneity persisted but no outliers were identified, the weighted median method was employed, as it can provide consistent estimates even when a portion of the instruments are invalid[26]. When the number of SNPs was limited, MR-PRESSO was applied directly to the original instrument set; subsequent analyses used either the outlier-corrected IVW result if outliers were present, or the weighted median method if none were detected. For additional robustness, simple mode and weighted mode estimators were also applied as supplementary methods.

All IVW results presented in this article were corrected via MR-PRESSO in cases of heterogeneity. A result was deemed statistically significant (P < 0.05) based on the primary method applied: fixed-effect IVW (no pleiotropy/heterogeneity), MR-Egger (pleiotropy present), or weighted median (heterogeneity present).

PPI network

Inflammatory cytokines significantly associated (P < 0.05) with each AD, as determined by relevant statistical methods, were screened and used to construct a PPI network through the Search Tool for the Retrieval of Interacting Genes (STRING, version 12.0; http://string-db.org)[12]. The PPI network was applied with a minimum interaction score threshold set to 0.4 (medium confidence).

Statistical analysis

All analyses were carried out via R (v4.2.1) statistical software. MR analysis was performed via the R-based package “TwoSampleMR”. The “MR_PRESSO” package (https://github.com/rondolab/MR-PRESSO) was used for multiplicity tests, and bootstrapping was set to replicate 3000 times.

Causal effect of AL amyloidosis on 21 ADs: primary MR analyses, robustness analyses and replication validation

From the initial set of 10 AL amyloidosis-associated SNPs (P < 5×10^− 8), 8 were retained as instrumental variables (IVs) after clumping for linkage disequilibrium (Supplementary Table 1). These 8 IVs were used to assess the causal relationships between AL amyloidosis and 21 autoimmune disease (AD) datasets, covering common AD types. Among these datasets, 20 represented specific AD subtypes, whereas the "universal ADs" dataset from FinnGen comprised a collection of 45 AD subtypes, including rheumatoid arthritis, pemphigoid, and others (Table 1).

The IVW method identified AL amyloidosis as a statistically significant risk factor for universal ADs (OR: 1.04, 95% CI: 1.00–1.07, P < 0.05) and celiac disease (OR: 1.13, 95% CI: 1.00–1.27, P < 0.05) (Fig. 2, Supplementary Table 2). In contrast, AL amyloidosis had a protective effect on IBD (OR: 0.90, 95% CI: 0.83–0.96, P < 0.05) and UC (OR: 0.87, 95% CI: 0.81–0.93, P < 0.05), as tested by IVW (Fig. 2, Supplementary Table 2). Scatter plots depict the effect of each SNP on AL amyloidosis and these four ADs (Fig. 3A).

Robustness analyses supported the validity of these findings. Cochran’s Q tests, MR‒Egger regression intercepts, and MR-PRESSO analyses revealed no evidence of significant heterogeneity or directional pleiotropy in the associations between AL amyloidosis and the four ADs (Supplementary Table 3). Leave-one-out analyses confirmed that no single SNP disproportionately influenced the causal estimates (Fig. 3B).

Replication validations further supported the causal effects of AL amyloidosis on celiac disease, IBD, and UC. These validations were performed via three independent datasets curated by external consortia, with the results consistently demonstrating the same direction of effect. As detailed in Supplementary Table 4, the IVW analysis revealed that the OR for AL amyloidosis in patients with celiac disease was 1.13 (95% CI: 1.00–1.28, P = 0.054), which, while slightly above the conventional significance threshold, suggests a potential association. Patients with AL amyloidosis were found to be less susceptible to IBD (OR: 0.83, 95% CI: 0.74–0.93, P < 0.05) and UC (OR: 0.99, 95% CI: 0.99–1.00, P < 0.05), which is consistent with the primary findings. No evidence of heterogeneity or pleiotropy bias was detected, further ensuring the robustness of these findings (Supplementary Table 4).

The Role of Inflammatory Cytokines in Universal ADs, Celiac Disease, IBD, and UC

To further investigate common inflammatory cytokine changes potentially contributing to the increased or decreased risk of these ADs, MR analyses were conducted by applying pQTLs for all 77 inflammatory cytokines, followed by robustness analyses to account for heterogeneity and pleiotropy (Supplementary Table 5). IL21 was found to reduce the risk of universal ADs (OR: 0.71, 95% CI: 0.64–0.78, P < 0.05) and celiac disease (OR: 0.20, 95% CI: 0.13–0.32, P < 0.05) while increasing the risk of inflammatory bowel disease (IBD; OR: 1.20, 95% CI: 1.00–1.45, P < 0.05) and ulcerative colitis (UC; OR: 1.17, 95% CI: 1.05–1.29, P < 0.05) (Supplementary Fig. 1, Supplementary Table 6). Additionally, interleukin 27 receptor subunit alpha (IL27RA) was identified as a risk factor for IBD (OR: 1.04, 95% CI: 1.0–1.07, P < 0.05) and UC (OR: 1.05, 95% CI: 1.0–1.09, P < 0.05) via the IVW method. Although no significant effect was observed on universal ADs or celiac disease in IVW analyses, other less powerful methods suggested a potential risk-enhancing effect, as shown in Supplementary Fig. 1.

Given that IBD encompasses Crohn's disease, UC, and indeterminate-type colitis, shared inflammatory cytokines influencing both IBD and UC have been identified. These genes included CCL3, CCL22, TGFB3, IL16, IL18R1, IL1R2, IL1RL1, and IL27RA (Supplementary Fig. 1). To further explore the distinct and overlapping causal effects of inflammatory cytokines across the four ADs, a bubble plot was generated to visualize the causal effects of each tested cytokine on these diseases individually (Fig. 4).

Mechanistic studies reveal that the occurrence of AL amyloidosis is associated with elevated levels of interleukin-6 (IL-6)[27] and C-C motif chemokine ligand 3 (CCL3)[28]. These cytokines/chemokines were further implicated in ADs through Mendelian randomization analysis, as supported by Supplementary Fig. 1. To further explore potential alterations in levels of inflammatory cytokines (outcome) induced by AL amyloidosis (exposure), Mendelian randomization analysis was performed. This approach aimed to elucidate causal associations while minimizing confounding factors inherent in observational studies. Among 77 tested inflammatory cytokines, a significant causal association was observed only for interleukin enhancer binding factor 3 (ILF3; beta: -0.09, 95% CI: -0.17 to -0.01, P < 0.05), suggesting that AL amyloidosis reduces circulating levels of ILF3. However, no significant evidence was found to confirm a relationship between ILF3 and any of the four ADs individually (Supplementary Fig. 1).

Reduced Circulating ILF3 in AL Amyloidosis Alters Cytokine Abundance via BAFF

Recent studies have shown that suppressed expression of ILF3 elevates circulating levels of B-cell activating factor (BAFF, also known as TNFSF13B), thereby increasing the risk of autoimmune diseases such as multiple sclerosis and systemic lupus erythematosus in the Sardinian population[29, 30]. This relationship has been validated experimentally: pulldown assays demonstrate that knockdown of ILF3 leads to increased BAFF expression in cell lines[30]. Furthermore, BAFF is known to interact with various inflammatory mediators and plays a central role in inflammation and immune regulation[31–33]. Based on these findings, we hypothesize that reduced ILF3 levels in AL amyloidosis may influence the expression of downstream inflammatory cytokines through the modulation of BAFF. To explore this potential pathway, we constructed PPI networks using medium confidence criteria for BAFF and the inflammatory cytokines implicated in each of the four previously mentioned autoimmune conditions: universal ADs, celiac disease, IBD, and UC (Fig. 5). In universal ADs, BAFF was associated with IL2, CCL5, IL1A, and IL21 (Fig. 5A). In celiac disease, BAFF is linked to IL6 and IL21 (Fig. 5B). In IBD, BAFF was correlated with CXCL11, CCL3, IL7R, and IL21 (Fig. 5C). In UC, BAFF was associated with CCL3 and IL21 (Fig. 5D). Notably, BAFF exhibited a consistent correlation with IL21 across all four diseases, suggesting a potential shared pathway involving these cytokines.

This study, for the first time, investigated the causal effects of AL amyloidosis on ADs and revealed that AL amyloidosis was associated with an increased risk of celiac disease and universal ADs but a reduced risk of IBD and UC, which was validated in a replication cohort. Despite the analysis of 77 inflammatory factors, none were identified as mediators between AL amyloidosis and the tested ADs, as the impact of AL amyloidosis on inflammatory cytokine levels was minimal. However, mapping the causal effects of these inflammatory factors across universal ADs, celiac disease, IBD, and UC provides valuable insights into the immunological pathways underlying these conditions.

Findings from Mendelian randomization studies are less prone to confounding and reverse causality bias, making a strictly conducted Mendelian randomization study, especially when validated by an external cohort, more reliable than conventional observational studies[34, 35]. An observational study reported that 21% of AL amyloidosis patients had concomitant immune disorders, with Sjögren’s syndrome being the most common disorder[6]. However, odds ratios relative to the general population were not calculated, and protective factors—which are also critical for deciphering underlying mechanisms—are often neglected. In contrast, well-conducted MR studies investigate both risk and protective factors while maintaining strong statistical power.

Human plasma pQTL data have shown tremendous potential in drug development[22, 36], particularly in identifying therapeutic targets for immune-mediated diseases[37]. A recent study integrating pQTL data for 2,920 plasma proteins identified 26 promising therapeutic targets with favorable safety profiles[36]. Consistent with this, our study revealed that AL amyloidosis is associated with reduced levels of interleukin enhancer binding factor 3 (ILF3, also known as NF90), a protein that selectively suppresses the translation of BAFF. Genetic variants in the 3’ untranslated region (UTR) of BAFF have been linked to the development of specific ADs[29], suggesting that the ILF3-BAFF regulatory axis may play a key role in autoimmune pathogenesis.

IL21, a cytokine with immunoregulatory activity, emerged as a key factor in this study. Dysregulation of IL21 is commonly observed in AD patients, and its effects vary across different ADs[38, 39]. While the role of IL21 has been extensively studied in systemic lupus erythematosus[40] and rheumatoid arthritis[41], the precise mechanisms underlying its dysregulation remain unclear. In this study, we observed that BAFF was consistently correlated with IL21 across all tested ADs, including universal ADs, celiac disease, IBD, and UC. This finding suggests a potential shared pathway involving BAFF and IL21, which may play a central role in AD progression or suppression. By integrating pQTL MR analyses and PPI networks, we propose the hypothesis that increased levels of BAFF, driven by reduced ILF3 expression in AL amyloidosis, regulate IL21 levels, ultimately influencing the risk of developing certain ADs. This ILF3-BAFF-IL21 axis provides a novel framework for understanding the immunological mechanisms linking AL amyloidosis to ADs.

One of the key strengths of this study is the use of MR to rigorously test and validate the causal effects of AL amyloidosis on common AD subtypes from a genetic perspective. Another strength lies in the mapping of the effects of inflammatory factors across universal ADs, celiac disease, IBD, and UC, which provides a comprehensive view of disease-specific inflammatory signaling. The integration of pQTL and PPI analyses further strengthens the findings by identifying a potential mechanistic axis linking AL amyloidosis to ADs.

In conclusion, this study provides critical insights into the causal relationships between AL amyloidosis and ADs and explores the underlying inflammatory factor interactions between them. Notably, the ILF3-BAFF-IL21 axis has emerged as a potential mechanistic pathway driving these effects, highlighting the need for further investigation.

However, several limitations should be acknowledged. First, no mediating factors were identified, as ILF3 was the only inflammatory factor associated with AL amyloidosis among the 77 tested, precluding the use of multivariable MR analysis. Second, while prior pull-down analyses between ILF3 and BAFF were conducted by Idda ML et al.[30], these findings were not experimentally validated in our study. Third, although PPI networks provide a useful framework for visualizing cytokine interactions, they rely on preexisting physical or text-mined associations, which may not fully capture the biological complexity of these pathways.

Despite these limitations, the consistent correlation observed between BAFF and IL21 suggests a meaningful relationship, warranting further experimental validation to confirm causal pathways and explore their clinical implications. Future studies should focus on validating the ILF3-BAFF-IL21 axis through functional experiments and investigating its potential as a therapeutic target for AD.

AL amyloidosis	Immunoglobulin light chain amyloidosis
ADs	autoimmune diseases
MR	Mendelian randomization
PPI	Protein‒protein interaction
OR	odds ratio
CI	confidence interval
IBD	inflammatory bowel disease
UC	ulcerative colitis
BAFF	B-cell-activating factor
IVs	instrumental variables
GWAS	genome-wide association study
pQTLs	protein quantitative trait loci
SNPs	single nucleotide polymorphisms
LD	linkage disequilibrium
MAF	minor allele frequency
IVW	inverse variance weighted
ILF3	Interleukin Enhancer Binding Factor 3
IL21	Interleukin 21
CCL3	C-C motif chemokine ligand 3
CCL22	C-C motif chemokine ligand 22
TGFB3	Transforming Growth Factor Beta 3
IL16	Interleukin 16
IL18R1	Interleukin 18 receptor 1
IL1R2	Interleukin 1 receptor type 2
IL1RL1	Interleukin 1 Receptor Like 1
IL27RA	Interleukin 27 receptor subunit alpha

Funding

This research was supported by the National Natural Science Foundation of China (No. 82270161), the Guangdong Basic and Applied Basic Research Foundation (No. 2024B1515020054), the Science and Technology Planning Project of Guangdong Province, China (No. 2023B1111050004), the NSFC Incubation Project of Guangdong Provincial People's Hospital (KY0120220024 and KY012021160), Project of Administration of Traditional Chinese Medicine of Guangdong Province of China (20242002) and the High-level Hospital Construction Project of Guangdong Provincial People's Hospital (DFJHBF202107).

Ethics Declaration

Publicly available GWAS summary statistics were used. No new data were collected, and no new ethical approval was not applicable.

Sanchorawala V. Systemic light chain amyloidosis. N Engl J Med. 2024;390:2295–307. https://doi.org/10.1056/NEJMra2304088.
Gertz MA. Immunoglobulin light chain amyloidosis: 2022 update on diagnosis, prognosis, and treatment. Am J Hematol. 2022;97:818–29. https://doi.org/10.1002/ajh.26569.
G M, Mj S. Dangerous small B-cell clones. Blood. 2006;108. https://doi.org/10.1182/blood-2006-03-001164.
Weiss BM, Hebreo J, Cordaro DV, Roschewski MJ, Baker TP, Abbott KC, et al. Increased serum free light chains precede the presentation of immunoglobulin light chain amyloidosis. J Clin Oncol Off J Am Soc Clin Oncol. 2014;32:2699–704. https://doi.org/10.1200/JCO.2013.50.0892.
Cobb JK, Shiver L, Russell CR, Chen B, Bassil C. Proteinase 3 antibody and anti-double-stranded DNA in a patient with immunoglobin light chain amyloidosis. Cureus. 2023;15:e47667. https://doi.org/10.7759/cureus.47667.
Basset M, Hummedah K, Kimmich C, Veelken K, Dittrich T, Brandelik S, et al. Localized immunoglobulin light chain amyloidosis: Novel insights including prognostic factors for local progression. Am J Hematol. 2020;95:1158–69. https://doi.org/10.1002/ajh.25915.
Yamaguchi Y, Ujiie H, Ohigashi H, Iwata H, Muramatsu K, Endou T, et al. Linear IgA bullous dermatosis associated with immunoglobulin light-chain amyloidosis. Acta Derm Venereol. 2017;97:528–9. https://doi.org/10.2340/00015555-2590.
Song Y, Li J, Wu Y. Evolving understanding of autoimmune mechanisms and new therapeutic strategies of autoimmune disorders. Signal Transduct Target Ther. 2024;9:263. https://doi.org/10.1038/s41392-024-01952-8.
Cho JH, Feldman M. Heterogeneity of autoimmune diseases: Pathophysiologic insights from genetics and implications for new therapies. Nat Med. 2015;21:730–8. https://doi.org/10.1038/nm.3897.
Sekula P, Del Greco M F, Pattaro C, Köttgen A. Mendelian randomization as an approach to assess causality using observational data. J Am Soc Nephrol JASN. 2016;27:3253–65. https://doi.org/10.1681/ASN.2016010098.
Burgess S, Scott RA, Timpson NJ, Davey Smith G, Thompson SG, EPIC- InterAct Consortium. Using published data in mendelian randomization: A blueprint for efficient identification of causal risk factors. Eur J Epidemiol. 2015;30:543–52. https://doi.org/10.1007/s10654-015-0011-z.
Szklarczyk D, Kirsch R, Koutrouli M, Nastou K, Mehryary F, Hachilif R, et al. The STRING database in 2023: Protein-protein association networks and functional enrichment analyses for any sequenced genome of interest. Nucleic Acids Res. 2023;51:D638–46. https://doi.org/10.1093/nar/gkac1000.
Wang K, Yi H, Wang Y, Jin D, Zhang G, Mao Y. Proteome-wide multicenter mendelian randomization analysis to identify novel therapeutic targets for lung cancer. Arch Bronconeumol. 2024;60:553–8. https://doi.org/10.1016/j.arbres.2024.05.007.
Burgess S, Davies NM, Thompson SG. Bias due to participant overlap in two‐sample mendelian randomization. Genet Epidemiol. 2016;40:597–608. https://doi.org/10.1002/gepi.21998.
da Silva Filho MI, Försti A, Weinhold N, Meziane I, Campo C, Huhn S, et al. Genome-wide association study of immunoglobulin light chain amyloidosis in three patient cohorts: Comparison with myeloma. Leukemia. 2017;31:1735–42. https://doi.org/10.1038/leu.2016.387.
Schönland SO, Hegenbart U, Bochtler T, Mangatter A, Hansberg M, Ho AD, et al. Immunohistochemistry in the classification of systemic forms of amyloidosis: A systematic investigation of 117 patients. Blood. 2012;119:488–93. https://doi.org/10.1182/blood-2011-06-358507.
Hemani G, Zheng J, Elsworth B, Wade KH, Haberland V, Baird D, et al. The MR-base platform supports systematic causal inference across the human phenome. eLife. 2018;7:e34408. https://doi.org/10.7554/eLife.34408.
MacArthur J, Bowler E, Cerezo M, Gil L, Hall P, Hastings E, et al. The new NHGRI-EBI catalog of published genome-wide association studies (GWAS catalog). Nucleic Acids Res. 2017;45:D896–901. https://doi.org/10.1093/nar/gkw1133.
Liu JZ, van Sommeren S, Huang H, Ng SC, Alberts R, Takahashi A, et al. Association analyses identify 38 susceptibility loci for inflammatory bowel disease and highlight shared genetic risk across populations. Nat Genet. 2015;47:979–86. https://doi.org/10.1038/ng.3359.
Anderson CA, Boucher G, Lees CW, Franke A, D’Amato M, Taylor KD, et al. Meta-analysis identifies 29 additional ulcerative colitis risk loci, increasing the number of confirmed associations to 47. Nat Genet. 2011;43:246–52. https://doi.org/10.1038/ng.764.
Zhou W, Nielsen JB, Fritsche LG, Dey R, Gabrielsen ME, Wolford BN, et al. Efficiently controlling for case-control imbalance and sample relatedness in large-scale genetic association studies. Nat Genet. 2018;50:1335–41. https://doi.org/10.1038/s41588-018-0184-y.
Folkersen L, Gustafsson S, Wang Q, Hansen DH, Hedman ÅK, Schork A, et al. Genomic and drug target evaluation of 90 cardiovascular proteins in 30,931 individuals. Nat Metab. 2020;2:1135–48. https://doi.org/10.1038/s42255-020-00287-2.
Sun BB, Maranville JC, Peters JE, Stacey D, Staley JR, Blackshaw J, et al. Genomic atlas of the human plasma proteome. Nature. 2018;558:73–9. https://doi.org/10.1038/s41586-018-0175-2.
Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: Effect estimation and bias detection through egger regression. Int J Epidemiol. 2015;44:512–25. https://doi.org/10.1093/ije/dyv080.
Burgess S, Thompson SG. Interpreting findings from mendelian randomization using the MR-egger method. Eur J Epidemiol. 2017;32:377–89. https://doi.org/10.1007/s10654-017-0255-x.
Bowden J, Davey Smith G, Haycock PC, Burgess S. Consistent estimation in mendelian randomization with some invalid instruments using a weighted median estimator. Genet Epidemiol. 2016;40:304–14. https://doi.org/10.1002/gepi.21965.
Bellofatto IA, Nikolaou PE, Andreadou I, Canepa M, Carbone F, Ghigo A, et al. Mechanisms of damage and therapies for cardiac amyloidosis: A role for inflammation? Clin Res Cardiol. 2024. https://doi.org/10.1007/s00392-024-02522-2.
Batlle C, de Groot NS, Iglesias V, Navarro S, Ventura S. Characterization of soft amyloid cores in human prion-like proteins. Sci Rep. 2017;7:12134. https://doi.org/10.1038/s41598-017-09714-z.
Steri M, Orrù V, Idda ML, Pitzalis M, Pala M, Zara I, et al. Overexpression of the cytokine BAFF and autoimmunity risk. N Engl J Med. 2017;376:1615–26. https://doi.org/10.1056/NEJMoa1610528.
Idda ML, Lodde V, McClusky WG, Martindale JL, Yang X, Munk R, et al. Cooperative translational control of polymorphic BAFF by NF90 and miR-15a. Nucleic Acids Res. 2018;46:12040–51. https://doi.org/10.1093/nar/gky866.
Fodor I, Serban O, Serban DE, Farcau D, Man SC, Dumitrascu DL. B cell-activating factor (BAFF) in children with inflammatory bowel disease. Pediatr Res. 2021;89:1798–803. https://doi.org/10.1038/s41390-020-01155-1.
Fu Y, Wang L, Xie C, Zou K, Tu L, Yan W, et al. Comparison of non-invasive biomarkers faecal BAFF, calprotectin and FOBT in discriminating IBS from IBD and evaluation of intestinal inflammation. Sci Rep. 2017;7:2669. https://doi.org/10.1038/s41598-017-02835-5.
Alexandrakis MG, Roussou P, Pappa CA, Messaritakis I, Xekalou A, Goulidaki N, et al. Relationship between circulating BAFF serum levels with proliferating markers in patients with multiple myeloma. BioMed Res Int. 2013;2013:389579. https://doi.org/10.1155/2013/389579.
Davies NM, Holmes MV, Davey Smith G. Reading mendelian randomisation studies: A guide, glossary, and checklist for clinicians. BMJ. 2018;362:k601. https://doi.org/10.1136/bmj.k601.
Lawlor DA, Tilling K, Davey Smith G. Triangulation in aetiological epidemiology. Int J Epidemiol. 2017;:dyw314. https://doi.org/10.1093/ije/dyw314.
Deng Y-T, You J, He Y, Zhang Y, Li H-Y, Wu X-R, et al. Atlas of the plasma proteome in health and disease in 53,026 adults. Cell. 2024;:S0092-8674(24)01268-6. https://doi.org/10.1016/j.cell.2024.10.045.
Zhao JH, Stacey D, Eriksson N, Macdonald-Dunlop E, Hedman ÅK, Kalnapenkis A, et al. Genetics of circulating inflammatory proteins identifies drivers of immune-mediated disease risk and therapeutic targets. Nat Immunol. 2023;24:1540–51. https://doi.org/10.1038/s41590-023-01588-w.
Ren HM, Lukacher AE, Rahman ZSM, Olsen NJ. New developments implicating IL-21 in autoimmune disease. J Autoimmun. 2021;122:102689. https://doi.org/10.1016/j.jaut.2021.102689.
Chen L, Ruan G, Cheng Y, Yi A, Chen D, Wei Y. The role of Th17 cells in inflammatory bowel disease and the research progress. Front Immunol. 2022;13:1055914. https://doi.org/10.3389/fimmu.2022.1055914.
Bocharnikov AV, Keegan J, Wacleche VS, Cao Y, Fonseka CY, Wang G, et al. PD-1hiCXCR5- T peripheral helper cells promote B cell responses in lupus via MAF and IL-21. JCI Insight. 2019;4:e130062, 130062. https://doi.org/10.1172/jci.insight.130062.
Rao DA, Gurish MF, Marshall JL, Slowikowski K, Fonseka CY, Liu Y, et al. Pathologically expanded peripheral T helper cell subset drives B cells in rheumatoid arthritis. Nature. 2017;542:110–4. https://doi.org/10.1038/nature20810.

No competing interests reported.

Download PDF

Reviewers agreed at journal
02 Oct, 2025
Reviewers invited by journal
02 Oct, 2025
Submission checks completed at journal
23 Sep, 2025
First submitted to journal
21 Sep, 2025

You are reading this latest preprint version

Causal Effects of AL Amyloidosis on Autoimmune Diseases and the Role of Inflammatory Cytokines: Insights from Mendelian Randomization

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

Introduction

Methods

Study design

Data sources

Selection of instrumental variables

Two-sample MR analysis

PPI network

Statistical analysis

Result

The Role of Inflammatory Cytokines in Universal ADs, Celiac Disease, IBD, and UC

Reduced Circulating ILF3 in AL Amyloidosis Alters Cytokine Abundance via BAFF

Discussion

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1