Author ORCID Identifier

https://orcid.org/0000-0002-8207-3822

Semester

Summer

Date of Graduation

2024

Document Type

Dissertation

Degree Type

PhD

College

School of Medicine

Department

Not Listed

Committee Chair

John Hollander

Committee Co-Chair

Paul Chantler

Committee Member

Paul Chantler

Committee Member

Ivan Martinez

Committee Member

Emidio Pistilli

Committee Member

Aaron Robart

Abstract

Introduction: The leading cause of mortality in patients with diabetes mellitus is heart failure. When mitochondrial health and function is disrupted, cardiac contractile function is compromised. Therefore, understanding mitochondrial regulation may benefit predicting and countering diabetic cardiomyopathy. MicroRNAs (miRNAs) play a crucial role in diabetic cardiac mitochondrial protein expression. Long noncoding RNAs (lncRNAs) have been shown to regulate miRNAs, but not in the mitochondrion. Additionally, it has yet to be confirmed if lncRNAs utilize the same mechanisms for mitochondrial import that miRNAs use, or if lncRNA presence in the mitochondrion fluctuates in diabetic mitochondria like miRNA presence does. The purpose of this work is to understand how the cardiac mitochondrial transcriptome is changed by diabetes mellitus, specifically how changes in lncRNA passage into the mitochondrion alters bioenergetic and cardiac function.

Methods and Results: Using both mouse and human cardiac tissue, we first examined the presence of lncRNAs within mitochondria using next-generation sequencing. After identifying mitochondrially-imported lncRNAs, we performed crosslinked immunoprecipitation-based sequencing for polynucleotide phosphorylase (PNPase), a known importer for noncoding RNAs including miRNAs. PNPase-bound lncRNA sequences were evaluated using machine learning algorithms to identify features that enhanced likelihood of binding. Those features were integrated into an artificial construct with high interaction potential for PNPase, which was experimentally verified using gel-shift and fluorescence assays. Next-generation sequencing was repeated on mitochondrial isolate from diabetic mouse and human cardiac tissue, and differences in lncRNA mitochondrial presence were assessed. Several lncRNAs showed a significant change in mitochondrial abundance, including metastasis associated lung adenocarcinoma transcript 1 (Malat1). Small non-coding RNA sequencing revealed that presence of miR-320a, a miRNA binding target for Malat1, was also altered in diabetic mitochondria. Both non-coding RNAs were proven to alter mitochondrial function as well as co-precipitate using in vitro experimentation. In vivo echocardiography of a Malat1 knockout mouse model on a high-fat diet presented with exacerbated systolic and diastolic dysfunction. Malat1 loss increased recruitment of miR-320a and its mitochondrial genome-encoded target mRNA MT-ND1 to the RNA induced silencing complex (RISC), suggesting that disruption of Malat1 presence in mitochondria increased miR-320a availability to downregulate MT-ND1 and ultimately lead to cardiac dysfunction.

Conclusions: The work summarized in the preceding experiments exposes lncRNAs as a major and dynamic component of the mitochondrial transcriptome that can act as an axis of regulation in coordination with miRNAs. Further, disruption of this axis can result in pathological changes similar to what is observed in diabetic cardiomyopathy. These studies also suggest that lncRNA components could be utilized to enhance mitochondrial import and targeting of miRNAs.

Download

Share

COinS