University of Southern California Dissertations and Theses

The intersection of mitochondrial biology and cancer: insights from mitochondrial microproteins and mtDNA alterations

(USC Thesis Other)

The intersection of mitochondrial biology and cancer: insights from mitochondrial microproteins and mtDNA alterations

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content The Intersection of Mitochondrial Biology and Cancer: Insights from Mitochondrial
Microproteins and mtDNA Alterations
by
Melanie Kristine Flores
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(MOLECULAR BIOLOGY)
August 2024

ii
ACKNOWLEDGMENTS
The completion of this dissertation would not have been possible without the
support and encouragement of many individuals to whom I am deeply grateful.
I would like to first acknowledge the members of my committee for their feedback
and consideration during this process. I would like to express my appreciation to my
advisor, Dr. Pinchas Cohen, for allowing me the independence to explore my own ideas
and learn through experience. His mentorship has been instrumental in shaping my
development as a researcher and scholar. I would like to thank the members of the
Cohen lab: Dr. Kelvin Yen, Hemal Mehta, Dr. Hiroshi Kumagai, Ricardo Ramirez II, Dr.
Junxiang Wan, Dr. Su Jeong Kim, and Dr. Brendan Miller. Thank you for making me a
better scientist, for listening and helping me overcome my failures, for your
encouragement (and tough love), and of course for your great company during all of our
Manas lab lunches.
I would like to specifically thank my lab mate, emotional support human, and
wonderful friend Ana Silverstein. You are a one of a kind scientist and friend, and I am
beyond grateful to have had you by my side for the past five years. Thank you so much
for your support in getting me to the finish line.
I would like to thank my family for their blind support in me during this process.
Even though you have no idea “what” I’m doing, you make it easier to find the “why”
behind it all.
Thank you to my dog Floyd, you will never read this, but your presence by my
side and unconditional love for me (with or without a PhD) makes each day a bit
brighter.

iii
Finally, to my husband Chris, you are my best friend, my rock, and my most
favorite human being. Without your support, (and lots of home cooked meals), I quite
literally would not have finished this. Knowing that you are my number one supporter
has given me everything I need to succeed.

iv
Chapter 1, in full, is a review article currently being prepared for the submission of the
material. The dissertation author was the primary investigator and author of this
material. Other authors include Ana Silverstein, Zeferino Reyna, and Pinchas Cohen.
Chapter 2, in full, has been submitted for publication of the material as it may appear in
Mitochondrial DNA copy number assessment is a potent predictor for prostate cancer in
White but not Black Individuals, Cancer Epidemiology, Biomarkers & Prevention. The
dissertation author was the primary investigator and author of this material. Other
authors include Jessica L. Janes, Mirajul Islam, Junxiang Wan, Jiali Liu, R. Renee
Reams, Li-Ming Su, Kelvin Yen, Hemal H. Mehta, Allison L. Reagan, Lauren E. Howard,
Emily Wiggins, Adriana C. Vidal, Stephen J. Freedland, and Pinchas Cohen.
Chapter 3, in full, is currently being prepared for the submission of the material. The
dissertation author was the primary investigator and author of this material. Other
authors include Thalida Arpawong, Susanne Henning, Pei Liang, Brendan Miller, Kelvin
Yen, Hemal Mehta, Hiroshi Kumagai, Junxiang Wan, William J. Aronson and Pinchas
Cohen

v
TABLE OF CONTENTS
Acknowledgements…………………………………………………………………………..….ii
List of Tables……………………………………………………………………………………vii
List of Figures………………………………………………………………………………….viii
Abstract …………………….……………………………………………………………………ix
Chapter 1: Introduction: The Mitochondrion and Mitochondrial Encoded Microproteins
are Regulators of Cancer Biology……………………………………………………………..1
ABSTRACT………………………………………………………………………………1
1.1 Mitochondrial Regulation of Cancer………………………………………………2
1.2 Mitochondrial Microproteins……………………………………………………….4
1.3 Mitochondrial microproteins are regulators of mitochondrial function………...6
1.4 Humanin involvement in cancer……………………………………………..……7
1.5 SHLPS and Potential Cancer Connections………………………………..…….8
1.6 Conclusions and Future Directions…………………………………...…………10
1.7 Figures……………………………………………………………………………...12
Chapter 2: Mitochondrial DNA Copy Number Assessment is a Potent Predictor for
Prostate Cancer………………………………………………………………………………..13
ABSTRACT…………………………………………………………………………….13
2.1 Introduction………………………………………………………………………...15
2.2 Materials and Methods……………………………………………………………17
2.3 Results……………………………………………………………………………...20
2.4 Discussion………………………………………………………………………….23
2.5 Tables and Figures………………………………………………………………..29

vi
Chapter 3: A Somatic Mutation in the Novel Mitochondrial Microprotein SCREAM is
Implicated in Cancer Biology…………………………………………………………………35
ABSTRACT…………………………………………………………………………….35
3.1 Introduction………………………………………………………………………...36
3.2 Materials and Methods……………………………………………………………37
3.3 Results……………………………………………………………………………...44
3.4 Discussion………………………………………………………………………….48
3.5 Figures……………………………………………………………………………...51
Future Directions………………………………………………………………………………55
References……………………………………………………………………………………..67
Supplemental Tables and Figures…………………………………………………………...66

vii
LIST OF TABLES
Chapter 2
Table 2-1: Demographic and clinical characteristics at the time of biopsy
stratified by prostate cancer status across all individuals and by race…………………..29
Table 2-2. Univariable and multivariable logistic regression models for
testing the association between mtDNA copy number and risk of overall
prostate cancer by race……………………………………………………………………….30
Table 2-3. Comparing model performance among multivariable logistic
regression models stratified by race…………………………………………………………31
Supplemental Tables
Supplemental Table 2-1. Demographic and clinical characteristics at
the time of biopsy stratified by prostate cancer grade for all individuals
and by race……………………………………………………………………………………..67
Supplemental Table 2-2. Univariable and multivariable multinomial
logistic regression models for testing the association between mtDNA
copy number and prostate cancer grade stratified by race……………………………….68

viii
LIST OF FIGURES
Chapter 1
Figure 1-1: Mitochondria and MDPs regulate aspects of cancer biology………………..12
Chapter 2
Figure 2-1: Comparison of the distribution of Mitochondrial DNA
copy number levels between cases (individuals with prostate cancer)
and controls (individuals without prostate cancer) for Black
and White individuals………………………………………………………………………….32
Figure 2-2: Prostate cancer classification by Mitochondrial DNA
copy number levels between cases (individuals with prostate cancer)
and controls (individuals without prostate cancer) for Black
and White individuals………………………………………………………………………….32
Figure 2-3. Comparing receiver operating characteristic (ROC) curves
and area under the curve (AUC) comparing multivariable logistic
regression models with and without mtDNA………………………………………………..34
Chapter 3
Figure 3-1: The somatic tumor mitochondrial SNV T72C changes
the smORF of the mitochondrial microprotein SCREAM………………………………….51
Figure 3-2: Endogenous Detection of SCREAM…………………………………………...52
Figure 3-2: The SCREAM variant Y18H exhibits pro-oncogenic effects………………...53
Figure 3-4: Tumor promoting effects of Y18H in mouse prostate cancer
models…………………………………………………………………………………………..54
Supplemental Figures
Supplemental Figure 2-1: Comparison of control Mitochondrial DNA copy
number levels in White and Black individuals………………………………………………66
Supplemental Figure 2-2: Comparison of the distribution of Mitochondrial
DNA copy number by prostate cancer grade for Black and White Individuals………….69
Supplemental Figure 3-1: Endogenous Detection of SCREAM in LnCaP cells………...70

ix
ABSTRACT
The intricate relationship between mitochondrial biology and cancer is a
burgeoning field of study with significant implications for understanding and treating the
disease. This dissertation explores the role of mitochondrial microproteins and
mitochondrial DNA (mtDNA) alterations in cancer biology. In chapter one, we begin by
reviewing the emerging class of mitochondrial-encoded microproteins, highlighting their
regulatory roles in apoptosis, reactive oxygen species (ROS) production, and
mitochondrial metabolism, with a focus on humanin and small humanin-like peptides
(SHLPs). These microproteins demonstrate potential as key modulators of cancer,
warranting further investigation into their mechanisms and therapeutic applications.
In chapter two, our research delves into racial disparities in prostate cancer by
examining mtDNA copy number as a biomarker. Our research reveals that elevated
mtDNA copy number is a potent predictor of prostate cancer in White individuals but not
in Black individuals, who exhibit higher baseline mtDNA levels. These findings suggest
that mtDNA copy number may contribute to racial health disparities in prostate cancer
and could inform racially tailored diagnostic approaches.
Finally, in chapter three, we identify a novel somatic mutation in the mitochondrial
microprotein SCREAM, linked to cancer progression. This mutation, a single nucleotide
variant in the D-loop region of mtDNA, results in an amino acid change that enhances
the pro-oncogenic properties of the SCREAM microprotein. The discovery of this
mutation and its functional consequences further underscores the potential of
mitochondrial microproteins as novel therapeutic targets across multiple cancer types.

x
Collectively, this dissertation underscores the critical role of mitochondrial biology
in cancer, with a focus on mitochondrial microproteins and mtDNA alterations as
promising avenues for future research and therapeutic development.

1
CHAPTER 1: Introduction: The Mitochondrion and Mitochondrial Encoded
Microproteins are Regulators of Cancer Biology
ABSTRACT:
Here we review the intricate relationship between mitochondrial function and cancer,
while specifically highlighting the role of an emerging class of microproteins derived from
the mitochondrion. Mitochondria play a crucial role in regulating metabolic reprogramming
as well as reactive oxygen species (ROS)-mediated signaling. Further, mutations in
mitochondrial DNA (mtDNA) are associated with increased oxidative stress and
alterations in mtDNA are associated across many different cancer types. Mitochondrialencoded microproteins regulate mitochondrial function and may play a further role in
regulating aspects of cancer biology. In particular, the mitochondrial-encoded
microproteins humanin and small humanin-like peptides (SHLPs), emerge as important
regulators with diverse effects on apoptosis, ROS production, and mitochondrial
metabolism. This concise review underscores the promising potential of mitochondrial
microproteins as key players in cancer regulation, urging further investigation into their
precise mechanisms and therapeutic implications.

2
1.1 Mitochondrial Regulation of Cancer
The mitochondrion, often referred to as the powerhouse of the cell, is an organelle with
diverse capabilities beyond just energy production. Mitochondria are key regulators of
integral cellular processes and proper mitochondrial function is essential for maintaining
cellular homeostasis both within healthy and cancerous cells. Indeed, mitochondrial
function and modulation of key signaling pathways plays a large role in the regulation of
cancer cells. In this review, we will discuss the multifaceted role of mitochondria in cancer
and, in particular, evaluate how a class of microproteins derived from the mitochondria
may play a role in the regulation of tumorigenicity and pathogenesis.
1.1.1 Metabolic reprogramming of cancer cells
Mitochondria and mitochondrial dysfunction have been at the center of tumor biology
since Otto Warburg’s observation that cancer cells produce elevated levels of lactate in
the presence of oxygen, a state termed aerobic glycolysis [1, 2]. However, evidence
suggests that this shift towards aerobic glycolysis is not the result of mitochondrial
dysfunction but rather a metabolic reprogramming that is required to support rapid cancer
cell proliferation and tumor progression [3]. Indeed, the reprogramming of cancer cell
metabolism towards aerobic glycolysis has many benefits in the context of tumor
progression and survival such as; increased survival in hypoxic conditions, tumor immune
evasion, and microenvironment conditioning [4-7]. However, the notion that cancer cells
rely exclusively on aerobic glycolysis has come under scrutiny in recent years and it is
now clear that cancer cells also rely on oxidative phosphorylation (OXPHOS) for tumor
progression. For example, mitochondrial biogenesis and respiration induced by

3
peroxisome proliferator-activated receptor-gamma coactivator (PGC-1α) has been shown
to increase tumor metastasis and cancer cell motility [8]. Additionally, enhanced reliance
on OXPHOS has been shown to contribute to drug resistance in cancer cells [9]. These
studies suggest that all cancer cells do not rely exclusively on aerobic glycolysis for
survival, and instead a more precise network of metabolic reprogramming is at play. Thus,
mitochondrial metabolic reprogramming and bioenergetic plasticity is vital to the health
and growth of cancer cells.
1.1.2 Mitochondrial mutations in cancer
Mitochondria are unique organelles in that they maintain their own genome containing 37
genes which encode for two rRNAs, 22tRNAs, and 13 large proteins that are subunits of
complexes involved in OXPHOS [10]. Alterations and mutations in the mitochondrial
genome (mtDNA) have been associated with various different tumors and across cancer
types [11]. These mutations have been associated with altered OXPHOS and oncogenic
signaling [12-14]. For example, in prostate cancer, mutations within mtDNA have been
shown to occur early in prostate cancer progression and are associated with increased
tumor aggressivity [15-17]. Similarly, studies have identified an increase in somatic
mtDNA mutations in leukemia as well as colorectal and pancreatic cancers [18-20]. While
the presence of mtDNA mutations is well established in numerous cancer models, precise
mechanisms for how these mutations impact tumor pathogenesis and cancer induction
have yet to be established.

4
1.1.3 Mitochondrial ROS and ROS mediated signaling
Mitochondria are important drivers of cellular communication and regulate pathways
involved in proliferation, apoptosis, and cellular stress responses. Notably, mitochondria
are sources of reactive oxygen species (ROS) within tumors and are involved in ROSmediated signaling within cancer cells [21]. Mitochondrial ROS are crucial regulators of
cancer development and overproduction of ROS/ROS signaling enhances tumorigenesis
by promoting proliferation, contributing to increased survival, and inducing genomic
instability that contributes to metastasis [22, 23]. More specifically, ROS signaling has
been shown to activate pathways such as ERK 1/2 and Akt, leading to increased cell
survival, growth and motility [24-26]. Additionally, mutations in both nuclear and mtDNA
can lead to impaired OXPHOS and result in further mitochondrial ROS production, which
in turn can cause further oxidative DNA damage and overall genomic instability resulting
in increased metastasis [13, 27, 28]. This process leads to a “vicious cycle” of ROSinduced mutations that contribute to cancer development. Overall, mitochondria are
significant contributors of oxidative stress and ROS that can regulate key cell survival,
proliferation, and tumor metastasis pathways in cancer cells.
1.2 Mitochondrial Microproteins
Mitochondrial microproteins, also known as mitochondrial derived peptides (MDPs), are
a subclass of naturally occurring, bioactive microproteins encoded by small open reading
frames (sORFs) in the mitochondrial genome. As proteins under 100 amino acids in
length, MDPs act as systemic regulators of disease. Since their initial discovery, MDPs
have shown dynamic biological effects across various physiological contexts. To date, a

5
total of nine MDPs have been discovered and characterized, including humanin,
mitochondrial open reading frame of the 12S rRNA-c (MOTS-c), small humanin-like
peptides (SHLP) 1-6, and most recently Small Human Mitochondrial ORF Over SErine
tRNA (SHMOOSE), however nearly 600 putative MDPs still await characterization [29].
As the first-ever discovered MDP, humanin has been widely characterized within many
biological systems. Initially discovered when screening for neuroprotective proteins
related to Alzheimer’s Disease (AD) [30], humanin has since demonstrated remarkable
effects across a number of age-related diseases [31-33], and has also shown promising
results as a modulator of healthspan and lifespan [34]. Since humanin’s discovery,
further exploration of the mitochondrial sORFome exposed additional MDPs with
regulatory effects in other biological contexts. While humanin was initially characterized
by its neuroprotective effects, MOTS-c is known as a novel modulator of metabolic
homeostasis [35]. The 16-amino-acid peptide has since demonstrated remarkable
protective effects in metabolic tissues, targeting muscle and primarily mediating insulin
and metabolic pathway activation [36]. In addition to humanin and MOTS-c, the six SHLPs
are an important collection of microproteins. As relatively recently discovered MDPs, they
remain the least characterized group, with various notable regulatory effects within many
disease contexts, including metabolic disorders and cancers. Most recently, SHLP-2 has
been indicated as an insulin sensitizer and potential novel therapeutic for metabolic
disorders [37]. As important regulators of biological processes, MDPs also show sharp
declines in expression with age as age-related diseases increase, demonstrating an
important relationship between MDPs and age-related disease pathophysiology. With
early classification in multiple age-related diseases, MDPs will continue to be

6
characterized in additional systems, showing clear effects as novel modulators of cellular
processes related to multiple cancers, which will be explored throughout this review.
1.3 Mitochondrial microproteins are regulators of mitochondrial function
While mitochondrial microproteins have proven to be unique and critical regulators of
intracellular communication, they are also important modulators of overall mitochondrial
function. As metabolic influencers, many of these MDPs play important roles in
maintaining mitochondrial homeostasis. Previous data suggests that SHLP2 and SHLP3
enhance mitochondrial metabolism [38], while humanin has demonstrated cytoprotective
effects against tissue-specific oxidative stress [39, 40]. Oxidative stress, caused by
reactive oxygen species (ROS), is a primary cause of mitochondrial dysfunction [41].
However, ROS at low levels are important signaling molecules. Thus, systematic
regulation of ROS production and cellular protection from ROS molecules plays an
important role in the health and maintenance of the cell. With observed localization to the
mitochondria in hRPE cells, humanin both protects against oxidative stress-induced cell
death and repairs mitochondrial function under stress [42]. In addition, humanin has
demonstrated cytoprotective effects in heart tissue, protecting against oxidative stressinduced mitochondrial dysfunction by decreasing complex I activity [40]. Similarly, SHLP2
and SHLP3 demonstrate unique cytoprotective effects, improving oxygen consumption
rate (OCR) and ATP production in NIT-1 beta-cells [38]. Thus, MDPs show tissuedependent protective effects of mitochondrial function, are key regulators of oxidative
stress effects within the cell, and act as potent modulators of overall mitochondrial

7
homeostasis. As mitochondria play key roles in healthy cell activity, such maintenance of
mitochondrial function is critical to understanding disease pathophysiology.
1.4 Humanin involvement in cancer
Humanin is a 24-amino acid mitochondria-derived peptide (MDP) from a small open
reading frame within the mitochondrially encoded 16S RNA. Humanin, in 2001 was the
first of now eight published MDPs and the first MDP found to have biological activity.
Humanin was simultaneously discovered in three independent laboratories [43].
Hashimoto et al. were the first group to publish in 2001 while searching for genes that
could protect against amyloid beta induced toxicity. They found Humanin to have
neuroprotective effects in Alzheimer’s cell models, including familial genetic-based
models and amyloid beta toxicity models [30]. Subsequently, Ikonen et al. found Humanin
interacted with Insulin-like growth factor-binding protein-3 (IGFBP3) and protected
primary neurons against amyloid beta toxicity [44]. Finally, Guo et al. found an
antagonistic interaction between Humanin and Bax (Bcl2-associated X protein) where
Humanin reduced Bax apoptotic effects in cells [45].
Humanin has shown various effects in multiple tissues, therefore it would be
beneficial to further explore its action in vivo with cancer being a promising avenue.
There are several reports that characterize the potential role of humanin in the
context of cancer. For instance, the humanin analogue (S14G-HNG) has been shown to
enhance suppression of cancer metastases in mouse models while simultaneously
protecting against chemotherapy induced damage to male germ cells and leukocytes [46].
Some studies suggest that humanin may play a role in ameliorating negative side effects

8
of cancer therapies. One such study showed that HNG prevented bortezomib induced
bone growth impairment without impairing the drugs anticancer effects. Interestingly, the
same study also reported that HNG suppressed the tumor doubling time in human
neuroblastoma mouse xenograft models [47]. Similarly, another study demonstrated that
humanin decreased chemotherapy-induced male germ cell apoptosis [48]. Further, HNG
has been shown to ameliorate chemotherapy-induced white blood cell and granulocyte
loss; germ cell apoptosis; and body and organ weight reductions [49]. A recent study has
demonstrated that higher humanin levels were associated with lower odds of both highgrade and overall prostate cancer incidence [50]. These studies suggest that humanin
not only mediates negative effects of chemotherapies through its cytoprotective effects,
but may also have anti-cancer properties. Conversely, another report has demonstrated
that humanin may have pro-tumoral effects in models of triple negative breast cancer [51].
Other groups have shown that humanin expression is elevated in gastric and bladder
cancers [52, 53]. These differences across cancer types may imply tissue specific
differences in the ability of humanin to modulate different cancers. Thus, the precise
impacts of humanin in the context of cancer and the specific mechanisms involved require
further exploration and study.
1.5 SHLPs and Potential Cancer Connections
Not very long after the discovery of humanin, Cobb et al. published on a set of novel
mitochondrial-derived peptides named small humanin-like peptides (SHLPs 1-6). They
were aptly named after humanin because several of them were found to have similar
effects in vitro and in vivo. The small open reading frames from these peptides were

9
likewise found within the mitochondrially encoded 16S RNA and were all between 20 and
38 amino acids long. Of particular interest were SHLP2 and SHLP3 due to their significant
impact on cell biology. In the same study by Cobb et al., both SHLP2 and SHLP3
considerably reduced apoptosis and the production of reactive oxygen species while
improving mitochondrial metabolism in Murine β-cells (NIT-1) and human prostate cancer
cells (22Rv1). Furthermore, SHLP2 enhanced insulin sensitivity in a rat model with ICV
infusion of SHLP2 by increasing the exogenous glucose infusion rate needed to sustain
euglycemia during insulin stimulation. SHLP3 was unable to prevent apoptosis in a
neuronal cell model of amyloid beta toxicity, however, SHLP2 was [38]. Diverse action in
varying tissues and cell models demonstrate that these microproteins are unique and
potent drivers of cellular biology requiring further investigation.
As with humanin, several studies have demonstrated associations between
SHLP2 and cancer. For instance, a study from Xiao et al. demonstrated that lowered
SHLP2 levels were linked with increased prostate cancer risk in white men specifically.
Further, the same study noted that higher serum levels of SHLP2 were predictive of a
negative biopsy in both white and black men [54]. Similarly, another study reported
SHLP2 values were lower in European American (EA) men with prostate cancer
compared to EA men without prostate cancer. The same report also illustrated that
baseline SHLP2 levels (SHLP2 levels in men without prostate cancer) were decreased in
African American (AA) men compared to EA men and proposed that this phenomenon
may contribute to the increased risk of prostate cancer in AA men compared to EA men
[50, 55]. These results indicate that higher endogenous SHLP2 may have protective
effects against the development of prostate cancer; and that these protective effects may

10
contribute to racial disparities in prostate cancer. One possible mechanism for why higher
SHLP2 levels are protective against cancer may be related to ROS regulation. As
previously discussed, elevated/dysregulated mitochondrial ROS leads to increased tumor
pathogenicity in many cancer types. Interestingly, SHLP2 has been shown to decrease
ROS production in the prostate cancer cell line 22Rv1 [38]. This may indicate that SHLP2
has the ability to ameliorate the effects of ROS signaling in cancer cells. Another MDP,
SHLP6 may also have protective effects against cancer progression. SHLP6 has been
shown to increase apoptosis in 22Rv1 cells, however the exact mechanism underlying
this finding are unknown [38, 56]. Overall, further research into the precise mechanisms
underlying the potentially protective effects of SHLP2 and SHLP6 in cancer and cancer
models.
1.6 Conclusions and Future Directions
Mitochondria regulate key aspects of cancer biology including cancer cell proliferation,
ROS signaling, metabolic reprogramming and plasticity, and metastasis. Mitochondrial
derived peptides have been shown to modulate aspects of mitochondrial function,
maintain mitochondrial homeostasis, and are regulators of general mitochondrial health.
It is therefore probable that these microproteins are also associated with aspects of
mitochondrial-mediated cancer regulation. Mitochondrial derived peptides such as
humanin and SHLP2 have clear associations with various cancer related phenotypes.
Both SHLP2 and humanin have been shown to regulate ERK 1/2 and STAT3 pathways,
both of which are relevant to cancer development and growth [24, 38, 56-58]. Further, the
ability of humanin and the SHLPs to reduce ROS production may contribute to the

11
regulation of cancer cells (Figure 1-1) [38, 42]. The exact mechanisms of humanin and
SHLP2/6 need further study to determine to what extent these microproteins are
protective or antagonistic towards cancer progression and tumorigenicity. Overall,
mitochondrial microproteins present a promising avenue of study in the field of cancer
biology.

12
1.7 Figures
Figure 1-1: Mitochondria and MDPs regulate aspects of cancer biology. Mitochondrial generated
reactive oxygen species (ROS) are upregulated in cancer cells. ROS activate ERK 1/2 and Akt signaling
pathways to enhance proliferation, cell survival and increased motility and metastasis. ROS stimulates
PGC1a which results in metabolic reprogramming towards OXPHOS, increased mitochondrial
biogenesis, and tumor metastasis. Increased ROS induces genomic instability in mtDNA, resulting in
mutations that lead to damaged OXPHOS and further increase ROS levels. Mitochondrial derived
peptides such as Humanin and various SHLPs help to reduce ROS overexpression and thus inhibit ROSinduced tumor progression.

13
CHAPTER 2: Mitochondrial DNA Copy Number Assessment is a Potent Predictor
for Prostate Cancer
ABSTRACT
Introduction: Black individuals are disproportionately burdened by prostate cancer
compared to White individuals. The mitochondrion is an untapped source for prostate
cancer (PCa) biomarkers, and previous work has shown altered mitochondrial DNA
(mtDNA) copy number is linked to mitochondrial dysfunction and tumorigenesis. We
assess whether mtDNA copy number is altered in patients with and without PCa in a
racially specific manner.
Methods: Circulating cell-free mtDNA copy number from plasma and mtDNA copy
number from white blood cells (WBCs) were measured in 199 patients undergoing
biopsy (50/50 White cases/controls and 50/49 Black cases/controls). MtDNA copy
number was determined via ddPCR. Logistic regressions tested associations between
mtDNA and PCa by race. The area under the curve (AUC) was compared between
covariate-only models and models with mtDNA.
Results: In both plasma and WBCs, mtDNA copy number was significantly increased in
cases compared to controls in White patients, but not in Black patients. Interestingly,
Black controls had higher mtDNA copy number levels than White controls. Multivariable
analysis revealed significant associations of Plasma mtDNA and WBC mtDNA with PCa
for White patients only. Elevated mtDNA copy number was more accurate in predicting
PCa in White patients than in Black patients.
Conclusions: Higher mtDNA copy number levels were associated with PCa in both
Black and White patients. Plasma mtDNA may be more accurate than WBC mtDNA in

14
predicting PCa incidence in Black men. Overall, Black controls had higher mtDNA copy
number levels than White controls, suggesting mtDNA copy number may be implicated
in PCa health disparities

15
2.1 Introduction
Prostate cancer (PCa) is the second most common cause of cancer death in United
States men, and globally, is the most common cancer in men with 1.4 million reported
cases in 2016 ([59, 60]1,2). In the United States, Black men make up a disproportionate
amount of the prostate cancer burden. The incidence rate of prostate cancer is 76%
higher than that of White men, and similarly, Black men have over double the mortality
rate compared to White men [61](3). Currently, screening for prostate cancer is largely
dependent on serum prostate-specific antigen (PSA), and aims to provide an
opportunity for early intervention and a decrease in mortality. However, PSA as an
effective screening tool has been met with increasing skepticism in recent years. A
recent meta-analysis of more than 700,000 men from five randomized controlled trials
found that while PSA screening may increase the detection of prostate cancer of any
stage, screening likely has no effect on all-cause mortality and little to no effect on
prostate cancer-specific mortality [62](4). Moreover, Black men are less likely to receive
screening related health benefits compared to White men [63-66](5,6,7,8). Taken
together, these observations indicate the need for additional screening methods and
robust PCa biomarkers beyond just PSA.
One such source of prostate cancer risk factors and potential biomarkers exists
within the mitochondria. Mitochondrial DNA (mtDNA) is routinely exposed to reactive
oxygen species (ROS) due to oxidative phosphorylation that occurs within the organelle
[67](9). As a result of this, mtDNA is prone to mutation and has a tenfold higher
mutation rate than that of nuclear DNA [67, 68](9, 10). This cycle of ROS induced
mutations is further perpetuated as mutations in mtDNA can lead to increased ROS

16
production as well as increased tumorigenicity and tumor growth [28](11). In prostate
cancer specifically, mutations in mtDNA occur early in PCa progression and are
associated with increased tumor aggression and unfavorable risk factors [15-17](12, 13,
14).
More recently, evidence has emerged that indicates mitochondrial DNA (mtDNA)
copy number may also be a potential biomarker for worse prognosis in prostate cancer
patients. One such study of 115 biopsies from prostate cancer patients revealed that
patients with biopsies with increased mtDNA content showed higher disease stage,
extracapsular extension, and trended toward increased Gleason score [69](15).
Similarly, a case-control study of a Chinese Han population found that mtDNA copy
number in peripheral blood leukocytes (PBLs) was significantly higher in PCa patients
compared to controls. Additionally, this study reported a significant dose-response
relationship between mtDNA copy number and PCa risk [70](16). Another study
conducted in non-Hispanic white men, reported that higher mtDNA copy number was
associated with increased risk of non-aggressive PCa as well as increased PSA levels
among controls [71](17).
While these studies identify mtDNA copy number as a potential important
biomarker for PCa outcomes, little work has been done to address how mtDNA copy
number may impact prostate cancer tumors in Black men specifically. Only one study
has evaluated whether mtDNA copy number is associated with prostate cancer in Black
men [72](18). In this analysis, mtDNA copy number was measured in PBLs from 217
Black PCa patients and revealed that high leukocyte mtDNA copy number was
associated with worse prognosis in Black PCa patients (18). While this study revealed

17
the potential importance of mtDNA copy number within Black patients, no previous work
has been done to address how mtDNA copy number differences across racial groups
may contribute to racial health disparities within prostate cancer patients. Here, we aim
to expand upon the current knowledge by both comparing mtDNA copy number levels
within cancer vs non-cancerous groups of White and Black men and by elucidating
differences in mtDNA copy number between racial groups. Additionally, we examine the
functionality of measuring mtDNA copy number from two sources; white blood cells
(WBCs) and circulating cell free mtDNA copy number isolated from plasma.
2.2 Materials and Methods
Study design and sample collection
This study utilized samples from a case-control study of men undergoing prostate
biopsy for suspicion of PCa at the Durham Veterans Affairs Health Care System
between 2007 and 2018. Suspicion of PCa was due to elevated PSA or suspicious
digital rectal exam (DRE). The study was approved by the Institutional Review Board
and all men were provided with written informed consent. Blood was drawn from all
patients at the time of biopsy, before the biopsy was taken. Patients with a biopsy
positive for PCa were considered “cases” while patients with a negative biopsy were
considered “controls.” Of the patients enrolled, 199 with blood samples available were
randomly selected (50 White controls, 50 White cases, 49 Black controls and 50 Black
cases). Men were excluded from the analysis if they were missing covariates of interest
(n=17 missing WBC mtDNA, n=1 with unknown DRE), leading to a final analysis cohort
of 181 patients.

18
Mitochondrial DNA copy number measurement
DNA was extracted from plasma or WBCs with a QIAmp Circulating Nucleic Acid Kit
(Qiagen) according to the manufacturer’s instructions. MtDNA copy number was
measured via ddPCR. Florescent probes and primers were combined with 2x Biorad
ddPCR super mix. DNA isolated from the previous step was then added to create PCR
ready samples. 20 μL of samples were loaded into individual wells of the 8-channel
disposable droplet generator cartridge along with 70 μL of droplet generator oil and
cartridge was loaded onto QX100 droplet generator to form droplets. 40 μL droplets
were transferred to a PCR plate and sealed for PCR. The data were analyzed by
QuantaSoftTM Analysis Pro and the absolute mtDNA copy numbers were reported.
Statistical analysis
Patient characteristics and mtDNA copy numbers were stratified by PC status across all
men and within both Black and White men. Wilcoxon rank-sum and chi squared tests
were used where appropriate to compare patient characteristics at time of biopsy
between cases and controls.
Univariable and multivariable logistic regression models were used to assess the
association between mtDNA copy number and odds of PCa (yes vs. no). mtDNA copy
number was treated as a continuous covariate in primary models. Odds ratios (OR) for
mtDNA were reported in units of 20 as opposed to units of 1 for easier interpretation of
model effects given large ranges of the scales. For clinical purposes, binary versions of
mtDNA were also assessed in separate models. Binary cutoffs for mtDNA were
determined using the maximum statistic approach or minimum p-value approach as
implemented with the %cutpoint SAS macro [73][REF]. Due to the possible influence of

19
outliers on cutoffs, the innermost 50% of the data was specified for cutoff determination.
In a sensitivity analysis (not shown), this was expanded to 90%, and the cutoffs
determined were nearly identical. The interaction between race and mtDNA was tested,
and due to significant interactions, separate cutoffs were determined for Black and
White men. Separate cutoffs were also determined for Plasma and WBC sample types
given different scale ranges. Models were then stratified by race and sample type. To
reduce the likelihood of overfitting, we used the rule of thumb to not include more than 1
covariate for every 10 PC events observed in multivariable models. Final models were
adjusted for age at biopsy, year of biopsy, and other known biomarkers of PC, including
PSA (ng/mL, log-transformed) and DRE (not suspicious vs. suspicious for cancer).
To assess whether mtDNA led to improvement in PC prediction above and
beyond other important clinical characteristics, the diagnostic performance of
multivariable models with and without mtDNA were compared. Sensitivity, specificity,
positive predictive value (PPV), negative predictive value (NPV), and their
corresponding 95% CIs were computed. Receiver operating characteristic (ROC)
curves for models with and without mtDNA (as continuous, units of 20) were plotted and
the area under the curve (AUC) was computed.
In a supplementary analysis, multinomial logistic regression models were used to
test the association between mtDNA copy number (continuous, units of 20) and lowgrade PC (Gleason<7) vs. no PC and high-grade PC (Gleason 7-10) vs. no PC.
Multivariable models were adjusted for the same covariates above. Statistical analyses
were performed using SAS 9.4 (SAS Institute Inc., Cary, NC) and R v4.3.1.

20
2.3 Results
Demographic & clinical characteristics
Among 181 included men (91 cases, 90 controls), 89 were Black (43 cases, 46
controls) and 92 were White (48 cases, 44 controls). PSA was significantly higher for
cases than controls across all men (<0.001), Black men (0.005), and White men
(0.013). No significant difference emerged between cases and controls in age at biopsy,
year of biopsy, BMI, or DRE across groups (Table 2-1).
MtDNA copy number is elevated in individuals with prostate cancer
In Plasma, median mtDNA copy number was higher in cases compared to
controls (p<0.001) (Table 2-1), although the size of this difference varied by race. In
White men, median Plasma mtDNA was substantially higher in cases than controls (714
vs. 115, p<0.001), whereas in Black men, the elevation in median mtDNA for cases
compared to controls was smaller and nonsignificant (504 vs. 289, p=0.098) (Figures 2-
1-2). Similarly, in WBC, median mtDNA copy number was increased in cases compared
to controls (p<0.001) (Table 2-1), and this association again differed by race. In White
men, median WBC was significantly higher for cases than controls (134 vs. 60,
p<0.001), whereas in Black men, median mtDNA copy number was lower for cases than
controls, but not significantly so (172 vs 207, p=0.684) (Figures 2-1-2). .
MtDNA copy number levels are elevated in control groups of Black men
We next compared the control groups of White and Black men to assess
differences between baseline mtDNA copy number levels. In WBC, Black controls had

21
significantly higher baseline mtDNA copy number levels than that of White men
(p<0.001). This phenomenon was similarly observed in plasma in which mtDNA levels
in the control group of Black men were elevated compared to White men.
Mitochondrial DNA copy number as a predictor of prostate cancer
When Plasma mtDNA was treated as a continuous covariate (per every 20 units),
it did not significantly predict PC in univariable (p=0.430) or multivariable (p=0.471)
analysis for Black men (Table 2-2). In contrast, increases in Plasma mtDNA (per every
20 units) were significantly associated with increased odds of PCa in both univariable
(OR=1.04, 95% CI=1.02-1.06, p<0.001) and multivariable (OR=1.03, 95% CI=1.01-1.06,
p=0.003) analysis for White men (Table 2-2). When Plasma mtDNA was treated as a
categorical covariate, a cutoff of 488 best separated Black cases from controls and a
cutoff of 247 best separated White cases from controls (Figure -23). Black men with
Plasma mtDNA ≥488 (vs. <488) had significantly higher odds of PCa in univariable
analysis (p=0.011) but not in multivariable analysis (p=0.063). White men with Plasma
mtDNA ≥247 (vs. <247) had significantly higher odds of PCa in both univariable
(p<0.001) and multivariable (p<0.001) analysis.
Regardless of whether Plasma mtDNA was treated as continuous or categorical,
the diagnostic performance of a covariate-only model (including age, year of biopsy,
PSA, and DRE) improved across measures with the addition of Plasma mtDNA to the
model for Whites but not Blacks (Table 2-3). The overall AUC for the covariate-only
model was 0.71 for both Blacks and Whites. This AUC remained unchanged with the
addition of Plasma mtDNA (continuous) to the model for Blacks whereas it increased to

22
0.83 with the addition of Plasma mtDNA (continuous) for Whites (Figure 2-4). A
likelihood ratio test comparing model deviances revealed that the addition of Plasma
mtDNA to the covariate-only model led to significant improvement in model fit in Whites
(p<0.001).
A similar but stronger pattern emerged when WBC mtDNA was predicting PCa.
When treated as a continuous covariate, WBC mtDNA (per every 20 units) was not
significantly associated with PCa in univariable (p=0.884) or multivariable (p=0.519)
analysis for Black men. In contrast, increases in WBC mtDNA (per every 20 units) were
significantly associated with increased odds of PCa in both univariable (OR=2.87, 95%
CI=1.94-4.82, p<0.001) and multivariable (OR=3.01, 95% CI=1.94-5.67, p=0.003)
analysis for White men (Table 2-2). When WBC mtDNA was treated as a categorical
covariate, a cutoff of 161 best separated Black cases from controls and a cutoff of 103
best separated White cases from controls (Figure 2-3). Black men with WBC mtDNA
≥161 (vs. <161) had significantly lower odds of PCa in multivariable (p=0.042) but not in
univariable (p=0.083) analysis. White men with WBC mtDNA ≥103 (vs. <103) had
significantly higher odds of PCa in both univariable (p<0.001) and multivariable
(p<0.001) analysis.
The diagnostic performance of a covariate-only model improved across
measures with the addition of WBC mtDNA (continuous or categorical) for White but not
Black men (Table 2-3). The AUC of a covariate-only model remained unchanged (0.71)
with the addition of WBC mtDNA (continuous) for Black men but increased to 0.96 for
White men.

23
When PCa grade (low-grade PCa vs. no PCa and high-grade PCa vs. no PCa)
was modeled, results were highly similar to those of PCa status (PCa vs. no PCa). For
Black men, neither Plasma nor WBC mtDNA copy number (per every 20 units)
predicted low- or high-grade PCa (vs. no PCa) in univariable or multivariable analyses.
For White men, both Plasma and WBC mtDNA copy number (per every 20 units)
predicted both low- and high-grade PCa (vs. no PCa) in univariable and multivariable
analyses. The effects were again stronger for WBC than Plasma mtDNA in Whites. In
multivariable analysis, a 20-unit increase in Plasma mtDNA was associated with an
odds ratio of 1.04 for both low- and high-grade PCa. A 20-unit increase in WBC mtDNA
was associated with an odds ratio of 4.34 for low-grade PCa and 4.17 for high-grade
PCa (Supplementary Tables 2-1 and 2-2, Supplementary Figure -21).
2.4 Discussion
In conclusion, it appears these measures are promising biomarkers for White but not
Black men, and that WBC is a stronger predictor than Plasma in White men specifically.
Mitochondrial dysfunction has long been closely linked to cancer cells since the
discovery that cancer cells exhibit altered respiration and rely more heavily on aerobic
glycolysis, a phenomenon known as the Warburg Effect [2](19). Since this discovery,
the role of mitochondrial dysfunction in various cancers has been more closely
evaluated. Newer evidence has revealed that mitochondrial oxidative stress increases
metastatic potential and promotes tumor progression in cancer cells [74](20).
Mitochondrial retrograde signaling via ROS, Ca2+, and oncometabolites released
from the mitochondria has been implicated in mitochondrial dysfunction-induced

24
tumorigenesis and cancer progression [75](21). Additionally, mitochondrial dysfunction
is closely linked to alterations in mtDNA copy number level. Changes in mtDNA copy
number are associated with reduced mitochondrial membrane potential as well as
oxidative stress [76](22). Reduction of mtDNA copy number has also been shown to
alter cell morphology and lower respiratory enzyme activities and ATP production
[77](23). Mitochondrial dysfunction is a hallmark of cancer, and it stands to reason that
this dysfunction may also be intertwined with mitochondrial DNA copy number levels.
Additionally, higher mtDNA copy number could be a result of an effort to compensate for
increased mitochondrial dysfunction.
In addition to disruption of normal mitochondrial function, lower mitochondrial DNA copy
number in tumor cells results in increased sensitivity to chemotherapeutics, an
increased rate of apoptosis, and elevated ROS levels [78](24). It has thus been
proposed that tumor cells may increase mtDNA copy number as a self-protective
mechanism to prevent apoptosis [79](25). Here, we sought to validate that changes in
mtDNA copy number are altered in PCa patients and samples from both WBC and
plasma may be an opportunity to detect and utilize mtDNA copy number as a novel PCa
biomarker. Indeed, we observed an increase in mtDNA copy number in the plasma of
both White and Black men with prostate cancer. Interestingly, while White men with
prostate cancer demonstrated elevated mtDNA copy number levels in WBCs, no such
significant difference in mtDNA copy number was observed in WBC isolated from Black
men with prostate cancer (Figure 2-1-2). This result may indicate the usefulness of
utilizing circulating cell free mtDNA copy number (plasma) over WBCs for detection of
elevated mtDNA copy number in Black men with prostate cancer specifically. Further, in

25
plasma, a mtDNA copy number ≥488 was associated with higher risk of PCa in Black
men, although only in univariable analyses. In contrast, in WBCs, a mtDNA copy
number ≥161 was associated with lower odds of PCa, although not significantly in
univariable analyses (Table 2-2) Taken together, these results may indicate a racial as
well as a sample source bias when measuring mtDNA copy number as a potential
biomarker of prostate cancer. In contrast to Black men, elevated mtDNA copy number
levels (≥247 in plasma and ≥103 in WBC) in White men were accurate in predicting
incidence of prostate cancer for patients undergoing biopsy (Figure 2-23). Of note,
increased mtDNA copy number (≥103) in WBCs of White men in our multivariable
model was highly accurate in predicting prostate cancer incidence (OR 139.06 95%
CI=25.06, >999), and the addition of mtDNA copy number to the covariate-only model
increased the AUC to 0.96. While these results are encouraging and indicate that
mtDNA copy number is a promising marker for predicting prostate cancer, we want to
interpret these results with caution, due to the extreme nature of our data and would
recommend validation in larger cohorts before making any overinterpretations.
[80]Overall, these results indicate that increased levels of mtDNA copy number in
patients may be a biomarker associated with prostate cancer incidence in a potentially
racially specific manner.
As previously described here, mtDNA is highly prone to mutations due to a high
concentration of ROS and relatively inefficient DNA repair mechanisms [81](26).
Elevated mtDNA mutations may result in an increase in mtDNA copy number as a way
to compensate for elevated mitochondrial dysfunction [78](24). It is also possible that an
increase in mtDNA copy number with a high number of mutations may be a self-

26
perpetuating cycle that could result in more aggressive prostate cancer tumors.
Evidence suggests that mtDNA copy number levels are dependent on maintenance of
mitochondrial genome stability [82](27), a process that may be exasperated by
increased mutational load. While elevated mtDNA copy number was less predictive of
prostate cancer in Black men compared to White men, we also observed significantly
higher baseline mtDNA copy number levels in both WBCs and plasma of Black men
with prostate cancer compared to White men. If an elevated mitochondrial DNA copy
number is associated with increased incidence of prostate cancer (as previously
discussed), it stands to reason that an elevated baseline in Black men who have not
developed PCa may be correlated, to some degree, with overall increased risk of
prostate cancer. If the baseline level of mtDNA copy number is already elevated in
Black men, this may also contribute to the inability of mtDNA copy number to distinguish
between cases and controls in our models. It is possible that these elevated levels of
mtDNA in Black men may indicate increased mitochondrial dysfunction that could result
in more aggressive prostate cancer presentation among Black men. Some evidence
suggests that mitochondrial dysfunction may be elevated in tumors of Black men
specifically. For instance, one such study has indicated that a cytochrome c deficiency
in Black tumors leads to mitochondrial dysfunction and promotes therapeutic resistance
and increased tumor aggressiveness in Black men (28). Interestingly, mtDNA copy
number has been shown to be depleted in some tumors relative to matched normal
tissue, specifically in bladder, breast, and kidney cancers [83, 84](29, 30). This may
indicate a tissue specific alteration of mtDNA copy number as it relates to different

27
cancers. Further research is therefore necessary to evaluate the underlying
mechanisms that lead to altered mtDNA copy number in cancer patients.
While this study is the first to compare mtDNA copy number levels across
ethnically diverse groups, there are several limitations including a small sample size and
retrospective design. This small sample size may potentially have led to biased OR
estimates although we attempted to minimize this effect by including the appropriate
number of covariates for the number events present. Additionally, the small sample size
may allow for our predicted cutoff values to be influenced by outliers within the dataset.
However, this was minimized by excluding outliers from the cutoff determination by
using the innermost 50 and 90% of the data. Future studies with larger sample sizes will
be needed to validate the cutoffs. Finally, due to the small sample size, we observed
near perfect prediction when WBC was binary, which lead to quasi complete separation
and very large ORs. Given that the continuous variable was a strong predictor as is, we
focus our interpretation on the continuous ORs, and hope that future studies with larger
sample sizes can test a variety of cutpoints and provide more reasonable estimates.[85,
86]
To summarize, mtDNA copy number was elevated in men with prostate cancer
compared to controls in both White and Black men. While overall, elevated mtDNA copy
number was more accurate at predicting prostate cancer incidence in White men,
circulating cell free mtDNA copy number (plasma) was more accurate for predicting
prostate cancer incidence in Black men. Remarkably, baseline mtDNA copy number
levels were higher in Black men without prostate cancer compared to White men.
Further studies with larger cohorts are needed to validate these findings and to

28
elucidate the mechanisms that result in differences in mtDNA copy number between
ethnic groups.

29
2.5 Tables and Figures
Table 2-1: Demographic and clinical characteristics at the time of biopsy stratified by prostate cancer
status across all individuals and by race
All Individuals Black Individuals White Individuals
Case
(N=91)
Control
(N=90) p value
Case
(N=43)
Control
(N=46) p value
Case
(N=48)
Control
(N=44) p value
Age (years) at biopsy 0.4281 0.5341 0.7571
Median 67.0 66.5 66.0 65.5 67.5 68.0
Q1, Q3 63.0, 70.0 62.0, 69.0 61.0, 69.0 61.0, 68.0 64.0, 70.0 63.5, 70.0
Year of biopsy 0.8121 0.4081 0.6651
Median 2017 2017 2017 2017 2017 2017
Q1, Q3 2016, 2017 2016, 2017 2016, 2017 2016, 2017 2016, 2017 2016, 2018
BMI (kg/m2) at biopsy 0.9932 0.7642 0.8222
<25 50 (50%) 50 (50%) 24 (45%) 29 (55%) 26 (55%) 21 (45%)
25-29.9 27 (51%) 26 (49%) 13 (54%) 11 (46%) 14 (48%) 15 (52%)
≥30 14 (50%) 14 (50%) 6 (50%) 6 (50%) 8 (50%) 8 (50%)
Digital Rectal Exam (DRE) 0.0512 0.1542 0.1972
Not suspicious for cancer 60 (46%) 71 (54%) 30 (44%) 38 (56%) 30 (48%) 33 (52%)
Suspicious for cancer 31 (62%) 19 (38%) 13 (62%) 8 (38%) 18 (62%) 11 (38%)
PSA (ng/mL) at biopsy <0.0011 0.0051 0.0131
Median 7.3 6.0 7.7 6.0 7.1 5.9
Q1, Q3 5.6, 11.3 4.3, 7.8 5.7, 14.9 4.5, 9.1 5.5, 9.3 3.6, 7.8
Plasma mtDNA <0.0011 0.0981 <0.0011
Median 553.3 162.2 503.7 289.2 714.2 115.1
Q1, Q3 201.3, 1154.0 77.8, 405.9 153.7, 1016.1 96.1, 487.5 282.9, 1263.1 63.8, 272.9
WBC† mtDNA <0.0011 0.6841 <0.0011
Median 150.4 104.2 172.3 207.4 134.1 59.7
Q1, Q3 117.4, 253.7 56.9, 207.8 122.3, 313.6 160.6, 285.9 108.8, 187.1 50.1, 85.5
1
Wilcoxon 2
Chi-Square
†
WBC=White Blood Cell count

30
Table 2-2. Univariable and multivariable logistic regression models for testing the association between
mtDNA copy number and risk of overall prostate cancer by race
Black Individuals White Individuals
OR (95% CI) p-value OR (95% CI) p-value
Plasma mtDNA
Univariable
Plasma mtDNA (continuous, units of 20) 1.00 (0.99, 1.01) 0.430 1.04 (1.02, 1.06) <0.001
Plasma mtDNA (binary)
< optimal cutoff value⸹ ref. ref.
≥ optimal cutoff value⸹ 3.23 (1.35, 8.08) 0.011 10.81 (4.28, 29.71) <0.001
Multivariable*
Plasma mtDNA (continuous, units of 20) 1.00 (0.99, 1.01) 0.471 1.03 (1.01, 1.06) 0.003
Plasma mtDNA (binary)
< optimal cutoff value⸹ ref. ref.
≥ optimal cutoff value⸹ 2.45 (0.99, 6.28) 0.063 7.28 (2.75, 20.83) <0.001
WBC† mtDNA
Univariable
WBC† mtDNA (continuous, units of 20) 0.99 (0.94, 1.05) 0.884 2.87 (1.94, 4.82) <0.001
WBC† mtDNA (binary)
< optimal cutoff value⸹⸹ ref. ref.
≥ optimal cutoff value⸹⸹ 0.46 (0.19, 1.09) 0.083 49.80 (15.68,
195.61) <0.001
Multivariable*
WBC† mtDNA (continuous, units of 20) 0.98 (0.92, 1.04) 0.519 3.01 (1.94, 5.67) <0.001
WBC† mtDNA (binary)
< optimal cutoff value⸹⸹ ref. ref.
≥ optimal cutoff value⸹⸹ 0.36 (0.13, 0.92) 0.042 139.06 (25.06,
>999) <0.001
†
WBC=White Blood Cell count
Each row represents a separate model
*Multivariable models adjusted for mtDNA variable, age at biopsy, year of biopsy, PSA (log-transformed), and
DRE
⸹
Optimal cutoff values for plasma mtDNA in black and white individuals are 488 and 247 on the original scale,
respectively
⸹⸹Optimal cutoff values for †
WBC mtDNA in black and white individuals are 161 and 103 on the original scale,
respectively

31
Table 2-3. Comparing model performance among multivariable logistic regression models stratified by race
Covariateonly modelβ
Covariate + Plasma mtDNA
models*
Covariate + WBC† mtDNA
models*
Plasma mtDNA
as continuous
per every 20
units
Plasma
mtDNA as
binary
WBC† mtDNA
as continuous
per every 20
units
WBC† mtDNA
as binary
Black Individuals
Sensitivity (%)
(95% CI)
53.49
(38.58, 68.40)
51.16
(36.22, 66.10)
62.79
(48.34, 77.24)
53.49
(38.58, 68.40)
60.47
(45.85, 75.08)
Specificity (%)
(95% CI)
78.26
(66.34, 90.18)
76.09
(63.76, 88.41)
71.74
(58.73, 84.75)
78.26
(66.34, 90.18)
67.39
(53.84, 80.94)
Positive Predictive Value (PPV)
(%)
(95% CI)
69.70
(54.02, 85.38)
66.67
(50.58, 82.75)
67.50
(52.99, 82.01)
69.70
(54.02, 85.38) 63.41
(48.67, 78.16)
Negative Predictive Value (NPV)
(%)
(95% CI)
64.29
(51.74, 76.84)
62.50
(49.82, 75.18)
67.35
(54.22, 80.48)
64.29
(51.74, 76.84) 64.58
(51.05, 78.11)
White Individuals
Sensitivity (%)
(95% CI)
60.42
(46.58, 74.25)
70.83
(57.97, 83.69)
81.25
(70.21, 92.29)
85.42
(75.43, 95.40)
91.67
(83.85, 99.49)
Specificity (%)
(95% CI)
61.36
(46.98, 75.75)
77.27
(64.89, 89.66)
75.00
(62.21, 87.79)
90.91
(82.41, 99.40)
95.45
(89.30, 100)
Positive Predictive Value (PPV)
(%)
(95% CI)
63.04
(49.09, 76.99)
77.27
(64.89, 89.66)
78.00
(66.52, 89.48)
91.11
(82.80, 99.43)
95.65
(89.76, 100)
Negative Predictive Value (NPV)
(%)
(95% CI)
58.70
(44.47, 72.92)
70.83
(57.97, 83.69)
78.57
(66.16, 90.98)
85.11
(74.93, 95.28)
91.30
(83.16, 99.45)
†
WBC = White blood cell count
β
Multivariable model was adjusted for age at biopsy, year of biopsy, PSA (log-transformed), and DRE
*Multivariable models adjusted for mtDNA variable, age at biopsy, year of biopsy, PSA (log-transformed), and DRE

32
Figure 2-1: Comparison of the distribution of Mitochondrial DNA copy number
levels between cases (individuals with prostate cancer) and controls (individuals
without prostate cancer) for Black and White individuals. For plasma mtDNA,
breaks in the y-axis were created to better visualize the majority of the data. Ranges of
plasma mtDNA shown include 0-2750, 3650-5400, and 7900-8500, which capture all
data points.

33
Figure 2-2: Prostate cancer classification by Mitochondrial DNA copy number
levels between cases (individuals with prostate cancer) and controls (individuals
without prostate cancer) for Black and White individuals. Plasma and WBC mtDNA
are both displayed on the original scale. For Plasma mtDNA, breaks in the y-axis were
created to better visualize the majority of the data. Ranges of Plasma mtDNA shown
include 0-2750, 3650-5400, and 7900-8500, which capture all data points. The Plasma
cutoffs of 488 and 247 for Blacks and Whites correspond to cutoffs of 24 and 12 on the
scale modeled (units of 20). The WBC cutoffs of 161 and 103 for Blacks and Whites
correspond to cutoffs of 8 and 5 on the scale modeled (units of 20).

34
Figure 2-3. Comparing receiver operating characteristic (ROC) curves and area
under the curve (AUC) comparing multivariable logistic regression models with
and without mtDNA
Plasma and WBC mtDNA were treated as continuous variables (per every 20 units) in
the models.

35
CHAPTER 3: A Somatic Mutation in the Novel Mitochondrial Microprotein
SCREAM is Implicated in Cancer Biology
ABSTRACT
Mitochondria are key regulators of tumorigenesis and cancer progression. Somatic
mutations in mitochondrial DNA (mtDNA) are prevalent across multiple tumor types.
Mutations in the control region (D-loop) of mtDNA have been linked to increased
aggressiveness and disease progression. Here we leverage data from The Cancer
Mitochondrial Atlas to identify a single nucleotide variant (SNV) in the D-loop region of
mtDNA that is a result of a somatic mutation in several different tumors types. This
somatic SNV alters the small open reading frame of the mitochondrial microprotein we
call SCREAM, and results in an amino acid change in the WT microprotein. Here we
show that the resulting mutant microprotein (Y18H) has potent pro-oncogenic effects in
both in vitro and in vivo systems. The identification of this somatic SNV and its resulting
alteration in the mitochondrial microprotein SCREAM could present a new therapeutic
target for multiple cancer types.

36
3.1 Introduction
The role of the mitochondrion in cancer is multi-faceted and extends far beyond the
regulation of metabolism in cancer cells. Indeed, mitochondria have been shown to
regulate several aspects of tumorigenesis and aid in promoting the plasticity required for
cancer growth [87]. Metabolic reprogramming by the mitochondria between oxidative
phosphorylation (OXPHOS) and aerobic glycolysis allows for increased flexibility of
cancer cells and has been shown to aid in increased cell proliferation and allow for
therapeutic resistance [3, 88]. In addition to metabolic reprogramming, mitochondria
play a crucial role in cellular communication and regulate key pathways such as
proliferation, apoptosis, and stress responses. Mitochondria are significant sources of
reactive oxygen species (ROS) within tumors, which are essential for cancer
development by promoting cell survival, growth, and genomic instability [23, 89].
Specifically, ROS signaling activates pathways like ERK 1/2 and Akt, enhancing
tumorigenesis and metastasis [24, 26]. Further, mutations in mitochondrial DNA
(mtDNA) have been associated with many different tumor types [11]. Mutations in
mtDNA can lead to altered OXPHOS, resulting in increased oxidative stress and tumorsignaling [90]. These mtDNA mutations have further been associated with increased
tumor aggression in prostate cancer specifically [15]. Despite the well-documented
presence of mtDNA mutations in various cancer models, the exact mechanisms by
which these mutations influence tumor development and drive cancer progression
remain elusive and unexplored.
In addition to metabolic reprogramming, ROS production and signaling, the
mitochondrion regulates key cellular processes through the utilization of mitochondrial

37
derived microproteins. Recently, advances in genomic, transcriptomic, and proteomic
analyses have revealed the presence of small open reading frames (smORFs) that
encode for numerous previously unannotated microproteins (<100 amino acids in
length) in both the nuclear and mitochondrial genome [10, 91, 92]. In the mitochondrial
genome, these smORFs encoding mitochondrial microproteins, or mitochondrial derived
peptides (MDPs), have been shown to have diverse biological roles across many
different disease states, including cardiovascular disease, diabetes, Alzheimer’s
disease, and cancer [36, 93-95]. Of note, several mitochondrial microproteins have
been studied in the context of cancer. For instance, humanin, the first discovered
mitochondrial microprotein has been shown to decrease metastases in mouse models
of melanoma [46]. Additionally, levels of humanin and another MDP, small-humanin-like
peptide 2 (SHLP2) have been reported to be lower in high-grade tumors of prostate
cancer patients [54]. This may indicate a protective effect for these microproteins in
prostate cancer progression. These findings suggest that MDPs could be an
underexplored resource for mitochondrial regulation in cancer. Here, we investigate the
impact of somatic mtDNA mutations in tumors on mitochondrial microproteins, and
identify a novel MDP that has direct effects on cancer progression and tumor growth.
3.2 Materials and Methods
Identification of Somatic Tumor Mutations in mtDNA
Somatic tumor mutations in mitochondrial DNA were previously identified, utilizing
matched tumor and normal tissue whole genome sequencing data aggregated from
2,658 cancers across 38 tumor types from The Cancer Genome Atlas (TCGA) as well

38
as the International Cancer Genome Consortium (ICGC) [11]. Briefly, variants were
initially called using VarScan2 with the same parameter settings as previously reported:
--strand-filter 1 (mismatches should be reported by both forward and reverse reads), --
min-var-freq 0.01 (minimum VAF 1%), --min-avg-qual 20 (minimum base quality 20), --
min-coverage × and --min-reads2 ×) [96]. First, germline polymorphisms and false
positive calls were filtered out. Next, DNA cross-contamination was examined.
Subsequently, the overall mtDNA substitution signatures were analyzed across the 96
possible mutation classes. Finally, the possibility of false positive calls due to mismapping was assessed. Somatic mtDNA tumor variant data was downloaded for further
analysis from The Cancer Mitochondrial Atlas at:
https://ibl.mdanderson.org/tcma/download.html. This data was then loaded into R
(1.4.1106) and mutation data files for all 38 tumor types, prostate tumors, and kidney
tumors were generated. For each dataset, (all tumor types, prostate tumors, and kidney
tumors), the frequency of somatic tumor variants was calculated. The position of each
variant was counted using the count function from the dplyr package. The frequencies
were then calculated as the count of each variant divided by the total number of
positions in the respective dataset. These frequencies were also converted to
percentages for further analysis. To generate mtDNA variant plots, the calculated
frequencies were visualized using the ggplot2 package. For each dataset, a scatter
plot was created to display the frequency of somatic mutations across different
mitochondrial DNA (mtDNA) positions. The plots were annotated to indicate the 13 large
mitochondrial encoded gene as well as the d-loop control region. A cutoff of 0.5% was
utilized to identify the most frequently mutated positions.

39
SCREAM and Y18H Structure Prediction
PEP-FOLD 4, a de novo peptide prediction server, was utilized to predict the structures
of both SCREAM and Y18H from their respective amino acid sequences, employing a
previously reported predictive algorithm [97]. PEP-FOLD 4 was chosen for its capability
to model small proteins (up to 40 amino acids in length), providing highly specific
peptide prediction models.
SCREAM Antibody Production
Rabbit anti-SCREAM sera were produced by YenZym Antibodies (Brisbane, CA).
SCREAM antibody was then purified using Antigen-Specific Affinity Purification utilizing
synthetic SCREAM peptide.
Synthesis of Synthetic SCREAM and Y18H
Synthetic SCREAM and Y18H were made by GenScript via solid-phase peptide
synthesis methods and Triflouracetic acid (TFA) was used to cleave synthesized
peptides from resin. Following synthesis, residual TFA was removed and peptides were
lyophilized and packaged in 1 mg sterile aliquots.
Endogenous SCREAM Detection via Mass Spectrometry
22Rv1 cells were grown on 100mm dishes in RPMI media with 10% FBS until full 95-
100% confluency was reached. Cells were then serum starved overnight at 37C. All
media was removed and cells were scraped and combined into one sample and stored
at -80C overnight. Cells were then lysed with RIPA Lysis and Extraction Buffer and
sonicated to obtain whole cell protein lysate. Whole cell lysate was then used for
immunoprecipitation of SCREAM protein via the Dynabeads™ Protein G

40
Immunoprecipitation Kit (Invitrogen) according to the manufacturer’s instructions. 10 ug
of SCREAM antibody was used for immunoprecipitation of endogenous SCREAM
protein and whole cell protein lysate was incubated overnight at 4C with the magnetic
bead-SCREAM antibody complex. Endogenous SCREAM protein was then eluted from
beads in elution buffer and 1M Tris-HCl and was sent for mass spectrometry analysis at
UCLA.
Detection of SCREAM via Western Blot
22Rv1 cells were plated at 1.5 million cells per well in a 6 well dish. Once ~90%
confluency was reached, cells were serum starved overnight and then lysed with RIPA
Lysis and Extraction Buffer and sonicated to obtain whole cell protein lysates. Samples
were then run on a BioRad 8%-16% Gel and transferred to a PVDF membrane using
the Trans-Blot Turbo Transfer System. The membrane was then blocked in 5% BSA in
TBS + 0.1% Tween 20 for 1 hour at RT. The membrane was washed and incubated
with primary SCREAM antibody at a 1:100 ratio overnight at 4C. The membrane was
then washed and incubated with 1:30,000 secondary anti-rabbit antibody conjugated to
HRP on the membrane for 1 hour at RT. The membrane was then excited using ECL
reagent for 2 minutes. For blocking with synthetic SCREAM, the membrane was
stripped using Restore™ PLUS Western Blot Stripping Buffer (Thermo Fisher) for 15
minutes at RT and was washed and blocked again with 5% BSA in TBS + 0.1% Tween
20 for 1 hour. The rest of the western blotting protocol was followed as previously
described, except for at the primary antibody incubation step where SCREAM antibody
and synthetic SCREAM protein (in a 1:5 ratio) was incubated on the membrane at 4C
overnight.

41
Immunocytochemistry
22Rv1 cells were seeded in RPMI medium containing 10%FBS at 0.08x106 cells/well in
a PureCol Coated 48-well plate and allowed to adhere for 24 hours. After reaching 50-
70% confluence, cells were fixed with 4% paraformaldehyde for 15 minutes at room
temperature (RT) and then washed 2X for 5 minutes with PBS at RT. Cells were then
permeabilized with 0.1% triton in PBS for 15 minutes at RT. Cells were then blocked
using 2% BSA in PSA for 1 hour at RT. After blocking, SCREAM antibody (1:50 dilution)
or a mixture of SCREAM antibody and synthetic SCREAM peptide (in a 5:1 ratio) and
Tom20 antibody (1:200 dilution) conjugated with Alexa-Fluor 594 were added to cells in
0.1% BSA PBS solution. The plate was then wrapped in parafilm, covered in aluminum
foil, and incubated overnight at 4C. After overnight incubation, all wells were washed 3X
for 5 minutes with PBS at RT. Secondary antibody (Alexa-Fluor 488 goat anti-rabbit)
was added in 0.1% BSA PBS at a 1:200 dilution for 45 minutes at RT. Cells were
washed 3X for 5 minutes with PBS at RT. In the second wash, nuclei were stained for 5
minutes at RT in PBS containing Hoechst 33258 at a 1:2000 dilution. Cells were
mounted in PBS and fluorescent staining was imaged using the AMG EVOS imaging
microscope. A minimum of 3 images per well were taken, and representative images
were utilized for each treatment.
Cell Culture
22Rv1 (CRL-2505) and LnCaP (CRL-1740) prostate cancer and A-498 kidney cancer
(HTB-44) cells used in this study were purchased from ATCC. 22Rv1 and LnCaP cells
were grown in RPMI 1640 supplemented with 10% FBS at 37C with 5% CO2 with media

42
renewal every 2-3 days. A-498 cells were grown in EMEM media supplemented with
10% FBS at 37C with 5% CO2 with media renewal twice per week.
Proliferation Assay
Proliferation was measured using the Invitrogen CyQUANT™ Direct Cell Proliferation
Assay. Briefly, 22Rv1 or A498 cells were plated in a 96 well plate a concentration of
30,000 cells per well and allowed to adhere overnight. When indicated, cells were then
treated with either UltraPure DI Water (vehicle control) or synthetically synthesized
SCREAM or Y18H at a concentration of 10 uM for 48 hours. When indicated, SCREAM
antibody or rabbit IgG were co-treated with synthetic Y18H or vehicle control for 48
hours at 2ug/mL concentration. After 48 hours, an equal volume of 2X Detection
reagent containing PBS, CyQUANT™ Direct Nucleic Acid Stain, and CyQUANT™
Direct Background Suppressor was added to cells. Cells were then incubated with 2X
Detection Reagent for 60 minutes at 37C. Fluorescence was read from the bottom of
the well using excitation/emission maxima 508/527 nm.
Apoptosis Assay
Apoptosis was measured using Promega’s RealTime-Glo™ Annexin V Apoptosis and
Necrosis Assay. As in the proliferation assay, 22Rv1 or A498 cells were plated in a 96
well plate a concentration of 30,000 cells per well and allowed to adhere overnight.
When indicated, cells were then treated with either UltraPure DI Water (vehicle control)
or synthetically synthesized SCREAM or Y18H at a concentration of 10 uM for 48 hours.
When indicated, SCREAM antibody or rabbit IgG were co-treated with synthetic Y18H
or vehicle control for 48 hours at 2ug/mL concentration. After 48 hours, cells were
treated with an equal volume of 2X Detection Reagent containing complete cell growth

43
media, Annexin V NanoBiT™ Substrate (1,000×), CaCl2 (1,000×), Necrosis Detection
Reagent (1,000×), Annexin V-SmBiT (1,000×), and Annexin V-LgBiT (1,000×). Relative
luminescent values (RLU) were then obtained using the Promega Glo-Max System.
Cell Growth Curve
22Rv1 cells were plated at 500,000 cells per well in 6 well dish. Cells were treated with
10uM of SCREAM, Y18H or UltraPure DI Water in triplicate for each time point. Cells
were retreated with peptide and given new media every 48 hours. Cells were counted
using the Invitrogen Countess 3 Automated Cell Counter every 48 hours for 8 days.
A498 Mutation Status
The SCREAM mutation status of the cell line A498 was determined by visualizing RNASequencing data of the A498 cell line obtained from a publicly available dataset in the
gene-expression omnibus (GEO) accession GSE83999. RNA-seq files were aligned to
the human genome assembly GRCh38 using the STAR alignment tools [98] and the
resulting bam files were loaded into the Integrative Genomics Viewer (v2.16.2) for
visualization of the SCREAM small open reading frame (chrM:21-95).
Mouse Prostate Cancer Allografts
Thirty-six eight-week-old FVB mice were purchased from Jackson Laboratory (Bar
Harbor, ME) and acclimated for 7 days on a standard AIN-93G diet (Dyets, Bethlehem,
PA). Post-acclimation, each mouse was subcutaneously injected in the flank with
approximately 0.5 million Myc-CaP cells (ATCC) in a 0.1 ml mixture of Matrigel and PBS
(Collaborative Biomedical Products, Bedford, MA) using a 26G5/8 needle. When the
tumor volume reached 30-50 mm³, mice were randomized into three treatment groups.
The first group served as the vehicle control and received 100 µl of intraperitoneal (IP)

44
water daily (n=12). The second group received SCREAM treatment at a dosage of 5
mg/kg body weight and was administered via IP daily (n=12). The third group received
Y18H treatment at the same dosage of 5 mg/kg body weight and was administered via
IP daily (n=12). Synthetic SCREAM and Y18H compounds at >99% purity (GenScript)
were dissolved in sterile water prior to administration. Body weight and average food
intake were monitored weekly, while the general health status of the mice was checked
daily. Tumor volume was measured twice a week using the following formula: tumor
volume (cm³) = 0.523 × [length (cm) × width (cm) × height (cm)]. Once tumor sizes
reached approximately 800-1200 mm³, mice were euthanized using isoflurane. Blood
was collected via cardiac puncture into EDTA tubes, and plasma was separated and
stored at -80°C. The primary outcomes measured were tumor volume and weight. All
biological results were analyzed using independent two-tailed Student’s t-test, a cutoff
of p < 0.05 was considered statistically significant.
3.3 Results
The somatic tumor mtDNA SNV T72C alters the smORF of the mitochondrial
microprotein SCREAM
Previous studies have shown that the control region of the mtDNA (D-loop) is
hypermutated in prostate cancer [11, 15]. However, this previous work completely
ignores the presence of microproteins in this region, and instead focuses on the impact
of single nucleotide variants (SNVs) on the 13 large protein coding genes. Here, we
hypothesized that somatic SNVs in mtDNA, and in the D-loop specifically, may impact
the small open reading frames of mitochondrial microproteins. In order to assess which

45
somatic mutations are the most prevalent across all cancer types, we leveraged tumor
SNV data deposited into The Cancer Mitochondrial Atlas which was first analyzed in
2020 by Yuan et al. We first determined the frequency of mitochondrial SNVs (mtSNVs)
across all 38 tumor types and used a frequency cut off of 0.5% to identify only the top
mutated positions. The only SNV that was above this 0.5% cutoff value (with a
frequency of 0.83%) across all tumor types was a T>C mutation at position 72 in the Dloop region of the mitochondria (Figure 3-1b). We then stratified the data by tumor type
to identify which tumors were the largest carriers of the T72C SNV. Of note, kidney
renal cell carcinomas and prostate adenocarcinomas contained this SNV at high
frequencies of 2.5% and 1.1% respectively (Figure 3-1b). As this mutation lies within the
D-loop, it does not result in an amino acid change in any of the large mitochondrial
genes, however mtSNV T72C does change the amino acid sequence of a novel
mitochondrial encoded by a smORF (Figure 3-1c-d), that we have named SCREAM
(Small Cancer Related d-Loop Associated Microprotein). Using the protein structure
prediction model PEP-FOLD 4, we modeled the predicted structure of the 24 amino acid
microprotein SCREAM, as well as the mutated version we have named SCREAM
Y18H, which is a result of the 18th amino acid modification from a tyrosine to a histidine
(Figure 3-1d).
SCREAM is endogenously expressed in the prostate cancer cell line 22Rv1
In order to assess the endogenous expression of SCREAM in cancer cells, we
developed an antibody against the wild type (WT) SCREAM microprotein and utilized
the prostate cancer cell line 22Rv1 for detection. Using whole cell lysates from 22Rv1
cells, we immunoprecipitated SCREAM for detection with mass spectrometry. Here we

46
were able to detect a SCREAM-derived peptide fragment from the C-terminus of the
microprotein, as well as confirm the expression of WT SCREAM in 22Rv1 cells
specifically (Figure 3-2a). The same fragment was detected in whole cell lysates from
LnCaP prostate cancer cells (Supplemental Figure 3-1). Interestingly, western blotting
from 22Rv1 whole cell lysates indicates endogenous SCREAM at ~14kDa and when
blocked with synthetic SCREAM protein, this 14kDa band was no longer present (Figure
3-2b). To further validate the endogenous expression of SCREAM and to investigate its
relative expression in different cellular compartments, we employed staining of fixed
22Rv1 cells (Figure 3-2c). Here we were able to visualize expression of SCREAM, in
which SCREAM staining is co-localized with the mitochondrial marker TOM20. After
blocking with synthetic SCREAM, staining is non-specific and not localized to cells.
The SCREAM variant Y18H promotes pro-oncogenic effects
Given that the mtSNV T72C is present in tumors but not normal tissue, we hypothesized
that the corresponding microprotein Y18H may impact cancer biology in a manner
beneficial to the overall tumor. To assess the activity of WT SCREAM and Y18H, we
again utilized the prostate cancer cell line 22Rv1 to determine the impact of these
microproteins on cancer cell growth. Interestingly, treatment with SCREAM modestly
induced proliferation of 22Rv1 cells, however only treatment with Y18H significantly
increased cellular proliferation and growth (Figure 3-3a, 3-3c). Unsurprisingly, a similar
trend was observed in the context of apoptosis where Y18H significantly decreased
apoptosis of 22Rv1 cells compared to the vehicle control, whereas WT SCREAM had
similar but much more limited effects (Figure 3-3b).

47
Inhibition of Y18H leads to tumor-suppressive effects
Next, we aimed to evaluate how the inhibition of SCREAM Y18H would impact similar
measures of tumor growth. Due to the difficulty involved in inducing specific and
widespread mitochondrial mutations [99], we first sought to identify a commercially
available cell culture line that possesses the T72C mtSNV and expresses Y18H
endogenously. Here, we utilized publicly available RNA sequencing data from both
prostate and kidney cancer cell lines, and visualized the SCREAM smORF in each cell
line using the Integrative Genomics Viewer, to identify a cell line with the T72C mtSNV
(Figure 3-3d). Using this method, we were able to identify the renal cell carcinoma cell
line A498 as having the T72C mtSNV. In order to inhibit endogenously expressed
Y18H, we treated A498 cells with the previously mentioned SCREAM antibody.
Treatment with the SCREAM antibody resulted in a significant and dramatic reduction in
A498 proliferation, as well as a significant increase in apoptosis (Figure 3-3e-f).
Interestingly, addition of synthetic Y18H did not dramatically change baseline
proliferation or apoptosis but was able to block the effects of the SCREAM antibody.
Mouse models of prostate cancer are stimulated by the Y18H variant
In order to assess the impact of WT SCREAM and Y18H on tumor growth in vivo, we
injected FVB mice with the murine prostate cancer cell line MyC-CaP to serve as an in
vivo allograft model of prostate cancer (Figure 3-4a). Mice were injected with either WT
SCREAM, Y18H, or saline control and we monitored tumor volume over time. We found
that injection with Y18H leads to significantly increased tumor volume, while WT

48
SCREAM did not significantly increase tumor volume (Figure 3-4c). Neither WT
SCREAM or Y18H impacted weight (Figure 3-4b).
3.4 Discussion
Here we assess how a specific frequently mutated mtSNV may directly impact
increased tumor progression through the alteration of the novel mitochondrial
microprotein SCREAM. Our frequency analysis allowed us to identify mtSNV T72C as
the most frequently mutated SNV across all 38 tumor types included in our analysis,
and moreover, indicated that this SNV may be of particular importance in kidney and
prostate tumors (Figure 3-1b).
The evaluation of somatic mutations in tumors is important for elucidating genes
that drive cancer progression. Cancer development and progression is largely driven by
the acquisition of mutations that confer a selective advantage, such as increased
proliferation rates or enhanced survival, to the mutant clone [100, 101]. Assuming that
some mutations (either germline or somatic) confer a selective advantage for the tumor,
it follows that these mutations would be more frequently detected in tumor samples due
to positive selection. Indeed, for somatic mutations reported in the Catalog of Somatic
Mutations in Cancer (COSMIC), it is widely acknowledged that the frequency of a
somatic mutation reflects its selective advantage in tumor development [101-103]. More
specifically, a recent study evaluating the functionality of somatic tumor mutations in
COSMIC found that mutations reported ≥10 times are likely to be functional and cancerrelevant [101]. While the D-loop of mtDNA has been recognized as being hypermutated
in different tumor types [104-107], our study examines how this mutational hotspot

49
directly impacts the smORFs of mitochondrial microproteins in this control region. Given
the increased frequency of the T72C mutation, we propose that this SNV confers a
selective advantage for tumors via the expression of SCREAM Y18H.
Indeed, in both our in vitro and in vivo models, exposure to synthetic Y18H
increased prostate cancer cell proliferation, decreased apoptosis, and promoted growth
of tumors (Figure 3-3a-c, Figure 3-4). Further, the development of an anti-SCREAM
antibody allowed us to elucidate the effects of the inhibition of endogenously expressed
Y18H. When endogenous Y18H was inhibited, we observed a dramatic reduction in cell
proliferation and an equally as dramatic increase in apoptosis (Figure 3-3e-f). When
taken together, these results reveal that when mutated, the SCREAM smORF may
result in a novel oncogene in multiple cancer types.
While this study spotlights the discovery and initial characterization of a novel
cancer-related mitochondrial microprotein, we acknowledge that there are several
limitations to this study that should be noted. In this study, we focus exclusively on the
characterization of the microprotein SCREAM, however, we ignore the possibility of
other mitochondrial microproteins that may also be impacted by tumor somatic mtSNVs.
In this study, we aimed to identify the mitochondrial microprotein that had the highest
chance of impacting tumor biology, and thus only characterized the microprotein
effected by the mtSNV with highest mutational frequency. It is certainly possible that the
smORFs of additional mitochondrial microproteins may be altered by mtSNVs not
highlighted in this study, and may also play a role in regulating tumor biology. Additional
studies are therefore required to examine the impact of other somatic mutations on
additional mitochondrial microproteins. In the context of SCREAM, this study primarily

50
focuses on biological effects in prostate and kidney cancers, however, the T72C mtSNV
is present in additional tumor types, although at lower frequencies. Further studies are
therefore required to assess the potency of the Y18H variant in other cancer types.
Another limitation of this study lies in the inability to directly correlate the T72C mtSNV
in humans with clinical outcomes such as tumor grade, tumor aggressivity, mortality,
and response to treatment. This was due to the lack of availability of clinical data in all
individuals with the T72C mtSNV, and the relative overall low frequency of the SNV.
Future studies that can make these correlations are therefore required to assess the
impact of this somatic mtSNV on important clinical parameters. Finally, in this initial
report, we characterize the effect of WT SCREAM and Y18H in regards to broad cancer
phenotypes but do not delve into the mechanism of action for the phenotypes observed.
We therefore hope that future studies can further elucidate the mechanisms responsible
for the potent actions of Y18H.
We propose that mitochondrial microproteins represent a novel mechanism by
which mitochondria contribute to the regulation of cancer and tumorigenesis. Here we
reveal the potential role for one such microprotein, SCREAM, and its corresponding
variant Y18H, in the regulation of cancer progression. Our results indicate that
mitochondrial microproteins such as SCREAM could serve as novel targets for cancer
therapeutics.

51
3.5 Figures
Figure 3-1: The somatic tumor mitochondrial SNV T72C changes the smORF of
the mitochondrial microprotein SCREAM. A) Schematic of methods used to identify
somatic mtSNVs. B) Frequencies of mtSNVs mapped to the mtDNA across all tumors
types and stratified by kidney and prostate cancer. C) mtSNV T72C lies within the
smORF of the novel mitochondrial microprotein SCREAM. D) The mtSNV T72C
changes the amino acid sequence of mitochondrial microprotein SCREAM, resulting in
a mutant version known as Y18H.
Tumor Type Frequency of T>C Mutation at position chrM:72 (%)
All Tumors 0.83
Kidney (RCC) 2.5
Prostate Adenocarcinoma 1.1
B
ASncmtRNA
D-loop
CytB lncND6
ND5
ND4
ND3
COX3
ATP8/ATP6
COX2
COX1
ND2 ND1 16S rRNA
12S rRNA
ND4L
T
F
V
L1
I
M
W
D K
G
R
H
S2
L2
S1
E
P lncCytB
A
N
C
Y
Q
lncND5
MDL1AS
ND6
143B
T>C SNV at chrM:72 A
One individual
(2,568 matched tumor vs
normal tissue samples)
Frequency of
tumor SNVs
Frequency of normal
tissue SNVs
Compare to get significant
SNVs for each individual
unique to the tumor
A
Identified somatic mtDNA SNVs that
impact mitochondrial microproteins
Small Cancer Related Effector d-Loop
Associated Microprotein: SCREAM
Amino Acid Sequence:
MNHSRELSMHLVFSSGGYARDSIA
C
MNHSRELSMHLVFSSGGYARDSIA MNHSRELSMHLVFSSGGHARDSIA
D
SCREAM (WT) Y18H (mutant)

52
Figure 3-2: Endogenous Detection of SCREAM. A) Unique mass spectrometry of a
SCREAM derived fragment detected in whole cell lysates of 22Rv1 cells. B) Western
blot of 22Rv1 whole cell lysates before and after blocking with synthetic SCREAM. C)
Fluorescent images showing SCREAM endogenous expression in 22RV1 cells. Top
panels from left to right: nucleus, mitochondria, SCREAM staining, merged; bottom
panels from left to right: nucleus, mitochondria, SCREAM staining blocked with
synthetic SCREAM, merged.
Ab only
Hoechst TOM 20 SCREAM Ab Merged
Blocked
14 kDa
14 kDa
water 22Rv1 WCL
SCREAM
Antibody
SCREAM Antibody
Blocked with
Synthetic SCREAM
B
C
A MNHSRELSMHLVFSSGGYARDSIA

53
Figure 3-2: The SCREAM variant Y18H exhibits pro-oncogenic effects. A)
Proliferation assay of 22Rv1 cells after treatment with water (control) and synthetic
SCREAM and Y18H. B) Apoptosis assay of 22Rv1 cells after treatment with water
(control) and synthetic SCREAM and Y18H. D) Sequence of the SCREAM smORF in
the A498 cell line from the Integrative Genomics Viewer. E) Proliferation assay of A498
cells after treatment with water, synthetic Y18H and cotreatments with SCREAM
antibody (Ab) and rabbit IgG (IgG). F) Apoptosis assay of A498 cells after treatment
with water, synthetic Y18H and cotreatments with SCREAM antibody (Ab) and rabbit
IgG (IgG).
A B
0 2 4 6 8 10
0
1000000
2000000
3000000
4000000
5000000
Days
Cell Count
control
SCREAM
Y18H
*
C
D E F

54
Figure 3-4: Tumor promoting effects of Y18H in mouse prostate cancer models.
A) Schematic of mouse prostate cancer allograft models. B) Weight changes over time
after treatment with synthetic SCREAM or Y18H compared to saline control. C) Tumor
volume measurements over time after treatment with synthetic SCREAM or Y18H
compared to saline control.
*
*
Tumor cell injection
Peptide injection
A B
C

55
FUTURE DIRECTIONS
Mitochondria are dynamic and essential organelles, whose function is deeply
intertwined with cancer cell biology. Here we describe how mitochondria regulate
cancer cell signaling and survival through metabolic reprogramming and the production
of ROS species. We catalog how previously published mitochondrial derived
microproteins (humanin and the SHLPs) regulate aspects of mitochondrial biology and
therefore impact aspects of cancer biology.
We propose mitochondrial DNA copy number as a novel biomarker for the
prediction of prostate cancer, and moreover, how this phenomenon occurs in a racially
specific manner. We observed elevated mtDNA copy number in Black individuals
without prostate cancer, suggesting a potential role for mtDNA copy number in racial
disparities in prostate cancer outcomes. However, further research is needed to
elucidate: (1) the precise mechanisms driving increased mtDNA copy number in
individuals with cancer, and (2) how these alterations specifically contribute to tumor
progression and cancer development. Additional research should also examine how
mtDNA copy number behaves as a biomarker across different tumor types.
Finally, we introduce the mitochondrial microprotein SCREAM and its
corresponding smORF as a novel onco-gene of interest. We illustrate how somatic
mtSNVs may play a crucial role in tumor development by impacting not just large
mitochondrial OXPHOS genes, but often overlooked mitochondrial microproteins. While
we demonstrated that the SCREAM Y18H variant has potent oncogenic effects in
different cancer types, more work is required to elucidate the mechanisms of action

56
involved in this process. Additionally, insights into how WT SCREAM regulates normal
cell biology processes is imperative in illuminating the native role of SCREAM.
Another area of interest lies within the potential connections between mtDNA
copy number and mitochondrial microproteins. With mtDNA copy number being
predictive of prostate cancer, studies correlating protective or oncogenic MDP levels—
such as SCREAM and SCREAM Y18H—and copy number may provide insight into how
copy number regulates MDPs (or vice versa). Future studies that connect these
observations may allow insight as to the mechanisms involved in mitochondrial
regulation of both copy number and MDPs in the context of cancer.
Overall, the field of mitochondrial microprotein research is an exciting opportunity
for unraveling the complex regulatory processes connecting the mitochondrion and
cancer. Mitochondrial microproteins hold great potential as targets for cancer
therapeutics, and further characterization of these small (but mighty) proteins could
open promising new avenues in mitochondrial research, with broad implications for the
cell biology field.

57
REFERENCES
1. Wallace, D.C., Mitochondria and cancer: Warburg addressed. Cold Spring Harb
Symp Quant Biol, 2005. 70: p. 363-74.
2. Warburg, O., On the origin of cancer cells. Science, 1956. 123(3191): p. 309-14.
3. Vaupel, P., H. Schmidberger, and A. Mayer, The Warburg effect: essential part of
metabolic reprogramming and central contributor to cancer progression. Int J
Radiat Biol, 2019. 95(7): p. 912-919.
4. Guo, D., et al., Aerobic glycolysis promotes tumor immune evasion by
hexokinase2-mediated phosphorylation of IkappaBalpha. Cell Metab, 2022.
34(9): p. 1312-1324 e6.
5. Kroemer, G. and J. Pouyssegur, Tumor cell metabolism: cancer's Achilles' heel.
Cancer Cell, 2008. 13(6): p. 472-82.
6. Pouyssegur, J., F. Dayan, and N.M. Mazure, Hypoxia signalling in cancer and
approaches to enforce tumour regression. Nature, 2006. 441(7092): p. 437-43.
7. Yang, J., et al., The enhancement of glycolysis regulates pancreatic cancer
metastasis. Cell Mol Life Sci, 2020. 77(2): p. 305-321.
8. LeBleu, V.S., et al., PGC-1alpha mediates mitochondrial biogenesis and
oxidative phosphorylation in cancer cells to promote metastasis. Nat Cell Biol,
2014. 16(10): p. 992-1003, 1-15.
9. Hirpara, J., et al., Metabolic reprogramming of oncogene-addicted cancer cells to
OXPHOS as a mechanism of drug resistance. Redox Biol, 2019. 25: p. 101076.
10. Mercer, T.R., et al., The human mitochondrial transcriptome. Cell, 2011. 146(4):
p. 645-58.
11. Yuan, Y., et al., Comprehensive molecular characterization of mitochondrial
genomes in human cancers. Nat Genet, 2020. 52(3): p. 342-352.
12. Larman, T.C., et al., Spectrum of somatic mitochondrial mutations in five
cancers. Proc Natl Acad Sci U S A, 2012. 109(35): p. 14087-91.
13. Park, J.S., et al., A heteroplasmic, not homoplasmic, mitochondrial DNA mutation
promotes tumorigenesis via alteration in reactive oxygen species generation and
apoptosis. Hum Mol Genet, 2009. 18(9): p. 1578-89.

58
14. Ralph, S.J., et al., The causes of cancer revisited: "mitochondrial malignancy"
and ROS-induced oncogenic transformation - why mitochondria are targets for
cancer therapy. Mol Aspects Med, 2010. 31(2): p. 145-70.
15. Hopkins, J.F., et al., Mitochondrial mutations drive prostate cancer aggression.
Nat Commun, 2017. 8(1): p. 656.
16. Jeronimo, C., et al., Mitochondrial mutations in early stage prostate cancer and
bodily fluids. Oncogene, 2001. 20(37): p. 5195-8.
17. Schopf, B., et al., OXPHOS remodeling in high-grade prostate cancer involves
mtDNA mutations and increased succinate oxidation. Nat Commun, 2020. 11(1):
p. 1487.
18. He, L., et al., Somatic mitochondrial DNA mutations in adult-onset leukaemia.
Leukemia, 2003. 17(12): p. 2487-91.
19. Jones, J.B., et al., Detection of mitochondrial DNA mutations in pancreatic
cancer offers a "mass"-ive advantage over detection of nuclear DNA mutations.
Cancer Res, 2001. 61(4): p. 1299-304.
20. Polyak, K., et al., Somatic mutations of the mitochondrial genome in human
colorectal tumours. Nat Genet, 1998. 20(3): p. 291-3.
21. Zampieri, L.X., et al., Mitochondrial Transfer in Cancer: A Comprehensive
Review. Int J Mol Sci, 2021. 22(6).
22. Sabharwal, S.S. and P.T. Schumacker, Mitochondrial ROS in cancer: initiators,
amplifiers or an Achilles' heel? Nat Rev Cancer, 2014. 14(11): p. 709-21.
23. Yang, Y., et al., Mitochondria and Mitochondrial ROS in Cancer: Novel Targets
for Anticancer Therapy. J Cell Physiol, 2016. 231(12): p. 2570-81.
24. McCubrey, J.A., et al., Roles of the Raf/MEK/ERK pathway in cell growth,
malignant transformation and drug resistance. Biochim Biophys Acta, 2007.
1773(8): p. 1263-84.
25. Roberts, P.J. and C.J. Der, Targeting the Raf-MEK-ERK mitogen-activated
protein kinase cascade for the treatment of cancer. Oncogene, 2007. 26(22): p.
3291-310.
26. Steelman, L.S., et al., Contributions of the Raf/MEK/ERK, PI3K/PTEN/Akt/mTOR
and Jak/STAT pathways to leukemia. Leukemia, 2008. 22(4): p. 686-707.
27. Ishikawa, K., et al., ROS-generating mitochondrial DNA mutations can regulate
tumor cell metastasis. Science, 2008. 320(5876): p. 661-4.

59
28. Petros, J.A., et al., mtDNA mutations increase tumorigenicity in prostate cancer.
Proc Natl Acad Sci U S A, 2005. 102(3): p. 719-24.
29. Miller, B., et al., Mitochondrial DNA variation in Alzheimer's disease reveals a
unique microprotein called SHMOOSE. Mol Psychiatry, 2022.
30. Hashimoto, Y., et al., A rescue factor abolishing neuronal cell death by a wide
spectrum of familial Alzheimer's disease genes and Abeta. Proc Natl Acad Sci U
S A, 2001. 98(11): p. 6336-41.
31. Chiba, T., et al., Colivelin prolongs survival of an ALS model mouse. Biochem
Biophys Res Commun, 2006. 343(3): p. 793-8.
32. Xu, X., et al., Humanin is a novel neuroprotective agent against stroke. Stroke,
2006. 37(10): p. 2613-9.
33. Muzumdar, R.H., et al., Acute humanin therapy attenuates myocardial ischemia
and reperfusion injury in mice. Arterioscler Thromb Vasc Biol, 2010. 30(10): p.
1940-8.
34. Yen, K., et al., The mitochondrial derived peptide humanin is a regulator of
lifespan and healthspan. Aging (Albany NY), 2020. 12(12): p. 11185-11199.
35. Reynolds, J.C., et al., MOTS-c is an exercise-induced mitochondrial-encoded
regulator of age-dependent physical decline and muscle homeostasis. Nat
Commun, 2021. 12(1): p. 470.
36. Lee, C., et al., The mitochondrial-derived peptide MOTS-c promotes metabolic
homeostasis and reduces obesity and insulin resistance. Cell Metab, 2015.
21(3): p. 443-54.
37. Mehta, H.H., et al., Metabolomic profile of diet-induced obesity mice in response
to humanin and small humanin-like peptide 2 treatment. Metabolomics, 2019.
15(6): p. 88.
38. Cobb, L.J., et al., Naturally occurring mitochondrial-derived peptides are agedependent regulators of apoptosis, insulin sensitivity, and inflammatory markers.
Aging (Albany NY), 2016. 8(4): p. 796-809.
39. Bachar, A.R., et al., Humanin is expressed in human vascular walls and has a
cytoprotective effect against oxidized LDL-induced oxidative stress. Cardiovasc
Res, 2010. 88(2): p. 360-6.
40. Thummasorn, S., et al., Humanin directly protects cardiac mitochondria against
dysfunction initiated by oxidative stress by decreasing complex I activity.
Mitochondrion, 2018. 38: p. 31-40.

60
41. Cui, H., Y. Kong, and H. Zhang, Oxidative stress, mitochondrial dysfunction, and
aging. J Signal Transduct, 2012. 2012: p. 646354.
42. Sreekumar, P.G., et al., The Mitochondrial-Derived Peptide Humanin Protects
RPE Cells From Oxidative Stress, Senescence, and Mitochondrial Dysfunction.
Invest Ophthalmol Vis Sci, 2016. 57(3): p. 1238-53.
43. Miller, B., et al., Peptides derived from small mitochondrial open reading frames:
Genomic, biological, and therapeutic implications. Exp Cell Res, 2020. 393(2): p.
112056.
44. Ikonen, M., et al., Interaction between the Alzheimer's survival peptide humanin
and insulin-like growth factor-binding protein 3 regulates cell survival and
apoptosis. Proc Natl Acad Sci U S A, 2003. 100(22): p. 13042-7.
45. Guo, B., et al., Humanin peptide suppresses apoptosis by interfering with Bax
activation. Nature, 2003. 423(6938): p. 456-61.
46. Lue, Y., et al., The Potent Humanin Analogue (HNG) Protects Germ Cells and
Leucocytes While Enhancing Chemotherapy-Induced Suppression of Cancer
Metastases in Male Mice. Endocrinology, 2015. 156(12): p. 4511-21.
47. Eriksson, E., et al., Protective role of humanin on bortezomib-induced bone
growth impairment in anticancer treatment. J Natl Cancer Inst, 2014. 106(3): p.
djt459.
48. Jia, Y., et al., The effects of humanin and its analogues on male germ cell
apoptosis induced by chemotherapeutic drugs. Apoptosis, 2015. 20(4): p. 551-
61.
49. Jia, Y., et al., The humanin analogue (HNG) prevents temozolomide-induced
male germ cell apoptosis and other adverse effects in severe combined immunodeficiency (SCID) mice bearing human medulloblastoma. Exp Mol Pathol, 2019.
109: p. 42-50.
50. Ramirez-Torres, A., et al., Racial differences in circulating mitochondria-derived
peptides may contribute to prostate cancer health disparities. Prostate, 2022.
82(13): p. 1248-1257.
51. Moreno Ayala, M.A., et al., Humanin Promotes Tumor Progression in
Experimental Triple Negative Breast Cancer. Sci Rep, 2020. 10(1): p. 8542.
52. Mottaghi-Dastjerdi, N., et al., Genome expression analysis by suppression
subtractive hybridization identified overexpression of Humanin, a target gene in
gastric cancer chemoresistance. Daru, 2014. 22(1): p. 14.

61
53. Omar, N.N., et al., Breaking the ritual metabolic cycle in order to save acetyl
CoA: A potential role for mitochondrial humanin in T2 bladder cancer
aggressiveness. J Egypt Natl Canc Inst, 2017. 29(2): p. 69-76.
54. Xiao, J., et al., Low circulating levels of the mitochondrial-peptide hormone
SHLP2: novel biomarker for prostate cancer risk. Oncotarget, 2017. 8(55): p.
94900-94909.
55. Gaines, A.R., et al., The association between race and prostate cancer risk on
initial biopsy in an equal access, multiethnic cohort. Cancer Causes Control,
2014. 25(8): p. 1029-35.
56. Miller, B., et al., Mitochondria-derived peptides in aging and healthspan. J Clin
Invest, 2022. 132(9).
57. Kim, S.J., et al., The mitochondrial-derived peptide humanin activates the
ERK1/2, AKT, and STAT3 signaling pathways and has age-dependent signaling
differences in the hippocampus. Oncotarget, 2016. 7(30): p. 46899-46912.
58. Lee, H., A.J. Jeong, and S.K. Ye, Highlighted STAT3 as a potential drug target
for cancer therapy. BMB Rep, 2019. 52(7): p. 415-423.
59. DeSantis, C.E., et al., Cancer statistics for African Americans, 2019. CA Cancer
J Clin, 2019. 69(3): p. 211-233.
60. Global Burden of Disease Cancer, C., et al., Global, Regional, and National
Cancer Incidence, Mortality, Years of Life Lost, Years Lived With Disability, and
Disability-Adjusted Life-Years for 29 Cancer Groups, 1990 to 2016: A Systematic
Analysis for the Global Burden of Disease Study. JAMA Oncol, 2018. 4(11): p.
1553-1568.
61. Bechis, S.K., P.R. Carroll, and M.R. Cooperberg, Impact of age at diagnosis on
prostate cancer treatment and survival. J Clin Oncol, 2011. 29(2): p. 235-41.
62. Ilic, D., et al., Prostate cancer screening with prostate-specific antigen (PSA) test:
a systematic review and meta-analysis. BMJ, 2018. 362: p. k3519.
63. Kelly, S.P., et al., Trends in the Incidence of Fatal Prostate Cancer in the United
States by Race. Eur Urol, 2017. 71(2): p. 195-201.
64. Kinlock, B.L., et al., Racial Disparity in Time Between First Diagnosis and Initial
Treatment of Prostate Cancer. Cancer Control, 2016. 23(1): p. 47-51.
65. Mahal, B.A., et al., Racial disparities in prostate cancer outcome among prostatespecific antigen screening eligible populations in the United States. Ann Oncol,
2017. 28(5): p. 1098-1104.

62
66. Woods-Burnham, L., et al., Physician Consultations, Prostate Cancer
Knowledge, and PSA Screening of African American Men in the Era of Shared
Decision-Making. Am J Mens Health, 2018. 12(4): p. 751-759.
67. Croteau, D.L. and V.A. Bohr, Repair of oxidative damage to nuclear and
mitochondrial DNA in mammalian cells. J Biol Chem, 1997. 272(41): p. 25409-
12.
68. Dong, J., L.J. Wong, and M.P. Mims, Mitochondrial inheritance and cancer.
Transl Res, 2018. 202: p. 24-34.
69. Kalsbeek, A.M.F., et al., Altered mitochondrial genome content signals worse
pathology and prognosis in prostate cancer. Prostate, 2018. 78(1): p. 25-31.
70. Zhou, W., et al., Peripheral blood mitochondrial DNA copy number is associated
with prostate cancer risk and tumor burden. PLoS One, 2014. 9(10): p. e109470.
71. Moore, A., et al., A prospective study of mitochondrial DNA copy number and the
risk of prostate cancer. Cancer Causes Control, 2017. 28(6): p. 529-538.
72. Xu, J., et al., Mitochondrial DNA copy number in peripheral blood leukocytes is
associated with biochemical recurrence in prostate cancer patients in African
Americans. Carcinogenesis, 2020. 41(3): p. 267-273.
73. Williams, B.A., et al., Finding Optimal Cutpoints for Continuous Covariates with
Binary and Time-to-Event Outcomes 2006, Mayo Clinic.
74. Chen, E.I., Mitochondrial dysfunction and cancer metastasis. J Bioenerg
Biomembr, 2012. 44(6): p. 619-22.
75. Hsu, C.C., L.M. Tseng, and H.C. Lee, Role of mitochondrial dysfunction in cancer
progression. Exp Biol Med (Maywood), 2016. 241(12): p. 1281-95.
76. Guha, M. and N.G. Avadhani, Mitochondrial retrograde signaling at the
crossroads of tumor bioenergetics, genetics and epigenetics. Mitochondrion,
2013. 13(6): p. 577-91.
77. Jeng, J.Y., et al., Maintenance of mitochondrial DNA copy number and
expression are essential for preservation of mitochondrial function and cell
growth. J Cell Biochem, 2008. 103(2): p. 347-57.
78. Castellani, C.A., et al., Thinking outside the nucleus: Mitochondrial DNA copy
number in health and disease. Mitochondrion, 2020. 53: p. 214-223.
79. Mei, H., et al., Reduced mtDNA copy number increases the sensitivity of tumor
cells to chemotherapeutic drugs. Cell Death Dis, 2015. 6(4): p. e1710.

63
80. Kumar, R., et al., Cytochrome c Deficiency Confers Apoptosome and
Mitochondrial Dysfunction in African-American Men with Prostate Cancer.
Cancer Res, 2019. 79(7): p. 1353-1368.
81. Mandavilli, B.S., J.H. Santos, and B. Van Houten, Mitochondrial DNA repair and
aging. Mutat Res, 2002. 509(1-2): p. 127-51.
82. Zhang, H. and K.K. Singh, Global genetic determinants of mitochondrial DNA
copy number. PLoS One, 2014. 9(8): p. e105242.
83. Reznik, E., et al., Mitochondrial DNA copy number variation across human
cancers. Elife, 2016. 5.
84. Yu, M., et al., Reduced mitochondrial DNA copy number is correlated with tumor
progression and prognosis in Chinese breast cancer patients. IUBMB Life, 2007.
59(7): p. 450-7.
85. Borboroglu, P.G., et al., Extensive repeat transrectal ultrasound guided prostate
biopsy in patients with previous benign sextant biopsies. J Urol, 2000. 163(1): p.
158-62.
86. Stewart, C.S., et al., Prostate cancer diagnosis using a saturation needle biopsy
technique after previous negative sextant biopsies. J Urol, 2001. 166(1): p. 86-
91; discussion 91-2.
87. Sejal Vyas, E.Z., Marcia C Haigis, Mitochondria and Cancer. Cell, 2016.
88. Ferro, F., et al., Autophagy and mitophagy in cancer metabolic remodelling.
Semin Cell Dev Biol, 2020. 98: p. 129-138.
89. Peng, C., et al., Mitochondrial ROS driven by NOX4 upregulation promotes
hepatocellular carcinoma cell survival after incomplete radiofrequency ablation by
inducing of mitophagy via Nrf2/PINK1. J Transl Med, 2023. 21(1): p. 218.
90. Chandra, D. and K.K. Singh, Genetic insights into OXPHOS defect and its role in
cancer. Biochim Biophys Acta, 2011. 1807(6): p. 620-5.
91. Khitun, A., T.J. Ness, and S.A. Slavoff, Small open reading frames and cellular
stress responses. Mol Omics, 2019. 15(2): p. 108-116.
92. Sandmann, C.L., et al., Evolutionary origins and interactomes of human, young
microproteins and small peptides translated from short open reading frames. Mol
Cell, 2023. 83(6): p. 994-1011 e18.

64
93. Miller, B., et al., Mitochondrial DNA variation in Alzheimer's disease reveals a
unique microprotein called SHMOOSE. Mol Psychiatry, 2023. 28(4): p. 1813-
1826.
94. Rochette, L., et al., Role of humanin, a mitochondrial-derived peptide, in
cardiovascular disorders. Arch Cardiovasc Dis, 2020. 113(8-9): p. 564-571.
95. Zuccato, C.F., et al., Mitochondrial-derived peptide humanin as therapeutic target
in cancer and degenerative diseases. Expert Opin Ther Targets, 2019. 23(2): p.
117-126.
96. Ju, Y.S., et al., Origins and functional consequences of somatic mitochondrial
DNA mutations in human cancer. Elife, 2014. 3.
97. Rey, J., et al., PEP-FOLD4: a pH-dependent force field for peptide structure
prediction in aqueous solution. Nucleic Acids Res, 2023. 51(W1): p. W432-W437.
98. Dobin, A., et al., STAR: ultrafast universal RNA-seq aligner. Bioinformatics, 2013.
29(1): p. 15-21.
99. Lim, K., Mitochondrial genome editing: strategies, challenges, and applications.
BMB Rep, 2024. 57(1): p. 19-29.
100. Bailey, M.H., et al., Comprehensive Characterization of Cancer Driver Genes and
Mutations. Cell, 2018. 173(2): p. 371-385 e18.
101. Gorlov, I.P., et al., Tumor somatic mutations also existing as germline
polymorphisms may help to identify functional SNPs from genome-wide
association studies. Carcinogenesis, 2020. 41(10): p. 1353-1362.
102. Forbes, S.A., et al., COSMIC: High-Resolution Cancer Genetics Using the
Catalogue of Somatic Mutations in Cancer. Curr Protoc Hum Genet, 2016. 91: p.
10 11 1-10 11 37.
103. Tate, J.G., et al., COSMIC: the Catalogue Of Somatic Mutations In Cancer.
Nucleic Acids Res, 2019. 47(D1): p. D941-D947.
104. Zhao, Y., et al., Significance of mutation in the D-loop region in cervical cancer.
West Indian Med J, 2010. 59(3): p. 291-4.
105. Zhao, Y.B., et al., Mutation in D-loop region of mitochondrial DNA in gastric
cancer and its significance. World J Gastroenterol, 2005. 11(21): p. 3304-6.
106. Ashtiani, Z.O., et al., Mitochondrial D-Loop polymorphism and microsatellite
instability in prostate cancer and benign hyperplasia patients. Asian Pac J
Cancer Prev, 2012. 13(8): p. 3863-8.

65
107. Wang, L., et al., Clinical Significance of the D-Loop Gene Mutation in
Mitochondrial DNA in Laryngeal Cancer. Onco Targets Ther, 2021. 14: p. 3461-
3466.

66
SUPPLEMENTAL FIGURES AND TABLES
Supplemental Figure 2-1: Comparison of control Mitochondrial DNA copy number
levels in White and Black individuals. Wilcoxon rank sum test was used to assess
differences between Black and White individuals without prostate cancer in both Plasma
(A) and White Blood Cells (B). For Plasma mtDNA, breaks in the y-axis were created to
better visualize the majority of the data. Ranges of plasma mtDNA shown include 0-
2750 and 3650-5400.
White Controls
Black Controls
500
1000
1500
2000
2500
200
4000
5000
Plasma mtDNA copy number
p=0.021
White Controls
Black Controls
0
200
400
600
800
WBC mtDNA copy number
p<0.001

67

68

69
Supplemental Figure 2-2: Comparison of the distribution of Mitochondrial DNA
copy number by prostate cancer grade for Black and White Individuals. Plasma
and WBC mtDNA copy number measures stratified by race and cancer-grade.

70
Supplemental Figure 3-1: Endogenous Detection of SCREAM in LnCaP cells.
Unique mass spectrometry of a SCREAM derived fragment detected in whole cell
lysates of LnCaP cells.
MNHSRELSMHLVFSSGGYARDSIA

Abstract (if available)

Abstract The intricate relationship between mitochondrial biology and cancer is a burgeoning field of study with significant implications for understanding and treating the disease. This dissertation explores the role of mitochondrial microproteins and mitochondrial DNA (mtDNA) alterations in cancer biology. In chapter one, we begin by reviewing the emerging class of mitochondrial-encoded microproteins, highlighting their regulatory roles in apoptosis, reactive oxygen species (ROS) production, and mitochondrial metabolism, with a focus on humanin and small humanin-like peptides (SHLPs). These microproteins demonstrate potential as key modulators of cancer, warranting further investigation into their mechanisms and therapeutic applications.
In chapter two, our research delves into racial disparities in prostate cancer by examining mtDNA copy number as a biomarker. Our research reveals that elevated mtDNA copy number is a potent predictor of prostate cancer in White individuals but not in Black individuals, who exhibit higher baseline mtDNA levels. These findings suggest that mtDNA copy number may contribute to racial health disparities in prostate cancer and could inform racially tailored diagnostic approaches.
Finally, in chapter three, we identify a novel somatic mutation in the mitochondrial microprotein SCREAM, linked to cancer progression. This mutation, a single nucleotide variant in the D-loop region of mtDNA, results in an amino acid change that enhances the pro-oncogenic properties of the SCREAM microprotein. The discovery of this mutation and its functional consequences further underscores the potential of mitochondrial microproteins as novel therapeutic targets across multiple cancer types.
Collectively, this dissertation underscores the critical role of mitochondrial biology in cancer, with a focus on mitochondrial microproteins and mtDNA alterations as promising avenues for future research and therapeutic development.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Discovery of mitochondria-DNA-encoded microproteins in neurodegeneration

PDF

The regulation, roles, and mechanism of action of mitochondrial-derived-peptides (MDPs) in aging

PDF

Air pollution, mitochondrial function, and growth in children

PDF

Cisplatin activates mitochondrial oxphos leading to acute treatment resistance in bladder cancer

PDF

Differences in the mutational landscape of clonal hematopoiesis of indeterminate potential among Races and between Male and Female patients with cancer.

PDF

Microenvironmental and biomechanical regulation of mitochondrial membrane potential in cancer cells

PDF

Ancestral inference and cancer stem cell dynamics in colorectal tumors

PDF

Molecular elucidation of nonhomologous DNA end-joining in the context of nucleosome core particles

PDF

DNA methylation as a biomarker in human reproductive health and disease

PDF

LINC00261 alters DNA repair and confers resistance to cisplatin independent of FOXA2 in lung adenocarcinoma

PDF

A multi-center prospective cohort study of the diagnostic yield and patient experience of multiplex gene panel testing for hereditary cancer risk

PDF

Exploration of the roles of cancer stem cells and survivin in the pathogenesis and progression of prostate cancer

PDF

Unraveling the role of iron-sulfur clusters in DNA repair enzymes

PDF

The association of prediagnostic metformin use with prostate cancer in the multiethnic cohort study

PDF

Sex differences in aging and the effects of mitochondria

PDF

Exploring the LAG-3/FGL-1 pathway: novel insights and therapeutic approaches in cancer immunotherapy

PDF

The influence of DNA repair genes and prenatal tobacco exposure on childhood acute lymphoblastic leukemia risk: a gene-environment interaction study

PDF

Monoamine oxidase (MAO) knock-out mouse models: Tools for studying the molecular basis of aggression, anxiety, autism and cancers

PDF

Dissecting novel roles for MAFR-1 in reproduction and metabolic homeostasis

PDF

Role of steroidogenic acute regulatory protein (STAR) and sterol carrier protein-x (SCPx) in the transport of cholesterol and other lipids

Asset Metadata

Creator Flores, Melanie Kristine (author)

Core Title The intersection of mitochondrial biology and cancer: insights from mitochondrial microproteins and mtDNA alterations

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Molecular Biology

Degree Conferral Date 2024-08

Publication Date 08/31/2024

Defense Date 08/22/2024

Publisher Los Angeles, California (original), University of Southern California (original), University of Southern California. Libraries (digital)

Tag cancer,mitochondria,mitochondrial DNA copy number,mitochondrial microproteins,mtDNA,somatic mutations

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Cohen, Pinchas (committee chair), Chen, Xiaojiang (committee member), Curran, Sean (committee member), Lenz, Heinz-Josef (committee member)

Creator Email melliekflo@gmail.com,mkflores@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC11399A0AA

Unique identifier UC11399A0AA

Identifier etd-FloresMela-13464.pdf (filename)

Legacy Identifier etd-FloresMela-13464

Document Type Dissertation

Format theses (aat)

Rights Flores, Melanie Kristine

Internet Media Type application/pdf

Type texts

Source 20240831-usctheses-batch-1205 (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu