Pharmaceutical Repurposing Strategies for Metabolic Disorders: Insights from Mendelian Randomization Studies
============================================================================================================

* Shixuan Zhang
* Xiaoxi Hu
* Dandan Hou
* Xiaoru Sun
* Dayan Sun

## Abstract

Metabolic disorders (MDs) are a group of medical conditions that impact the metabolism. These complex processes may have common characteristics among various diseases, thereby suggesting the potential of drug repurposing. Employing Mendelian Randomization (MR), we constructed a causal network between 2,478 targetable drug-gene expressions (eQTLs) and 30 broadly reported metabolic disorders. Our study identified 499 drug target genes significantly associated with 27 metabolic disorders (|MR coefficient| > 0.2). Pathway enrichment analysis of drug target genes indicates that regulation of response stimulus may serve as a common pathway across 14 diseases. Based on 53 commonly used clinical drugs for 18 diseases, we elucidated novel therapeutic mechanisms of some drugs, such as the potential of Valproic Acid to treat Schizophrenia by affecting key genes in Alcoholism, *SLC29A1*, and *HDAC4*. Furthermore, we identified potential for drug repurposing in four diseases, with Manic Episode and Type 1 Diabetes sharing four novel drugs: Cannabidiol, Doxorubicin, Genistein, and Propylthiouracil. Additionally, we predicted 189 potential therapeutic drugs affecting the causal genes of diseases. Overall, we established a causal network between metabolic disorders and drug target genes, explored possible pathways for drug action on disease treatment, and proposed drug repurposing strategies for four diseases.

**Highlights**

1.  To the best of our understanding, the manuscript we have submitted delineates the inaugural drug-gene causal network tailored for Metabolic Disorders (MDs), covering an array of 30 diseases.

2.  This investigation unveils potential novel drug mechanisms aimed at treating MDs, marking a pioneering exploration in this domain.

3.  We introduce strategies for the repurposing of drugs targeting four specific MDs, suggesting innovative approaches to treatment.

4.  The findings are presented through an easily navigable web interface, enabling thorough examination by users ([https://quantlifebits.shinyapps.io/mrmetabopharm/](https://quantlifebits.shinyapps.io/mrmetabopharm/)).

Keywords
*   Metabolic disorders
*   Mendelian Randomization
*   Drug repurposing

## Introduction

Metabolic disorders are emerging as a significant global health concern, with their prevalence experiencing a marked surge1. This growing interest is attributed to their extensive systemic complexity and widespread occurrence. The excessive variation in the human body’s internal environment is a primary trigger for these diseases, such as endocrine disruptions in hormones (diabetes2, hypertension3,4, dyslipidemia5, obesity6, Polycystic Ovary Syndrome6, etc.), disruptions in circadian rhythms (depression7, Alzheimer’s disease8, etc.), and dysregulation of cell growth regulation, all of which may be associated with chronic systemic diseases involving multiple organ systems.

Metabolic disorders may exhibit potential bidirectional interactions, thus offering ample opportunities for drug repurposing. Notably, diabetes is concurrently associated with several diseases in a bidirectional manner, such as cardiovascular and cerebrovascular diseases9, behavioral disorders10, cancer11, and dementia12, with diabetes medication metformin being utilized to assist in treating these conditions13–15. The accumulation of metabolic disorder states may lead to organ system damage, a process akin to aging16, affecting systems such as the cardiovascular17, nervous18, and endocrine systems19, with more severe accumulations potentially leading to various cancers20. Furthermore, Wray NR et al. suggested that the pathogenesis of metabolic diseases is complex21, and different diseases may share common pathogenetic mechanisms. Further research indicated that comorbidities of many diseases might be related to genetic factors22, suggesting the possibility of common effective intervention mechanisms between different diseases. Therefore, employing Mendelian randomization to identify common causal drug target genes may be an effective approach for developing new therapeutic strategies for diseases23.

The potential efficacy of classic drugs such as metformin in treating metabolic disorders other than diabetes underscores the significance of the interconnectedness among metabolic disorders. In 1988, Black J. proposed drug repurposing schemes24,25, which hold significant advantages in shortening the drug development cycle, mitigating disease resistance sooner26, and avoiding safety issues associated with new drug development27. While numerous treatment options exist for diseases classified as metabolic disorders 1,28,29, the need for further refinement of treatment strategies is underscored by the increasing prevalence rates published by the World Health Organization (WHO)1,30. Consequently, the broadening of existing drugs’ applications to encompass various diseases through shared genetic effects and causal genes, alongside investigating the mechanisms underlying drug actions, has become a focal point of research31. This approach is vital for circumventing issues such as drug resistance, adverse effects, and safety concerns associated with new medications in managing metabolic disorders. It holds significant promise for enhancing therapeutic strategies.

In this study, we explored potential drug-regulatory genes for diseases classified as metabolic disorders based on blood-derived targetable drug genes (expression Quantitative Trait Loci, eQTLs). The research design is depicted in **Fig. 1**. Initially, we utilized Genome-Wide Association Study (GWAS) data from FinnGen, eQTLs from the eQTLGen Consortium, and the Drug-Gene Interaction database to establish a complex causal network between drug genes and multiple diseases. Subsequently, methods such as reverse causation detection, pathway enrichment analysis, and drug prediction were employed to validate the primary findings. Furthermore, we constructed an effective causal network of ‘drug-gene-disease’ for metabolic disorder diseases and performed external validation for some of the diseases included in the analysis using GWAS data from the UK Biobank cohort to enhance the level of evidence. Finally, we have defined the network we constructed as “Metabolic disorders targetable drug causal network,” publicly available on an interactive website ([https://quantlifebits.shinyapps.io/mrmetabopharm/](https://quantlifebits.shinyapps.io/mrmetabopharm/)) to serve as a navigational map for subsequent research.

## Methods

### Data inclusion

#### eQTL of drug gene collection

For the assembly of drug gene sets, 4,464 genes associated with known or potential drugs were selected from the Drug-Gene Interaction database. Significant SNP loci within +100 kb of the start and end positions of each gene were extracted to form a set of genetic variations. Using peripheral blood meta-analysis cis-eQTL data (within ±1Mb of each probe and p value < 0.05 after false discovery rate correction) from 31,684 samples via the eQTLGen Consortium platform, SNPs significantly associated with drug genes within the genetic variation set (p value < 5X 10-8) were identified as genetic instrumental variables. The corresponding eQTL results represent summary statistical data between genetic variants and genes. SNPs with a Minor Allele Frequency (MAF) greater than 1% and an F-statistic greater than 10 (based on MAF) were retained to mitigate the bias from weak instrumental variables. These variants were then clumped with an R2 of 0.1% and a window size of 10,000 kb, based on the reference data from the European population in the 1000 Genomes Project. The research ultimately incorporated eQTLs from 2,478 drug-targetable genes for further investigation.

#### Disease set screening

The disease collection was selected based on keywords including metabolic disorders, endocrine metabolism, circadian rhythm disorders, cell growth regulation disorders, etc., through a search of the PubMed database. This search identified approximately 30 common disease types, most of which are related to environmental stress and adaptability. Using the OpenGWAS database and FinnGen32 (R9), the aforementioned diseases were further investigated, excluding those for which complete summary statistics could not be obtained. The study ultimately included 30 diseases (**Table 1**). Summary statistics from FinnGen and the UK Biobank, for diseases with available data from both datasets, were used for cross-validation. Diseases with a low number of SNPs and those lacking MAF in GWAS statistical data were excluded, resulting in 18 disease sets being included in the validation (**Table 1**).

### Mendelian randomization analysis

In this study, we employed a two-sample Mendelian randomization (MR) approach based on GWAS summary statistics to examine the causal relationship between plasma gene expression and the selected 30 diseases. For cases where a single SNP served as the genetic instrument (i.e., only one eQTL available for a given drug gene), the Wald ratio was used to estimate the causal effect. When two or more genetic instruments were available, inverse variance weighted MR (MR-IVW) was employed, assuming a zero intercept and conducting a weighted regression of the SNP-gene effects with the SNP-disease effects. Additionally, MR-Egger regression was utilized to test for horizontal pleiotropy between genetic instruments and diseases, and Cochran’s Q statistic was used to assess heterogeneity among the chosen genetic instruments. Both the MR regression beta and the odds ratio (log10(beta)) were obtained. Plasma gene expressions with p value<0.05 and consistent beta sign across the same disease were considered to have a causal relationship with the corresponding disease. Statistical analyses were conducted using R version 4.2.3, utilizing the TwoSampleMR packages.

### Reverse causation detection

To investigate potential reverse causality, plasma gene expressions identified as having a causal effect were included in bidirectional Mendelian randomization (MR) analyses. The analyses employed multiple methods to estimate the effects, including MR-IVW, MR-Egger, weighted median, simple mode, and weighted mode. Additionally, Steiger filtering was conducted to ensure the directionality of the association between gene expression and multiple diseases was correctly identified (*p* value < 0.05).

### Prediction of potential drugs for genes with causal effects

#### Drug prediction based on causal networks

Based on the causal gene network corresponding to each disease, two sets of opposing protein lists, one with beta greater than 0 and the other with beta less than 0, were utilized. Predictions for targetable drugs were made based on the SignatureSearch database, with criteria set at *p* value (up) < 0.01, *p* value (down) < 0.01, and Z score (sum) < −3. This method aims to explore a drug network that modulates the overall causal proteins in an antagonistic manner33. (**Table S1**)

#### Screening of drugs for clinical treatment of diseases

To explore targetable drugs for independent causal genes associated with diseases, the study identified a collection of 1,100 drugs used in clinical treatments from the Comparative Toxicogenomics Database (CTD), detailed in **Table S2**. Considering the potential adverse effects of non-clinical medications, the inclusion criteria for drugs were strictly controlled. Initially, drugs without targeted gene research were excluded, followed by those lacking a Chemical Abstracts Service Registry Number (CAS RN). Drugs were then selected based on the presence of ‘Direct Evidence’ marked as ‘T’ (True). Lastly, given the complexity of drug actions, only the top five drugs per disease, based on the number of references, were selected. This process ultimately included 53 drugs across 17 diseases, as shown in **Table S3**.

#### Determination of target genes for clinical therapeutic drugs

Data for 53 drug genes from the Comparative Toxicogenomics Database (CTD) has been accessed, focusing on their target proteins. Given eQTLs provide insights into gene expression, with GeneForms specified as mRNA, excluding instances of cotreatment, and the selection was narrowed to species including “Homo sapiens,” “Mus musculus,” “Peromyscus californicus,” “Rattus,” “Cricetulus griseus,” “Cavia porcellus,” “Cavia,” “Sus scrofa,” and “Rattus norvegicus.” Finally, to avoid drugs having a negative impact on diseases, our study only included drugs that, as the dosage increases, lead to changes in gene expression that have a positive effect on the disease (i.e., the relationship between the protein and the drug should be opposite to the relationship between the gene and the disease.

### Pathway enrichment studies

The study concentrated on drug genes and carried out Gene Ontology (GO) terms enrichment analysis via the Enrichment Map framework. This process entailed identifying significantly enriched pathways for individual disease sets (adjusted *p* value < 0.05) and utilized a clustering method with a CoSE Layout for categorizing and characterizing the pathways. The analyses were efficiently conducted leveraging the capabilities of Cytoscape’s AutoAnnotate and EnrichmentMap functions.

## Results

### Causal network of drug genes and multiple MDs

Leveraging known drug target genes, we aimed to explore effective causal genes across multiple diseases underscores the potential therapeutic targets for various conditions. Initially, our study utilized clinical research related to metabolic disorders, extracting GWAS summary data for 30 diseases from 377,277 individuals (210,870 females and 166,407 males) in the FinnGen database to perform Mendelian Randomization (MR, **Table 1**). We identified disease-related causal genes by examining blood eQTLs for 2,478 known drug target genes. Through sensitivity analysis (MR Egger) and consistency tests between IVW and Egger’s Q, along with verification of the causal effect direction (Steiger), MR incorporated 499 drug target genes with significant causal associations with 27 diseases (MR p value < 0.05, Q IVW and Q Egger >0.05, MR Egger> 0.05, Steiger *p* value< 0.05, |MR coefficient|> 0.2, **Fig. 1**). Among these, the *FN1* gene had the most disease connections, including 5 diseases (**Table 2**), where an increase in *FN1* protein was identified as a risk factor for Type 1 Diabetes (T1D)34, Polycystic Ovary Syndrome (PCOS)35, and a protective factor for Non-Alcoholic Fatty Liver Disease (NAFLD), Hypothyroidism/Myxedema, and Hyperthyroidism. The OBSCN gene was also linked to 5 diseases, with increased protein expression posing a risk for NAFLD, Colorectal Cancer36, Parkinson’s Disease, and serving as a protective factor for Preeclampsia and Non-Small Cell Lung Cancer37.

As shown in **Fig S1**, PCOS was associated with a greater number of causal genes, such as *CDKN1A*38, *TFRC*39, *MAPK8*40. Additionally, causal gene targets for T1D comprised *ICAM1*41, *NFKB1*42, and others. The study also uncovered potential causal targets for some cancers, for example, Small Cell Lung Cancer had 52 causal gene targets, including *VIM*43 and *TLR5*44. For psychiatric disorders, Schizophrenia included causal genes such as *NOS3*45, *ULK2*46, and *ALDH1A1*47 (**Table 2**). Using the UK Biobank (UKB) datasets, causal connections for only 10 diseases were verified, involving 31 genes (**Table S4**), due to the heterogeneity between the UKB and FinnGen populations.

### Functional clustering of key causal genes for multiple MDs

To explore the related functions of drug target genes with significant causal effects on diseases and the shared causal characteristics among diseases, our study conducted pathway function annotation for genes. 14 diseases demonstrated significant pathway enrichment (|MR coefficient| > 0.2, adjusted p value in pathway enrichment < 0.01), spanning 13 distinct pathway cluster categories. The causal genes for 13 diseases were annotated to pathways related to response stimulus regulation (**Fig. 1, Fig. S1B, Table S5**), including T1D, suggesting that intervening in the body’s stress response could be key to disease treatment. Uniquely in Alzheimer’s Disease (AD), our study focused on the ITGAV-VTN-ITGB5 pathway cluster, which includes various pathways related to the Integrin Subunit complex, such as ITGAV-ITGB5-VTN. The regulatory role of Integrin Subunit receptors is considered a key marker of aging. Additionally, the regulatory role of neurons was independently observed in AD. Lastly, NAFLD and AD were clustered into Positive peroxidase activity, highlighting potential therapeutic targets and the importance of oxidative stress pathways in these conditions.

### Exploration of clinical drugs for common key genes in multiple MDs

The regulation of drug target genes’ expression plays a pivotal role in the treatment of diseases. Our analysis revealed that key genes shared across multiple diseases may be regulated by a wider range of drugs (**Table 2** and **Table S6**), based on 53 clinical drugs. Within the response stimulus regulation pathway cluster, five genes were crucial for endocrine disorder diseases. The *FN1* gene, with causal connections to five diseases, was influenced by four drugs. Metformin, Troglitazone, and Zinc could potentially decrease *FN1* expression, offering therapeutic benefits for PCOS, while Genistein may increase *FN1* expression, offering a therapeutic effect for NAFLD. Additionally, the *ICAM1* gene was modulated by six drugs, showing therapeutic effects for Anxiety and NAFLD. The *EDN1* gene, also regulated by six drugs, presented therapeutic potential for Parkinson’s Disease, Anxiety, and Non-Small Cell Lung Cancer. Furthermore, the *CDKN1A* gene was affected by five drugs, indicating therapeutic effectiveness for PCOS. Lastly, significant causal associations with two nervous system diseases were observed with the Vimentin (*VIM*) gene, whose expression is co-downregulated by Sirolimus and Estradiol. This comprehensive analysis underscored the complex interplay between drug action and gene expression regulation, highlighting the potential for repurposing drugs across a spectrum of diseases by targeting key regulatory genes.

### Discovery of potential therapeutic mechanisms of MD clinical drugs

Diseases can be influenced by the expression of multiple genes and simultaneously modulated by various drugs. Hence, exploring the targetable drug-gene causal network, we investigated the KEGG pathway functions of drug target genes. Among 53 clinical drugs for multiple diseases, we identified 230 ‘drug - gene expression - disease’ connections (**Table 3, Fig. S7**). Valproic Acid regulated the expression of 20 genes and had a causal connection with Schizophrenia, where *SLC29A1* and *HDAC4* were enriched in Alcoholism, potentially involved in neuroexcitation regulation. Doxorubicin’s modulation of 13 gene expressions was causally linked to Colorectal Cancer, with *IL7R*, *STAT3*, *FLT1* enriched in HIF-1 signaling pathway and FoxO signaling pathway, involved in the activation regulation of cellular HIF factors. Additionally, NAFLD had a causal effect with 12 genes, enriched in the AGE-RAGE signaling pathway in diabetic complications (adjusted p value < 0.05), covering three genes, with Genistein potentially increasing *FN1* and *NOS3* expression to prevent liver fibrosis and cirrhosis. In PCOS, 12 genes had causal links with drugs, where *FN1*, *ICAM1*, *NOS3* were modulated by Ethinyl Estradiol’s gene expression, related to Legionellosis, and possibly involved in pro-inflammatory response and cell chemotaxis post bacterial infection. Moreover, genes like *MAPK8*, *CASP3*, *CXCL1*, *HSP90AA1*, regulated by Zinc, were associated with the progression of PCOS and participate in IL-17 signaling. Lastly, we observed associations among eight genes in Parkinson’s Disease, where *ITGAM* and *HLA-DRB1* significantly indicated the activation process of LPS, related to inflammatory responses. This comprehensive analysis highlighted the intricate mechanisms through which drugs modulated gene expression and their potential therapeutic effects on various diseases through specific pathway involvements.

### Disease-based causal proteins discover new uses for old drugs

Based on the influence of drugs on gene expression and the significant causal association between gene expression and disease occurrence (|MR coefficient| > 0.2), our findings suggested strategies for repurposing existing drugs for new therapeutic uses. The study identified 1,111 new ‘Drug-Gene-Disease’ causal connections (**Table S8**). Notably, drug repurposing opportunities were observed in four diseases: Bipolar Disorder, Manic Episode, Small Cell Lung Cancer, and T1D. Specifically, 41 genes were found to have a causal effect on T1D, with KEGG functional enrichment (**Table S9**) indicating potential new therapeutic pathways, including the PI3K-Akt signaling pathway and ECM-receptor interaction (**Table S10, Fig. S2 A(b) (c)**). Nine drugs were involved in regulating genes within the PI3K-Akt signaling pathway, with Aspirin recently reported as a protective drug for diabetes48 (**Fig. S2 A(a)**).

Currently, there are no targeted drugs for Manic Episode, but our research suggested 61 genes have a causal effect on Manic Episode. Functional enrichment also indicated that ECM-receptor interaction and Kaposi sarcoma-associated herpesvirus infection could be potential new therapeutic pathways, involving genes such as *COL6A1*, *COL6A2*, *VWF*, *THBS3*, *CCR1*, *EP300*, *HLA-G*, *PLCG2*, *FAS*, *PDGFB*, with 28 drugs modulating the Kaposi sarcoma-associated herpesvirus infection pathway (**Fig. S2 B(b) (c)**). Notably, based on the shared pathway of ECM-receptor interaction between Manic Episode and T1D, four common new drugs were identified: Cannabidiol, Doxorubicin, Genistein, Propylthiouracil.

Furthermore, predictive analyses based on existing drugs were conducted across 499 causal genes associated with metabolic disorders in different diseases, leading to a comprehensive drug prediction targeting these genes. We identified 189 drugs (p (up) < 0.01, p (down) < 0.01, Z score (sum) < −3, **Table S1**). Our results highlighted the potential collective modulation effects of some drugs on causal gene sets across multiple diseases (**Fig. S1 C**), with Reserpine showing potential therapeutic capabilities for Alzheimer’s Disease and NAFLD. Additionally, Bortezomib might have potential therapeutic effects on PLOS, NAFLD, and Schizophrenia, suggesting a broad spectrum of action and the possibility of drug repurposing for complex disorders.

## Discussion

To our knowledge, this study represents the pioneering effort to integrate eQTL analysis for uncovering pivotal drug-gene causal relationships in metabolic disorders, employing two-sample Mendelian Randomization (MR) alongside drug prediction techniques. Herein, we identified multiple potential drug target genes with causal associations to metabolic disorders, such as *FN1*, *ICAM1*, *SNCA*, *SLC29A1*, and *VIM*. The causal associations of some genes with diseases were validated in the UK Biobank (UKB) dataset, including *ICAM1* and *SLC29A1*. Additionally, we provided potential therapeutic mechanisms for various diseases through clinical drugs, such as the treatment of Schizophrenia with Valproic Acid, potentially based on the neuroexcitation regulation by *SLC29A1* and *HDAC4*. Genistein may increase the expression of *FN1* and *NOS3* to prevent liver fibrosis and cirrhosis, offering a treatment for NAFLD. Lastly, based on the MR causal network, we proposed drug repurposing strategies for diseases, focusing on four conditions. Manic Episode and T1D share common drug action pathways and four new common drugs, highlighting the potential for cross-disease therapeutic applications.

This study highlights several key genes with robust associations across metabolic disorder-related diseases, suggesting *FN1* (fibronectin 1), *ICAM1* (Intercellular Adhesion Molecule 1), and SNCA (Synuclein Alpha) as potential critical targets for these conditions. Research indicates *FN1* as an adverse marker for PCOS35,49 and T1D34, consistent with the observed increased prevalence of metabolic features (diabetes, hypertension, etc.) in women with PCOS50–52. Our findings suggested a negative correlation between *FN1* and NAFLD, where upregulation of *FN1* expression may suppress excessive inflammation and repair liver function in many NAFLD patients. Some studies have hinted that thyroid hormones could increase fibronectin 1 expression through the PI3K/Akt/HIF-1 pathway53, though research on its feedback effects on thyroid function is yet to be reported. Our results also show a strong causal effect of *ICAM1* with T1D, T1DW, and NAFLD. Located within a diabetes linkage region41, *ICAM1* primarily participates in integrin receptor activation. Increased expression of the *ICAM1* gene has been associated with all-cause mortality and cardiovascular disease incidence in T1D patients54,55. Moreover, an elevated concentration of *ICAM1* protein serves as a significant diagnostic marker for NAFLD56. Although the protective effect of *ICAM1* on Anxiety is a novel possible causality, it has been validated by analysis using UK Biobank data, with no previous studies reported. Furthermore, our analysis suggested that SNCA offers protection against PCOS but may pose a risk factor for AD. While some studies have reported correlations between PCOS and AD50, definitive answers remain elusive. Increases in SNCA protein have been suggested as molecular markers for AD57,58, while positively correlating with protective HSP90 proteins in PCOS patients59. Additionally, we found opposing causal relationships of *APP* (Amyloid Beta Precursor Protein) with AD and PCOS. Lastly, the protective role of *INSR* for AD60 and its risk association with Gestational Diabetes (GD)61 have been confirmed, with UK Biobank validating the association of *INSR* with Hypothyroidism. We also noted a significant association between *ITGAM* (CR3) and Parkinson’s Disease (PD)62, adding to the understanding of the genetic underpinnings and potential therapeutic targets across these metabolic and neurological disorders.

Our pathway analysis extensively focused on response stimulus regulation. Previous literature indicated that metabolic disorders are associated with resistance to stimuli, both external63 and internal29. Meta-analyses suggested possible concurrent mechanisms among some of these diseases64–67, leading us to believe that metabolic dysregulation induced by response stimulus regulation is significant evidence of the association among metabolic disorders. Additionally, in the context of Alzheimer’s Disease (AD), significant enrichment was found in the PTD domain neuronal, where APP and INSR were identified as the main causal genes, consistent with previous research68,69. Furthermore, in AD, we discovered evidence related to the itgav vtn itgb cluster (*ITGAV*, *ITGB5*, *LTBP2*), though such associations have not been reported in existing literature. However, the role of integrin receptors in AD has been confirmed70, suggesting our findings may point to novel therapeutic pathways for AD treatment. This underscores the importance of investigating the complex interplay between genetic factors and metabolic processes, paving the way for innovative approaches to understanding and treating metabolic disorders and related diseases.

Based on the constructed causal network for metabolic disorders, we explored the potential therapeutic mechanisms of known clinical drugs for each disease. The reduction in *SLC29A1* (Equilibrative nucleoside transporter type 1) expression, which can affect glutamate transport71, has been causally associated with Schizophrenia, a finding validated in the analysis using UK Biobank datasets. Current research suggests that the effect of Valproic Acid on Schizophrenia also involves modulation of glutamate transport, with our results indicating a novel pathway involving Acute ethanol through the *SLC29A1* and *HDAC4* genes. Moreover, the effect of Doxorubicin on Colorectal Cancer has been extensively reported72. We discovered that Doxorubicin’s reduction of *STAT3* and *FLT1* expression within the HIF-1 signaling pathway might effectively inhibit the activation of *HIF1A* and angiogenesis, potentially representing an important therapeutic approach for Colorectal Cancer (**Fig. S3 B**). This aligns with research suggesting Doxorubicin’s inhibition of HIF as an effective treatment for breast cancer73, consistent with our findings. Additionally, studies showed Genistein treats NAFLD by targeting Thromboxane A2. Our findings regarding its effect on the expression regulation of genes related to Apoptosis (*NOS3*), Thrombogertesis inflammation atherosclerosis74 (*ICAM1*56), and Mesangial matrix expansion (*FN1*) may suggest new drug treatment mechanisms for NAFLD (**Fig. S3 C**). We also noted that Copper might modulate the phase in Circadian entrainment through downregulation of *PRKG2* and *CALM3*, indicating a possible treatment pathway for Colorectal Cancer. Consistently, research suggested that disruptions in circadian rhythms could affect the likelihood of Colorectal Cancer75,76 (**Fig. S3 D**), highlighting the complex interplay between genetic expressions, drug effects, and disease outcomes.

Drug repurposing is considered a safer approach because it leverages existing knowledge about drug dosage, safety, and side effects, significantly enhancing the efficacy of clinical trials77. Exploring drug repurposing based on drug databases and clinical drugs used in metabolic disorders, we identified potential new uses for existing drugs in Manic Episode and T1D. The mechanism of action for drug repurposing in both conditions involves a common pathway: ECM-receptor interaction. Research indicated that the interaction between ECM and receptors can alleviate diabetic nephropathy78, and cell therapy for T1D also underscores the importance of ECM79. For T1D, the gene expression in this pathway could be potentially modulated by six drugs, with Cannabidiol, Doxorubicin, Genistein, and Propylthiouracil being of interest in the same pathway for Manic Episode. Clinical trials have suggested the potential adjunctive therapeutic effects of Cannabidiol for Manic Episode80 and T1D81. Besides, Genistein has been reported as a potential alleviator for both T1D and T2D82. Additionally, it is noteworthy that we have identified a potential association between the recently disclosed diabetes83 therapeutic drug and T1D. However, the mechanisms of action for these drugs have yet to be fully elucidated. Leveraging the existing clinical drugs combined with the causal network of multiple diseases, we identified potential therapeutic drugs for Manic Episode and T1D and substantiated the feasibility of these drugs based on literature evidence. Our research proposed the potential for repurposing existing drugs to provide new therapeutic options for complex diseases, thereby accelerating the path to effective treatments.

Inevitably, our study comes with its set of limitations, including the constrained validation of diseases within the UK Biobank and an emphasis on European populations. Such specificity may lead to a variation in findings across different ethnic groups, a reflection of the vast spectrum of genetic diversity. Moreover, while our strategy for drug repurposing in metabolic disorders is informed by initial insights, the comprehensive molecular mechanisms and the safety profiles of the proposed medications demand further in-depth investigation. Looking forward, we are optimistic that our work could help pave the way for more inclusive and holistic studies in the future, ultimately leading to the identification of treatments with universal efficacy.

## Supporting information

FigS1 [[supplements/305070_file05.pdf]](pending:yes)

FigS2 [[supplements/305070_file06.pdf]](pending:yes)

FigS3 [[supplements/305070_file07.pdf]](pending:yes)

Table1 [[supplements/305070_file08.xlsx]](pending:yes)

Table2 [[supplements/305070_file09.xlsx]](pending:yes)

Supplement Table [[supplements/305070_file10.xlsx]](pending:yes)

## Data Availability

All data generated are available online

[https://quantlifebits.shinyapps.io/mrmetabopharm/](https://quantlifebits.shinyapps.io/mrmetabopharm/) 

## Ethics approval and consent to participate

Not applicable.

## Consent for publication

All authors agree to publish.

## Availability of data and materials

We publish all the causal network and drug network information in the form of website on: [https://quantlifebits.shinyapps.io/mrmetabopharm/](https://quantlifebits.shinyapps.io/mrmetabopharm/).

## Competing interests

No conflict of interest.

## Funding

This study received financial support from National Natural Science Foundation of China grants (#82300574); National Natural Science Foundation of China (#62205065); China Postdoctoral Science Foundation Funded Project (#2022M720771). China Postdoctoral Science Foundation (#2023M730628).

## Authors’ contributions

SZ, XH designed most of the studies. DS, XS and DH carried out much of the work together. SZ, XH and XS analyzed the data. SZ and DH provided revision of the figure. XH and XS complete methodology-related content. DS provided insights into the discussion section. All authors wrote up the manuscript and approved the final manuscript.

![Figure1](http://medrxiv.org/http://medrxiv.stage.highwire.org/content/medrxiv/early/2024/03/30/2024.03.29.24305070/F1.medium.gif)

[Figure1](http://medrxiv.org/content/early/2024/03/30/2024.03.29.24305070/F1)

![Figure2](http://medrxiv.org/http://medrxiv.stage.highwire.org/content/medrxiv/early/2024/03/30/2024.03.29.24305070/F2.medium.gif)

[Figure2](http://medrxiv.org/content/early/2024/03/30/2024.03.29.24305070/F2)

![Figure3](http://medrxiv.org/http://medrxiv.stage.highwire.org/content/medrxiv/early/2024/03/30/2024.03.29.24305070/F3.medium.gif)

[Figure3](http://medrxiv.org/content/early/2024/03/30/2024.03.29.24305070/F3)

## Acknowledgements

Thanks to FinnGen and UK Biobank for providing data support.

*   Received March 29, 2024.
*   Revision received March 29, 2024.
*   Accepted March 30, 2024.


*   © 2024, Posted by Cold Spring Harbor Laboratory

The copyright holder for this pre-print is the author. All rights reserved. The material may not be redistributed, re-used or adapted without the author's permission.

## Reference

1.  Agus, A., Clément, K. & Sokol, H. Gut microbiota-derived metabolites as central regulators in metabolic disorders. Gut 70, 1174–1182, doi:10.1136/gutjnl-2020-323071 (2021).
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NjoiZ3V0am5sIjtzOjU6InJlc2lkIjtzOjk6IjcwLzYvMTE3NCI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDI0LzAzLzMwLzIwMjQuMDMuMjkuMjQzMDUwNzAuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 

2.  Blaner, W. S. Vitamin A signaling and homeostasis in obesity, diabetes, and metabolic disorders. Pharmacol Ther 197, 153–178, doi:10.1016/j.pharmthera.2019.01.006 (2019).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.pharmthera.2019.01.006&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=30703416&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305070.atom) 

3.  Ferrannini, E. & Natali, A. Essential hypertension, metabolic disorders, and insulin resistance. Am Heart J 121, 1274–1282, doi:10.1016/0002-8703(91)90433-i (1991).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/0002-8703(91)90433-I&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=2008856&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305070.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=A1991FF21600003&link_type=ISI) 

4.  Larsson, S. C., Butterworth, A. S. & Burgess, S. Mendelian randomization for cardiovascular diseases: principles and applications. Eur Heart J 44, 4913–4924, doi:10.1093/eurheartj/ehad736 (2023).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1093/eurheartj/ehad736&link_type=DOI) 

5.  Fu, Z. et al. Dyslipidemia in retinal metabolic disorders. EMBO Mol Med 11, e10473, doi:10.15252/emmm.201910473 (2019).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.15252/emmm.201910473&link_type=DOI) 

6.  Aron-Wisnewsky, J., Warmbrunn, M. V., Nieuwdorp, M. & Clément, K. Metabolism and Metabolic Disorders and the Microbiome: The Intestinal Microbiota Associated With Obesity, Lipid Metabolism, and Metabolic Health-Pathophysiology and Therapeutic Strategies. Gastroenterology 160, 573–599, doi:10.1053/j.gastro.2020.10.057 (2021).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1053/j.gastro.2020.10.057&link_type=DOI) 

7.  Gold, A. K. & Kinrys, G. Treating Circadian Rhythm Disruption in Bipolar Disorder. Curr Psychiatry Rep 21, 14, doi:10.1007/s11920-019-1001-8 (2019).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1007/s11920-019-1001-8&link_type=DOI) 

8.  Li, Y., Shao, L., Mou, Y., Zhang, Y. & Ping, Y. Sleep, circadian rhythm and gut microbiota: alterations in Alzheimer’s disease and their potential links in the pathogenesis. Gut Microbes 13, 1957407, doi:10.1080/19490976.2021.1957407 (2021).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1080/19490976.2021.1957407&link_type=DOI) 

9.  Bjornstad, P. et al. Long-Term Complications in Youth-Onset Type 2 Diabetes. N Engl J Med 385, 416–426, doi:10.1056/NEJMoa2100165 (2021).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1056/NEJMoa2100165&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305070.atom) 

10. Ducat, L., Rubenstein, A., Philipson, L. H. & Anderson, B. J. A review of the mental health issues of diabetes conference. Diabetes Care 38, 333–338, doi:10.2337/dc14-1383 (2015).
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoiZGlhY2FyZSI7czo1OiJyZXNpZCI7czo4OiIzOC8yLzMzMyI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDI0LzAzLzMwLzIwMjQuMDMuMjkuMjQzMDUwNzAuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 

11. Lega, I. C. & Lipscombe, L. L. Review: Diabetes, Obesity, and Cancer-Pathophysiology and Clinical Implications. Endocr Rev 41, doi:10.1210/endrev/bnz014 (2020).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1210/endrev/bnz014&link_type=DOI) 

12. Singh, D. D. et al. Development of Dementia in Type 2 Diabetes Patients: Mechanisms of Insulin Resistance and Antidiabetic Drug Development. Cells 11, doi:10.3390/cells11233767 (2022).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.3390/cells11233767&link_type=DOI) 

13. Coyle, C., Cafferty, F. H., Vale, C. & Langley, R. E. Metformin as an adjuvant treatment for cancer: a systematic review and meta-analysis. Ann Oncol 27, 2184–2195, doi:10.1093/annonc/mdw410 (2016).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1093/annonc/mdw410&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=27681864&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305070.atom) 

14. Calkin, C. V. et al. Treating Insulin Resistance With Metformin as a Strategy to Improve Clinical Outcomes in Treatment-Resistant Bipolar Depression (the TRIO-BD Study): A Randomized, Quadruple-Masked, Placebo-Controlled Clinical Trial. J Clin Psychiatry 83, doi:10.4088/JCP.21m14022 (2022).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.4088/JCP.21m14022&link_type=DOI) 

15. Goodwin, P. J. et al. Effect of Metformin vs Placebo on Invasive Disease-Free Survival in Patients With Breast Cancer: The MA.32 Randomized Clinical Trial. Jama 327, 1963–1973, doi:10.1001/jama.2022.6147 (2022).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1001/jama.2022.6147&link_type=DOI) 

16. Fan, R. et al. Importance of Bmal1 in Alzheimer’s disease and associated aging-related diseases: Mechanisms and interventions. Aging Cell 21, e13704, doi:10.1111/acel.13704 (2022).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1111/acel.13704&link_type=DOI) 

17. Kadowaki, T. et al. Interconnection between cardiovascular, renal and metabolic disorders: A narrative review with a focus on Japan. Diabetes Obes Metab 24, 2283–2296, doi:10.1111/dom.14829 (2022).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1111/dom.14829&link_type=DOI) 

18. Pitsillou, E., Liang, J., Hung, A. & Karagiannis, T. C. The circadian machinery links metabolic disorders and depression: A review of pathways, proteins and potential pharmacological interventions. Life Sci 265, 118809, doi:10.1016/j.lfs.2020.118809 (2021).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.lfs.2020.118809&link_type=DOI) 

19. van Tienhoven-Wind, L. J. & Dullaart, R. P. Low-normal thyroid function and the pathogenesis of common cardio-metabolic disorders. Eur J Clin Invest 45, 494–503, doi:10.1111/eci.12423 (2015).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1111/eci.12423&link_type=DOI) 

20. Sharma, B. R. & Kanneganti, T. D. NLRP3 inflammasome in cancer and metabolic diseases. Nat Immunol 22, 550–559, doi:10.1038/s41590-021-00886-5 (2021).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41590-021-00886-5&link_type=DOI) 

21. Wray, N. R., Wijmenga, C., Sullivan, P. F., Yang, J. & Visscher, P. M. Common Disease Is More Complex Than Implied by the Core Gene Omnigenic Model. Cell 173, 1573–1580, doi:10.1016/j.cell.2018.05.051 (2018).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.cell.2018.05.051&link_type=DOI) 

22. Cole, J. B. & Florez, J. C. Genetics of diabetes mellitus and diabetes complications. Nat Rev Nephrol 16, 377–390, doi:10.1038/s41581-020-0278-5 (2020).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41581-020-0278-5&link_type=DOI) 

23. Ference, B. A. Interpreting the Clinical Implications of Drug-Target Mendelian Randomization Studies. J Am Coll Cardiol 80, 663–665, doi:10.1016/j.jacc.2022.06.007 (2022).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.jacc.2022.06.007&link_type=DOI) 

24. Black, J. Drugs from emasculated hormones: the principle of syntopic antagonism. Science 245, 486–493, doi:10.1126/science.2569237 (1989).
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6Mzoic2NpIjtzOjU6InJlc2lkIjtzOjEyOiIyNDUvNDkxNy80ODYiO3M6NDoiYXRvbSI7czo1MDoiL21lZHJ4aXYvZWFybHkvMjAyNC8wMy8zMC8yMDI0LjAzLjI5LjI0MzA1MDcwLmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 

25. Ashburn, T. T. & Thor, K. B. Drug repositioning: identifying and developing new uses for existing drugs. Nat Rev Drug Discov 3, 673–683, doi:10.1038/nrd1468 (2004).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/nrd1468&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=15286734&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305070.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000223005000018&link_type=ISI) 

26. Mikubo, M., Inoue, Y., Liu, G. & Tsao, M. S. Mechanism of Drug Tolerant Persister Cancer Cells: The Landscape and Clinical Implication for Therapy. J Thorac Oncol 16, 1798–1809, doi:10.1016/j.jtho.2021.07.017 (2021).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.jtho.2021.07.017&link_type=DOI) 

27. Pushpakom, S. et al. Drug repurposing: progress, challenges and recommendations. Nat Rev Drug Discov 18, 41–58, doi:10.1038/nrd.2018.168 (2019).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/nrd.2018.168&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=30310233&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305070.atom) 

28. Sweeney, P., Gimenez, L. E., Hernandez, C. C. & Cone, R. D. Targeting the central melanocortin system for the treatment of metabolic disorders. Nat Rev Endocrinol 19, 507–519, doi:10.1038/s41574-023-00855-y (2023).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41574-023-00855-y&link_type=DOI) 

29. Ruan, W., Yuan, X. & Eltzschig, H. K. Circadian rhythm as a therapeutic target. Nat Rev Drug Discov 20, 287–307, doi:10.1038/s41573-020-00109-w (2021).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41573-020-00109-w&link_type=DOI) 

30. Cho, N. H. et al. IDF Diabetes Atlas: Global estimates of diabetes prevalence for 2017 and projections for 2045. Diabetes Res Clin Pract 138, 271–281, doi:10.1016/j.diabres.2018.02.023 (2018).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.diabres.2018.02.023&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=29496507&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305070.atom) 

31. Lin, J., Zhou, J. & Xu, Y. Potential drug targets for multiple sclerosis identified through Mendelian randomization analysis. Brain 146, 3364–3372, doi:10.1093/brain/awad070 (2023).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1093/brain/awad070&link_type=DOI) 

32. Kurki, M. I. et al. FinnGen provides genetic insights from a well-phenotyped isolated population. Nature 613, 508–518, doi:10.1038/s41586-022-05473-8 (2023).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41586-022-05473-8&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=36653562&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305070.atom) 

33. Evangelista, J. E. et al. SigCom LINCS: data and metadata search engine for a million gene expression signatures. Nucleic Acids Res 50, W697–w709, doi:10.1093/nar/gkac328 (2022).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1093/nar/gkac328&link_type=DOI) 

34. Pujar, M., Vastrad, B., Kavatagimath, S., Vastrad, C. & Kotturshetti, S. Identification of candidate biomarkers and pathways associated with type 1 diabetes mellitus using bioinformatics analysis. Sci Rep 12, 9157, doi:10.1038/s41598-022-13291-1 (2022).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41598-022-13291-1&link_type=DOI) 

35. Karakaya, C. et al. Further delineation of familial polycystic ovary syndrome (PCOS) via whole-exome sequencing: PCOS-related rare FBN3 and FN1 gene variants are identified. J Obstet Gynaecol Res 48, 1202–1211, doi:10.1111/jog.15187 (2022).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1111/jog.15187&link_type=DOI) 

36. Perry, N. A. et al. Loss of giant obscurins promotes breast epithelial cell survival through apoptotic resistance. Faseb j 26, 2764–2775, doi:10.1096/fj.12-205419 (2012).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1096/fj.12-205419&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=22441987&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305070.atom) 

37. Shriver, M. et al. Loss of giant obscurins from breast epithelium promotes epithelial-to-mesenchymal transition, tumorigenicity and metastasis. Oncogene 34, 4248–4259, doi:10.1038/onc.2014.358 (2015).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/onc.2014.358&link_type=DOI) 

38. Geng, X. et al. lnc-MAP3K13-7:1 Inhibits Ovarian GC Proliferation in PCOS via DNMT1 Downregulation-Mediated CDKN1A Promoter Hypomethylation. Mol Ther 29, 1279–1293, doi:10.1016/j.ymthe.2020.11.018 (2021).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.ymthe.2020.11.018&link_type=DOI) 

39. Zhang, L., Wang, F., Li, D., Yan, Y. & Wang, H. Transferrin receptor-mediated reactive oxygen species promotes ferroptosis of KGN cells via regulating NADPH oxidase 1/PTEN induced kinase 1/acyl-CoA synthetase long chain family member 4 signaling. Bioengineered 12, 4983–4994, doi:10.1080/21655979.2021.1956403 (2021).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1080/21655979.2021.1956403&link_type=DOI) 

40. Kumariya, S., Ubba, V., Jha, R. K. & Gayen, J. R. Autophagy in ovary and polycystic ovary syndrome: role, dispute and future perspective. Autophagy 17, 2706–2733, doi:10.1080/15548627.2021.1938914 (2021).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1080/15548627.2021.1938914&link_type=DOI) 

41. Gu, H. F., Ma, J., Gu, K. T. & Brismar, K. Association of intercellular adhesion molecule 1 (ICAM1) with diabetes and diabetic nephropathy. Front Endocrinol (Lausanne*)* 3, 179, doi:10.3389/fendo.2012.00179 (2012).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.3389/fendo.2012.00179&link_type=DOI) 

42. Katarina, K. et al. HLA, NFKB1 and NFKBIA gene polymorphism profile in autoimmune diabetes mellitus patients. Exp Clin Endocrinol Diabetes 115, 124–129, doi:10.1055/s-2007-949589 (2007).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1055/s-2007-949589&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=17318773&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305070.atom) 

43. Katsarou, S. D. et al. Detyrosinated α-Tubulin, Vimentin and PD-L1 in Circulating Tumor Cells (CTCs) Isolated from Non-Small Cell Lung Cancer (NSCLC) Patients. J Pers Med 12, doi:10.3390/jpm12020154 (2022).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.3390/jpm12020154&link_type=DOI) 

44. Marshall, N. A. et al. Immunotherapy with PI3K inhibitor and Toll-like receptor agonist induces IFN-γ+IL-17+ polyfunctional T cells that mediate rejection of murine tumors. Cancer Res 72, 581–591, doi:10.1158/0008-5472.Can-11-0307 (2012).
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NjoiY2FucmVzIjtzOjU6InJlc2lkIjtzOjg6IjcyLzMvNTgxIjtzOjQ6ImF0b20iO3M6NTA6Ii9tZWRyeGl2L2Vhcmx5LzIwMjQvMDMvMzAvMjAyNC4wMy4yOS4yNDMwNTA3MC5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=) 

45. Fattakhov, N. et al. Haplotype analysis of endothelial nitric oxide synthase (NOS3) genetic variants and metabolic syndrome in healthy subjects and schizophrenia patients. Int J Obes (Lond*)* 42, 2036–2046, doi:10.1038/s41366-018-0124-z (2018).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41366-018-0124-z&link_type=DOI) 

46. Sumitomo, A. et al. Ulk2 controls cortical excitatory-inhibitory balance via autophagic regulation of p62 and GABAA receptor trafficking in pyramidal neurons. Hum Mol Genet 27, 3165–3176, doi:10.1093/hmg/ddy219 (2018).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1093/hmg/ddy219&link_type=DOI) 

47. Bai, X., Fermandez, E., Gould, G. & Strong, R. Homozygous Deletion of Glutathione Peroxidase 1 and Aldehyde Dehydrogenase 1a1 Genes Is Not Associated with Schizophrenia-Like Behavior in Mice. J Biochem Pharmacol Res 1, 228–235 (2013).
    
    

48. Bowman, L. et al. Effects of Aspirin for Primary Prevention in Persons with Diabetes Mellitus. N Engl J Med 379, 1529–1539, doi:10.1056/NEJMoa1804988 (2018).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1056/NEJMoa1804988&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305070.atom) 

49. Liao, M. et al. Effects and plasma proteomic analysis of GLP-1RA versus CPA/EE, in combination with metformin, on overweight PCOS women: a randomized controlled trial. Endocrine 83, 227–241, doi:10.1007/s12020-023-03487-4 (2024).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1007/s12020-023-03487-4&link_type=DOI) 

50. Butler, A. E., Moin, A. S. M., Sathyapalan, T. & Atkin, S. L. A Cross-Sectional Study of Protein Changes Associated with Dementia in Non-Obese Weight Matched Women with and without Polycystic Ovary Syndrome. Int J Mol Sci 25, doi:10.3390/ijms25042409 (2024).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.3390/ijms25042409&link_type=DOI) 

51. Sathyapalan, T. & Atkin, S. L. Recent advances in cardiovascular aspects of polycystic ovary syndrome. Eur J Endocrinol 166, 575–583, doi:10.1530/eje-11-0755 (2012).
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6MzoiZWplIjtzOjU6InJlc2lkIjtzOjk6IjE2Ni80LzU3NSI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDI0LzAzLzMwLzIwMjQuMDMuMjkuMjQzMDUwNzAuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 

52. Armanini, D., Boscaro, M., Bordin, L. & Sabbadin, C. Controversies in the Pathogenesis, Diagnosis and Treatment of PCOS: Focus on Insulin Resistance, Inflammation, and Hyperandrogenism. Int J Mol Sci 23, doi:10.3390/ijms23084110 (2022).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.3390/ijms23084110&link_type=DOI) 

53. Taglieri, L. et al. Thyroid hormone regulates fibronectin expression through the activation of the hypoxia inducible factor 1. Biochem Biophys Res Commun 493, 1304–1310, doi:10.1016/j.bbrc.2017.09.169 (2017).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.bbrc.2017.09.169&link_type=DOI) 

54. Astrup, A. S. et al. Markers of endothelial dysfunction and inflammation in type 1 diabetic patients with or without diabetic nephropathy followed for 10 years: association with mortality and decline of glomerular filtration rate. Diabetes Care 31, 1170–1176, doi:10.2337/dc07-1960 (2008).
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoiZGlhY2FyZSI7czo1OiJyZXNpZCI7czo5OiIzMS82LzExNzAiO3M6NDoiYXRvbSI7czo1MDoiL21lZHJ4aXYvZWFybHkvMjAyNC8wMy8zMC8yMDI0LjAzLjI5LjI0MzA1MDcwLmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 

55. Lin, J. et al. Inflammation and progressive nephropathy in type 1 diabetes in the diabetes control and complications trial. Diabetes Care 31, 2338–2343, doi:10.2337/dc08-0277 (2008).
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoiZGlhY2FyZSI7czo1OiJyZXNpZCI7czoxMDoiMzEvMTIvMjMzOCI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDI0LzAzLzMwLzIwMjQuMDMuMjkuMjQzMDUwNzAuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 

56. Ito, S., Yukawa, T., Uetake, S. & Yamauchi, M. Serum intercellular adhesion molecule-1 in patients with nonalcoholic steatohepatitis: comparison with alcoholic hepatitis. Alcohol Clin Exp Res 31, S83–87, doi:10.1111/j.1530-0277.2006.00292.x (2007).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1111/j.1530-0277.2006.00292.x&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=17331172&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305070.atom) 

57. Twohig, D. & Nielsen, H. M. α-synuclein in the pathophysiology of Alzheimer’s disease. Mol Neurodegener 14, 23, doi:10.1186/s13024-019-0320-x (2019).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1186/s13024-019-0320-x&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305070.atom) 

58. Beatino, M. F. et al. α-synuclein as an emerging pathophysiological biomarker of Alzheimer’s disease. Expert Rev Mol Diagn 22, 411–425, doi:10.1080/14737159.2022.2068952 (2022).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1080/14737159.2022.2068952&link_type=DOI) 

59. Butler, A. E., Moin, A. S. M., Sathyapalan, T. & Atkin, S. L. A Cross-Sectional Study of Alzheimer-Related Proteins in Women with Polycystic Ovary Syndrome. Int J Mol Sci 25, doi:10.3390/ijms25021158 (2024).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.3390/ijms25021158&link_type=DOI) 

60. Kellar, D. & Craft, S. Brain insulin resistance in Alzheimer’s disease and related disorders: mechanisms and therapeutic approaches. Lancet Neurol 19, 758–766, doi:10.1016/s1474-4422(20)30231-3 (2020).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/s1474-4422(20)30231-3&link_type=DOI) 

61. Wang, J., Wang, K., Liu, W., Cai, Y. & Jin, H. m6A mRNA methylation regulates the development of gestational diabetes mellitus in Han Chinese women. Genomics 113, 1048–1056, doi:10.1016/j.ygeno.2021.02.016 (2021).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.ygeno.2021.02.016&link_type=DOI) 

62. Cedarbaum, J. M., Breck, L., Kutt, H. & McDowell, F. H. Controlled-release levodopa/carbidopa. I. Sinemet CR3 treatment of response fluctuations in Parkinson’s disease. Neurology 37, 233–241, doi:10.1212/wnl.37.2.233 (1987).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1212/wnl.37.2.233&link_type=DOI) 

63. Celebi Sozener, Z., et al. Epithelial barrier hypothesis: Effect of the external exposome on the microbiome and epithelial barriers in allergic disease. Allergy 77, 1418–1449, doi:10.1111/all.15240 (2022).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1111/all.15240&link_type=DOI) 

64. Escobar-Morreale, H. F. & Roldán-Martín, M. B. Type 1 Diabetes and Polycystic Ovary Syndrome: Systematic Review and Meta-analysis. Diabetes Care 39, 639–648, doi:10.2337/dc15-2577 (2016).
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoiZGlhY2FyZSI7czo1OiJyZXNpZCI7czo4OiIzOS80LzYzOSI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDI0LzAzLzMwLzIwMjQuMDMuMjkuMjQzMDUwNzAuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 

65. Vella, K., Vella, S., Savona-Ventura, C. & Vassallo, J. Thyroid dysfunction in pregnancy - a retrospective observational analysis of a Maltese cohort. BMC Pregnancy Childbirth 22, 941, doi:10.1186/s12884-022-05266-x (2022).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1186/s12884-022-05266-x&link_type=DOI) 

66. Lin, Y. H., Chen, K. J., Peng, Y. S., Chen, P. C. & Yang, Y. H. Type 1 diabetes impairs female fertility even before it is diagnosed. Diabetes Res Clin Pract 143, 151–158, doi:10.1016/j.diabres.2018.07.010 (2018).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.diabres.2018.07.010&link_type=DOI) 

67. Oulis, P. et al. Oxcarbazepine as monotherapy of acute mania in insufficiently controlled type-1 diabetes mellitus: a case-report. Ann Gen Psychiatry 6, 25, doi:10.1186/1744-859x-6-25 (2007).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1186/1744-859X-6-25&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=17922898&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305070.atom) 

68. Su, Y. et al. Radix Rehmanniae Praeparata (Shu Dihuang) exerts neuroprotective effects on ICV-STZ-induced Alzheimer’s disease mice through modulation of INSR/IRS-1/AKT/GSK-3β signaling pathway and intestinal microbiota. Front Pharmacol 14, 1115387, doi:10.3389/fphar.2023.1115387 (2023).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.3389/fphar.2023.1115387&link_type=DOI) 

69. Saito, T. et al. Single App knock-in mouse models of Alzheimer’s disease. Nat Neurosci 17, 661–663, doi:10.1038/nn.3697 (2014).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/nn.3697&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=24728269&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305070.atom) 

70. Zenaro, E. et al. Neutrophils promote Alzheimer’s disease-like pathology and cognitive decline via LFA-1 integrin. Nat Med 21, 880–886, doi:10.1038/nm.3913 (2015).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/nm.3913&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=26214837&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305070.atom) 

71. Shan, D., Haroutunian, V., Meador-Woodruff, J. H. & McCullumsmith, R. E. Expression of equilibrative nucleoside transporter type 1 protein in elderly patients with schizophrenia. Neuroreport 23, 224–227, doi:10.1097/WNR.0b013e3283500987 (2012).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1097/WNR.0b013e3283500987&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=22314683&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305070.atom) 

72. Liu, J., Li, L., Zhang, B. & Xu, Z. P. MnO(2)-shelled Doxorubicin/Curcumin nanoformulation for enhanced colorectal cancer chemo-immunotherapy. J Colloid Interface Sci 617, 315–325, doi:10.1016/j.jcis.2022.02.132 (2022).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.jcis.2022.02.132&link_type=DOI) 

73. Fourie, C., du Plessis, M., Mills, J. & Engelbrecht, A. M. The effect of HIF-1α inhibition in breast cancer cells prior to doxorubicin treatment under conditions of normoxia and hypoxia. Exp Cell Res 419, 113334, doi:10.1016/j.yexcr.2022.113334 (2022).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.yexcr.2022.113334&link_type=DOI) 

74. Balta, S. Atherosclerosis and Non-Alcoholic Fatty Liver Disease. Angiology 73, 701–711, doi:10.1177/00033197221091317 (2022).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1177/00033197221091317&link_type=DOI) 

75. Huisman, S. A. et al. Colorectal liver metastases with a disrupted circadian rhythm phase shift the peripheral clock in liver and kidney. Int J Cancer 136, 1024–1032, doi:10.1002/ijc.29089 (2015).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1002/ijc.29089&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=25045881&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305070.atom) 

76. Li, X. M. et al. A circadian clock transcription model for the personalization of cancer chronotherapy. Cancer Res 73, 7176–7188, doi:10.1158/0008-5472.Can-13-1528 (2013).
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NjoiY2FucmVzIjtzOjU6InJlc2lkIjtzOjEwOiI3My8yNC83MTc2IjtzOjQ6ImF0b20iO3M6NTA6Ii9tZWRyeGl2L2Vhcmx5LzIwMjQvMDMvMzAvMjAyNC4wMy4yOS4yNDMwNTA3MC5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=) 

77. Abbruzzese, C. et al. Drug repurposing for the treatment of glioblastoma multiforme. J Exp Clin Cancer Res 36, 169, doi:10.1186/s13046-017-0642-x (2017).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1186/s13046-017-0642-x&link_type=DOI) 

78. Wang, J. et al. Low molecular weight fucoidan alleviates diabetic nephropathy by binding fibronectin and inhibiting ECM-receptor interaction in human renal mesangial cells. Int J Biol Macromol 150, 304–314, doi:10.1016/j.ijbiomac.2020.02.087 (2020).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.ijbiomac.2020.02.087&link_type=DOI) 

79. Zamboni, F. & Collins, M. N. Cell based therapeutics in type 1 diabetes mellitus. Int J Pharm 521, 346–356, doi:10.1016/j.ijpharm.2017.02.063 (2017).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.ijpharm.2017.02.063&link_type=DOI) 

80. Ashton, C. H., Moore, P. B., Gallagher, P. & Young, A. H. Cannabinoids in bipolar affective disorder: a review and discussion of their therapeutic potential. J Psychopharmacol 19, 293–300, doi:10.1177/0269881105051541 (2005).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1177/0269881105051541&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=15888515&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305070.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000228999800012&link_type=ISI) 

81. Lehmann, C. et al. Experimental cannabidiol treatment reduces early pancreatic inflammation in type 1 diabetes. Clin Hemorheol Microcirc 64, 655–662, doi:10.3233/ch-168021 (2016).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.3233/ch-168021&link_type=DOI) 

82. Jain, R., Bolch, C., Al-Nakkash, L. & Sweazea, K. L. Systematic review of the impact of genistein on diabetes-related outcomes. Am J Physiol Regul Integr Comp Physiol 323, R279–r288, doi:10.1152/ajpregu.00236.2021 (2022).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1152/ajpregu.00236.2021&link_type=DOI) 

83. Ostrominski, J. W. & Aroda, V. R. Aspirin and diabetes prevention among healthy older adults-practice-changing or hypothesis-generating? Lancet Diabetes Endocrinol 12, 84–85, doi:10.1016/s2213-8587(23)00363-7 (2024).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/s2213-8587(23)00363-7&link_type=DOI)