Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Role of steroidogenic acute regulatory protein (STAR) and sterol carrier protein-x (SCPx) in the transport of cholesterol and other lipids
(USC Thesis Other)
Role of steroidogenic acute regulatory protein (STAR) and sterol carrier protein-x (SCPx) in the transport of cholesterol and other lipids
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
ROLE OF STEROIDOGENIC ACUTE REGULA TORY PROTEIN (ST AR)
AND STEROL CARRIER PROTEIN-X (SCPX) IN THE TRANSPORT OF
CHOLESTEROL AND OTHER LIPIDS
by
Melanie Galano
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 PHARMACOLOGY AND TOXICOLOGY)
August 2022
Copyright 2022 Melanie Galano
ii
Dedication
To my father, the motivation behind all my aspirations, for his unending support.
This is for you and because of you.
iii
Acknowledgements
My utmost gratitude goes to my PI and mentor, Dr. Vassilios Papadopoulos for his belief
in my ability to learn and succeed. I am grateful for his unwavering support throughout this journey
and for his guidance through the challenges that these past five years have brought. To have a
mentor as inspirational and a scientist as knowledgeable and passionate as him as a PI has truly
been an honor.
I also appreciate Dr. Martine Culty for being one of my committee members and for being
one of my closest advisors throughout my PhD. Her input, suggestions, and expertise in the field
have greatly shaped my development as a scientist as well as the work discussed here. I would like
to thank Dr. Barry Zirkin for his continuous support, advice, and encouragement as my committee
member and I am thankful for the time that was put into helping my work become stronger.
I would also like to thank other members of the Papadopoulos and Culty labs for their help
and support throughout these years. To Chantal Sottas, for being the best lab manager we could
ask for and for always being willing to help with experiments, orders, or any other issue I run into
regardless of its complexity. To Dr. Yuchang Li, for being another mentor to me in the lab and for
always being willing to share your scientific knowledge with me in every way from going over
new concepts with me, to teaching me new techniques, to helping me troubleshoot experiments.
To the other students in the Papadopoulos and Culty labs for always being supportive of me and
for helping me on a daily basis through the ups and downs of the PhD journey.
Lastly, I thank my family and closest friends, especially my parents Rogelio and Marilou
and my brother Ryan, for all the love and support they’ve given me throughout every stage of my
life, and for all the sacrifices they’ve made to help get me to this point today. I wouldn’t be the
person or the scientist I am today without each of you.
iv
Acknowledgement of Contributions of Authors to Manuscripts
Manuscript I: Role of STAR and SCP2/SCPx in the Transport of Cholesterol and Other Lipids.
Galano M. and Papadopoulos V. 2022. Manuscript to be submitted to the International Journal of
Molecular Sciences.
• Melanie Galano: writing
• Vassilios Papadopoulos: writing
Manuscript II: Role of Constitutive STAR in Leydig Cells. Galano M, Li Y, Li L, Sottas C,
Papadopoulos V. International Journal of Molecular Sciences. 2021 Feb 18;22(4):2021. doi:
10.3390/ijms22042021.
• Melanie Galano: Conceptualization, methodology, data curation, investigation, formal
analysis, writing
• Yuchang Li: methodology
• Lu Li: methodology
• Chantal Sottas: methodology
• Vassilios Papadopoulos: Conceptualization, methodology, data curation, investigation,
formal analysis, writing, resources, funding acquisition
Manuscript III: Role of Constitutive STAR in Mitochondrial Structure and Function in MA-10
Leydig Cells. Galano M. and Papadopoulos V. 2022. Endocrinology. Manuscript under revision.
• Melanie Galano: Conceptualization, methodology, data curation, investigation, formal
analysis, writing
• Vassilios Papadopoulos: Conceptualization, methodology, data curation, investigation,
formal analysis, writing, resources, funding acquisition
v
Manuscript IV: SCPx deficiency caused by novel heterozygous SCP2 variant leads to alterations
in lipid metabolism. Galano M., Ezzat S., Lerner-Ellis J., Papadopoulos V. 2022. Manuscript to be
submitted to the American Journal of Human Genetics.
• Melanie Galano: Conceptualization, methodology, data curation, investigation, formal
analysis, writing
• Shereen Ezzat: Investigation, writing, formal analysis
• Jordan Lerner-Ellis: Investigation, writing, formal analysis
• Vassilios Papadopoulos: Conceptualization, methodology, data curation, investigation,
formal analysis, writing, resources, funding acquisition
vi
Table of Contents
Dedication ....................................................................................................................................... ii
Acknowledgements ........................................................................................................................ iii
List of Tables ............................................................................................................................... viii
List of Figures ................................................................................................................................ ix
Abstract .......................................................................................................................................... xi
Chapter I: Role of STAR and SCP2/SCPx in the Transport of Cholesterol and Other Lipids ....... 1
Abstract ....................................................................................................................................... 1
1. Introduction ......................................................................................................................... 1
1.1. Synthesis ..................................................................................................................... 3
1.2. Distribution ................................................................................................................. 4
1.3. Intracellular trafficking: vesicular and non-vesicular cholesterol transport ............... 6
2. Model cholesterol transport proteins .................................................................................. 7
2.1. STAR .......................................................................................................................... 7
2.2. SCP2/SCPx ............................................................................................................... 12
3. Conclusion ........................................................................................................................ 19
Chapter II: Role of Constitutive STAR in Leydig Cells ............................................................... 21
Abstract ..................................................................................................................................... 21
1. Introduction ....................................................................................................................... 22
2. Results .............................................................................................................................. 23
3. Discussion ........................................................................................................................ 37
4. Materials and Methods ..................................................................................................... 42
Chapter III: Role of Constitutive STAR in Mitochondrial Structure and Function in MA-10
Leydig Cells .................................................................................................................................. 47
Abstract ..................................................................................................................................... 47
1. Introduction ....................................................................................................................... 47
2. Materials and Methods ...................................................................................................... 49
3. Results ............................................................................................................................... 53
4. Discussion ......................................................................................................................... 67
vii
Chapter IV: SCPx Deficiency Caused by Novel Heterozygous SCP2 Variant Leads to Severe
Alterations in Lipid Metabolism ................................................................................................... 73
Abstract ..................................................................................................................................... 73
1. Introduction ....................................................................................................................... 74
2. Materials and Methods ...................................................................................................... 76
3. Results ............................................................................................................................... 82
4. Discussion ....................................................................................................................... 101
Chapter V: Summary and Discussion ......................................................................................... 108
References ................................................................................................................................... 118
viii
List of Tables
Table 1. Table of information about antibodies .......................................................................... 64
Table 2. Mitochondria-related differentially expressed gene list between wild-type MA-10
and STARKO1 cells ................................................................................................................... 67
Table 3. Quantification of normal versus damaged mitochondria in TEM images .................... 73
Table 4. Patient plasma fatty acid panel ..................................................................................... 95
Table 5. Summary of RNA-sequencing-derived pathways ...................................................... 105
Table 6. Summary of lipidomics-derived pathways ................................................................. 107
ix
List of Figures
Figure 1. Function of hormone-induced STAR and constitutive STAR .................................... 23
Figure 2. SCP2/SCPx function in the transport and metabolism of cholesterol and other
lipids ............................................................................................................................................ 30
Figure 3. STAR expression and levels in various models in the absence and presence of
hormonal stimulation .................................................................................................................. 37
Figure 4. Screening and validation of CRISPR/Cas9-mediated STAR KO in MA-10 cells ...... 38
Figure 5. Progesterone production by wild-type MA-10 and STAR KO cell lines in response
to stimulus ................................................................................................................................... 41
Figure 6. Progesterone production by wild-type MA-10 and STAR KO cell lines in response
to TSPO drug ligands .................................................................................................................. 43
Figure 7. Alterations in lipid droplet content between wild-type MA-10 and STARKO1
cells ............................................................................................................................................. 44
Figure 8. Role of DAG signaling in progesterone production by wild-type MA-10 and
STARKO1 cells .......................................................................................................................... 48
Figure 9. STAR KO results in alterations in mitochondrial structure and function ................... 68
Figure 10. Reintroduction of STAR in STARKO1 cells ............................................................ 71
Figure 11. Mitochondrial ultrastructure in STARKO-OE1 and STARKO-OE2 cells ............... 73
Figure 12. Characterization of mitochondrial function in STARKO-OE1 and STARKO-OE2
cells ............................................................................................................................................. 75
Figure 13. Localization of STAR to the mitochondria ............................................................... 78
Figure 14. Characterization of STAR overexpression in wild-type MA-10 cells ...................... 80
Figure 15. Characterization of SCP2 mutation in patient fibroblasts ......................................... 98
Figure 16. Quantification of peroxisome abundance in NHDF and WESP cells ..................... 100
Figure 17. Western blot analyses of β-oxidation enzymes ....................................................... 102
Figure 18. Confirmation of differentially expressed genes identified by RNA sequencing ..... 104
Figure 19. Lipidomic analyses of various lipid groups between NHDF and WESP cells ........ 108
x
Figure 20. SCPx protein levels in WESP cells in response to fenofibrate and
4-hydroxytamoxifen treatment .................................................................................................. 112
Figure 21. Fatty acid profile of NHDF and WESP cells following treatment .......................... 114
xi
Abstract
The trafficking of cholesterol between cellular compartments is tightly regulated, as
cholesterol must be properly distributed to carry out its various key functions. One mechanism
through which cholesterol is trafficked in the cell is via cholesterol transport proteins, involving
the movement of cholesterol between membranes via binding to specific proteins. These include
the steroidogenic acute regulatory protein (STAR), which is classically known to facilitate
cholesterol transport to the mitochondria for steroid biosynthesis, and sterol carrier protein 2/sterol
carrier protein-x (SC2/SCPx), which are non-specific lipid transfer proteins suggested to function
in the transport of many lipids including cholesterol. Although much work has been done to
investigate the roles of these proteins in cholesterol transport, many aspects of their functions are
still unclear.
In order to elucidate the function of constitutive STAR, or STAR that is present under basal
conditions as opposed to hormone-induced STAR, we used CRISPR/Cas9 technology to generate
a STAR knockout (KO) MA-10 mouse tumor Leydig cell line (STARKO1). STAR KO cells had
significantly increased levels of cholesteryl ester, diacylglycerol, and phosphatidylcholine
compared to wild-type (WT) MA-10 cells. Additionally, many lipid-related genes were
differentially expressed between WT MA-10 and STARKO1 cells. Further, we showed that
mitochondrial structure and function were altered by STAR knockout and that reintroduction of
STAR into STARKO1 cells exacerbated, rather than recovered mitochondrial structure and
function. Additionally, the processing of STAR into its mature form was inhibited in STARKO1
cells with STAR overexpression suggesting that mitochondrial dysfunction alters STAR
processing. Taken together, these data indicate that constitutive STAR may have roles in lipid
xii
metabolism and mitochondrial function which are independent of the role of hormone-induced
STAR in cholesterol transport for steroidogenesis.
We also investigated the role(s) of SCPx in a patient with a mutation in SCP2 presenting
with brainstem neurodegeneration and testicular defects. The patient’s SCP2 mutation led to
decreased levels of SCPx, but normal levels of SCP2. RNA sequencing identified many
differentially expressed genes that were altered between patient and control fibroblasts. Lipidomic
analyses identified many species of free fatty acids, acylcarnitines, sterols, phospholipids, and
sphingolipids that had altered levels between patient and control fibroblasts. Pathway analyses
using transcriptomic and lipidomic data identified several metabolic pathways that were affected
by the patient’s SCP2 mutation such as PPAR signaling, cholesterol metabolism, and fatty acid
biosynthesis. We also identified two compounds, fenofibrate and 4-hydroxytamoxifen, that
recovered SCPx levels and improved fatty acid levels in patient fibroblasts. Collectively, these
data suggest that SCPx plays a role in the transport and/metabolism of many lipids that function
in several key pathways, which may explain many of the clinical presentations associated with
SCPx deficiency.
Thus, the data presented here suggest that classic examples of intracellular cholesterol
transport proteins play a more general role in lipid transport and metabolism in addition to their
respective roles in cholesterol trafficking.
1
Chapter I: Role of STAR and SCP2/SCPx in the Transport of
Cholesterol and Other Lipids
Abstract
Cholesterol is a lipid molecule essential for several key cellular processes including
steroidogenesis. As such, the trafficking and distribution of cholesterol is tightly regulated by
various pathways that include vesicular and non-vesicular mechanisms. Non-vesicular
mechanisms include the binding of cholesterol to cholesterol transport proteins, which facilitates
the movement of cholesterol between cellular membranes. Classic examples of cholesterol
transport proteins are the steroidogenic acute regulatory protein (STAR), which facilitates
mitochondrial cholesterol transport for acute steroidogenesis, and SCP2/SCPx which are non-
specific lipid transfer proteins suggested to be involved in the transport and metabolism of many
lipids including cholesterol between several cellular compartments. This review discusses the
roles of STAR and SCP2/SCPx in cholesterol transport as model cholesterol transport proteins,
as well as more recent findings that support the role of these proteins in the transport and/or
metabolism of other lipids.
1. Introduction
Cholesterol is a lipid molecule that has been associated with many diseases such as
cardiovascular disease, atherosclerosis, Alzheimer’s disease, and different types of cancers [1-3].
While much work centered on cholesterol is focused on its dysregulation and role in the
pathology of these diseases, cholesterol is indeed an essential lipid molecule critical in the
maintenance of a variety of indispensable functions and pathways. Cholesterol is a 27-carbon
2
molecule containing four fused rings and an 8-carbon tail, making it a hydrophobic molecule [4].
As such, cholesterol is a major component of cellular membranes, where cholesterol interacts
with other membrane lipids to regulate the permeability of the membrane to certain ions and
solutes, rigidity of the membrane to act as a scaffold for membrane proteins, and fluidity of the
membrane to allow for rapid diffusion, budding to form vesicles, or fusion with other membranes
for trafficking [5]. Furthermore, cholesterol is the precursor for all steroid hormones, which play
vital roles in reproduction, salt and water balance, and stress response [4]. Cholesterol is also
metabolized into various oxysterols that then give rise to bile acids, which are critical for the
digestion and absorption of lipids [4, 6]. Because of its role in several key cellular processes,
cholesterol must be properly trafficked throughout the cell, which can occur via vesicular or non-
vesicular pathways, the latter of which involves cholesterol transport proteins [7]. Cholesterol is
suggested to interact with a variety of cholesterol transport proteins for its intracellular
trafficking, such as the steroidogenic acute regulatory protein (STAR; STARD1), a key protein
in acute steroidogenesis, and sterol carrier protein 2/sterol carrier protein-x (SCP2/SCPx), which
are non-specific lipid transfer proteins suggested to play a role in cholesterol transport and
metabolism [8, 9]. While STAR is classically known to transport cholesterol for mitochondrial
steroidogenesis, recent data suggest that STAR may function in the transport and/or metabolism
of other lipids [10]. Additionally, since the discovery of STAR as the key cholesterol transporter
for steroid biosynthesis, rather than SCP2, there has been less emphasis on the general role of
SCP2 in the transport of cholesterol and other lipids [11-13]. In this review, we discuss
intracellular cholesterol distribution and current evidence of the multifunctional roles of STAR
and SCP2/SCPx as model cholesterol transport proteins.
3
1.1.Synthesis
Cholesterol is derived from two sources: de novo synthesis or dietary intake. While all
mammalian cells are able to synthesize cholesterol, steroidogenic cells of the adrenals, testis,
ovaries, and brain synthesize cholesterol at the highest rates [14, 15]. De novo cholesterol
biosynthesis occurs in the endoplasmic reticulum (ER) by a complex process involving several,
tightly regulated enzymes of the mevalonate pathway [4, 16, 17]. Briefly, this process involves
the subsequent reactions of the condensation of three molecules of acetyl-CoA forming
hydroxymethylglutaryl-CoA (HMG-CoA), the reduction of HMG-CoA to mevalonate, the
conversion of mevalonate to isoprenoids, the polymerization of the isoprenoids to squalene,
which is cyclized to lanosterol, and finally the conversion of lanosterol to cholesterol. The rate-
limiting step of the mevalonate pathway is the conversion of HMG-CoA to mevalonate, which is
catalyzed by HMG-CoA reductase enzyme (HMGCR) [18]. HMGCR is the target of the class of
cholesterol lowering drugs known as statins [19].
In addition to de novo cholesterol synthesis, cholesterol is also supplied to the cell via
dietary intake. Cholesterol derived from food is absorbed by the small intestine, packed into
chylomicrons, and transported to the liver, where cholesterol is processed into very low-density
lipoproteins (VLDL) [20]. Through circulation, VLDL become low-density lipoproteins (LDL),
which bind to the LDL receptor (LDLR) on the cell surface. This complex is endocytosed
through clathrin-coated pits into the cytoplasm which then fuse to endosomal compartments for
processing of the LDL, while the LDLR is recycled back to the cell surface. The cholesteryl
esters derived from LDL are then de-esterified to free cholesterol by lysosomal acid lipase [4,
21]. Another mechanism of cholesterol uptake into the cell occurs via a non-endocytic pathway
involving the uptake of high-density lipoprotein (HDL) by the scavenger receptor class B, type I
4
(SR-BI), which is mostly expressed in the steroidogenic cells of the testis, ovaries, and adrenals
[22]. HDL particles bind SR-BI, which transfers HDL cholesterol to the plasma membrane [23].
The cholesteryl esters derived from HDL are then hydrolyzed by hormone sensitive lipase,
forming free cholesterol [24].
1.2.Distribution
Regulation of cellular cholesterol levels occurs at many levels. Cholesterol biosynthesis
can be regulated by the sterol regulatory element-binding protein 2 (SREBP2), which regulates
transcription of genes important for cholesterol biosynthesis, and by HMGCR [1]. Additionally,
cholesterol homeostasis may be regulated by controlling LDLR-mediated uptake of cholesterol
or by controlling cholesterol efflux by cholesterol efflux transporters such as the ATP-binding
cassette subfamily A member 1 (ABCA1) [1]. Further, levels of cholesterol can be controlled by
cholesterol esterification through the regulation of acyl-coenzyme A: cholesterol acyltransferases
(ACAT1 and ACAT2) [25]. While active, free cholesterol is most commonly found in
membranes or trafficked to other organelles for other functions, cholesterol esterification
prevents the accumulation of free cholesterol and primes esterified cholesterol for storage in lipid
droplets [26].
Lipids of biological mammalian membranes include sphingolipids, glycerol-based lipids,
and cholesterol and the lipid composition of a specific membrane is determined in part by its
subcellular location [27]. In the cell, about 65-80% of cellular cholesterol is present at the plasma
membrane, and the Golgi apparatus and endosomal recycling compartment (ERC), which are
closely associated with the plasma membrane, have intermediate cholesterol levels [28].
However, the ER has as low as 0.1-2% of total cellular cholesterol despite this being the site
cholesterol biosynthesis [29]. This has led to the notion that newly synthesized cholesterol in the
5
ER is rapidly transported to other organelles for other functions or stored in lipid droplets as
cholesteryl esters. Additionally, mitochondria are also cholesterol-poor, despite mitochondria
being the site in which steroidogenesis is initiated [30, 31]. Therefore, the low concentrations of
mitochondrial cholesterol necessitate the rapid transport of cholesterol from intracellular stores
into mitochondria for steroid production [32].
In addition to the variability in cholesterol distribution between the membranes of different
organelles, cholesterol distribution within individual membranes is also heterogeneous. So called
cholesterol-rich domains in membranes form partly based on the distribution of proteins inserted
into the membrane [27]. For example, if a protein does not interact with cholesterol, the protein
is likely to stabilize a cholesterol-poor region of the membrane, which leads cholesterol to
accumulate at other regions of the membrane that are then enriched in cholesterol and become
more tightly packed [33]. Conversely, instead of being excluded from cholesterol enriched
regions, some proteins are found to preferentially interact with cholesterol-rich domains. Some
general characteristics of proteins that would favor interaction with cholesterol-rich domains
include lipidation or the presence of certain peptide segments. Types of lipidation that encourage
association of certain proteins with these domains include the covalent attachment of cholesterol,
lipid anchoring by glycosylphosphatidylinositol (GPI), and acylation of cysteine residues with
palmitic acid or of the N-terminal group with myristic acid [34-37]. In addition to lipidation,
there are also certain segments of integral membrane proteins at the membrane surface that are
able to recognize cholesterol. Evaluation of all proteins known to interact with cholesterol led to
the identification of the cholesterol recognition/interaction amino acid consensus (CRAC) motif,
-L/V-(X)1-5-Y-(X)1-5-R/K-, where (-X-)1-5 is one to five residues of any amino acid [38]. An
example of a protein with a CRAC motif is the translocator protein (TSPO), which is an 18 kDa
6
outer mitochondrial membrane protein shown to facilitate the transport of cholesterol to the inner
mitochondrial membrane (IMM) [39]. It is likely that the presence of a CRAC domain in TSPO
plays a role in its positioning at a cholesterol-rich membrane domain [27].
1.3. Intracellular trafficking: vesicular and non-vesicular cholesterol transport
The transport of cholesterol between organelles or from intracellular stores is vital for many
cellular processes, therefore, proper cholesterol trafficking is finely regulated. Because
cholesterol itself is insoluble, it is trafficked in the cell in two general ways: vesicular pathways
and non-vesicular pathways. Vesicular cholesterol transport involves the delivery of membrane
cholesterol via a vesicle that is formed by budding off of the donor membrane followed by the
vesicle fusing with another membrane [40]. Vesicular cholesterol transport, such as the
trafficking of cholesterol via internalization of LDL derived cholesterol, vesicle formation, and
movement of cholesterol through the endosomal and lysosomal pathways, requires an intact
cytoskeleton and metabolic energy [29, 41].
Although much work has been done elucidating mechanisms of vesicular cholesterol
trafficking, evidence shows that inhibiting vesicular transport by genetic or pharmacological
means does not inhibit intracellular cholesterol transport [42-44]. Furthermore, work has shown
that cholesterol trafficking between the ER to the plasma membrane occurs much faster than
what is seen with vesicular trafficking via membrane proteins [45]. Non-vesicular cholesterol
trafficking involves the transport of cholesterol from a donor membrane to an acceptor
membrane, which does not require metabolic energy. This process can occur spontaneously, via
passive diffusion, the rate of which is dependent on the donor membrane’s lipid composition, by
cholesterol transport proteins, either soluble or membrane-bound, or by membrane contact sites,
which involve brief interactions between membranes [46]. Although passive diffusion of
7
cholesterol is a slow process, membrane contact sites or cholesterol transport proteins can
accelerate cholesterol transport [41, 46]. This review focuses on the role of cholesterol transport
proteins in intracellular cholesterol trafficking for various cellular processes.
2. Model cholesterol transport proteins
There are several proteins and protein families that have been indicated to facilitate non-
vesicular cholesterol transport between membranes. Here, we will discuss in depth model
cholesterol transport proteins: STAR, which is known to be critical for cholesterol transport into
the mitochondria for hormone-induced steroidogenesis and SCP2/SCPx, which are non-specific
lipid transfer proteins suggested to play a role in the transport and/or metabolism of many lipids
including cholesterol.
2.1. STAR
One important transport route of cholesterol is the delivery of cholesterol from intracellular
stores into mitochondria for steroid biosynthesis. Hormonal stimulation of steroidogenic cells
results in the rapid delivery of cholesterol, the precursor of all steroids, to the IMM where
cholesterol is first converted to pregnenolone [47]. One of the proteins indispensable for steroid
biosynthesis is STAR, which is known to facilitate the transfer of cholesterol to mitochondria
upon hormonal stimulation [48]. In addition to STAR, there are many other proteins and
enzymes involved in steroid biosynthesis, spanning several cellular compartments.
2.1.1. Cholesterol transport in steroid biosynthesis
The rate-limiting step of acute steroid biosynthesis is the transport of cholesterol derived
from intracellular stores at the outer mitochondrial membrane (OMM) to the matrix side of the
IMM where the cytochrome P450 side chain cleavage enzyme CYP11A1 resides [47]. This
process is initiated by stimulation of steroidogenic cells by the pituitary trophic hormones
8
luteinizing hormone (LH), follicle stimulating hormone (FSH), and adrenocorticotropic hormone
(ACTH), which induce cyclic AMP (cAMP) production [49]. Previous work in our lab has
shown that hormonal stimulation and cAMP induction initiates the formation of a multi-protein
complex called the transduceosome, consisting of cytosolic and OMM proteins that together
facilitate cholesterol transport across the OMM [50]. The cytosolic components of the
transduceosome include the acyl-CoA binding domain-containing 3 (ACBD3), protein kinase A
regulatory subunit 1 (PKA-RIα), and hormone induced STAR, which are anchored to the OMM
components of the transduceosome: TSPO and voltage dependent anion channel 1 (VDAC1).
While the exact mechanism by which cholesterol moves through the transduceosome is still
unclear, disruption of the interactions between these proteins and knockdown or deletion studies
demonstrate the importance of each of these proteins in cholesterol transport for steroid
biosynthesis [10, 50-53]. Once at the OMM, proteins of the steroidogenic metabolon, which
include TSPO and VDAC along with the IMM proteins ATPase family AAA domain-containing
protein 3A (ATAD3A) and CYP11A1, facilitate the transport of cholesterol to the matrix side of
the IMM where cholesterol is converted to the first steroid, pregnenolone, by CYP11A1 [54].
The direct interaction between STAR and several of these proteins such as TSPO and VDAC1
has shown to be critical for steroidogenesis [32, 55].
2.1.1. Role of STAR in steroidogenesis
Because steroidogenic cells store very low amounts of steroids, steroids must be
synthesized rapidly in response to hormonal stimulation. While chronic steroidogenesis involves
a slow process of the transcription of steroidogenic enzymes, acute steroidogenesis involves the
rapid transfer of cholesterol to CYP11A1 [56]. STAR was first discovered when it was shown
that the acute steroidogenic response paralleled the synthesis of a 37 kDa phosphoprotein, or
9
STAR [57, 58]. Overexpression of STAR in MA-10 mouse tumor Leydig cells was found to
induce steroidogenesis to a similar extent as cAMP and introduction of STAR into non-
steroidogenic COS-1 cells transfected with the CYP11A1 system induced steroidogenesis 6-fold
[48, 59-61]. Further, the key role of STAR in cholesterol transport for acute steroidogenesis is
most evident through findings showing that STAR mutations in humans cause congenital lipoid
adrenal hyperplasia (lipoid CAH), a disease characterized by a severe deficiency in steroid
production and an accumulation of cholesterol in steroidogenic cells [62, 63]. Star knockout
(KO) mice have a similar phenotype as humans with STAR mutations, however gonadal function
was less affected in mice [64].
2.1.2. STAR protein activity
STAR is synthesized as a 37 kDa cytosolic preprotein that is composed of an N-terminal
mitochondrial targeting sequence and a C-terminal cholesterol-binding STAR-related lipid
transfer (START) domain [59, 65, 66]. While STAR is constitutively expressed under basal
conditions, hormonal stimulation parallels a rapid increase in STAR levels and leads to the
translocation of STAR to the OMM [57-59]. STAR is synthesized from pre-existing mRNA, as
inhibition of gene transcription did not affect induction of steroidogenesis by cAMP, and only
newly synthesized STAR protein is active [67, 68]. 37 kDa STAR is rapidly processed at the
OMM to an inactive 30 kDa mature protein which is imported to the mitochondrial matrix where
it is degraded [55, 69]. 37 kDa active STAR has a half-life in the cytoplasm of 3-5 minutes,
while mature 30 kDa STAR in the matrix has an average half-life of 2-4 hours [70, 71]. Deletion
of the mitochondrial targeting sequence of STAR (N-62 STAR) resulted in no change in its
activity, although N-62 STAR inserted cholesterol into other membranes, suggesting that
STAR’s targeting sequence is vital for confining its activity to the mitochondria [13, 72]. STAR
10
functions solely at the OMM and does not need to enter the mitochondria for its activity [73, 74].
Further work has also shown that the residence time of STAR at the OMM is proportional to its
steroidogenic activity [75]. STAR activity has been shown to be tightly regulated by various
mechanisms including phosphorylation at Ser-194, which induces its cholesterol transfer activity
by 50% and interaction with 14-3-3γ, which negatively regulates steroid production by blocking
phosphorylation of STAR at Ser-194 [67, 76-78].
2.1.3. STAR function in cholesterol transport and the START domain
While the exact mechanism by which STAR facilitates cholesterol transport into the
mitochondria is unknown, several studies utilizing cell-free systems have suggested that STAR
binds cholesterol and that STAR can transfer cholesterol between membranes. Kallen et al
showed that N-62 STAR stimulates cholesterol transfer from liposomes to heat-treated
mitochondria, while mutant STAR unable to induce steroid production did not stimulate
cholesterol transfer [13]. Petrescu et al also studied cholesterol binding properties of STAR and
showed that fluorescent NBD cholesterol binds N-62 STAR with a Kd of 32 nM [12]. This group
also showed that STAR enhanced mitochondrial cholesterol transport 100-fold and that STAR
was 67-fold more effective in transporting cholesterol from mitochondria of MA-10 cells than
from mitochondria of human fibroblasts [12]. Further work by Tuckey et al showed that N-62
STAR stimulated cholesterol transfer from donor phospholipid vesicles to acceptor vesicles
containing the P450scc system by 5-10-fold without other mitochondrial protein components
[79]. Additionally, Baker et al showed that various STAR mutants including N-62 STAR can
bind NBD cholesterol and transfer cholesterol between liposomes in vitro [80].
As mentioned above, STAR contains a C-terminal cholesterol binding START domain, a
hydrophobic sterol-binding pocket composed of four α-helices and nine antiparallel β-sheets,
11
which is common among other closely related lipid transfer proteins known as START proteins
[66, 80-82]. Crystal structure analysis of the START domain of STAR suggests that STAR binds
cholesterol with 1:1 stoichiometry in vitro [66]. One current model of STAR function is that the
C-helix of STAR’s START domain interacts with protonated phospholipid head groups at the
OMM, inducing the C-helix to swing open [83]. According to this molten globule model, this
conformational change allows STAR to bind and release cholesterol [83]. In addition to its
proposed role in binding and releasing cholesterol itself, another model suggests that STAR can
mobilize cholesterol bound to the cholesterol-binding domain of TSPO for import to the IMM
[83].
12
2.1.4. Other roles of STAR
Whereas most studies investigating the function of STAR have focused on the role of
hormone-induced STAR in cholesterol transport for steroid biosynthesis, we developed a STAR
KO MA-10 mouse tumor cell line (STARKO1) to investigate the role of constitutive STAR [10].
We showed that the absence of constitutive STAR, or STAR protein present under basal
conditions independent of hormonal stimulation, altered lipid droplet content, specifically
leading to dramatic alterations in the amount of cholesteryl ester, diacylglycerol, and
phosphatidylcholine in STARKO1 cell lipid droplets [10]. Alterations in lipid droplet content
paralleled alterations in the levels of many lipid-related genes. These data suggest that STAR
may function in the transport and/or metabolism of various other lipids, independent of its role in
cholesterol transport for steroidogenesis. Further, our recent data suggest that absence of
constitutive STAR leads to alterations in mitochondrial structure and function, which are
exacerbated by reintroduction of STAR into STARKO1 cells (unpublished data). Taken together
these data show that STAR may have distinct functions in addition to its known classical role in
cholesterol transport for acute steroidogenesis (Figure 1).
2.2.SCP2/SCPx
Unlike STAR, SCP2/SCPx have broad specificity for various lipids, and for this reason, it
is also referred to as a non-specific lipid transfer protein (nsLTP) [9]. SCP2/SCPx have been
Figure 1. Function of hormone-induced STAR and constitutive STAR. Hormonal
stimulation induces STAR to localize to the OMM where it interacts and works with
other transduceosome proteins to transport cholesterol to CYP11A1 at the IMM for acute
steroidogenesis. CYP11A1 converts cholesterol to the first steroid, pregnenolone. Upon
reaching the OMM, the mitochondrial targeting sequence of STAR is cleaved,
inactivating the protein, and inducing its import into the matrix. Constitutive STAR plays
a role in the transport and/or metabolism of cholesteryl esters, diacylglycerol, and
phosphatidylcholine as knockout of STAR results in the accumulation of these lipids.
Abbreviations: Ch: cholesterol; P5: pregnenolone; DAG: diacylglycerol; CE: cholesteryl
ester; PC: phosphatidylcholine.
13
suggested to function in intracellular cholesterol transport between several sites such as
mitochondria, ER, and plasma membrane [11, 84, 85]. In addition to cholesterol, previous
studies have suggested that SCP2/SCPx are involved in the transport and metabolism of other
lipids including cholesteryl esters, fatty acids, fatty acyl-CoAs, and phospholipids [86].
2.2.1. SCP2/SCPx gene and protein products
The SCP2 gene contains two distinct transcription initiation sites and encodes a 15 kDa
pro-SCP2 and the 58 kDa SCPx, which are both identical at the C-termini [87]. Upon further
processing, the 15 kDa pro-SCP2 is cleaved to form the mature 13 kDa SCP2. The 58 kDa SCPx
is also post-transcriptionally processed into 46 kDa SCPx as well as the mature 13 kDa SCP2
[88, 89]. SCP2 and SCPx are highly expressed in adrenals, testis, ovaries, liver, and intestine,
which are all tissues with high rates of cholesterol metabolism [90]. While SCP2 and SCPx have
both been indicated to have lipid transfer activity, SCPx has been shown to be a critical enzyme
in peroxisomal β-oxidation [91]. In line with these functions, SCP2 and SCPx are mostly
localized to peroxisomes, owing to the peroxisomal targeting AKL sequence at the C-termini
[92-94]. Additionally, there is evidence suggesting that these proteins contain a predicted
mitochondrial targeting sequence at the N-termini, suggesting dual targeting [86, 95].
2.2.2. Role of SCP2/SCPx in cholesterol transport
Because SCP2 and SCPx are both synthesized via expression of a single gene (SCP2),
most genetic manipulation studies described here are non-specific as to whether the findings
pertain to SCP2 alone, SCPx alone, or both proteins. Hence, when describing studies involving
SCP2 and SCPx non-specifically, “SCP2/SCPx” is used. Several lines of evidence have
supported the role of SCP2/SCPx in intracellular cholesterol trafficking. Studies have shown that
recombinant human SCP2 binds cholesterol at a single binding site, however, the reported Kd
14
values between these studies are drastically different with one reporting a Kd of 4.2 nM and
another reporting a Kd of 0.3 μM [96, 97]. Further, work suggesting that SCP2/SCPx functions in
cholesterol transport include in vitro studies showing that SCP2/SCPx are effective in enhancing
sterol trafficking ~27-fold from plasma membranes to microsomal membranes and ~12-fold
from plasma membranes to mitochondria [98]. Additionally, since cholesterol transport between
lysosomes to plasma membrane occurs within two minutes in intact cells, which is inconsistent
with vesicular transport, a cholesterol transport protein-mediated mechanism involving
SCP2/SCPx was postulated [99]. It was shown that cholesterol transport from lysosomal
membranes to plasma membranes that were isolated from L-fibroblasts was enhanced 364-fold
by SCP2/SCPx [84]. This study also showed that, in plasma membrane of L-cells with Scp2
overexpressed, cholesterol levels/mg protein decreased by 38%, consistent with other data
showing that SCP2 is involved in cholesterol distribution away from plasma membrane and to
other cellular sites such as lipid droplets [84, 100]. Additionally, cholesterol/mg protein
decreased by 17% in lysosomal membranes isolated from these L-cells, while there was a 2.2-
fold increase in cholesterol/mg protein in ER membranes from L-cells overexpressing Scp2 [84,
99]. In intact cells, transfection with Scp2 increased exogenous cholesterol uptake by 1.9-fold
and total cholesterol mass by 1.4-fold [88]. Further, SCP2/SCPx enhanced cholesterol transport
from plasma membrane to ER for esterification by ACAT in L-cells and enhanced intracellular
cholesterol cycling in hepatoma cells [88, 101, 102]. SCP2 also has been shown to play a role in
cholesterol efflux in that transfection of L-cells with Scp2 inhibited HDL-mediated cholesterol
efflux from lipid droplets to plasma membrane through lipid rafts [100].
In addition to these studies utilizing intact cells to elucidate a role for SCP2/SCPx in
cholesterol transport, studies using genetic manipulation of Scp2 in animal models have also
15
been done. In mice overexpressing Scp2, there was an increase in plasma LDL cholesterol, a
decrease in plasma HDL cholesterol and a 70% increase in hepatic total cholesterol [103]. In
Scp2 gene ablated mice, total hepatic cholesterol decreased by 15%, likely due to decreases in
cholesteryl esters [104]. In addition to these findings, animal models have also suggested a role
of SCP2 in biliary cholesterol secretion. In rats with Scp2 overexpression, total hepatic
cholesterol content and total bile acid content increased [103]. Conversely, in rats treated with
Scp2 antisense oligonucleotides which led to a 60% reduction in SCP2 levels in the liver, there
was a delay in biliary cholesterol secretion [105]. Taken together, these data support the role of
SCP2/SCPx in intracellular cholesterol transport between a variety of membranes and for a
variety of critical cellular processes.
2.2.3. Role of SCP2 in steroidogenesis
In addition to these proposed functions in cholesterol trafficking by SCP2/SCPx, much
work has also been done to delineate a potential role of SCP2 in cholesterol transport to the
mitochondria for steroidogenesis. While STAR is known to mediate the acute, rapid transport of
cholesterol from the OMM to the IMM thereby depleting the OMM of cholesterol, many studies
have been done to investigate whether SCP2 plays a role in replenishing the OMM with
cholesterol from intracellular stores. In isolated steroidogenic adrenal cells, SCP2 enhanced
radiolabeled cholesterol transport from lipid droplets to mitochondria and also significantly
increased pregnenolone production [106]. Introduction of SCP2 and the components of the
cholesterol side chain cleavage system in non-steroidogenic COS-7 cells also led to increased
steroid production [107]. Furthermore, Scp2 is most highly expressed in the steroidogenic tissues
(adrenals, testis, and ovaries), and hormonal stimulation of the steroidogenic cells of these tissues
leads to an increase in SCP2 mRNA expression and protein levels by a cAMP dependent
16
pathway and increased its association with mitochondria [108-111]. Additionally, in Scp2
overexpressing L-cells, it was shown that sterol transport from isolated lysosomal to
mitochondrial membranes was enhanced by SCP2 [112].
However, although there is much indirect evidence that SCP2 plays a role in
steroidogenesis, the function of SCP2 in cholesterol transport to the mitochondria for steroid
production has been called into question. Firstly, it was shown that SCP2 enhanced cholesterol
transfer to mitochondrial membranes regardless of whether the mitochondria were isolated from
MA-10 cells or from fibroblasts showing that the cells need not be steroidogenic for SCP2 to
transport cholesterol to mitochondria [12]. Secondly, it was found that Scp2 gene-ablated mice
had normal serum testosterone, progesterone, and corticosteroid levels [104]. It may be the case
that, although SCP2 is not indispensable for steroidogenesis, it may play a role in cholesterol
transport to the mitochondria, which is supplemented by other compensatory mechanisms.
2.2.4. Role of SCP2/SCPx in the transport of other lipids
While SCP2/SCPx are classically known as sterol transport proteins, these proteins also
possess high binding affinities for many other lipid classes and have been suggested to play a
role in the transport and/or metabolism of many other lipids. For example, SCP2 has high
affinity for fatty acids with a reported Kd of 234 nM, similar to other fatty acid binding proteins,
and previous reports show that SCP2/SCPx enhance the cellular uptake and intracellular
transport of fatty acids [113-116]. Further studies have shown that SCP2/SCPx function in fatty
acid transport to peroxisomes for oxidation and to ER for phospholipid incorporation [116, 117].
Another lipid group in which SCP2 has high affinity for are fatty acyl CoAs with reported Kds in
the range of what has been cited for acyl CoA binding protein [118, 119]. In vitro studies and
studies in intact cells have shown that SCP2/SCPx stimulate the incorporation of microsomal
17
fatty acyl CoA into phosphatidic acid [117, 120]. Additionally, SCP2 has high affinity for
phosphatidylinositol (PI) and may play a role in PI transport to the plasma membrane based on
data showing that Scp2 overexpression results in the significant redistribution of PI from
mitochondrial and ER membranes to plasma membranes [121]. Lastly, SCP2 has been shown to
have nanomolar affinity to all sphingolipid classes and in vitro studies using liver homogenates
suggest that SCP2/SCPx increase sphingomyelin transport [122-124].
2.2.5. Role of SCPx in peroxisomal β-oxidation
In addition to the function of SCPx as a lipid transporter, the 46 kDa SCPx protein that
arises from 58 kDa SCPx exhibits 3-ketoacyl-CoA thiolase activity [125]. The classical 3-
ketoacyl-CoA thiolase is specific for catalyzing the final step of the peroxisomal β-oxidation of
straight-chain fatty acids, while the 46 kDa SCPx is responsible for catalyzing the final step of
the peroxisomal β-oxidation of branched-chain fatty acids and the metabolism of cholesterol for
bile acid synthesis [126] [91]. Indeed, Scp2 gene-ablated mice had defects in the metabolism of
branched-chain fatty acids with a ten-fold accumulation of phytanic acid in knockout mice [104].
Further, mice null of SCPx, but normal levels of SCP2, had altered levels of hepatic fatty acids,
suggesting that branched-chain fatty acid oxidation requires SCPx independent of SCP2 [127].
The indispensable role of SCPx in peroxisomal β-oxidation was further exemplified when a
homozygous 1-nucleotide insertion in SCP2 in a patient, which resulted in the complete absence
of SCPx protein, led to leukoencephalopathy with dystonia and motor neuropathy, hyposmia,
18
azoospermia and an accumulation of branched-chain fatty acids [128]. While deficiencies in
several peroxisomal enzymes and/or proteins leading to neurological diseases such as X-linked
adrenoleukodystrophy and Refsum disease had been previously reported across many patients,
this was the first report of a patient with SCPx deficiency [128]. The second report of SCPx
Figure 2. SCP2/SCPX function in the transport and metabolism of cholesterol and other
lipids. SCP2/SCPX is suggested to transport cholesterol between various cellular membranes
and organelles such as the plasma membrane, mitochondria, lipid droplets, ER, lysosomes,
and peroxisomes. Cholesterol intermediates also undergo oxidation in peroxisomes via SCPX
for bile acid synthesis. In addition to cholesterol, SCP2/SCPX have been indicated to play a
role in fatty acid, phospholipid, and sphingolipid transport. SCPX is also a key enzyme in the
peroxisomal β-oxidation of fatty acids. Abbreviations: Ch: cholesterol; ER: endoplasmic
reticulum.
19
deficiency was caused by a compound heterozygous mutation in SCP2, again leading to
undetectable levels of SCPx and neurodegenerative symptoms [129]. Recently, our lab worked
on the characterization of a third patient with SCPx deficiency and the first caused by a
heterozygous mutation in SCP2, leading to low, but detectable levels of SCPx (unpublished
data). Similar to previous reports, our patient presented with severe neurological symptoms.
However, in contrast to the previous studies, the patient’s pristanic and phytanic acid levels were
normal, indicating that the patient’s low levels of SCPx were sufficient for branched-chain fatty
acid metabolism. Despite normal pristanic and phytanic levels, levels of many other lipid species
among various lipid classes including fatty acids, acylcarnitines, sterols, phospholipids, and
sphingolipids were altered in patient fibroblasts, indicating a role of SCPx in the transport and/or
metabolism of these lipids (unpublished data). Taken together, these data exemplify the critical
and multifunctional roles of SCP2/SCPx as non-specific lipid transporters and a key enzyme in
peroxisomal oxidation (Figure 2).
3. Conclusion
Cholesterol transport proteins play vital roles in the intracellular distribution of
cholesterol for several key cellular processes such as steroidogenesis. Although STAR is
classically known to function in cholesterol transport for acute steroidogenesis, recent studies
suggest that it may have additional roles in the transport and metabolism of other lipids.
Furthermore, while SCP2/SCPx were first recognized as sterol transfer proteins, an accumulation
of data ranging from cell-free systems to patients with SCP2 mutations have similarly suggested
a broader role of these proteins in lipid transport and metabolism. Thus, the data presented here
suggest that these classic examples of intracellular cholesterol transport proteins play a more
20
general role in lipid transport and metabolism in addition to their respective roles in cholesterol
trafficking.
21
Chapter II: Role of Constitutive STAR in Leydig Cells
Abstract
Leydig cells contain significant amounts of constitutively produced steroidogenic acute
regulatory protein (STAR; STARD1). Hormone-induced STAR plays an essential role in inducing
the transfer of cholesterol into the mitochondria for hormone-dependent steroidogenesis. STAR
acts at the outer mitochondrial membrane, where it interacts with a protein complex, which
includes the translocator protein (TSPO). Mutations in STAR cause lipoid congenital adrenal
hyperplasia (lipoid CAH), a disorder characterized by severe defects in adrenal and gonadal steroid
production; in Leydig cells, the defects are seen mainly after the onset of hormone-dependent
androgen formation. The function of constitutive STAR in Leydig cells is unknown. We generated
STAR knockout (KO) MA-10 mouse tumor Leydig cells and showed that STAR KO cells failed
to form progesterone in response to dibutyryl-cAMP and to TSPO drug ligands, but not to 22(R)-
hydroxycholesterol, which is a membrane-permeable intermediate of the CYP11A1 reaction.
Electron microscopy of STAR KO cells revealed that the number and size of lipid droplets were
similar to those in wild-type (WT) MA-10 cells. However, the density of lipid droplets in STAR
KO cells was drastically different than that seen in WT cells. We isolated the lipid droplets and
analyzed their content by liquid chromatography–mass spectrometry. There was a significant
increase in cholesteryl ester and phosphatidylcholine content in STAR KO cell lipid droplets, but
the most abundant increase was in the amount of diacylglycerol (DAG); DAG 38:1 was the
predominantly affected species. Lastly, we identified genes involved in DAG signaling and lipid
metabolism which were differentially expressed between WT MA-10 and STAR KO cells. These
results suggest that constitutive STAR in Leydig cells is involved in DAG accumulation in lipid
22
droplets, in addition to cholesterol transport. The former event may affect cell functions mediated
by DAG signaling.
1. Introduction
Steroidogenesis begins with the transport of cholesterol from intracellular stores into the
mitochondria. This is the hormone sensitive and rate-limiting step. The steroidogenic acute
regulatory protein (STAR; STARD1) plays a critical role in cholesterol transport to the
mitochondria for steroidogenesis. In Leydig cells, STAR is constitutively expressed under basal
conditions, independent of hormonal stimulation, which parallels the formation of low steroid
levels [130]. Hormonal stimulation causes a rapid induction of STAR, which is coupled to an
increase in cholesterol transfer to mitochondria and increased steroid formation [131]. It has been
suggested that STAR, along with other cytosolic and outer mitochondrial membrane proteins such
as the translocator protein (TSPO; 18 kDa), form the transduceosome complex upon hormonal
stimulation to move cholesterol to the inner mitochondrial membrane, where the CYP11A1
enzyme resides and converts cholesterol into pregnenolone [54].
STAR belongs to a family of structurally related proteins that contains the STAR-related
lipid transfer (START) domain, of which STAR is the only member shown to be involved in
cholesterol transfer for steroidogenesis [132]. STAR is synthesized as a 37 kDa active cytosolic
protein, and upon hormonal stimulation, it moves to the outer mitochondrial membrane where it
functions in steroidogenesis and where its mitochondrial targeting sequence is cleaved, yielding
an inactive 30 kDa protein [48]. Although the exact mechanism by which STAR induces
steroidogenesis remains heavily debated, its critical role in cholesterol transport and steroid
formation is widely accepted due to observations from STAR knockout (KO) mouse models,
which accumulate lipids in adult adrenals and gonads and require adrenal steroid replacement for
23
survival [64]. These findings support the phenotype seen in patients with lipoid congenital adrenal
hyperplasia (lipoid CAH), a disease caused by mutations in the human Star gene and characterized
by severe deficiency in steroid production and lipid accumulation in steroidogenic cells [133].
Despite current evidence revealing the importance of STAR’s function in cholesterol
transfer and steroidogenesis, no studies have been conducted specifically examining the role of
constitutive STAR. However, previous studies have suggested a dual functionality for STAR, with
one role independent of cholesterol transport [80, 134]. To study the potential roles of constitutive
STAR, we developed STAR KO MA-10 mouse tumor Leydig cells. STAR KO cells do not
produce steroids in response to hormonal stimulation but do respond to treatment with 22(R)-
hydroxycholesterol. Moreover, TSPO drug ligands, which induce steroidogenesis independent of
hormonal stimulation, do not induce progesterone production in STAR KO cells [135, 136]. Our
structural and functional studies revealed alterations in lipid droplet content in STAR KO cells.
The increase in cholesterol esters (CE) found in STAR KO lipid droplets and the accumulation of
diacylglycerol (DAG) and phosphatidylcholine (PC) suggest that basal STAR may function
independent of cholesterol transport, including lipid droplet biogenesis. Collectively, our results
show that basal STAR has a function distinct from hormone-induced STAR, which involves DAG
metabolism or signaling in lipid droplets.
2. Results
2.1. Induction of constitutive Star following hormonal stimulation
Steroidogenic activity of MA-10 cells, and their responsiveness to hormones vary over
time. Therefore, it is essential to our work to assess steroidogenic capacity and the responsiveness
of our cells to hormonal stimulation, and in this case, STAR expression levels. To do so, we
compared STAR expression levels and steroidogenic activity in MA-10 cells to those of adult
24
mouse and rat Leydig cells. Leydig cells were isolated from adult mouse and rat testis. Figure 3A
shows Star mRNA expression in mouse and rat Leydig cells and in WT MA-10 cells under basal
conditions and upon hormonal stimulation. In all models, Star mRNA is present under basal
conditions, and stimulation with 50 ng/µl hCG causes an induction of Star mRNA. STAR protein
levels were also analyzed from mouse and rat Leydig cells and in WT MA-10 cells. In correlation
with the quantitative real time PCR (qRT-PCR) data, the immunoblots in Figure 3B show the
protein levels of basal STAR in all models, which are all increased after stimulation with hCG.
Testosterone production by isolated adult mouse and rat Leydig cells and progesterone production
by MA-10 cells under basal conditions and in response to 50 ng/µl hCG were also measured,
showing an increase in steroid production following hormonal stimulation (Fig. 3C). In addition
to characterizing the responsiveness of our MA-10 cells, these data show that, in Leydig cells,
hormonal stimulation parallels an increase in STAR expression and in steroid production, but that
STAR expression and low levels of steroid production are also present under basal conditions.
This suggests a possible role of constitutively expressed STAR independent of hormonal
stimulation.
25
Basal hCG
- STAR
- β -
actin
Mouse
Adult Leydig
Cells
1.04 1.62
- STAR
- β -
actin
Rat
Adult Leydig Cells
0.56 1.43
- STAR
- β -
actin
MA-10 Cells
0.65 1.57
A
B
C
Mouse Rat MA-10
0.000
0.001
0.002
0.003
0.004
0.005
0.1
0.2
0.3
0.4
Leydig Cell Type
Relative Expression
Star
Basal
hCG
**
***
**
Control hCG
0
100
200
300
400
500
50 ng/ml hCG for 2 hours
MA-10 cells
Progesterone (ng/mg protein)
***
Mouse Rat
0
5
10
15
20
50 ng/ml hCG for 2 hours
Adult Leydig Cells
Testosterone (ng/10
5
cells)
Basal
hCG
*
*
Figure 3. STAR expression and levels in various models in the absence and presence of hormonal
stimulation. A. qRT-PCR analyses of Star mRNA expression in mouse, rat, and MA-10 Leydig cells
under basal conditions and upon hCG stimulation. Gapdh was used as the housekeeping gene. Data are
shown as mean ± SEM (n=3). B. Western blot analyses of STAR protein levels in mouse, rat, and MA-
10 Leydig cells under basal conditions and upon hCG stimulation. Quantification was done by
calculating the ratio of the density of the STAR band to that of β -actin through ImageJ. Quantification
is shown below each immunoblot. C. ELISA analyses of testosterone production by mouse and rat
Leydig cells (left) and progesterone production by MA-10 cells (right) in cell media following
stimulation with 50 ng/ml hCG for 2 hours. Data are shown as mean ± SEM (n=3 for mouse; n=2 for
rat). * p < 0.05; ** p < 0.01; *** p < 0.001.
26
STAR 1
STAR 2
STAR 3
STAR 4
0.000
0.002
0.004
0.006
0.008
Star
STAR Primer
Relative Expression
MA-10
STARKO1
***
***
***
***
STAR 1
STAR 2
STAR 3
STAR 4
0.000
0.002
0.004
0.006
0.008
0.010
Star
STAR Primer
Relative Expression
MA-10
STARKO2
***
***
***
**
STARKO1: GLRHQAAGHWPRAQLESTGGFQSRVDGSSSTSELSAW
WT MA-10: GLRHQAVLAIGQELNWRALGDSSPGWMGQVRRRSSLLG
WT MA-
10:
STARK
O1:
STARKO2:
GLRHQAALAKSSTGEHWGIPVPGGWVKFDVGALCLVNK
WT MA-10:
A
B
C
D
STARK
WT MA-
10:
WT MA-
10:
STARK
O1:
WT MA-
10:
STARK
O1:
WT MA-
10:
STARK
O2:
WT MA-
10:
STARK
O2:
WT MA-
10:
STARK
O2:
WT MA-
10:
STARK
O2:
27
2.2. Gene deletion of Star by CRISPR/Cas9
To develop stable STAR KO cell lines from MA-10 cells, guide RNAs (gRNAs) were
designed to specifically target part of exon2 of Star, which corresponds to STAR’s N-terminal
mitochondrial targeting sequence. Two gRNAs were generated using the following single-
stranded (ss) oligonucleotides: (1) F: 5’-ATTAAGGCACCAAGCTGTGCGTTTT-3’ and R: 5’-
GCACAGCTTGGTGCCTTAATCGGTG-3’ (2) F: 5’-CACCAAGCTGTGCTGGCCATGTTTT-
3’ and R: 5’-ATGGCCAGCACAGCTTGGTGCGGTG-3’. The gRNAs were each introduced into
plasmids containing the Cas9 endonuclease and orange fluorescent protein (OFP). These plasmids
were then transfected into MA-10 cells. OFP positive cells were fluorescence activated cell (FAC)
sorted into single cells per well and cultured until enough protein could be extracted from the cells
to screen for STAR KO by immunoblot. Various protein samples from single colonies that were
probed for the presence of STAR are shown in Figure 4A. This analysis showed that the STAR
protein was absent in samples 5 and 8, which were subsequently named STARKO2 cells and
STARKO1 cells, respectively.
Figure 4. Screening and validation of CRISPR/Cas9-mediated STAR KO in MA-10 cells. (A)
Western blot screening for STAR KO cell line following CRISPR/Cas9-mediated STAR KO
and FACS. Samples in lane 5 (STARKO2) and lane 8 (STARKO1) show no STAR band and,
therefore, are STAR KOs. (B) qRT-PCR analyses of STARKO1 and STARKO2 where Gapdh
was used as the housekeeping gene. STAR 1, STAR 2, STAR 3, and STAR 4 are primers for
various regions of the mouse Star gene. Data are shown as mean ± SD (n=3). * p < 0.05; ** p
< 0.01; *** p < 0.001. (C) BLAST DNA sequence of Star in STARKO1 cells (top) compared
to wild type mouse Star (bottom). Dashes represent nucleotide deletions. Amino acid sequence
of wild type STAR (top) compared to that of STARKO1 (bottom). Highlighted amino acids
represent sequence changes. (D) BLAST DNA sequence of Star in STARKO2 cells (top)
compared to wild type mouse Star (bottom). Dashes represent nucleotide deletions. Amino
acid sequence of wild type STAR (top) compared to that of STARKO2 (bottom). Underlined
amino acids represent sequence changes.
28
Lack of gene expression of Star in STARKO1 and STARKO2 KO cell lines was further
validated by qRT-PCR, the results of which are shown in Figure 4B. Star mRNA expression was
significantly decreased in both STARKO1 and STARKO2 for all primers used (STAR 1-4), which
were specific for different regions of the Star gene. Next-generation sequencing (NGS)-based
amplicon sequencing and Basic Local Alignment Search Tool (BLAST) results revealed that
CRISPR/Cas9 introduced a frameshift mutation in both STAR KO cell lines that disrupted the Star
gene (Fig. 4C and 4D). We found that STARKO1 had a deletion of two nucleotides in exon2,
resulting in a frameshift mutation and complete disruption of the Star gene sequence following the
deletions. This corresponded with a change in the amino acid sequence of STAR following these
mutations (Fig. 4C). Similarly, using the NGS results and BLAST to align STARKO2 with WT
Star, we found that there were seven nucleotides deleted from the Star gene in STARKO2, again
resulting in frameshift mutations and changes in the amino acid sequence of STAR (Fig. 4D).
These results suggest that STARKO1 and STARKO2 are true STAR KO cell lines, where the Star
gene is mutated beginning at exon2.
29
2.3. STAR KO inhibits hormone-induced steroidogenesis
To determine that STAR activity was disrupted, we used ELISA to measure progesterone
levels produced by STARKO1 and STARKO2 cells in response to hormonal stimulation. Neither
STARKO1 nor STARKO2 could be induced to produce progesterone in response to hCG
stimulation (Fig. 5A). This suggests that the function of STAR in cholesterol transport, which is
M A -1 0
S T A R K O 1
S T A R K O 2
0
1 0
2 0
3 0
3 0 0
3 5 0
4 0 0
4 5 0
5 0 0
5 0 n g /m l h C G fo r 2 h o u rs
P ro g e s te ro n e (n g /m g p ro te in )
Control
hCG
M A -1 0
S T A R K O 1
S T A R K O 2
0
2 0
4 0
6 0
8 0
1 0 0
1 0 0 0
1 5 0 0
2 0 0 0
1 m M d b c A M P fo r 2 h o u r s
P ro g e s te ro n e (n g /m g p ro te in )
Control
dbcAMP
M A -1 0
S T A R K O 1
S T A R K O 2
0
2 0
4 0
6 0
8 0
1 0 0
2 5 0 0
3 0 0 0
3 5 0 0
4 0 0 0
4 5 0 0
5 0 0 0
5 0 µ M 22(R )-h y d r o x y c h o le s te r o l fo r 2 h o u r s
P ro g e s te ro n e (n g /m g p ro te in )
Control 22(R )-hydroxycholesterol
A
B
C
Figure 5. Progesterone production by wild-type MA-10 and STAR KO cell lines in
response to stimulus. (A) ELISA analyses of progesterone levels in cell media following
stimulation with 50 ng/ml hCG for 2 hours. (B) ELISA analyses of progesterone levels
in cell media following stimulation with 1 mM dbcAMP for 2 hours. (C) ELISA analyses
of progesterone levels in cell media following stimulation with 50 μM 22(R)-
hydroxycholesterol for 2 hours. Data are shown as means ± SEM (n=3).
30
critical in hormone-induced steroidogenesis, is disrupted as a result of our STAR KO. Similar
trends were seen for both STAR KO cell lines following stimulation with dbcAMP (Fig. 5B).
To ensure that the steroidogenic pathway downstream of STAR was not affected as a result
of STAR KO, we treated the STAR KO cells with 22(R)-hydroxycholesterol, a membrane
permeable intermediate of the CYP11A1 (P450scc) reaction, which is downstream of STAR
activity. Treatment with 22(R)-hydroxycholesterol in STARKO1 and STARKO2 produced
progesterone at levels similar to WT MA-10 cells (Fig. 5C). These results suggest that the
steroidogenic pathway downstream of STAR activity was still intact in both STAR KO cell lines.
Because STARKO1 and STARKO2 act identically in response to these various treatments, we
chose to carry out subsequent experiments using only STARKO1.
2.4. STAR KO inhibits TSPO ligand-mediated steroidogenesis
We next sought to determine if STAR KO affects TSPO-mediated steroidogenesis. TSPO
drug ligands such as XBD173 and FGIN-1-27 have been shown to induce steroidogenesis in the
absence of hormonal stimulation, but in the presence of basal levels of STAR [10, 11]. Since STAR
KO cells lack constitutive STAR, we sought to determine if TSPO-ligand induced steroidogenesis
acts independently of STAR. MA-10 and STAR KO cells were treated with 50 μM XBD173 or 50
μM FGIN-1-27 for 2 hours and progesterone production was measured by ELISA. Results showed
that XBD173 stimulates progesterone production in MA-10 cells, but that progesterone production
is not induced to levels similar to WT MA-10 cells when STAR is absent (Fig. 6A). Although
progesterone production was not induced in STARKO1 cells to the extent of WT MA-10 cells,
XBD173 caused a small but significant increase in progesterone produced by STARKO1 cells
compared to control STARKO1 cells (Fig. 6A). Treatment with FGIN-1-27 also showed that
progesterone production is not induced to levels similar to WT MA-10 cells when STAR is absent
31
(Fig. 6B). However, there was no difference in progesterone produced between control and FGIN-
1-27-treated STARKO1 cells. These results suggest that constitutive STAR, or STAR that is
present independent of hormonal stimulation in MA-10 cells but absent in STAR KO cells, plays
a role in TSPO-mediated steroidogenesis.
2.5. STAR KO results in alterations in lipid droplet content
A distinctive characteristic of Leydig cells is the abundance of large lipid droplets. Bodipy
493/503 staining showed no difference in lipid droplet number per cell between WT MA-10 and
STARKO1 cells (Fig. 7A). Electron microscopy of STAR KO cells also revealed that the number
and size of lipid droplets were similar to those in WT MA-10 cells (Fig. 7B). The lack of lipid
droplet accumulation in STARKO1 cells is in contradiction to the proposed role of STAR in lipoid
CAH. Our data suggests that basal STAR expression may have a function different than hormone-
induced STAR since unstimulated STAR KO cells do not accumulate lipid droplets. While there
was no difference in lipid droplet number, the density of the lipid droplets in STAR KO cells was
drastically different than that seen in WT cells (Fig. 7B). Lower magnification electron
MA-10
STARKO1
0
100
200
300
50 µM XBD173 for 2 hours
Progesterone (ng/mg protein)
Control
XBD173
***
*
MA-10
STARKO1
0
100
200
300
50 µM FGIN-1-27 for 2 hours
Progesterone (ng/mg protein)
Control
FGIN-1-27
**
A B
Figure 6. Progesterone production by wild-type MA-10 and STAR KO cell lines in response
to TSPO drug ligands. A. ELISA analyses of progesterone levels in cell media following
stimulation with 50 μM XBD173 for 2 hours. B. ELISA analyses of progesterone levels in cell
media following stimulation with 50 μM FGIN-1-27 for 2 hours. Data are shown as means ±
SEM (n=3). * p < 0.05; ** p < 0.01; *** p < 0.001.
32
micrographs are shown in Fig. S1 to provide a better overview of changes in density between the
lipid droplets of WT MA-10 and STAR KO cells.
Liquid-chromatography-mass spectrometry (LC-MS) data showed differences in the lipid
profiles of MA-10 and STARKO1 cell lipid droplets (Fig. 7C). There was a significant increase in
CE and PC in STAR KO cell lipid droplets, but the most abundant increase was in the amount of
DAG (Fig. 7D). There were no significant differences in DAG and CE content between WT and
Mitotracke
r
Bodipy
DAPI Merge
MA-10
Mitotracker Bodipy
DA
PI
Merge
STARKO1
MA-10 STARKO1
0
50
100
150
200
Bodipy 493/503
# Lipid Droplets/Cell
A
B
C
MA-10 STARKO1
33
D
E
F
34
STAR KO cells upon hCG stimulation, but there was a significant change in PC content (Fig. 7D).
The profile of individual DAG species revealed that DAG 38:1 is the species that is primarily
increased in STARKO1 lipid droplets compared to WT (Fig. 7E, F). These results suggest that
DAG accumulation in lipid droplets and signaling may play a role in the pathology of lipoid CAH,
including loss of the ability to form steroids as seen with STAR mutations in humans. The
differences in lipid droplet content between WT and STAR KO MA-10 cells in the absence and
presence of hormonal stimulation also suggest that STAR may have a function in lipid metabolism
that is independent of its role in cholesterol transport.
2.6. DAG signaling may play a role in progesterone production
To explore the effects of DAG accumulation when STAR is knocked out, we treated MA-
10 and STARKO1 cells with the DAG analog, 1-oleoyl-2-acetyl-sn-glycerol (OAG) for two hours
in the absence and presence of hormonal stimulation. Treatment with OAG led to a dose-dependent
increase in basal progesterone production but inhibition of hormone-induced progesterone
production in MA-10 cells (Fig. 8A). However, OAG had no effect on either the basal or hormone-
stimulated progesterone formation by STARKO1 cells (Fig. 8A). This further suggests that the
increase in DAG accumulation seen in the lipid droplets of STAR KO cells may play a role in the
Figure 7. Alterations in lipid droplet content between WT MA-10 and STARKO1 cells. (A)
Bodipy 493/503 staining of lipid droplets in MA-10 (left) and STARKO1 (right) cells with
quantification shown below. Scale bar: 10 μm. (B) Electron microscopy images of lipid droplets
in WT MA-10 cells (left) and STARKO1 cells (right). Arrows point to lipid droplets. (C) Peak
intensities of various lipid classes identified in lipid droplets of wild-type MA-10 and
STARKO1 cells in the absence and presence of hormonal stimulation. (D) Peak intensities of
lipid classes shown to be significantly altered in lipid droplets of wild-type MA-10 and
STARKO1 cells in the absence and presence of hormonal stimulation. (E) Peak intensities of
individual diacylglycerol (DAG) species identified in lipid droplets of MA-10 and STARKO1
cells. (F) Peak intensities of DAG 38:1 in lipid droplets of MA-10 and STARKO1 cells in the
absence and presence of hormonal stimulation. Data are shown as means ± SEM. * p < 0.05;
** p < 0.01; *** p < 0.001.
35
inability of STAR KO cells to produce steroids. To further investigate the effects of DAG
alterations, we inhibited phospholipase C (PLC) and protein kinase C (PKC), which are upstream
and downstream of DAG, respectively, in the DAG signaling pathway. Surprisingly, treatment
with the PLC inhibitor U73122 in the absence and presence of 50 ng/ml hCG for two hours
significantly increased progesterone production at 10 and 100 μM in STARKO1 cells (Fig. 8B).
In WT MA-10 cells, treatment with these concentrations of U73122 in combination with 50 ng/ml
hCG significantly decreased progesterone production (Fig. 8B). However, treatment with the PKC
inhibitor calphostin C (1 μM) alone or in combination with 50 ng/ml hCG and/or 50 μM OAG for
two hours, did not alter progesterone production in STARKO1 cells (Fig. 8C), suggesting that
DAG itself, rather than its effect on PKC, may inhibit steroid production in STARKO1 cells.
Collectively, these data suggest that DAG may play a role in the inhibition of steroidogenesis when
STAR is knocked out, which may be attributed to the activation of PLC but is independent of DAG
activation of PKC.
To gain further insight into the function of STAR in lipid metabolism or DAG signaling,
we carried out RNA sequencing to identify differentially expressed genes between MA-10 and
STARKO1 cells related to lipids and/or DAG (data not shown). Among the differentially
expressed genes identified by Ingenuity Pathway Analysis, some genes were part of lipid and DAG
interaction networks. These genes include Plin1, Bscl2, Pip5k1b, Stard12, and Plpp3. These genes
were confirmed to be significantly differentially expressed by RT-qPCR except Bscl2, which was
decreased in STARKO1 cells, but not significantly (Fig. 8D). Perilipin 1 (PLIN1) and Berardinelli-
Seip congenital lipodystrophy 2/Seipin (BSCL2) are lipid droplet-associated proteins [137, 138].
STARD12 (DLC-1) is a member of the START protein family; phosphoinositide 5-kinase 1 β
(PIP5K1β) is important in phospholipase D (PLD) activation; and phospholipid phosphatase 3
36
(PLPP3/LPP3) plays a role in DAG formation [139-141]. These alterations in gene expression
among genes involved in lipid metabolism and DAG signaling further demonstrate that STAR in
MA-10 cells seems to have a function independent of cholesterol transport for steroidogenesis.
0
2 0
4 0
6 0
8 0
U 7 3 1 2 2
P ro g e s te ro n e (n g /m g p ro te in )
*
* *
* * *
U 7 3 1 2 2 (u M ) 0 0 .1 1 1 0 1 0 0 0 0 .1 1 1 0 1 0 0
M A -1 0 S T A R K O 1
0
2 0
4 0
6 0
8 0
1 0 0
2 0 0
3 0 0
4 0 0
5 0 0
6 0 0
U 7 3 1 2 2 + h C G
P ro g e s te ro n e (n g /m g p ro te in )
M A -1 0 S T A R K O 1
h C G - + + + + + - + + + + +
U 7 3 1 2 2 (u M ) - - 0 .1 1 1 0 1 0 0 - - 0 .1 1 1 0 1 0 0
*
* *
* * *
* * *
0
1 0
2 0
3 0
4 0
5 0
2 0 0
4 0 0
6 0 0
8 0 0
C a lp h o s tin C
P ro g e s te ro n e (n g /m g p ro te in )
*
*
* * *
M A -1 0 S T A R K O 1
h C G - + - - + + + - + - - + + +
c a lp h o s tin C - - + - + - + - - + - + - +
O A G - - - + - + + - - - + - + +
M A-1 0 S T AR K O 1
0
2 0
4 0
6 0
8 0
1 0 0
M A -1 0 a n d S T A R K O 1 c e lls tre a te d w ith
O A G
P ro g e s te ro n e (n g /m g p ro te in )
h C G - - - - - - - - - -
O A G (µ M ) 0 1 1 0 5 0 1 0 0 0 1 1 0 5 0 1 0 0
M A-1 0 S T AR K O 1
0
2 0 0
4 0 0
6 0 0
M A -1 0 a n d S T A R K O 1 c e lls tre a te d w ith
O A G a n d 5 0 n g /m l h C G
P ro g e s te ro n e (n g /m g p ro te in )
h C G - + + + + + - + + + + +
O A G (µ M ) 0 0 1 1 0 5 0 1 0 0 0 0 1 1 0 5 0 1 0 0
*
*
*
*
*
*
A
B
C
D
Plin1
Bscl2
Pip5k1b
Stard12
Plpp3
0.00000
0.00002
0.00004
0.005
0.010
0.015
Lipid-Related Differentially Expressed Genes
Relative Expression
MA-10
STARKO1
***
**
*
*
37
3. Discussion
In Leydig cells, basal levels of STAR correlate with low levels of steroidogenesis [130].
Hormonal stimulation causes a rapid increase in STAR mRNA expression and protein levels, which
parallels an induction of steroidogenesis [131]. Multiple lines of evidence have shown that STAR
plays a critical function in cholesterol transport into the mitochondria for hormone-induced
steroidogenesis [59-62]. Indeed, STAR overexpression in MA-10 cells induces progesterone
production to similar levels as those seen in response to cAMP stimulation [59]. Furthermore,
transient transfection of Star cDNA, in addition to that of the CYP11A1 system, into non-
steroidogenic cells increased pregnenolone synthesis greater than 4-fold [60]. The importance of
STAR in cholesterol transport is most evident through the presence of STAR mutations in humans
which causes lipoid CAH [61]. Despite evidence supporting the role of STAR in cholesterol
transport and hormone-induced steroidogenesis, the exact function of constitutively expressed
STAR remains unknown. Previous work in our laboratory identified small molecule inhibitors of
the cholesterol binding domain of STAR, where even the most active compound only inhibited
steroidogenesis by about 60% [134]. Additionally, a study that blocked phosphorylation of an
important phosphorylation site on the cholesterol binding domain of STAR only resulted in 50%
inhibition of its activity [80]. These findings suggest a cholesterol-independent function of STAR,
since blocking its cholesterol binding domain does not completely abolish steroidogenic activity.
Figure 8. Role of DAG signaling in progesterone production by WT MA-10 and STARKO1
cells. (A) ELISA analyses of progesterone levels in cell media following treatment with
varying concentrations of the DAG analog OAG in the absence (left) and presence (right) of
50 ng/ml hCG for 2 hours. (B) ELISA analyses of progesterone levels in cell media following
treatment with varying concentrations of PLC inhibitor U73122 in the absence (left) and
presence (right) of 50 ng/ml hCG for 2 hours. (C) ELISA analyses of progesterone levels in
cell media following treatment with 1 μM of PKC inhibitor calphostin C in combination with
50 ng/ml hCG and/or 50 μM OAG for 2 hours. (D) qRT-PCR analyses of lipid-related
differentially expressed genes. Data are shown as means ± SEM (n=3). * p < 0.05; ** p <
0.01; *** p < 0.001.
38
The current study provides evidence that constitutive STAR may have a function which is
distinct from that of hormone-induced STAR. It is important to note that the data presented here
suggest that constitutive STAR may have a role distinct from hormone-induced STAR in MA-10
mouse tumor Leydig cells and, at present, it is unknown whether this role may be consistent among
normal Leydig cells and those from other species. Further investigation should be done to elucidate
the role of constitutive STAR in steroidogenic cells of other species, such as in Leydig cells of
rats, since the importance of lipid droplets as a source of cholesterol for steroidogenesis varies
between species. To study the potential function(s) of constitutive STAR, we used CRISPR/Cas9
to develop a STAR KO MA-10 cell line (STARKO1), which does not respond to hormonal
stimulation or treatment with TSPO drug ligands. TSPO drug ligands have previously been shown
to induce steroidogenesis independent of hormonal stimulation, but in the presence of constitutive
STAR [135, 136]. Since TSPO drug ligands are unable to induce steroid formation in STAR KO
cells, this suggests that basal STAR functions in TSPO-mediated steroidogenesis. The limited
increase in progesterone produced by STARKO1 cells treated with XBD173 may be due to STAR-
independent mechanisms of steroidogenesis.
It is currently thought that mutant STAR in humans inhibits acute steroidogenesis and that
continued hormonal stimulation leads to the accumulation of cholesterol as CE in the lipid droplets
of steroidogenic cells, causing cell damage and an inhibition of basal steroidogenesis [62]. Our
data suggest that STAR KO cells do not accumulate lipids, but that the contents of their lipid
droplets are different than WT cells. This contradicts the proposed role of STAR in lipoid CAH,
suggesting that basal STAR may have a function different than hormone-induced STAR, since the
absence of constitutive STAR in STAR KO MA-10 cells does not alter lipid droplet number.
Additionally, while STAR KO MA-10 cells do not accumulate lipid droplets, previous work
39
utilizing a single-cell CRISPR/Cas9 approach to delete STAR in Y-1 adrenal cells showed that
transfected (CRISPR positive) cells accumulate lipid droplets 24 hours and 48 hours after
transfection [142]. The discrepancy between our data and this previous study may be due to tissue-
specific differences, evident through work showing that lipid droplet accumulation in the adrenals
of STAR KO mice is significantly more severe compared to lipid droplet accumulation seen in
STAR KO mouse testis [64]. It is also possible that the generation of a stable STAR KO cell line
as described here may give more insight to the long-term effects of STAR KO compared to a
transient transfection.
We investigated the contents of the lipid droplets and found that the lipid profiles of the
lipid droplets between WT MA-10 and STARKO1 cells were different. The most abundant change
was an increase in the amount of DAG in STARKO1 cell lipid droplets, which suggests that STAR
may play a role in lipid metabolism and/or DAG signaling. The levels of CE and PC in STARKO1
cell lipid droplets were also significantly higher than those seen in WT MA-10 lipid droplets.
Accumulation of CE in STARKO1 lipid droplets was expected as the absence of STAR would
increase free cholesterol to be converted into CE and stored in lipid droplets. The increase in PC
may explain the increase in DAG levels, as PC can be converted to DAG through a PLC-dependent
mechanism in addition to a PLD-dependent mechanism [143].
To explore possible implications of DAG accumulation in STARKO1 lipid droplets, we
treated MA-10 and STARKO1 cells with various molecules involved in DAG signaling. The dose-
dependent inhibition of hormone-induced steroid production in MA-10 cells by OAG demonstrates
that DAG accumulation may play a role in the lack of steroid production when STAR is knocked
out. Previous work has demonstrated that STAR overexpression in high fat diet induced non-
alcoholic fatty liver disease (NAFLD) in mice decreased intracellular DAG levels, indicating a
40
protective role for STAR in NAFLD through reduction of DAG levels [144]. Since this shows that
STAR overexpression may lead to a reduction in DAG levels, it is consistent with our data which
demonstrate that STAR KO leads to an accumulation of DAG. Next, we found that inhibition of
PLC, which is an upstream activator of DAG, by U73122 inhibited progesterone production in
MA-10 cells, but significantly increased progesterone production in STARKO1 cells. This is
consistent with previous findings that showed that inhibition of PLC by U73122 in R2C rat tumor
Leydig cells, which constitutively produce steroids and have high basal levels of STAR, was
inhibitory to steroid production and decreased STAR protein levels [77]. The effects of DAG and
the PLC inhibitor on MA-10 cells are in agreement with previous studies on the role of PKC and
PLC in steroidogenic cells [145-147]. The increase in steroid production in STARKO1 cells in
response to PLC inhibition suggests that increased activation of DAG by PLC may contribute to
the lack of steroid production in the absence of STAR. This also suggests a mechanism for
steroidogenesis that is STAR independent but PLC dependent. We also found that inhibition of
PKC, which is activated by DAG, did not affect steroid production, suggesting that the role of
STAR in DAG signaling is independent of PKC activation. This may be explained by previous
data that suggested only 1,2-diacyl-sn-glycerols (1,2-DAG) can activate PKCs, whereas other
DAG stereoisomers (1,3-diacyl-sn-glycerols and 2,3 diacyl-sn-glycerols) cannot [148, 149].
Further, it has been shown that hormone sensitive lipase (HSL) preferentially hydrolyzes sn-1(3)
ester bonds, which then would not form 1,2-DAG and therefore not activate PKC [150]. These
data further implicate STAR in DAG signaling and/or metabolism, independent of its function in
cholesterol transport for hormone-induced steroidogenesis.
Lastly, we performed RNA sequencing to identify differentially expressed genes between
MA-10 and STARKO1 cells. Interestingly, our RNA sequencing analysis did not show that
41
steroidogenic factor 1 (SF-1) was differentially expressed, although a previous report showed that
knockout of SF-1 in Leydig cells led to lipid accumulation in part through the suppression of STAR
levels [151]. However, we did identify other differentially expressed genes that are associated with
lipid metabolism. PLIN1 is localized to the surface of lipid droplets and has been shown to play
an essential role in regulating the accumulation and hydrolysis of triacylglyceride (TAG) and DAG
in lipid droplets [138]. Upon phosphorylation by protein kinase A (PKA), PLIN1 recruits lipolytic
proteins, such as HSL, which has hydrolytic activity against DAG and TAG, to the lipid droplet
surface [152]. However, previous studies have demonstrated that reduction of PLIN1 levels
increased basal lipolysis but decreased PKA-stimulated lipolysis, showing that PLIN1 may
function in both repressing basal lipolysis and in enhancing PKA-stimulated lipolysis [152]. Here,
we see that under basal conditions, Plin1 mRNA expression is increased when STAR is knocked
out, which may contribute to decreased DAG metabolism. We also identified other genes
associated with DAG formation (PLPP3 and PIP5K1β) that were found to be increased in
STARKO1 cells compared to WT MA-10 cells. PLPP3 dephosphorylates phosphatidic acid (PA)
to form DAG [153]. PIP5K1β is necessary for phosphatidylinositol 4,5-bisphosphate (PIP2)
activity, which is not only metabolized to DAG, but is also an essential cofactor for PLD, which
hydrolyzes PC to PA, and which then can be converted to DAG [140]. Increased expression of
PLPP3 and PIP5K1β therefore gives insight into possible mechanisms of DAG accumulation that
we see in STARKO1 lipid droplets. Lastly, we found that expression of Stard12 was also increased
in STARKO1 cells. STARD12 was considered initially to function mainly as a tumor suppressor
[139]. However, further studies have shown that STARD12 binds PLC-𝛿1 and activates hydrolysis
of PIP2, which functions in DAG formation [154]. In addition, it has been shown that STARD12
is localized to the mitochondria in Huh-7 cells [155]. It is possible that in the absence of STARD1,
42
STARD12 expression increases to facilitate a compensatory mechanism. Collectively, these data
are consistent with DAG accumulation seen in STARKO1 lipid droplets and further suggest a role
of constitutive STAR in MA-10 cells.
4. Materials and Methods
4.1. Primary Leydig Cell Isolation
Leydig cells were isolated from adult Sprague-Dawley rats and C57BL/6 mice (Charles
River Laboratories) as previously described [25]. The rats and mice were bred and maintained in
accordance with protocols approved by the Institutional Animal Care and Use Committee of the
University of Southern California (Protocol # 20791; approved on 8/30/2018). Testis were
decapsulated, dissociated in 0.25 mg/ml collagenase, and shaken at 80 cycles/min at 34°C for 15
min. Once dissociated, the seminiferous tubules were removed, and supernatant-containing cells
were centrifuged at 800xg for 20 min. Pellets were applied to a Percoll density gradient and
centrifuged at 14,000 RPM for 45 min at 4°C. The Leydig cell-containing fraction was layered
onto a BSA density gradient and centrifuged at 50xg for 10 min, which yielded 85% pure Leydig
cells as revealed by 3β-hydroxysteroid dehydrogenase staining. For hCG stimulation, Leydig cells
were treated with 50 ng/ml hCG for 2 hours. Steroid production was measured using the
Testosterone ELISA Kit (Cayman Chemical). Steroid production data were normalized to cell
number. qRT-PCR and immunoblot analyses were carried out as described here.
4.2. Cell Culture
MA-10 cells were kindly provided by Dr. Mario Ascoli (University of Iowa, Iowa City,
IA). MA-10 cells were found to be negative for mycoplasma by the Mycoplasma PCR Detection
Kit (Applied Biological Materials Inc.). WT MA-10 mouse tumor Leydig cells [156] and STAR
KO cells were maintained in Dulbecco’s modified Eagle medium/F-12medium + Glutamax
43
supplemented with 5% heat-inactivated fetal bovine serum, 2.5% heat-inactivated horse serum,
and 1% penicillin/streptomycin at 37°C and 3.5% CO2.
4.3. CRISPR/Cas9-mediated gene deletion of Star in MA-10 cells
Two sets of ss DNA oligonucleotides specifically targeting exon2 of Star were designed
after using Synthego’s CRISPR single gRNA Design Tool (https://www.synthego.com). The ss
oligonucleotides were annealed and cloned into the GeneArt® CRISPR Nuclease Vector which
contains an OFP reporter (Thermo Fisher Scientific). Plasmid extraction was done using the Zyppy
Plasmid Miniprep Kit (Zymo Research). Sequencing of the cloned plasmids were done to confirm
the insertion of the gRNAs. Plasmids were transfected into MA-10 cells using Lipofectamine 3000
and Opti-MEM according to the manufacturer’s recommendations. 24 hours post-transfection,
FACS was used to sort the OFP positive cells into single colonies.
STAR KO was determined using NGS, qRT-PCR, and immunoblotting. NGS was done
through GENEWIZ Next Generation Sequencing Services. For Star mRNA determination, total
RNA was extracted using the Quick-RNA MiniPrep Plus kit (Zymo Research). 500 ng of isolated
RNA was applied to amplify cDNA using the PrimeScript RT Master Mix (Takara Bio). qRT-
PCR with PowerUP SYBR Green Master Mix (Applied Biosystems) detection was performed
using the qTOWER³ (Analyik Jena AG). This method was used for all qRT-PCR data shown here.
Primers for different regions of Star are as follows: STAR1 F:
5’TCCTCGCTACGTTCAAGCTG-3’ R: 5’-AGCTCCGACGTCGAACTT-3’, STAR2 F: 5’-
TCGCTACGTTCAAGCTGTGTG-3’ R: 5’-GGCTCCGACGTCGAACTTGA-3’, STAR3 F: 5’-
AGAGGTGGCTATGCAGAAGG-3’ R: 5’-CATGCGGTCCACAAGTTCTT-3’, STAR4 F: 5’-
GGAGCAGAGTGGTGTCATCA-3’ R: 5’-TGGCGAACTCTATCTGGGTC-3’. Gapdh was
used as the housekeeping gene.
44
4.4. Immunoblot analysis
Protein was extracted from mouse and rat primary Leydig cells, WT MA-10 cells, and
STAR KO cells using RIPA buffer. Protein concentration was determined after centrifugation
using the Pierce BCA Protein Assay Kit (Thermo Scientific). Sodium dodecyl sulfate-
polyacrylamide gel electrophoresis was performed using 7.5 μg of protein extract and a 4%-20%
Tris-glycine gradient gel (Bio-Rad) and the resulting bands electro-transferred to a polyvinylidene
fluoride membrane. Blocking of the membranes was done using 5% bovine serum albumin before
incubating with STAR antibody (1:1000; Cell Signaling #8449) and secondary antibody,
WesternSure® Goat anti-Rabbit HRP Secondary Antibody (1:5000; LI-COR). Membranes were
stripped using the Restore Western Blot Stripping Buffer (Thermo Scientific) and reprobed using
anti-β-actin (1:5000). The immunoreactive proteins were visualized using Radiance Peroxide and
Radiance Plus (Azure Biosystems) and imaged using the Azure c600 (Azure Biosystems).
4.5. Quantification of steroid production
WT MA-10 and STAR KO cells (1×10
4
per well) were plated on 96-well plates in triplicate
for 24 hours. Before stimulation, medium was removed, each well was washed with phosphate-
buffered saline, and serum-free medium was added with one of the following: 50 ng/ml human
chorionic gonadotropin (hCG; NIDDK), 1 mM dibutyryl cyclic AMP (dbcAMP; Sigma), 50 μM
22(R)-hydroxycholesterol (Sigma), 50 μM XBD173 (Sigma), 50 μM FGIN-1-27 (Cayman
Chemical), 1-oleoyl-2-acetyl-sn-glycerol (Cayman Chemical), U73122 (Cayman Chemical), or 1
mM calphostin C (Cayman Chemical). After 2 hours incubation at 37°C, the media was collected
to measure steroid production. The remaining cells were lysed with 0.1 N sodium hydroxide for
protein measurements. Steroid production was measured using the Progesterone ELISA Kit
(Cayman Chemical). Steroid production data were normalized to protein contents.
45
4.6. Lipid droplet imaging, isolation, and analysis
Lipid droplets were imaged by transmission electron microscopy (TEM) at the Doheny
Eye Institute (Los Angeles, USA). Lipid droplet isolation was done using Cell Biolabs, Inc.’s Lipid
Droplet Isolation Kit. 50 million cells were pelleted and processed according to the manufacturer’s
recommendations. Purity of the isolated lipid droplets were assessed through immunoblots using
GAPDH (1:1000; Cell Signaling) and PLIN1 (1:1000; Cell Signaling) antibodies. Western blot
analyses were performed as described above. The content of the isolated lipid droplets was then
analyzed using LC-MS. Samples for LC-MS were prepared by adding 200 µL of water to each
sample and vortexing. 1000 µL of methyl tert-butyl ether (MTBE) was then added to each sample,
vortexed for 10 seconds and left to settle for 10 minutes. 800 µL of the top layers were transferred
to 10x75 mm glass culture tubes. These steps were repeated. The MTBE extracts were then
evaporated at room temperature in a Thermo Scientific™ Savant™ SPD131DDA SpeedVac™
Concentrator. Extracts were reconstituted with 250 µL 10 mM ammonium acetate in 50:50 (v:v)
dichloromethane:methonol. Reconstituted extracts were then transferred to injection vials for flow
infusion (no column) LC-MS analysis. LC was done using the Shimadzu Nexera XR (Shimadzu
Corporation) with an injection volume of 50 µL, autosampler temperature of 15°C, and an infusion
flow rate of 7.0 µL/min. MS was done using the 5600+ TripleTOF quadrupole-time of flight
system (AB SCIEX). Scanning of the positive mode was done using TOF-NS1 from m/z 200 to
1250. Scanning of the negative mode was done through MS/MS, using unit mass resolution in the
quadrupole (0.7 amu bandpass) at each nominal mass from 200-1250. MS data was analyzed using
the LipidView Software (AB SCIEX).
4.7. Mitochondria and lipid droplet staining and confocal microscopy
46
To stain mitochondria and lipid droplets, cells were plated at a density of 200,000 cells per
well for 24 hours and stained with 0.1 μM MitoTracker Red CMXRos (Invitrogen) for 30 min at
37°C, washed with PBS 3 times, and stained with 2 μM Bodipy 493/503 (Invitrogen) for 15 min
at 37°C. Cells were fixed in 4% PFA for 15 min at room temperature, stained with DAPI, and
observed by Zeiss LSM 880 confocal microscopy.
4.8. Statistical analysis
All data are presented as mean ± standard error of the mean from three independent
experiments unless indicated otherwise. Moreover, all experiments were conducted in triplicates
unless indicated otherwise. GraphPad Prism (version 7) was used for graphic presentation and
statistical analyses of data were performed using the Student t test. Means were considered
statistically different when p < 0.05.
5. Conclusion
Taken together, these results provide evidence for a function of constitutive STAR in
Leydig cells that is distinct from its role in hormone-induced steroidogenesis. Constitutive STAR
may function in TSPO-mediated steroidogenesis and may have a role in DAG accumulation in
lipid droplets when STAR is mutated or knocked out. DAG accumulation may contribute to the
inability to produce steroids in the absence of functional STAR.
47
Chapter III: Role of Constitutive STAR in Mitochondrial Structure
and Function in MA-10 Leydig Cells
Abstract
The steroidogenic acute regulatory protein (STAR) is critical for the transport of
cholesterol into the mitochondria for hormone-induced steroidogenesis. Steroidogenic cells
express STAR under control conditions (constitutive STAR). Upon hormonal stimulation, STAR
localizes to the outer mitochondrial membrane (OMM) where it facilitates cholesterol transport
and where it is processed to its mature form. Here, we show that knockout of STAR in MA-10
mouse tumor Leydig cells (STARKO1) causes defects in mitochondrial structure and function
under basal conditions. We also show that overexpression of STAR in STARKO1 cells
exacerbates, rather than recovers, mitochondrial structure and function, which further disrupts the
processing of STAR at the OMM. Our findings suggest that constitutive STAR is necessary for
proper mitochondrial structure and function and that mitochondrial dysfunction leads to defective
STAR processing at the OMM.
1. Introduction
Steroidogenesis is a tightly controlled process involving a highly intricate protein complex
and multiple steroidogenic enzymes that are localized to various cellular compartments such as the
cytoplasm, mitochondria, and endoplasmic reticulum in steroidogenic cells [40, 54, 157].
Mitochondria play an indispensable role in steroidogenesis, as the first committed step of steroid
production, the conversion of cholesterol to pregnenolone by cytochrome P450 cholesterol side
chain cleavage enzyme (CYP11A1), is in the mitochondrial matrix [8]. Additionally, the hormone-
48
sensitive and rate-limiting step of steroidogenesis is the transfer of cholesterol from the outer
mitochondrial membrane (OMM) to the inner mitochondrial membrane (IMM), which occurs
through the transduceosome, an intricate and multifaceted protein complex that includes multiple
cytosolic and OMM proteins such as the translocator protein (TSPO) and the voltage-dependent
anion channel 1 (VDAC1) [50, 158]. Upon hormonal stimulation, the transduceosome complex
forms at the OMM, increasing the transport of cholesterol into the mitochondria and initiating
steroidogenesis [50].
The steroidogenic acute regulatory protein (STAR; STARD1) is a transduceosome protein
that facilitates cholesterol transport into the mitochondria and is critical for acute hormone-induced
steroid production [59]. Many studies have demonstrated that an increase in STAR protein
expression is linked to an increase in steroidogenesis [56, 159, 160]. Furthermore, it was shown
that the role of STAR in steroidogenesis is indispensable, since mutations in the human STAR gene
were found to cause congenital lipoid adrenal hyperplasia (lipoid CAH) [60]. Infants with lipoid
CAH have a severe deficiency of gonadal and adrenal steroid production and must receive steroid
hormone replacement to avoid death by salt loss, hyperkalemia, acidosis, or dehydration [60, 133,
161]. However, although these data show clear evidence of the importance of STAR in
steroidogenesis, the exact mechanism by which STAR induces the transport of cholesterol into the
mitochondria is still being studied.
STAR is synthesized as a 37 kDa cytosolic preprotein that contains an N-terminal
mitochondrial targeting sequence [72]. Hormonal stimulation causes a rapid induction of STAR
protein expression and also causes the translocation of STAR to the mitochondria [73, 162]. STAR
functions solely in cholesterol transport at the OMM and the time it spends at the OMM directly
determines the rate of steroidogenesis [73-75, 163]. Moreover, STAR is active at the OMM only
49
transiently, and its mitochondrial targeting sequence is rapidly cleaved to form a mature, inactive
30 kDa protein that is then imported into the mitochondrial matrix, but its exact role there is still
speculative [73].
Previously, we developed STAR-knockout (KO) MA-10 mouse tumor Leydig cells
(STARKO1), and showed that constitutive STAR, or STAR that is expressed independent of
hormonal stimulation, plays a role in lipid metabolism independent of its role in cholesterol
transport [10]. Since STAR functions at mitochondria and because previous work has shown that
steroidogenesis affects mitochondrial ultrastructure and function, we were interested in
determining whether STAR KO affects mitochondrial ultrastructure and/or function [164-166].
Here, we found that STAR KO did severely alter mitochondrial ultrastructure, leading to breaks
in the membrane and a reorganization of cristae structure. We also found that STAR KO altered
mitochondrial bioenergetics, leading to alterations in the rates of mitochondrial respiration, ATP
production, and glycolysis. To determine whether reintroducing STAR would recover
mitochondrial structure and/or function in STARKO1 cells, we generated STAR overexpressing
STARKO1 cells (STARKO-OE1 and STARKO-OE2). Unexpectedly, we found that STAR
overexpression in these cells exacerbated, rather than recovered, mitochondrial dysfunction.
Lastly, by utilizing STARKO-OE1 and STARKO-OE2, as well as MA-10 cells with STAR
overexpression, we showed that these alterations in mitochondrial structure and function inhibit
the processing of STAR to its mature form. Collectively, our findings suggest that constitutive
STAR is necessary for mitochondrial structure and function and that alterations in mitochondrial
structure and function inhibit the processing of STAR at the OMM.
2. Materials and Methods
Cell Culture
50
MA-10 cells were kindly provided by Dr. Mario Ascoli (University of Iowa, Iowa City,
IA, USA). We previously used CRISPR/Cas9 to develop STAR KO MA-10 cell lines (STARKO1
and 2) [10]. In this study, we used only the STARKO1 cell line. STAR KO cells do not produce
steroids in response to hormonal stimulation but do respond to treatment with 22(R)-
hydroxycholesterol [10]. WT MA-10 cells and STARKO1 cells were cultured at 37°C and 3.5%
CO2 in Dulbecco’s modified Eagle medium/F-12medium + Glutamax supplemented with 5% heat-
inactivated fetal bovine serum, 2.5% heat-inactivated horse serum, and 1%
penicillin/streptomycin. STAR overexpressing STARKO1-OE1, STARKO1-OE2, and wild-type
(WT) MA-10 cells overexpressing STAR (MA-10+STAR) cell lines (described below) were
maintained in the same media, supplemented with 400 μg/mL of G418 (Thermo Fisher Scientific,
Waltham, MA, USA).
STAR Overexpression
STARKO1 or WT MA-10 cells were transfected with STAR plasmid with a myc tag
(Origene, Rockville, MD, USA, Catalog # MR203861) using Lipofectamine 3000 and Opti-MEM
according to the manufacturer’s recommendations (Thermo Fisher Scientific, Waltham, MA,
USA). Positively transfected cells were selected using 400 μg/mL of G418. STAR overexpression
was determined via immunoblot as described below.
Transmission Electron Microscopy
TEM was done at the Core Center of Excellence Nano Imaging at the University of
Southern California. Primary fixative contained 2.5% glutaraldehyde, 2% formaldehyde, 0.1 M
HEPES, and 0.115 M Sucrose. Cells were washed with 0.1 M cacodylate. Secondary fixative
contained 1% osmium tetroxide. Cells were stained with uranyl acetate, followed by a series of
30-100% ethanol washes done to dehydrate the cells. Propylene oxide was used to transition the
51
cells to a microfuge tube, and the cells were then infiltrated by increasing concentrations of
polybed 812 epoxy resin. A Lecia EM UC6 Ultramicrotome was used to section the cell-containing
block, which was examined using an FEI Talos F200C G2 Biological Transmission Electron
Microscope (Thermo Fisher Scientific, Waltham, MA, USA).
Seahorse Analysis
Cells were plated at a density of 10,000 cells per well in Seahorse XF Cell Culture
microplates overnight in complete media (Agilent Technologies, Santa Clara, CA, USA). Assay
media consisted of Agilent Seahorse XF DMEM supplemented with 1 mM glucose, 1 mM
pyruvate, and 2 mM glutamine. Mitochondrial respiration was measured using the Seahorse
Agilent Mito Stress Test Kit. Concentrations of 2.5 μM oligomycin, 2 μM FCCP, and 0.5 μM
Rot/AA were used for this assay. ATP produced via mitochondrial oxidative phosphorylation or
via glycolysis, the two major pathways of ATP production, was measured using the Seahorse XF
Real-Time ATP Rate Assay Kit. ATP produced by oxidative phosphorylation was calculated using
the oxygen consumption rate inhibited by oligomycin while ATP produced via glycolysis was
calculated using the proton efflux rate derived from glycolysis. Glycolysis was measured using the
Agilent Seahorse XF Glycolytic Rate Assay Kit. All test kits were used according to
manufacturer’s recommendations. The assays were run on a Seahorse XFe96 Analyzer and
analyzed in Wave 2.6.1.
Immunoblot Analysis
Immunoblot analyses was done as described previously [10]. Briefly, protein extraction
was done using RIPA buffer supplemented with protease inhibitor and protein concentration was
done using the Pierce BCA Protein Assay Kit (Thermo Scientific, Waltham, MA, USA). SDS-
PAGE was conducted using a 4%-20% Tris-glycine gradient gel. Proteins were transferred to a
52
PVDF membrane, which was then blocked using 5% BSA for 30 minutes. The list and descriptions
of antibodies are given in Table 1. Proteins were visualized using Radiance Peroxide and Radiance
Plus and imaged using an Azure c600 (Azure Biosystems, Dublin, CA, USA).
Steroid Production Measurements
Cells were plated at a density of 10,000 cells per well in 96-well plates overnight for 24
hours. Culture medium was removed, each well was washed twice with PBS, and 50 ng/mL human
chorionic gonadotropin (hCG; National Hormone and Peptide Program, Harbor-UCLA Medical
Center, Torrance, CA, USA) in serum-free media was used to stimulate cells at 37°C for 2 hours.
The media were collected to measure steroid production using a Progesterone ELISA Kit
according to the manufacturer’s recommendations (Cayman Chemical, Ann Arbor, MI, USA;
catalog no.: 582601; RRID: AB_2811273). For data in Figure 2B, the intra-assay coefficient of
variation was 9.1% and for data in Figure 6A, the intra-assay coefficient of variation was 8.5%.
53
All samples were assayed in triplicate. All %B/B0 values were between 20-80%. Progesterone
production was normalized to protein concentration.
Immunocytochemistry
Cells were plated at a density of 200,000 cells per well on coverslips for 24 hours. For
mitochondria, cells were stained with 0.1 μM MitoTracker Red CMXRos at 37°C for 30 minute
and washed 3 times with PBS (Thermo Scientific, Waltham, MA, USA). Cells were fixed in 4%
PFA in PBS for 10 minutes at room temperature, permeabilized with 0.1% Triton-X 100, blocked
with 5% donkey serum, and incubated with primary antibodies overnight at 4°C (Table 1). Cells
were washed and incubated with secondary antibodies for 30 minutes at room temperature. Cells
were stained with DAPI and observed using a Zeiss LSM 880 confocal microscope. Quantification
of colocalization was done using ImageJ analysis of 10 images per sample.
Statistical analysis
All data are presented as mean ± standard error of the mean from three independent
experiments unless indicated otherwise. Moreover, all experiments were conducted in triplicate
unless indicated otherwise. GraphPad Prism (version 8) was used for graphical presentation, and
statistical analyses were performed using the Student’s t test. Means were considered statistically
different when p < 0.05.
3. Results
Star knockout causes alterations in mitochondrial structure and mitochondrial respiration
Previously, we conducted RNA sequencing of wild type (WT) MA-10 and STARKO1
cells, generated a differentially-expressed gene (DEG) list, and identified many lipid-related genes
that were differentially expressed between the two cell types [10]. These data are available at the
NCBI Gene Expression Omnibus database (GEO); accession no. GSE165392.
54
As our first step to determine whether there were any potential changes in mitochondria
with STAR knock out, we identified genes from the DEG list that were related to mitochondrial
structure and/or function; the function or pathway of the gene, the gene names, and the fold change
are listed in Table 2 with genes with the highest fold change shown in gray. There were many
genes in this list that pointed to alterations in mitochondrial morphology and mitochondrial
function. For example, myoferlin (Myof), which is a regulator of mitochondrial fission, and O-
GLcNAcase (Oga), which has been shown to regulate mitochondrial ultrastructure and membrane
potential, were found to be differentially expressed between WT MA-10 cells and STARKO1
cells, suggesting potential changes in mitochondrial structure [167-169]. To determine whether
mitochondrial structure was actually altered as a result of STAR KO, we used TEM to visualize
mitochondrial ultrastructure. In WT MA-10 cells, mitochondria have a well-defined double
membrane and distinct tubulovesicular cristae structure (Figure 9A). However, in STARKO1 cells,
most mitochondria lacked a defined membrane structure or had clear breaks in the membrane, and
the cristae structure was no longer clearly visible, indicating mitochondrial damage (Figure 9A).
Together, these data suggest that STAR depletion in MA-10 cells leads to alterations in
mitochondrial structure and that constitutive STAR plays a role in mitochondrial morphology.
Next, we investigated changes in mitochondrial function, since our RNA-sequencing data
also identified many differentially expressed genes related to mitochondrial function (Table 2).
We used Seahorse XF analyses to study various parameters of mitochondrial function. Using the
Mito Stress Test kit, we could identify any alterations in mitochondrial respiration. However, we
found that basal respiration and total ATP production were not different between WT MA-10 and
STARKO1 cells (Figure 9B). These parameters measure a cell’s energetic demand under baseline
conditions and the ATP produced by mitochondria to meet that energetic demand, respectively
55
[170]. However, we did find that maximal respiration, spare respiratory capacity, and proton leak
of STARKO1 cells were significantly lower than that of WT cells (Figure 9B). Maximal
respiration measures the maximum rate of respiration that a cell can achieve in response to
energetic demand, while spare respiratory capacity measures a cell’s ability to respond to energetic
demand and may indicate cell fitness [170]. These data suggest that STARKO1 cells are less
capable of responding to energetic demands than WT cells, indicating defects in mitochondrial
56
respiration. Additionally, proton leak measures basal respiration that is not coupled to ATP
production, which may be used to regulate mitochondrial ATP production [170]. Collectively,
57
these data demonstrate that STAR KO causes alterations in mitochondrial function as seen by
decreases in spare respiratory capacity and proton leak.
Development of STAR overexpressing Star knockout cells
After showing that STAR KO causes mitochondrial dysfunction, we investigated whether
reintroducing STAR would be sufficient to recover mitochondrial structure and function, a typical
approach when using knock out cell lines. We overexpressed STAR in STARKO1 cells using a
plasmid with a myc tag. Immunoblot analysis showed that control WT MA-10 cells contained the
30 kDa mature STAR protein, while STARKO1 did not (Figure 10A). Following transfection,
selection, and immunoblot analyses, we identified two samples that successfully re-expressed
STAR, which we named STARKO-OE1 and STARKO-OE2 (Figure 10A). Although both samples
express STAR under basal conditions, our data showed that the induction of STAR levels
following hormonal stimulation varied between STARKO-OE1 and STARKO-OE2. Specifically,
STARKO-OE1 had low basal levels of STAR, while STARKO-OE2 had basal levels of STAR
that were higher than that of untreated WT MA-10 cells, consistent with overexpression. In
response to hormonal stimulation, there was a significant induction of STAR in STARKO-OE1
cells, but there was no change in the levels of STAR between basal and stimulated STARKO-OE2
samples (Figure 10A). Additionally, STARKO-OE1 and STARKO-OE2 STAR protein ran at a
higher molecular weight than WT MA-10 cells, which suggests a defect in the processing of STAR
to the 30 kDa mature, inactive form in these samples. To investigate potential defects in STAR
Figure 9. STAR KO results in alterations in mitochondrial structure and function. (A) Electron
microscopy images of mitochondria in WT MA-10 cells (left) and STARKO1 cells (right).
Arrows indicate alterations in the membrane and cristae structure of STARKO1 mitochondria.
(B) Seahorse XF analyses of basal respiration, ATP production, maximal respiration, spare
respiratory capacity, and proton leak of WT MA-10 cells compared to STARKO1 cells. Data are
shown as mean ± SEM (n=3); *** P < 0.001.
58
processing, we also developed MA-10 cells with STAR overexpression (MA-10+STAR). We
address the processing of STAR in STARKO-OE1 and STARKO-OE2 in comparison to STAR
processing in MA-10 cells with STAR overexpression later in the Results section.
To determine whether re-expressing STAR in STARKO1 cells is sufficient to recover
steroidogenic activity in these cells, we measured steroid production by ELISA. We found that
STARKO-OE1 cells had a steroid production profile similar to that of WT MA-10 cells, with low,
but detectable basal progesterone production and significantly higher progesterone production
following hormonal stimulation with hCG (Figure 10B). In contrast, STARKO-OE2 had high
basal progesterone production that was only slightly increased in response to hCG (Figure 10B).
These data show that STAR re-expression in STARKO1 cells is sufficient to recover steroidogenic
activity, though STARKO-OE1 and STARKO-OE2 produce different levels of basal steroids
consistent with their respective basal levels of STAR.
Star overexpression in Star knockout cells is insufficient to recover mitochondrial structure and
function
We also wanted to characterize STARKO-OE1 and STARKO-OE2 in relationship to
mitochondria by determining whether STAR overexpression in these cells was sufficient to
recover mitochondrial structure and function. We imaged mitochondrial ultrastructure by TEM,
and mitochondria in STAR-overexpressing STARKO1 cells looked significantly different than
those in WT MA-10 or STARKO1 cells. Specifically, STARKO-OE1 mitochondria had less
59
ruptures in the mitochondrial membrane compared to STARKO1 cells, but the cristae still looked
more linear and less defined than those seen in WT MA-10 cells (Figure 11A). Additionally,
mitochondrial membrane and matrix structure looked to be even more disrupted in STARKO-OE2
cells compared to STARKO1 (Figure 11A). To quantify the number of normal versus damaged
mitochondria, we counted 200+ mitochondria in each sample and measured the percentage of
Figure 10. Reintroduction of STAR in STARKO1 cells. (A) Immunoblot analyses of STAR-
overexpressing STARKO1 cells (STARKO-OE1 and STARKO-OE2) under basal (left) and
hormone-stimulated (right) conditions. β-actin is used as the normalizing protein. (B) ELISA
analyses of progesterone levels in cell media under basal and hormone-stimulated conditions.
Data are shown as mean ± SEM (n=3); * P < 0.05; ** P < 0.01; *** P < 0.001.
60
mitochondria that were damaged (Table 3). We found that 5.5% of mitochondria in WT MA-10
cells were damaged compared to 81.9% in STARKO1 cells. Interestingly, STARKO-OE1 cells
had a slightly lower percentage of damaged mitochondria than STARKO1 cells at 68.9%, while
STARKO-OE2 cells had a slightly higher percentage at 85.7%. Therefore, STAR overexpression
in STARKO1 cells was insufficient to recover mitochondrial structure. Furthermore, alterations in
mitochondrial structure were exacerbated in STARKO-OE2 cells, which express higher basal
levels of STAR and basal steroid production.
Next, to further characterize the mitochondrial effects of STAR overexpression in STAR
KO cells, we repeated the Seahorse analyses with STARKO-OE1 and STARKO-OE2 cells. Using
the Mito Stress Test kit, we measured the various parameters of mitochondrial respiration and
found that, while there were no changes in basal respiration or ATP production between WT MA-
10 and STARKO1 cells, the rates of both of these parameters were significantly lower in
STARKO-OE1 and STARKO-OE2 cells compared to WT MA-10 cells, but not significantly
different from STARKO1 cells (Figure 12A). Additionally, maximal respiration, spare respiratory
capacity, and proton leak were significantly lower in STARKO-OE1 and STARKO-OE2 than WT
MA-10 cells, but similar to STARKO1 cells (Figure 12A).
Because we saw differences in total ATP production between WT MA-10 and STARKO-
OE1 and STARKO-OE2 cells, we used the Seahorse XF Real-Time ATP Rate Assay to look
further into cellular ATP production. The ATP Rate Assay measures the rates of ATP production
derived specifically via glycolysis and via oxidative phosphorylation, the two main pathways of
cellular ATP production. When measuring rates of ATP production, we found that glycolysis-
derived ATP (glycoATP) was significantly lower in STARKO-OE1 and STARKO-OE2 compared
to WT, which was similar to what was seen in STARKO1 cells (Figure 12B). However, while ATP
61
produced by oxidative phosphorylation (mitoATP) was significantly higher in STARKO1 cells, it
was significantly lower in STARKO-OE1 cells compared to WT (Figure 12B). These alterations
resulted in total ATP production by STARKO-OE1 and STARKO-OE2 that was significantly
lower than total ATP produced by WT MA-10 cells (Figure 12B). Together, these data suggest
that the alterations seen in mitochondrial structure and/or function when STAR is knocked out are
exacerbated by the reintroduction of STAR and that STAR KO leads to alterations in ATP
production.
62
Next, because we saw differences in glycolysis-derived ATP production, we used the
Seahorse XF Glycolytic Rate Assay to measure rates of glycolysis. The Glycolytic Rate Assay,
which defines glycolysis as the process of converting glucose to lactate, can specifically measure
the number of protons exported by cells into the assay media over time by glycolysis. When we
measured the glycolytic rate of these cells, we found that basal glycolysis and the proton efflux
rate derived from glycolysis were significantly lower in STARKO-OE1 and STARKO-OE2 cells
compared to WT, similar to what was seen for STARKO1 cells (Figure 12C). Also, in general, the
parameters measured for each assay was lowest for STARKO-OE2, and the rate of compensatory
glycolysis, which is the rate of glycolysis after oxidative phosphorylation is inhibited that then
drives the cell to use glycolysis to meet energy demands, was significantly lower in STARKO-
OE2 cells compared to all other samples (Figure 12C). Taken together, these data show that STAR
re-expression in STARKO1 cells does not recover mitochondrial function but exacerbates
mitochondrial dysfunction and rates of glycolysis.
Processing of STAR is disrupted as a result of mitochondrial dysfunction
Lastly, we were interested in determining whether STAR in STAR-overexpressed
STARKO1 cells was still able to localize to the mitochondria. We co-stained STARKO-OE1 and
STARKO-OE2 cells with Mitotracker and anti-STAR antibodies and visualized the cells by
Figure 11. Mitochondrial ultrastructure in STARKO-OE1 and STARKO-OE2 cells. (A)
Electron microscopy images of mitochondria in STAR-overexpressing STARKO1 cells
(bottom panels) compared to WT MA-10 and STARKO1 mitochondria (top panels).
63
confocal microscopy (Figure 13A). We quantified the colocalization of mitochondria and STAR
and found that STAR in STARKO-OE1 cells colocalized to mitochondria under both basal and
Figure 12. Characterization of mitochondrial function in STARKO-OE1 and STARKO-
OE2 cells. (A) Seahorse XF analyses of mitochondrial respiration including basal
respiration, ATP production, maximal respiration, spare respiratory capacity, and proton
leak. (B) Seahorse XF analyses of ATP produced via glycolysis (glycoATP) or via
oxidative phosphorylation (mitoATP) and total ATP production. (C) Seahorse XF
analyses of glycolytic rates including basal glycolysis, basal proton efflux rate (PER), %
PER from glycolysis, and compensatory glycolysis. Data are shown as mean ± SEM.
(n=3) * P < 0.05; ** P < 0.01; *** P < 0.001.
64
hormone-induced conditions similar to WT MA-10 cells. We also found that colocalization
between STAR and mitochondria in STARKO-OE2 was significantly higher than what was seen
for WT MA-10 cells, likely due to the higher basal expression of STAR in these cells.
Although we showed that STAR in STAR-overexpressing STARKO1 cells localizes to
mitochondria, we wanted to determine whether STAR localized specifically to the OMM in these
samples. Previous work showed that STAR requires contact with the OMM protein VDAC1 for
proper activity [55]. We co-stained STARKO-OE1 and STARKO-OE2 cells with anti-VDAC1
and anti-STAR to investigate whether STAR was localized to the OMM in these samples (Figure
5B). Our data showed that in WT MA-10 cells, hCG stimulation caused a significant decrease in
STAR-VDAC1 colocalization compared to control, likely due to the rapid processing of STAR at
the OMM and subsequent translocation to the matrix following hormonal stimulation (Figure
13B). However, in STARKO-OE1 and STARKO-OE2 cells, there was no difference in STAR-
VDAC1 colocalization between basal and hormone-stimulated cells. These data suggest that there
is a defect in the processing of 37 kDa STAR, which acts at the OMM, to the 30 kDa mature form
that is targeted to the mitochondrial matrix in STARKO-OE1 and STARKO-OE2 cells, potentially
due to the alterations in mitochondrial structure and/or function described above.
Finally, we determined whether STAR overexpression in WT MA-10 cells using the same
plasmid introduced in STARKO-OE1 and STARKO-OE2 would also result in defects in the
processing of STAR. We overexpressed STAR in MA-10 cells (MA-10+STAR) and measured
steroid production by ELISA. MA-10+STAR cells had a similar profile as STARKO-OE2 cells,
with high basal and hormone-induced levels of progesterone production (Figure 14A). Next, we
used immunoblotting to determine whether there was also a defect in the processing of STAR in
MA-10+STAR cells, where mitochondria are still intact. We found that all STAR-overexpressing
65
cell lines had a band at the same molecular weight as full-length STAR in WT MA-10 cells (Figure
14B). However, in addition to the band for 37 kDa STAR, WT MA-10 and MA-10+STAR cells
had a strong 30 kDa band, indicative of proper processing to the mature 30 kDa STAR.
Furthermore, we used a myc antibody to probe for myc-tagged STAR. We found that the intensity
of the myc band for STARKO-OE1 and STARKO-OE2 was consistent with those seen in the
immunoblot for STAR shown in figure 14B, where myc-tagged STARKO-OE1 levels are low
under basal conditions and high under hormonal stimulation, while myc-tagged STARKO-OE2
levels are high under both basal and hormone-stimulated conditions (Figure 14C). However, we
found that following hCG stimulation, there was a decrease in the intensity of the MA-10+STAR
band, suggesting that STAR with the myc tag was processed correctly following hormonal
stimulation (Figure 14C). These data show that when mitochondria are intact and functional, as in
MA-10+STAR cells, STAR is processed correctly. Overall, our data suggest a role for constitutive
STAR in mitochondrial structure and function since STAR KO in MA-10 cells causes
mitochondrial dysfunction that cannot be recovered by STAR re-expression and may lead to the
inability of STAR to be processed at the OMM.
66
67
4. Discussion
Previously, our laboratory developed STAR KO MA-10 cells (STARKO1) and showed
that constitutive STAR plays a role in lipid metabolism independent of its role in cholesterol
transport for steroidogenesis [10]. Here, we worked on further characterizing STARKO1 cells and
potential roles of constitutive STAR. Our data showed that constitutively expressed STAR plays
a role in mitochondrial structure and function, evidenced by changes in mitochondrial morphology
and rates of mitochondrial respiration in the absence of STAR.
Mitochondria are double membrane organelles with an OMM and IMM separated by an
aqueous intermembrane space, which together surround the mitochondrial matrix [171]. -
Characteristic of steroid-producing cells and in addition to containing a significant number of lipid
droplets, Leydig cells contain numerous mitochondria with tubulovesicular cristae [172, 173]. Our
TEM images of WT MA-10 cell mitochondria were consistent with normal mitochondrial
ultrastructure, with a well-defined double membrane and tubulovesicular cristae. However,
mitochondria in STARKO1 cells were characterized by many ruptures in the membranes.
Ruptured mitochondria were also seen in adrenal glands of STAR KO mice [165]. Structural
damage to the mitochondria may be caused by loss of membrane integrity or organelle swelling,
both of which are indicative of cellular stress [174]. Additionally, STARKO1 mitochondria had
an undefined cristae structure with many electron poor areas, indicating a reorganization of lipid
Figure 13. Localization of STAR to the mitochondria. (A) Colocalization analysis between
STAR and Mitotracker under basal and hormone-induced conditions in various samples.
Quantification is shown below images. (B) Colocalization analysis between STAR and
VDAC1 under basal and hormone-induced conditions in various samples. Quantification is
shown to the right of the images. Data are shown as mean ± SEM. * P < 0.05; ** P < 0.01;
*** P < 0.001.
68
Figure 14. Characterization of STAR overexpression in WT MA-10 cells. (A) ELISA
analyses of progesterone levels in cell media under basal and hormone-stimulated
conditions. (B) Immunoblot analyses of STAR overexpression in WT MA-10 cells (MA-
10+STAR) under basal (left) and hormone-stimulated (right) conditions. (C) Immunoblot
analyses of myc tag in STARKO1-OE1, STARKO-OE2, and MA-10+STAR cells under
basal (left) and hormone-stimulated (right) conditions. β-actin is used as the normalizing
protein. Data are shown as mean ± SEM (n=3) ** P < 0.01; *** P < 0.001.
69
and protein composition in the matrix, which has been seen in a variety of pathologic conditions
[175]. Furthermore, while mitochondria in STARKO-OE1 cells were less damaged than those seen
in STARKO1 cells, many mitochondria had electron poor areas in the matrix and cristae that were
more linearly shaped. Linearization of cristae with enhanced electron density suggests a substantial
reorganization of membrane lipid organization [171]. Lastly, STARKO-OE2 cells had the most
damaged mitochondria, characterized by breaks in mitochondrial membranes and reorganization
and linearization of the cristae structure. Since mitochondrial membrane potential is vital for ATP
synthesis and the inner membrane that forms the cristae contains proteins critical for the electron-
transport chain and ATP synthesis, alterations in membrane and cristae structure may explain the
changes seen in mitochondrial respiration in these cells [176].
Previous work has suggested a relationship between steroidogenesis and mitochondrial
function, bioenergetics, and dynamics. For example, studies have shown that steroidogenesis in
MA-10 cells is highly dependent on mitochondrial ATP and that inhibition of the electron transport
chain in primary rat Leydig cells blocks hormone-stimulated steroid production [177, 178].
Additionally, a previous study has shown that the luteinizing hormone-cAMP pathway is critical
for the regulation of mitochondrial dynamics and energetics [164]. Our work supports these
findings, as we have shown that KO of STAR impairs steroidogenesis and causes defects in
mitochondrial respiration, ATP production, and glycolysis. Since MA-10 cells are tumor Leydig
cells, they rely heavily on glycolysis to produce ATP, which is necessary for rapid growth and
proliferation [179]. Our data showed that STARKO1 cells had significantly lower rates of
glycolysis and ATP produced via glycolysis, but higher rates of ATP produced via oxidative
phosphorylation compared to WT MA-10 cells. Similar to our data, a previous study showed that
knockdown of GATA4, a transcription factor expressed in mammalian Leydig cells and a regulator
70
of steroidogenesis, led to impaired glycolysis [180]. Previous work has also shown that glycolysis-
suppressed cancer cells are reprogrammed to use mitochondrial oxidative phosphorylation via
autophagy, which may explain why total ATP production was not changed between WT MA-10
and STARKO1 cells despite a significant reduction in glycoATP in STARKO1 cells [179].
To determine whether these alterations in mitochondrial structure and function could be
rescued by STAR re-expression, we reintroduced STAR into STARKO1 cells. Our data showed
that reintroduction of STAR exacerbated rather than rescued mitochondrial function, likely due to
irreversibly damaged mitochondria in STARKO1 cells, as seen by membrane ruptures and
unfolding of the cristae. Alterations in mitochondrial function in STARKO1-OE1 and STARKO1-
OE2 cells included changes in total ATP production. Damaged mitochondria can lead to decreased
ATP production which may result in swelling, reduced enzyme activity, and reduced protein
synthesis [181, 182]. Additionally, damaged mitochondria have impaired oxidative
phosphorylation leading to increased reactive oxygen species, which can lead to lipid, protein, and
nucleic acid damage [182-184]. Thus, KO of STAR may have severe downstream consequences
involving ATP production because STAR plays a role in proper mitochondrial structure and
function.
Despite irreversible changes in mitochondrial structure and function, STARKO-OE1 and
STARKO-OE2 cells did recover steroidogenic activity. Contrary to this, several reports associate
decreased steroidogenesis with damaged mitochondria, consistent with our data with STARKO1
cells. For example, knockdown of AAA domain-containing protein 3A (ATAD3A), a protein
suggested to play a role in OMM and IMM contact sites for hormone-induced steroidogenesis and
VDAC1, in MA-10 cells both resulted in decreased acute steroidogenesis accompanied by
alterations in mitochondrial cristae morphology [54]. Though this may occur in some cases, other
71
studies have shown that steroidogenic activity is spared despite alterations in mitochondrial
structure. Indeed, ablation of optic atrophy 1 (Opa1), a mitochondrial fusion protein required for
cristae maintenance, in BeWo cells resulted in increased steroidogenesis despite cristae
remodeling [185]. Further, mice in which mitochondrial dysfunction was induced maintained
normal testosterone production [186]. In the case of STAR overexpressing STARKO1 cells, it may
be that despite mitochondrial damage, STAR can still act on the OMM to mobilize stored
cholesterol and thus, induce steroid formation.
Lastly, we found that the processing of STAR was defective in STARKO-OE1 and
STARKO-OE2 cells. STAR is synthesized as a 37 kDa cytosolic protein and upon hormonal
stimulation is localized to the OMM where it facilitates cholesterol transport into the mitochondria
and where afterwards it is rapidly cleaved to a 30 kDa inactive protein that is targeted to the matrix
[73]. Our data are consistent with previous work showing that STAR acts at the OMM and that the
time STAR spends at the OMM is directly correlated with the rate of steroid production. We
showed that when mitochondria are damaged, with membrane breaks and cristae restructuring,
STAR can no longer be processed to its 30 kDa form, though steroidogenesis can still occur. This
may explain why STARKO1-OE2 cells, which had the most damaged mitochondria, also have the
highest basal levels of steroid production, perhaps because STAR cannot be processed or
translocated into the mitochondrial matrix. Collectively, our data show that STAR KO in MA-10
cells causes irreversible damage to mitochondrial structure and function that results in defects in
STAR processing, which is then worsened by reintroduction of STAR.
Conclusion
Our data show that STAR KO in MA-10 Leydig cells alters mitochondrial structure and
function, which is not recovered, but rather, further impaired, by the reintroduction of STAR.
72
Additionally, we show that when mitochondrial structure and/or function is impaired, STAR
processing is defective, resulting in the inability of STAR to move from the OMM to the
mitochondrial matrix.
73
Chapter IV: SCPx Deficiency Caused by Novel Heterozygous SCP2
Variant Leads to Severe Alterations in Lipid Metabolism
Abstract
The detoxification of very-long chain and branched-chain fatty acids and the metabolism
of cholesterol to form bile acids occurs largely through a process called peroxisomal β-oxidation.
Mutations in several peroxisomal proteins involved in β-oxidation have been reported, resulting in
diseases characterized by neurological defects. The final step of the peroxisomal β-oxidation
pathway is catalyzed by sterol carrier protein-x (SCPx) which is encoded by the SCP2 gene.
Previously, there have been two reports of SCPx deficiency, which resulted from a homozygous
and compound heterozygous SCP2 mutation, respectively. We report herein the first patient with
a heterozygous SCP2 mutation leading to late onset SCPx deficiency. Clinical presentations of the
patient included progressive brainstem neurodegeneration, cardiac dysrhythmia, muscle wasting,
and azoospermia. Plasma fatty acid analysis revealed abnormal values of medium-, long-, and
very-long chain fatty acids. Protein expression of SCPx and other enzymes involved in β-oxidation
were altered between patient and normal fibroblasts. RNA sequencing and lipidomic analyses
identified metabolic pathways including PPAR signaling, serotonergic signaling, steroid
biosynthesis, and fatty acid degradation. Treatment with fenofibrate or 4-hydroxytamoxifen
increased SCPx levels and certain fatty acid levels in patient fibroblasts. These findings suggest
that the patient’s SCP2 mutation resulted in decreased protein levels of SCPx, which altered many
metabolic pathways. Increasing SCPx levels through pharmacological interventions may reverse
some effects of SCPx deficiency. Collectively, this work provides insight into many of the clinical
consequences of SCPx deficiency and provides evidence for potential treatment strategies.
74
1. Introduction
Mitochondria and peroxisomes are the two sites of fatty acid β-oxidation, with the
metabolism of short-, medium-, and long-chain fatty acids occurring in mitochondria, and the
chain shortening of very-long chain fatty acids (VLCFA), long-chain dicarboxylic acids, 2-
methyl-branched fatty acyl-CoAs, eicosanoids, and bile acid precursors taking place in
peroxisomes [187]. Peroxisomal β-oxidation of straight-chain fatty acyl-CoAs occurs through
the function of acyl-CoA oxidase 1 (ACOX1), L-bifunctional protein (LBP), and 3-ketoacyl-
CoA thiolase (ACAA1), while the oxidation of 2-methyl-branched fatty acyl-CoAs is catalyzed
by acyl-CoA oxidase 2 (ACOX2), D-bifunctional protein (DBP), and sterol carrier protein-x
(SCPx) [187]. In addition to these enzymes, other proteins involved in peroxisomal β-oxidation
include ABCD1, which functions in the import of straight chain VLCFA into peroxisomes, and
2-methylacyl-CoA racemace (AMACR), which converts fatty acids with a methyl group in the
(R)-configuration to the (S)-configuration to become substrates for peroxisomal β-oxidation
[188].
Human diseases caused by a defect in one of these peroxisomal proteins are called single
peroxisomal enzyme deficiencies (PEDs). Patients with PED typically present with severe
neurological symptoms. The most common PED is called X-linked adrenoleukodystrophy (X-
ALD), which is caused by mutations in the ABCD1 gene, resulting in elevated levels of VLCFA,
and most frequently presents with severe cerebral ALD or adrenomyeloneuropathy (AMN)
[188]. Other PEDs include ACOX1 deficiency, caused by mutations in the ACOX1 gene and has
been reported in about 30 patients, DBP deficiency which is caused by mutations in the
HSD17B4 gene encoding DBP and has been reported in over 100 patients, and AMACR
deficiency caused by mutations in the AMACR gene and reported in about 10 patients [188].
75
Lastly, mutations in the SCP2 gene cause SCPx deficiency, which has previously been reported
in only two patients. The first patient reported with SCPx deficiency had a homozygous 1-
nucleotide insertion and presented with leukoencephalopathy with motor and peripheral
neuropathy, dystonia, hyposmia, nystagmus, and azoospermia [128, 129]. The second patient
with SCPx deficiency was reported to be a compound heterozygote with spinocerebellar ataxia
and brain iron accumulation [129].
The SCP2 gene contains two transcription start sites, resulting in the 58 kDa SCPx and the
15 kDa pro-sterol carrier 2 protein (SCP2) [86]. SCPx is post-translationally cleaved with its N-
terminal resulting in a 45 kDa thiolase and its C-terminal resulting in a 13 kDa mature SCP2
protein. Pro-SCP2 is also processed to form the 13 kDa SCP2 protein. SCPx and SCP2 have been
implicated to be involved in the transport and metabolism of various lipids including sterols [84,
189, 190], fatty acids [113, 191, 192], phospholipids [99, 190], and fatty acyl-CoAs [104, 193].
SCPx and SCP2 are most highly expressed in tissues involved in the oxidation and trafficking of
cholesterol: adrenals, ovary, testis, liver, and intestine [194]. While SCPx is almost exclusively
localized to peroxisomes, about 50% of SCP2 is peroxisomal and about 50% is extraperoxisomal
(mitochondria, endoplasmic reticulum, and cytosol) [194, 195]. The 45 kDa SCPx exhibits lipid-
transfer and sterol-carrier activities and is a 3-ketoacyl-CoA thiolase enzyme that has been shown
to be involved in the peroxisomal oxidation of branched chain fatty acids, straight chain fatty acids,
and the branched side chain of cholesterol [194].
Here, we report the third patient with SCPx deficiency and the first resulting from a
missense heterozygous mutation in SCP2. The patient is a 59-year-old male presenting with
brainstem neurodegeneration and azoospermia. Whole genome sequencing revealed a c.572A>G
heterozygous mutation resulting in a His191Arg substitution. Patient plasma fatty acid analysis
76
revealed abnormalities in the levels of several fatty acids including g- and a-linoleic acid and
arachidonic acid. Primary dermal fibroblasts from the patient and control normal human dermal
fibroblasts (NHDF) were used to investigate a link between the patient’s SCP2 mutation and its
clinical consequences. Western blot analysis revealed a significant reduction in the patient’s SCPx
levels compared to NHDF as well as alterations in the levels of many other proteins involved in
peroxisomal and mitochondrial β-oxidation. RNA sequencing identified several lipid related genes
that were differentially expressed between patient and normal fibroblasts. Lipidomic analysis
identified alterations in several free fatty acids, sterols, acylcarnitines, sphingolipids and
phospholipids involved in metabolic pathways such as steroid and primary bile acid biosynthesis,
linoleic metabolism, and fatty acid degradation. Finally, we identified two compounds that have
shown to increase SCPx levels and certain fatty acid levels in patient fibroblasts compared to
untreated cells. Collectively, our data suggest that the patient’s novel heterozygous SCP2 mutation
causes alterations in a significant number of genes and lipids that are involved in several critical
lipid metabolic pathways, which manifest as neurodegeneration and as the testicular defects seen
in the patient.
2. Materials and Methods
Genome Sequencing
Whole genome sequencing was performed using next generation sequencing (NGS) on
the Illumina NovaSeq 6000 platform. 98% percent of the exome was sequenced at greater or
equal to 10X depth coverage. Library preparation was performed using the Illumina TruSeq
PCR-free library preparation kit according to the manufacturer’s instructions. Seven hundred
micrograms of DNA were fragmented to 400 bp by acoustic shearing using a Covaris LE220
instrument (Covaris Inc., Woburn, MA). Library size range was assessed using a Bioanalyzer
77
(Agilent Technologies, Santa Clara, USA) and quantified with Kapa library quantification kit
(Roche). Genomic libraries were loaded on an Illumina patterned flowcell and followed by
cluster generation and 150-bases paired-end sequencing on Illumina HiSeq X platform to
generate 90-100 gbases of raw data per library. Sequencing of the DNA was performed on a
research basis at The Center for Applied Genomics (TCAG; Hospital for Sick Children, Toronto,
ON).
Data processing and analysis was performed using 1) the Franklin Genoox Platform
(genoox.com) and 2) local GATK best practices (GATK 3.7). 1) Reads were aligned using bwa
against hg19 reference. Duplicate reads are removed. Variant calling was performed using
GATK (version 4.1) and FreeBayes (version 1.1.0). 2). Base calling is performed using bcl2fastq
2.20 (for HiSeq 2500) or HiSeq Analysis Software (for HiSeq X). Reads are mapped to the b37
reference sequence using the bwa-mem algorithm and duplicate reads are marked using Picard
Tools. Local realignment and base quality score recalibration using GATK follows. Variants are
called using GATK HaplotypeCaller. For whole genome sequencing (WGS), variant quality
score recalibration (VQSR) is performed for filtering variants.
Variants were identified for further analysis using the following filtering parameters to
capture substitutions, small or large insertions, deletions, duplications or indels. Human Phenotype
Ontology (HPO) was used to query for genes related to the phenotypic presentation including
terms for neurological, neuromuscular, and musculoskeletal search terms.
Clinical Samples
This individual was enrolled in the Adults with Undiagnosed Rare Disease genome
sequencing research study approved by the Mount Sinai Hospital Research Ethics Board (#12-
0222-E) [196].
78
Cell Culture
Patient fibroblasts (WESP) were maintained in minimum essential medium α (MEM α)
(Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% fetal bovine serum.
Adult normal human dermal fibroblasts (NHDF) were acquired from PromoCell and grown in
Fibroblast Growth Medium 2 with SupplementMix (PromoCell GmbH, Heidelberg, Germany).
Fibroblasts were grown at 37°C and 5% CO2.
Cells were treated with 1, 10, and 25 µM fenofibrate (Sigma, St. Louis, MO, USA) or
0.1, 0.5 and 1 µM (Z)-4-hydroxytamoxifen (Sigma, St. Louis, MO, USA) for 24 hours prior to
cell pellet collection for downstream analyses. For combination treatments, concentrations of 25
µM fenofibrate and 0.5 µM (Z)-4-hydroxytamoxifen were used to treat the cells for 24 hours
prior to cell pellet collection.
Next Generation Sequencing
Amplicon-EZ sequencing was used to confirm the c.572A>G mutation in WESP cells
(GENEWIZ from Azenta Life Sciences, South Plainfield, NJ, USA). Basic Local Alignment
Search Tool (BLAST) was used to align the wild type human SCP2 gene with next generation
sequencing results for WESP samples (National Center for Biotechnology Information, Bethesda,
MD, USA).
Quantitative Real-Time Polymerase Chain Reaction
The Quick-RNA MiniPrep Plus kit (Zymo Research, Irvine, CA, USA) was used to extract
total RNA for all qRT-PCR data shown here. 500 ng of extracted RNA per sample was used for
reverse transcription using PrimeScript RT Master Mix (Takara Bio, Mountain View, CA, USA).
qRT-PCR was done using the Applied Biosystems PowerUP SYBR Green Master Mix (Thermo
Scientific, Waltham, MA, USA) and the qTOWER
3
(Analyik Jena AG, Jena Germany). Primer
79
information is detailed in the supplementary materials. Gene expression data were normalized to
GAPDH.
Immunoblot Analysis
Immunoblot analysis was carried out as previously described [10]. Briefly, total protein
was extracted from fibroblasts using RIPA buffer with Pierce Protease Inhibitor and protein
concentration was measured using the Pierce BCA Protein Assay kit (Thermo Scientific,
Waltham, MA, USA). 7.5 µg of protein per sample and a 4%–20% Tris–glycine gradient gel was
used for SDS-PAGE (Bio-Rad, Hercules, CA, USA). Protein was transferred to a polyvinylidene
fluoride membrane. 5% bovine serum albumin in PBST was used for all blocking steps. All
primary antibodies were incubated in blocking solution overnight at 4°C. Information about
antibodies is detailed in the supplementary materials. Secondary antibodies were incubated for 1
hour at room temperature. Restore Western Blot Stripping Buffer was used to strip the
membranes (Thermo Scientific, Waltham, MA, USA). Membranes were visualized by Radiance
Peroxide and Radiance Plus (Azure Biosystems, Dublin, CA, USA). Immunoreactive proteins
were imaged using the Azure c600 (Azure Biosystems).
Immunofluorescence
Fibroblasts were plated at a density of 200,000 cells per well on coverslips for 24 hours.
Cells were stained with 0.1 µM Mitotracker Red CMXRos for 30 min at 37°C, washed with PBS
3 times, and fixed with 4% PFA for 10 min at room temperature (Thermo Scientific, Waltham,
MA, USA). Cells were washed, incubated in .1% Triton X-100 for 10 minutes at room temperature
and washed. Samples were blocked with 5% donkey serum, washed, and incubated with primary
antibodies overnight at 4°C. After washing, secondary antibodies were added for 30 minutes at
80
room temperature and then washed again. Cells were stained with DAPI and observed using the
Zeiss LSM 880 confocal microscope. Quantification was done using ImageJ.
RNA sequencing
Three replicates per sample were sent for total cell RNA sequencing. RNA extraction and
preparation was done according to protocol using the Qiagen Allprep RNA/DNA isolation kit
(Qiagen, Valencia, CA, USA). For library preparation, the Clontech Takara SMARTer Total RNA
Stranded Pico V2 Library prep kit was used (Takara Bio, Mountain View, CA, USA). Libraries
were dual index 8bp and Illumina Adapters were used. The libraries were sequenced on the
Nextseq500 at 2x75 cycles. Partek Genomics Suite was used to analyze the gene expression data
and to identify differentially expressed genes between WESP and NHDF samples. Significantly
differentially expressed genes were identified using FDR < 0.05 and fold change < -2 or > 2 as
cutoffs.
Lipidomic Analyses
Five replicates of WESP cells and five replicates of NHDF cells with 4 million cells per
sample were used for lipidomic analysis, which was performed at the UCSD Lipidomics Core
[197]. For free fatty acid analysis, 500 µl MeOH was added to 50 µl of sample for extraction. This
was followed by the subsequent addition of 25 µl 1N HCl and 100 µl 0.1 ng internal standard mix.
The samples were vortexed, and iso-octane was added. The samples were vortexed again,
centrifuged, and the iso-octane layer was collected. The sample was re-extracted with iso-octane,
combined with the initial extract, and solvent was removed. Samples were derivatized with PFBB
and DIPEA. Solvent was removed and the samples were reconstituted in 50 µl iso-octane and
transferred to the MS vial with insert. GC-MS analysis was done using the Agilent 5975 GC/Mass
81
Spectrometer using an injection volume of 1 µl. Analysis was done using Mass Hunter and
Multiquant.
Acylcarnitines were extracted by adding 250 µl butanol/methanol (3:1) and 100 µl 1X
internal standard (1ng/µl) to 50 µl of sample. The samples were vortexed and 250 µl heptane/ethyl
acetate (3:1) and 250 µl acetic acid 1% in H2O were added. The samples were vortexed and
centrifuged and the upper layer was collected. Solvent was removed. The samples were
reconstituted in 50 µl NP Buffer A (59/40/1 IPA/HEX, H2O containing 10 mM NH4oAC. For LC-
MS analysis, the samples were reconstituted in 50 µl 90% MeOH and 0.1% formic acid. 10 µl per
sample was injected into the Waters Acquity UPLC System using a Phenomenex Kinetics C18
2.1x150mm 1.7um column. Analysis was done using Sciex 6500 Qtrap.
Sterols were extracted by adding 500 µl butanol/methanol (3:1) and 250 µl 1N NaOH to
100 µl of sample. Samples were sonicated for 10 minutes, vortexed, and incubated for 1.5 hours
at 37°C. 20 °C 1X internal standard, 500 µl heptane/ethyl acetate (3:1), and 250 µl 1N HCl were
added. The samples were vortexed and centrifuged. The upper layer was transferred, and the
solvent was removed in Speed Vac. LC-MS analysis was done similar to acylcarnitine analysis.
Phospholipids were extracted as described for acylcarnitines. Analysis was done by NP-
UPLC/MS. Chromatography was done using the Waters Acquity UPLC System with a
Phenomenex, Luna 3 µm Silica, 150x2 mm column and 10 µl injection volume. Analysis was done
as described for acylcarnitines. Ceramide and sphingomyelin analysis was derived from
phospholipid extraction and analysis.
Statistical Analysis
GraphPad Prism 9 was used for statistical analyses and graphical presentation. Statistical
analyses of data were performed using the student t test and means were considered statistically
82
different when p < 0.05. All experiments were conducted in triplicate unless otherwise stated and
all data are presented as mean ± standard error of the mean.
3. Results
Clinical characteristics of patient
The patient described here is a 59-year-old Greek Canadian male with a seven-year history
of progressive degenerative neurological disease. This included loss of cranial nerve function with
progressive inability to swallow, chew, speak, breathe, and control head movement. Craniospinal
imaging studies are negative, and the patient shows no cerebellar problems clinically or
radiographically. Taken together, these neurological phenotypes are consistent with brainstem,
rather than cerebellar, neurodegeneration. In addition to these neurological defects, the patient also
has history of cardiac dysrhythmia without associated risk factors, and primary hypogonadism
with azoospermia and reduced testosterone with elevated follicle-stimulating hormone levels.
Because of the known role of SCPx in fatty acid metabolism, the patient’s plasma fatty
acid levels were measured, and the results are displayed in Table 4. The patient’s fatty acid profile
shows that there were many abnormalities in the levels of medium-, long-, and very-long chain
fatty acids. Specifically, the patient’s levels of the medium-chain fatty acids octanoic acid (8:0)
and decenoic acid (10:0) and the long-chain fatty acids hexadecadienoic acid (16:2), g-linolenic
acid (18:3w6), a-linolenic acid (18:3w3), and arachidic acid (20:0) were below normal.
Contrastingly, the patient’s levels of the long chain fatty acids arachidonic acid (20:4w6) and mead
acid (20:3w9) and the VLCFA DHA (22:6w3) and nervonic acid (24:1w9) were above normal
(Table 4). These abnormalities in the levels of medium-, long-, and very-long chain fatty acids are
83
consistent with the role of SCPx in fatty acid β-oxidation and suggest that both mitochondrial and
peroxisomal fatty acid β-oxidation are affected by the patient’s SCP2 mutation.
Identification of a novel heterozygous SCP2 variant
Table 4. Patient plasma fatty acid profile
84
Whole genome sequencing identified a novel c.572A>G heterozygous variant resulting in
a p.His191Arg substitution. This p.His191Arg was not identified in the literature, nor was it
identified in ClinVar or LOVD 3.0 databases. The variant was identified in dbSNP (rs372168791).
The variant was identified in control databases in 11 of 282548 chromosomes at a frequency of
0.00003893 (Genome Aggregation Database March 6, 2019, v2.1.1). The variant was observed in
the following populations: Other in 2 of 7204 chromosomes (freq: 0.000278), European (non-
Finnish) in 9 of 129022 chromosomes (freq: 0.00007), but was not observed in the African, Latino,
Ashkenazi Jewish, East Asian, European (Finnish), or South Asian populations. The p.His191
residue is conserved in mammals and other organisms, and 8 of 8 computational analyses (SIFT,
FAHTMM, DANN, MT, MetaLR, Revel, PolyPhen, MutationTaster) suggest that the variant may
impact the protein: however, this information is not predictive enough to assume pathogenicity.
The variant occurs outside of the splicing consensus sequence and in silico or computational
prediction software programs (SpliceSiteFinder, MaxEntScan, NNSPLICE, GeneSplicer) do not
predict a difference in splicing. In summary, the clinical significance of this variant cannot be
determined with certainty at this time. This variant is classified as a variant of uncertain
significance.
Characterization of SCPx and SCP2 in NHDF and WESP cells
Primary fibroblasts (WESP) were generated from the patient and the heterozygous
c.572A>G variant was confirmed by next-generation sequencing (NGS)-based amplicon
sequencing. NGS and Basic Local Alignment Search Tool (BLAST) results revealed that about
50% of the reads contained the wild type (WT) nucleotide while about 50% of the reads had the
c.572A>G mutation, confirming the heterozygous variant in SCP2 identified via whole genome
sequencing (Figure 15A). Normal human dermal fibroblasts (NHDF) were used as control to
85
compare levels of SCPx in WESP cells. Quantitative real-time polymerase chain reaction (qRT-
PCR) was used to measure expression of the SCPx-coding region of the SCP2 gene and showed
that there was about 50% lower mRNA expression of this region of SCP2 in WESP cells compared
to, consistent with a heterozygous mutation (Figure 15B). There was no difference in the
expression of the SCP2-coding region of the SCP2 gene between the two samples since this region
has its own promoter (Figure 15B). Protein levels of SCPx were examined by immunoblot which
showed that there was a significant reduction in the amount of 58 kDa and 45 kDa SCPx in WESP
cells compared to NHDF (Figure 15C). Consistent with qRT-PCR results, there was no difference
in levels of 13 kDa SCP2 between the two samples (Figure 15C). These data confirm the presence
of the heterozygous variant in the SCPx-coding region of the SCP2 gene and show that it leads to
a significant decrease in mRNA expression and protein levels.
SCPx deficiency alters peroxisome number
Next, we were interested in identifying potential changes in the abundance of peroxisomes.
Alterations in peroxisome abundance have been reported in fibroblasts of patients with other
peroxisomal disorders such as Zellweger’s disease [198]. To determine any changes in the number
of peroxisomes in the patient WESP cells, we used immunocytochemistry using the peroxisomal
membrane protein 70 (PMP70), which is a commonly used peroxisomal marker. We found that
patient WESP cells had significantly less peroxisomes than control cells (Figure 16A-16C). As a
control, we stained for mitochondria using Mitotracker and found that there was no difference in
the number of mitochondria between the two cell types (Figure 16A-16C). We also stained for
SCPx, which had significantly decreased levels in patient cells compared to control, consistent
with the data in Figure 1 (Figure 16A-16C). These data show that a decrease in SCPx levels leads
86
to a decrease in the number of peroxisomes, which is consistent with what is seen in other
peroxisomal disorders associated with neurodegenerative phenotypes.
87
Alterations in levels of other peroxisomal and mitochondrial β-oxidation enzymes
We next wanted to identify any potential changes in other peroxisomal β-oxidation
enzymes along the straight-chain fatty acyl-CoAs pathway. Figure 17A demonstrates an increase
in ACOX1 but decrease in ACAA1 in patient WESP cells. In contrast oxidation of 2-methyl-
branched fatty acyl-CoAs was not affected as demonstrated by lack of difference in DBP levels
(Figure 17A). These data show that the patient’s SCPx variant results in alterations in the levels of
other peroxisomal β-oxidation enzymes, which may affect fatty acids beyond those directly
metabolized by SCPx.
As mitochondria are the other main site of fatty acid β-oxidation, we were also interested
in measuring the levels of mitochondrial β-oxidation enzymes. Carnitine palmitoyltransferase 1
(CPT1A) is a key enzyme in the transport of long-chain fatty acids across the inner mitochondrial
membrane. In particular, the long chain-specific acyl-coA dehydrogenase (LCAD) catalyzes the
first step of mitochondrial fatty acid β-oxidation [199]. Immunoblot data showed that both CPT1A
and LCAD levels were increased in patient fibroblasts compared to control (Figure 17B). These
data further show that the patient’s SCPx variant are accompanied by alterations in mitochondrial
fatty acid β-oxidation in addition to its effects on peroxisomal β-oxidation.
Figure 15. Characterization of SCP2 mutation in patient fibroblasts. (A) BLAST DNA
sequence of amplicon-based NGS results of SCP2 in patient fibroblasts (WESP) compared
to the human wild type (WT) sequence. Red box shows A>G mutation present in half of the
reads (top) and WT nucleotide in half the reads (bottom). (B) qRT-PCR analyses of the
SCPx coding region of SCP2 (left) and the SCP2 coding region of SCP2 (right) where data
is normalized to GAPDH. Data are shown as mean ± SEM (n=3). ***p < 0.001. (C) Western
blot analyses of SCPx and SCP2 in control fibroblasts (NHDF) and WESP cells.
88
Figure 16. Quantification of peroxisome abundance in NHDF and WESP cells. (A)
Confocal microscopy image of a NHDF stained with Mitotracker, PMP70 (peroxisomes),
SCPx, and DAPI. Scale bar: 20 µm. (B) Confocal microscopy images of a WESP
fibroblast stained with Mitotracker, PMP70 (peroxisomes), SCPx, and DAPI. Scale bar:
20 µm. (C) Quantification of confocal images. The area of staining with each marker was
quantified in 10 images per sample using ImageJ. Data are shown as mean ± SEM (n=10).
**p < 0.01; ***p < 0.001.
89
To assess other putative pathways influenced by SCPx, we next examined levels of
peroxisome proliferator-activated receptor α (PPARα), a nuclear receptor that regulates the
expression of genes involved in fatty acid β-oxidation [200]. Immunoblotting data showed that
PPARα was increased in patient fibroblasts compared to control cells, consistent with the increased
expression of PPARα-responsive genes in Scp2 gene-ablated mice (Figure 17C) [201].
Additionally, given the recognized role of SCPx in bile acid synthesis we next measured levels of
cholesterol 7α-hydroxylase (CYP7A1), a rate-limiting step in this process [202]. Interestingly, we
found that levels of CYP7A1 were increased in patient fibroblasts compared to control cells, which
is in line with a previous report citing CYP7A1 repression in SCP2 overexpressing human
hepatocytes (Figure 17C) [203].
RNA-sequencing identified differentially expressed genes and altered pathways related to lipid
trafficking and metabolism
The observed effects on fatty acid β-oxidation and on bile acid synthesis prompted us to
seek a broader view of alterations in the patient’s transcriptome. To this end, we sought to identify
differential expression of genes governing lipid transport and metabolism. RNA sequencing data
were confirmed by RT-qPCR of several lipid-related genes (Figure 18). We uploaded our DEG
list generated through Partek Genomic Flow into Ingenuity Pathway Analysis (IPA). Overlaying
the DEG list with the “cholesterol” interaction network identified significantly decreased
expression of: FABP4, ANGPTL4, CYP27A1, ABCA1, LIPG, NPC2, GPAT3, OSBP2, PLA2G4A,
PLIN2, PSEN2, and KCNJ. In contrast, CYP27A1 and INHBA were significantly increased in
WESP cells compared to NHDF (Figure 18A). We pursued a similar strategy in addressing the
“lipid” interaction network. This strategy yielded significantly reduced expression of: AR, ESR1,
NR0B1, and SYT1 in WESP cells compared to NHDF (Figure 18B). This process was also done
90
with other interaction networks including that of phospholipids, SCP2, testosterone, and
steroidogenic acute regulatory protein (STAR) which is known to play a critical role in cholesterol
transport for steroidogenesis. These yielded the identification of many genes that were also found
through our cholesterol and lipid interaction searches, in addition to PPARG for the phospholipid
interaction network, CACNA1A for the SCP2 interaction network, MAOA for the testosterone
91
interaction network, and PTERG2 for the STAR interaction network, which each decreased in
WESP cells compared to NHDF (Figure 18C).
To identify pathways that may be altered due to changes in the patient’s transcriptome, we
uploaded the DEG list that were confirmed by RT-qPCR into MetaboAnalyst 5.0. This online
platform mapped the DEG list into affected pathways (Table 5). The top affected pathway
identified by MetaboAnalyst 5.0 was cholesterol metabolism, which aligns with known functions
of SCPx in metabolism of cholesterol for bile acid synthesis and in cholesterol transport. The genes
that are part of the cholesterol metabolism pathway which were confirmed to be differentially
expressed between NHDF and WESP cells are: ABCA1, LIPG, ANGPTL4, NPC2, and CYP27A1.
Another pathway that may be affected due to alterations in the expression of several genes is PPAR
signaling, which is an important regulator of many genes critical for peroxisomal β-oxidation.
DEG in this pathway included: FABP4, CYP27A1, ANGPTL4, PLIN2, and PPARG. Additionally,
another pathway with many genes shown to be differentially expressed is that of serotonergic
synapse signaling. DEG in this pathway include: CACNA1A, PLA2G4A, and MAOA.
Alterations in free fatty acid levels and related pathways
The identification of many altered fatty acids in the patient’s plasma fatty acid profile and
of many lipid-related pathways from our RNA sequencing analysis prompted us to investigate
other potential changes in the patient’s lipidome. We conducted lipidomic analysis using NHDF
Figure 17. Western blot analyses of β-oxidation Enzymes. (A) Western blot analyses of
protein levels of ACOX1, ACAA1 (thiolase), HSD17B4 (DBP), and SCPx mapped according
to each protein’s role in the peroxisomal β-oxidation pathway. Left side of the pathway is the
classical pathway and right shows the alternative pathway. (B) Western blot analyses of
protein levels of CPT1A and LCAD which are important mitochondrial β-oxidation enzymes.
(C) Western blot analyses of protein levels of PPARα and CYP7A1.
92
93
and WESP cells for different classes of lipids including free fatty acids, acylcarnitines, sterols,
phospholipids, and sphingolipids. Significant differences in the levels of many free fatty acid
species identified through this lipidomic analysis was consistent with the abnormalities in several
fatty acid species derived from the patient’s plasma. For example, arachidonic acid (20:4), mead
acid (20:3w9), and DHA (22:6) were out of normal range or altered in both the patient’s plasma
and in his fibroblasts, respectively (Table 4 and Figure 19A). However, there were also several
fatty acids that were significantly altered in WESP cells that were not detected in the patient’s
plasma fatty acid profile. These included: lauric acid (12:0), eicosatrienoic acid (20:3), EPA (20:5),
DTA (22:4), and DPA (22:5) (Table 4 and Figure 19A). Additionally, we noted that nervonic acid
Figure 18. Confirmation of differentially expressed genes identified by RNA sequencing. (A) qRT-
PCR analyses of differentially expressed genes part of the “cholesterol” interaction network. (B)
qRT-PCR analyses of differentially expressed genes part of the “lipid” interaction network. (C). qRT-
PCR analyses of differentially expressed genes part of other interaction networks. Data are shown as
mean ± SEM (n=3). *p < 0.05; **p < 0.01; ***p < 0.001.
Table 5. Summary of RNA sequencing-derived pathways
94
(24:1) was higher than normal in the patient’s plasma profile, but significantly lower than that seen
in NHDF in our fibroblast lipidomic analysis. The discrepancies between the patient’s plasma fatty
acid profile and the cellular lipidomic data may be attributed to differing extra- and intra-cellular
levels of these lipids, respectively. Nevertheless, the significant differences in many species of free
fatty acid levels between NHDF and WESP cells further point to the pivotal role of SCPx in fatty
acid metabolism.
To further identify potential pathways that could be affected as a result of perturbed free
fatty acids, we used MetaboAnalyst 5.0 to map the lipidomic data into pathways (Table 6). One of
the pathways that involves several free fatty acids which were significantly altered between NHDF
and WESP cells is the biosynthesis of unsaturated fatty acids. This pathway includes: EPA (20:5),
DPA (22:5), DHA (22:6), eicosatrienoic acid (20:3 N9), arachidonic acid (20:4), adrenic acid
(22:4), and icosenoic acid (20:1). Additionally, the linoleic acid metabolism pathway was
identified to be altered with changes in the levels of arachidonic acid (20:4) and eicosatrienoic acid
(20:3 N9). In addition to these lipids, PLA2G4A, which is part of our DEG list, is also part of this
pathway. These data further confirm that the metabolism of several fatty acids is affected by SCPx.
Alterations in acylcarnitine levels and related pathways
Since SCPx is known to play a critical role in fatty acid β oxidation, and because we noted
differences in levels of mitochondrial fatty acid β oxidation enzymes between NHDF and WESP
cells, we next turned our attention to acylcarnitine species. We found that the levels of several
acylcarnitine species were significantly increased in WESP cells compared to NHDF (Figure 19B).
Consistent with this observation, MetaboAnalyst 5.0 identified that CAR 16:0 is part of the fatty
acid degradation pathway (Table 6). These data further show that, in addition to its role in
peroxisomal β oxidation, SCPx is also important for mitochondrial fatty acid β oxidation.
95
Table 6. Summary of lipidomics-derived pathways
96
Figure 19. Lipidomic analyses of various lipid groups between NHDF and WESP cells. (A) Fatty
acid profile. (B) Acylcarnitine profile. (C) Sterol profile. (D) Phospholipid profile. (E)
Sphingolipid profile: Ceramides (left) and Sphingomyelins (right). Data are shown as mean ±
SEM (n=5). *p < 0.05; **p < 0.01; ***p < 0.001.
97
Alterations in sterol levels and related pathways
As a known sterol transfer protein and important enzyme in cholesterol metabolism, we
were also interested in identifying potential changes in sterol levels between NHDF and WESP
cells. Lipidomic analysis identified that the levels of 14-demthyl-lanosterol, 24-
hydroxycholesterol, 25-hydroxycholesterol, 7α-hydroxycholesterol, 7-ketocholesterol, 8-
dehydrocholesterol, 24,25-dehidrolanosterol, 7α-hydroxy-4-cholesten-3-one, 4,4-dimethyl-
cholest-8(9)-en3β-ol, zymostenol, and zymosterol were all significantly lower, while 25-
hydroxycholesterol was significantly higher in WESP cells compared to NHDF (Figure 19C).
These alterations in sterol levels are consistent with the critical role of SCPx in sterol transport and
metabolism.
We also identified pathways that may be altered due to the changes in sterol levels in the
patient fibroblasts (Table 6). Interestingly, one of the affected pathways was steroid biosynthesis,
which included 24,25-dehidrolanosterol, lanosterol, 14-demthyl-lanosterol, zymostenol, and
zymosterol. The alteration in the levels of these sterols, that are part of the steroid biosynthetic
pathway, may provide a biochemical explanation behind the hypogonadism seen in this patient.
Another pathway that shows to be significantly altered by the changes in sterol levels is primary
bile acid biosynthesis. This pathway involves 25-hydroxycholesterol, 7α-hydroxycholesterol, and
7α-hydroxy-4-cholesten-3-one. CYP27A1, which is part of our DEG list is also part of the primary
bile acid biosynthesis pathway. These data are consistent with the role of SCPx in cholesterol
metabolism for bile acid synthesis and suggest that this process may be disrupted due to the face
of SCPx reduction.
Alterations in phospholipid levels and related pathways
98
As indicated earlier, SCPx has also been previously suggested to play a role in the transport
and/or metabolism of phospholipids [99, 190]. Our lipidomic analysis found that the amount of
phosphatidylglycerol (PG) and phosphatidylserine (PS) were significantly higher in WESP cells
compared to NHDF (Figure 19D). These phospholipids are part of the glycerophospholipid
metabolism pathway, in addition to PLA2G4A and GPAT4 from our DEG list (Table 6). These
data are aligned with the putative role of SCPx in phospholipid metabolism.
Alterations in sphingolipid levels
We were also interested in identifying alterations in levels of sphingolipids since certain
sphingolipids are highly enriched in the central nervous system and play critical roles in neuronal
growth and differentiation. Additionally, abnormal levels of sphingolipids such as ceramides are
often found in patients with neurodegeneration [204]. We found that many ceramide species had
varying levels between control NHDF fibroblasts and patient WESP fibroblasts including CER
d18:0/24:1, CER d18:1/22:5, CER d18:1/24:0, CER d18:1/24:1, CER d18:1/26:0, and CER
d18/26:1 (Figure 19E). Each of these ceramide species had significantly lower levels in patient
fibroblasts compared to control. Because ceramide levels are determined in part by the
ceramide/sphingomyelin cycle in which ceramide can be both produced from sphingomyelin
metabolized to sphingomyelin, we also measured sphingomyelin levels in both cell types [205].
Interestingly, while there were many species of ceramides that were decreased in WESP cells
compared to NHDF, there were only two sphingomyelin species that had significantly different
levels: SM 32:2 and SM 44:2 (Figure 19E). This suggests that, although the
ceramide/sphingomyelin cycle may be functioning normally, there may be alterations in other
pathways of ceramide production, such as from de novo synthesis or from the ceramide salvage
pathway [205]. The varying levels in sphingolipids between control and patient cells also suggest
99
that the altered ceramide levels may play a role in the patient’s neurodegenerative phenotype.
Collectively, our lipidomic data show that SCPx deficiency is accompanied by reorganization of
the patient’s lipidome, which may explain many of the patient’s clinical presentations.
Identification of compounds that increase SCPx expression
Because the patient’s SCP2 mutation results in decreased SCPx expression and leads to
alterations in lipid metabolism, we screened for compounds that could restore SCPx levels.
Prompted by our differential expression profiling pointing to possible PPAR involvement, we
treated patient and control fibroblasts with fenofibrate, a widely adopted PPARα agonist for
treatment of dyslipidemia [206]. Unlike in normal fibroblasts which were relatively unaffected,
patient WESP cells showed a dose-dependent increase in SCPx levels following fenofibrate
treatment (Figure 20A). As a previous report also showed that the selective estrogen receptor
modulator 4-hydroxytamoxifen (4-OHT), also known as afimoxifene, can increase SCPx
expression in DL23 cells [207] we also tested this compound. We found that 4-OHT treatment
significantly and selectively increased SCPx levels compared to untreated patient cells (Figure
20B). These data suggest that 4OHT may also be a viable strategy for increasing SCPx levels in
conditions characterized by SCPx deficiency.
Since fenofibrate and 4-OHT have separate mechanisms of action, we hypothesized that a
combination treatment of the two may have synergistic effects in increasing SCPx expression.
However, treating patient fibroblasts with fenofibrate and 4-OHT resulted in no additional
increases from those observed with either compound alone (Figure 20C). This shows that, while
these drugs can independently increase SCPx levels, they do not work synergistically on this target.
Fenofibrate and 4-OHT improve the patient’s fatty acid profile
100
Finally, we wanted to investigate whether increasing SCPx expression through fenofibrate
or 4-OHT is sufficient to restore the patient’s free fatty acid levels. We analyzed free fatty acid
levels in NHDF and WESP cells treated with fenofibrate alone, 4-OHT alone, or fenofibrate and
4-OHT in combination for 24 hours. We confirmed that many fatty acid species in the patient’s
cells were increased following treatment with fenofibrate including 16:0, 17:0, 18:0, 18:2, 18:3
N3, 18:3 N6, 18:4, 20:5, and 22:4 (Figure 21). Further, total free fatty acids in patient cells were
also significantly increased by fenofibrate treatment. Similarly, 4-OHT treatment demonstrate a
positive effect on fatty acid levels including: 16:0, 18:2, 18:3 N6, and 22:6 (Figure 21). However,
in general, 4-OHT was not as effective in increasing fatty acid levels as with fenofibrate.
101
Additionally, the combination of drugs did not exhibit a synergistic effect on fatty acid levels.
Collectively, these data show that treatment with fenofibrate, or 4-OHT alone, can improve the
fatty acid profile in patient cells and provide evidence that these compounds maybe beneficial in
the treatment of SCPx deficiency.
4. Discussion
Here, we report the third patient with SCPx deficiency and the first resulting from a
heterozygous variant in SCP2. Both previously reported patients with SCPx deficiency were
similar to each other, displaying cerebellar presentations and hypointensity in the pons and
thalamus [128, 129]. Previously reported patients had mutations on both alleles of the SCP2 gene
resulting in premature stop codons and a total absence of SCPx as visualized by western blotting.
Unlike these previous cases, the patient described herein lacks a cerebellar phenotype and all
imaging studies were unremarkable. Instead, our patient exhibited striking brainstem
neurodegeneration with testicular spermatogenesis defect. Although the patient has a distinct
neurological phenotype, our data show that his heterozygous mutation in SCP2 led to a decrease,
rather than absence, of SCPx. Additionally, although previous cases of severe SCPx deficiency are
associated with accumulation of the branched chain fatty acids pristanic and phytanic acid our
patient with reduced SCPx does not exhibit this feature [91, 104, 127-129, 208, 209]. This suggests
Figure 20. SCPx protein levels in WESP cells in response to fenofibrate and 4-hydroxytamoxifen
treatment. (A) Western blot analyses of samples treated with various concentrations of fenofibrate for 24
hours (left) and quantification of the blots (right). (B) Western blot analyses of samples treated with
various concentrations of 4-hydroxytamoxifen (4-OHT) for 24 hours (left) and quantification of the blots
(right). (C). Western blot analyses of samples treated with 25 µM fenofibrate, 0.5 µM 4-OHT, and the
fenofibrate + 4-OHT combination for 24 hours (left) and the quantification of the blots (right). All western
blots were conducted in triplicate. Quantification was done using ImageJ analyses after normalizing to
the untreated band for each group followed by normalizing to GAPDH. Data are shown as mean ± SEM
(n=3). *p < 0.05; **p < 0.01.
102
103
104
105
that reduced SCPx that is produced, albeit at lower levels, maintains some activity, which may
explain some differences between our patient and the previously reported cases.
We found that, in addition to SCPx, other fatty acid β oxidation proteins had altered levels
between control and patient fibroblasts. Consistent with this, previous work in SCPx-null mice
have shown an increase in β-oxidation enzymes in the absence of SCPx [127, 187]. Additionally,
PPARα is known to regulate many genes in the fatty acid oxidation pathway [127, 210]. The
increase in PPARα levels in WESP cells shown here may be responsible for the observed increases
in ACOX1, CPT1A, and LCAD. The latter being recognized targets of PPARα action [211-215].
However, we also noted a decrease in ACAA1 which is consistent with its negative regulation by
PPARα [216]. Because ACAA1 is responsible for catalyzing the same step of the β-oxidation
pathway as SCPx (but with different substrates), the decrease in ACAA1 levels may further explain
the severity in the phenotype caused by SCPx deficiency since both peroxisomal β-oxidation
pathways are affected [91]. In addition to PPARα and the β-oxidation enzymes, we also found that
the levels of CYP7A1 is increased in WESP cells compared to NHDF, which is consistent with
what was seen in Scp2 deficient mice [217].
Our RNA sequencing and lipidomic data identified several metabolic pathways that have
genes and/or lipids that are altered between NHDF and WESP fibroblasts. Some of these pathways
include fatty acid degradation, steroid biosynthesis, and serotonergic signaling which are relevant
to the patient’s neurological and testicular phenotypes. One of the key differentially expressed
genes that was shown to play a role in many altered pathways is CYP27A1, which our pathway
analyses found to function in cholesterol metabolism, primary bile acid biosynthesis, and PPAR
Figure 21. Fatty acid profile of NHDF and WESP cells following treatment. Measurement of various
species of free fatty acids after treatment with 25 µM fenofibrate, 0.5 µM 4-hydroxytamoxifen, or a
combination of the two compounds for 24 hours. Data are shown as mean ± SEM (n=3). *p < 0.05; **p
< 0.01; ***p < 0.001.
106
signaling. Indeed, CYP27A1 catalyzes the hydroxylation of cholesterol at C27, regulates
cholesterol biosynthesis, is involved in reverse cholesterol transport, catalyzes steps in both the
classical and alternative pathways of bile acid synthesis, and is regulated by PPAR activators [218-
220]. Additionally, CYP27A1 was shown to be increased in SCPx-null male mice, consistent with
the upregulation we found in our study [221]. Evidently, our data show that the identified genes
affected by the patient’s SCP2 mutation align with the known function of SCPx in many critical
pathways. Furthermore, our data show that the patient’s SCP2 mutation itself leads to a variety of
alterations in genes, lipids, and pathways. While some of these perturbations are directly related
to reduced SCPx, several others are likely the consequence of additional downstream effects.
Taken together, our gene expression, lipidomic, and pathway analyses provide further insight into
the biochemical basis of the patient’s striking clinical presentation.
To begin to translate our findings to the bedside we sought to identify compounds that
could restore SCPx levels. Previous reports have shown that SCP2 is a PPARα responsive gene
and that some PPARα ligands can increase SCP2 expression [222-224]. Additionally, based on
our findings that the patient’s SCP2 mutation leads to a decrease in peroxisome abundance and to
alterations in PPAR signaling, we hypothesized that treatment of patient fibroblasts with a PPARα
agonist may be a viable intervention. Indeed, treatment with fenofibrate increased SCPx levels in
patient WESP fibroblasts. Moreover, we identified that 4-OHT, a selective estrogen receptor
modulator, also increased SCPx expression. While unclear, our patient’s cell responsiveness may
be afforded by their endogenously downregulated ESR1 gene expression. Lastly, we identified that
treatment with fenofibrate or 4-OHT improves the fatty acid profile of patient fibroblasts, further
pointing towards the use of these compounds as potential treatments for SCPx deficiency.
107
In conclusion, our work describes the first patient with a heterozygous SCP2 mutation
resulting in reduced protein expression. The patient presented with azoospermia, cardiac
dysrhythmia, muscle wasting, and brainstem neurodegeneration. The decreased SCPx levels led
to widespread gene alterations resulting in dysregulated lipid synthesis and metabolism. Our
identification that fenofibrate and 4-OHT can restore SCPx protein levels and related functions
should prompt future studies into their role for neurodegenerative peroxisomal disorders.
108
Chapter V: Summary and Discussion
Summary
Cholesterol is a hydrophobic lipid molecule that functions in a variety of key cellular
processes spanning several cellular compartments. Cholesterol is synthesized in the endoplasmic
reticulum, though this organelle stores less than 1% of total cellular cholesterol. Additionally,
steroid hormone biosynthesis is initiated in mitochondria, yet mitochondria also store very little
cholesterol for steroid production. Thus, there must be mechanisms of rapid cholesterol transport
to move cholesterol from its site of synthesis to other cellular compartments where it functions.
The intracellular distribution and trafficking of cholesterol involves several mechanisms of
cholesterol transport that are tightly regulated, including binding to cholesterol transport proteins.
This dissertation expands on the currently available data about the cholesterol transport proteins,
STAR and SCP2/SCPx, and investigates their functions in lipid metabolism in general.
STAR
STAR is classically known to function when hormonal stimulation causes a rapid
increase in the protein levels of hormone-induced STAR, which facilitates the transport of
cholesterol to the mitochondria for acute steroidogenesis. However, while the role of hormone-
induced STAR in cholesterol transport for steroidogenesis has been heavily studied, no studies
were previously done to investigate the function of constitutive STAR. Our data showed that in
mouse and rat primary Leydig cells and in MA-10 mouse tumor Leydig cells, STAR is present
constitutively under basal conditions, and hormonal stimulation leads to a rapid increase in
STAR levels. We further showed that low levels of constitutive STAR are consistent with low
levels of steroidogenesis in all models, which are all increased after hormonal stimulation. Since
our data showed that STAR is expressed constitutively in all models, albeit at low levels, we
109
hypothesized that constitutively expressed STAR may have a distinct function. To address this,
we developed STAR KO MA-10 mouse tumor Leydig cells (STARKO1 and STARKO2) using
CRISPR/Cas9. STARKO1 and STARKO2 cells are completely absent of STAR protein and do
not produce steroids in response to hCG or dbcAMP but do respond to 22(R)-
hydroxycholesterol. STARKO1 cells do not respond to treatment with TSPO ligands, suggesting
that constitutive STAR plays a role in TSPO-mediated steroidogenesis.
Because accumulation of lipid droplets is one of the phenotypes of patients with LCAH,
we were interested in characterizing the lipid droplets of STARKO1 cells. Contrary to what is
seen in humans with STAR mutations, STARKO1 cells do not accumulate lipid droplets.
However, we found that the lipid droplet content of STARKO1 cells were drastically different
than those seen in WT MA-10 cells. There were significant increases in the amount of CE and
PC in STARKO1 cell lipid droplets, however, the most dramatic increase was in the amount of
DAG. Since the absence of constitutive STAR led to an accumulation of DAG, we investigated
the role of DAG signaling in progesterone production. We found that the DAG analog, OAG, in
combination with hCG, dose-dependently inhibited progesterone production in WT MA-10 cells,
suggesting that DAG accumulation is inhibitory to steroidogenesis. We also found that inhibition
of PLC, which is upstream of DAG production, increases steroid production in STARKO1 cells,
but that treatment with a PKC inhibitor had no effect of steroid production by STARKO1 cells.
These data suggest that accumulation of DAG may contribute to the inhibition of steroidogenesis
in STARKO1 cells, which may be associated with PLC activation of DAG but is independent of
PKC signaling. Taken together, these data show that constitutive STAR may have a function
independent of cholesterol transport for steroidogenesis, specifically that STAR may play a role
110
in the metabolism and/or signaling of other lipids including DAG since lack of constitutive
STAR led to alterations in the levels of several lipids in STARKO1 cell lipid droplets.
In Chapter III of this dissertation, we looked further into the role of constitutive STAR. In
addition to containing a significant number of lipid droplets, steroidogenic cells also contain
numerous mitochondria. RNA sequencing identified many differentially expressed genes
between WT MA-10 and STARKO1 cells that were related to mitochondrial structure and
function. We also found that mitochondria in STARKO1 cells had altered ultrastructure
compared to WT mitochondria, with ruptures in the membrane and reorganization of cristae
structure. Further, Seahorse XF analyses showed that certain parameters of mitochondrial
respiration including maximal respiration, spare respiratory capacity, and proton leak were
significantly lower in STARKO1 cells compared to WT cells, suggesting alterations in
mitochondrial function.
Next, we developed STAR-overexpressing STARKO1 cells (STARKO-OE1 and
STARKO-OE2), which each had STAR successfully re-expressed, to determine if reintroducing
STAR into STARKO1 cells, could recover mitochondrial structure and/or function. STAR re-
expression in STARKO1 cells recovered steroidogenic activity, albeit to different levels in
STARKO-OE1 and STARKO-OE2. While STAR expression and steroidogenic function were
recovered in STARKO1 cells by reintroducing STAR, we found that mitochondrial structure and
function were not. Although mitochondria in STARKO-OE1 cells had less ruptures in the
mitochondrial membrane compared to STARKO1 cells, the cristae structure was still less
defined than those seen in WT MA-10 cells. Interestingly, mitochondrial membrane and matrix
structure were even more damaged in STARKO-OE2 cells compared to STARKO1 cells.
111
Mitochondrial function was also not recovered in STARKO-OE1 and STARKO-OE2
cells. STARKO-OE1 and STARKO-OE2 cells had significantly lower basal respiration, ATP
production, maximal respiration, spare respiratory capacity, and proton leak compared to WT
MA-10 cells showing that STAR overexpression in STARKO1 cells exacerbated defects in
mitochondrial respiration. When measuring ATP production, we showed that glycoATP and
mitoATP were significantly lower in STARKO-OE1 and STARKO-OE2 cells compared to WT
MA-10 cells. We also found that the rate of glycolysis in STARKO-OE1 and STARKO-OE2
cells were significantly lower compared to WT MA-10 cells. Together, these data indicate that
STAR re-expression in STARKO1 cells exacerbates, rather than recovers, defects in
mitochondrial function and glycolysis.
Our data also showed that hCG stimulation of WT MA-10 cells resulted in a reduction in
STAR at the OMM likely due to its rapid processing at the OMM and subsequent translocation
to the matrix following hormonal stimulation. However, there was no difference in STAR levels
at the OMM in STAR overexpressing STAR KO cells, suggesting that there may be a defect in
the processing of STAR in these samples. To further investigate this, we overexpressed STAR in
MA-10 cells (MA-10+STAR) and found that all STAR-overexpressing cell lines had 37 kDa
STAR, but that only WT MA-10 cells and MA-10+STAR cells had 30 kDa STAR. These data
suggest that when mitochondria are intact and functional, as in WT MA-10 and MA-10+STAR
cells, STAR is correctly processed. However, in the presence of damaged and dysfunctional
mitochondria, STAR processing is defective. Collectively, these data indicate that constitutive
STAR plays a role in mitochondrial structure and function and that reintroduction of STAR into
STAR KO cells is insufficient to recover defects in mitochondria. These data also show that
impairment of mitochondrial structure and function leads to defects in STAR processing.
112
SCP2/SCPx
SCP2 and SCPx are encoded by the single SCP2 gene, and as such, most early studies
investigating the function of these proteins are non-specific to SCP2 or SCPx. SCP2/SCPx were
initially suggested to bind and transport cholesterol. Subsequent studies indicated that these
proteins may be non-specific for their substrates, with many cell-free assays suggesting that
SCP2/SCPx may bind a variety of lipids, including fatty acids, phospholipids, and sphingolipids.
Additionally, SCPx was found to be a key enzyme in peroxisomal β-oxidation. Using a human
model of SCPx deficiency, we sought to further delineate the roles of these proteins in the
transport and/or metabolism of certain lipids.
In Chapter IV of this dissertation, we describe a patient with a heterozygous mutation in
SCP2 who presented with progressive brainstem neurodegeneration and testicular defects. We
found that the patient’s SCP2 mutation led to a reduction in SCPx levels in patient fibroblasts
compared to normal human fibroblasts. We showed that patient fibroblasts had significantly less
peroxisomes than normal fibroblasts. In addition to SCPx, we also found that the levels of other
peroxisomal and mitochondrial β-oxidation enzymes were altered between patient and normal
fibroblasts, indicating that both peroxisomal and mitochondrial β-oxidation are affected by the patient’s
SCP2 mutation. RNA sequencing identified several genes related to cholesterol and other lipids that were
differentially expressed between patient and normal fibroblasts. Using the RNA sequencing data, we
identified several pathways that could be affected in the patient and may explain some of the patient’s
clinical presentations including cholesterol metabolism, PPAR signaling, and serotonergic signaling.
Since SCPx has been indicated to play a role in the transport and/or metabolism of many lipids
and because SCPx is critical in peroxisomal β-oxidation, we were interested in characterizing the
lipid profile of the patient fibroblasts. We found that there were several species of free fatty acids
that had altered levels between patient and control fibroblasts, some of which were found to be
113
part of the biosynthesis of unsaturated fatty acids pathway and linoleic acid metabolism pathway.
These data show that SCPx functions in the metabolism of several fatty acids. We also found
acylcarnitine and sterol species that had altered levels between patient and control fibroblasts and
are part of the fatty acid degradation pathway and steroid biosynthesis pathway, respectively.
Alterations in the levels of phospholipids that are part of the glycerophospholipid metabolism
pathway as well as in several species of ceramides were found through our lipidomic analyses.
Together, these data indicate that SCPx functions in the transport and/or metabolism of several
lipid classes and that alterations in the identified pathways may account for certain aspects of the
patient’s clinical phenotype.
To restore SCPx levels, we treated patient and control fibroblasts with fenofibrate, 4-
OHT, or a combination of the two. We found that each compound alone significantly increased
SCPx levels in patient fibroblasts compared to untreated patient fibroblasts, however, the two
compounds did not work synergistically on SCPx. We also found that each of these compounds
alone significantly improved the patient’s fatty acid profile compared to untreated patient
fibroblasts, suggesting that these compounds may be beneficial in treating SCPx deficiency.
Collectively, our data show that decreased SCPx levels leads to widespread gene alterations and
alterations in lipid metabolism and confirm that SCPx plays a role in the transport and
metabolism of several lipids including fatty acids, acylcarnitines, sterols, phospholipids, and
sphingolipids. Our data also suggest that fenofibrate and 4-OHT may restore SCPx protein levels
and that these compounds may be beneficial for the treatment of neurodegenerative peroxisomal
disorders.
Cholesterol Transport Proteins
114
Taken together, these data show that both specific and non-specific lipid transfer proteins
play a fundamental role in lipid trafficking and metabolism in general, in addition to their
respective specialized functions. STAR is classically known to function specifically in
cholesterol transport in hormone-induced steroidogenesis. However, our data suggest that STAR
also plays a more general role in lipid transfer and metabolism since the absence of STAR results
in alterations in lipids other than cholesterol such as DAG and PC. Similarly, SCPx is classically
known to function in cholesterol and fatty acid metabolism for peroxisomal β-oxidation.
However, our data suggest that SCPx has broader functions in lipid transfer and metabolism
since SCPx deficiency results in a complete reorganization of the lipid profile, including changes
in acylcarnitine, phospholipid, and sphingolipid levels.
Future Directions
STAR
Future studies involving STAR could build upon investigating the effects of having low
versus high levels of constitutive STAR. Our data show that STARKO-OE1 cells and STARKO-
OE2 cells have low and high levels of constitutive STAR, respectively, and that hormonal
stimulation leads to a significant increase in STAR protein levels and steroid production in
STARKO-OE1 cells but leads to only a slight increase in STARKO-OE2 cells. Further, our data
show that STARKO-OE2 cells, which have high constitutive STAR, have the most damaged
mitochondria and the lowest rates of mitochondrial respiration, ATP production, and glycolysis
among all models tested. Studies can be designed to investigate whether the more pronounced
defects in mitochondrial structure and respiration may be due to the high basal levels of STAR.
One way this could be studied is through utilizing MA-10+STAR cells, which also have high
levels of basal STAR. It would be interesting to determine if continued passaging of MA-
115
10+STAR cells eventually leads to mitochondrial ultrastructure damage or mitochondrial
dysfunction, similar to what is seen with STARKO-OE2 cells. Another potential outcome could
be that STAR overexpression in WT MA-10 cells may not lead to mitochondrial dysfunction
because mitochondria are intact prior to STAR overexpression, but that STAR overexpression as
seen in STARKO-OE2 cells exacerbates mitochondrial dysfunction because mitochondria were
already damaged prior to reintroducing STAR. This outcome would necessitate further
investigation into the differences between STARKO-OE1 and STARKO-OE2.
Our current data suggesting functions of constitutive STAR have all been done utilizing
MA-10 cell models. Another direction that this project could take in the future is to confirm all
data obtained using the STAR KO cell model through in vivo studies using STAR KO rats.
Another laboratory previously developed STAR KO mice, however, these are no longer
available [64]. Further, the mouse testis has low levels of androgen binding protein (ABP),
meaning that, even if steroidogenesis is impaired, the majority of testosterone that is produced is
free and active [225]. Therefore, when studying androgen production and its effects on the body,
rats are better models to use than mice since they have high levels of ABP, comparable to that of
humans [225]. With the STAR KO rat model, we could conduct histological analyses of the
adrenal glands and testes, measure steroid hormone levels in response to hormone stimulation
and TSPO ligands, characterize lipid droplet content, and characterize mitochondrial structure
and function. These data would be important in further delineating the function of constitutive
STAR in a more relevant model and in strengthening our results found in the cell model.
SCP2/SCPx
Our data show that SCPx deficiency affects several pathways related to lipid metabolism.
As such, it is likely that targeting one specific pathway is insufficient to ameliorate the patient’s
116
unique phenotype. While fenofibrate and 4-OHT showed to be effective in increasing SCPx
levels individually, the combination treatment had no synergistic effect on SCPx levels. Thus,
future studies could be designed to investigate whether targeting other genes or pathways that
were identified to be altered through our gene expression or lipidomic analyses, may work
synergistically with either fenofibrate or 4-OHT in increasing SCPx levels and helping to
improve the patient’s phenotype. For example, another pathway we identified to have several
genes differentially expressed between NHDF and WESP cells is serotonergic synapse signaling.
Increasing serotonergic signaling and developing therapies that increase serotonin concentration
in the synaptic cleft has been suggested to be a potential strategy for preventing
neurodegeneration in Alzheimer’s disease [226]. It is possible that treating our patient with
fenofibrate or 4-OHT in combination with a compound targeting serotonergic synapse signaling
may increase SCPx levels, improve the patient’s lipid profile, and ameliorate aspects of his
neurological phenotype.
Lastly, because of the patient’s current state, our studies described above utilize patient
fibroblasts to investigate potential roles of SCPx and molecular mechanisms underlying the
patient’s clinical conditions. However, utilizing a tissue more relevant to the patient’s phenotype
and to the function of SCPx may be more informative. A previous study reported that mice null
in SCPx but normal levels of SCP2 had altered hepatic fatty acid levels [127]. Another direction
this project could take is to develop an animal model of SCPx deficiency and investigate tissues
with high rates of cholesterol metabolism such as liver, adrenals, or testis to determine if our
current findings are also seen in these relevant tissues. Furthermore, studying neuronal cells in
this animal model may also provide more data behind the neurodegenerative effects of SCPx
117
deficiency and may lead to the identification of potential therapeutic strategies to treat SCPx
deficiency.
Overall Conclusions
The data presented in this thesis are consistent with the notion that the classic cholesterol
transport proteins, STAR and SCP2/SCPx, participate in functions independent of cholesterol
transport. STAR may play a role in the transport and/or metabolism of other lipids, including
DAG, and in the maintenance of mitochondrial structure and function. SCPx is involved in the
transport and/or metabolism of several lipid classes including fatty acids, acylcarnitines, sterols,
phospholipids, and sphingolipids.
118
References
1. Luo, J., H. Yang, and B.L. Song, Mechanisms and regulation of cholesterol homeostasis.
Nat Rev Mol Cell Biol, 2020. 21(4): p. 225-245.
2. Shibuya, Y., C.C. Chang, and T.Y. Chang, ACAT1/SOAT1 as a therapeutic target for
Alzheimer's disease. Future Med Chem, 2015. 7(18): p. 2451-67.
3. Kuzu, O.F., M.A. Noory, and G.P. Robertson, The Role of Cholesterol in Cancer. Cancer
Res, 2016. 76(8): p. 2063-70.
4. Cortes, V.A., et al., Physiological and pathological implications of cholesterol. Front
Biosci (Landmark Ed), 2014. 19(3): p. 416-28.
5. Maxfield, F.R. and I. Tabas, Role of cholesterol and lipid organization in disease. Nature,
2005. 438(7068): p. 612-21.
6. Luu, W., et al., Oxysterols: Old Tale, New Twists. Annu Rev Pharmacol Toxicol, 2016.
56: p. 447-67.
7. Wong, L.H., A.T. Gatta, and T.P. Levine, Lipid transfer proteins: the lipid commute via
shuttles, bridges and tubes. Nat Rev Mol Cell Biol, 2019. 20(2): p. 85-101.
8. Midzak, A., et al., Mitochondrial protein import and the genesis of steroidogenic
mitochondria. Mol Cell Endocrinol, 2011. 336(1-2): p. 70-9.
9. Kriska, T., et al., Sterol carrier protein-2 (SCP-2) involvement in cholesterol
hydroperoxide cytotoxicity as revealed by SCP-2 inhibitor effects. J Lipid Res, 2010.
51(11): p. 3174-84.
10. Galano, M., et al., Role of Constitutive STAR in Leydig Cells. Int J Mol Sci, 2021. 22(4).
11. Pfeifer, S.M., et al., Sterol carrier protein 2: a role in steroid hormone synthesis? J
Steroid Biochem Mol Biol, 1993. 47(1-6): p. 167-72.
119
12. Petrescu, A.D., et al., Steroidogenic acute regulatory protein binds cholesterol and
modulates mitochondrial membrane sterol domain dynamics. J Biol Chem, 2001.
276(40): p. 36970-82.
13. Kallen, C.B., et al., Steroidogenic acute regulatory protein (StAR) is a sterol transfer
protein. J Biol Chem, 1998. 273(41): p. 26285-8.
14. Charreau, E.H., et al., Hormonal modulation of 3-hydroxy-3-methylglutaryl coenzyme A
reductase activity in gonadotropin-stimulated and -desensitized testicular Leydig cells. J
Biol Chem, 1981. 256(24): p. 12719-24.
15. Rainey, W.E., J.W. Shay, and J.I. Mason, ACTH induction of 3-hydroxy-3-methylglutaryl
coenzyme A reductase, cholesterol biosynthesis, and steroidogenesis in primary cultures
of bovine adrenocortical cells. J Biol Chem, 1986. 261(16): p. 7322-6.
16. Goldstein, J.L. and M.S. Brown, Regulation of the mevalonate pathway. Nature, 1990.
343(6257): p. 425-30.
17. BLOCH, K., Biological synthesis of cholesterol. Harvey Lect, 1952. 48: p. 68-88.
18. Sato, R., J.L. Goldstein, and M.S. Brown, Replacement of serine-871 of hamster 3-
hydroxy-3-methylglutaryl-CoA reductase prevents phosphorylation by AMP-activated
kinase and blocks inhibition of sterol synthesis induced by ATP depletion. Proc Natl Acad
Sci U S A, 1993. 90(20): p. 9261-5.
19. Wong, W.W., et al., HMG-CoA reductase inhibitors and the malignant cell: the statin
family of drugs as triggers of tumor-specific apoptosis. Leukemia, 2002. 16(4): p. 508-
19.
20. Cartwright, I.J., D. Plonné, and J.A. Higgins, Intracellular events in the assembly of
chylomicrons in rabbit enterocytes. J Lipid Res, 2000. 41(11): p. 1728-39.
120
21. Chang, T.Y., et al., Cholesterol sensing, trafficking, and esterification. Annu Rev Cell
Dev Biol, 2006. 22: p. 129-57.
22. Hu, J., et al., Cellular cholesterol delivery, intracellular processing and utilization for
biosynthesis of steroid hormones. Nutr Metab (Lond), 2010. 7: p. 47.
23. Rigotti, A., H.E. Miettinen, and M. Krieger, The role of the high-density lipoprotein
receptor SR-BI in the lipid metabolism of endocrine and other tissues. Endocr Rev, 2003.
24(3): p. 357-87.
24. Leiva, A., et al., Mechanisms regulating hepatic SR-BI expression and their impact on
HDL metabolism. Atherosclerosis, 2011. 217(2): p. 299-307.
25. Lange, Y. and T.L. Steck, Active membrane cholesterol as a physiological effector. Chem
Phys Lipids, 2016. 199: p. 74-93.
26. Steck, T.L. and Y. Lange, Cell cholesterol homeostasis: mediation by active cholesterol.
Trends Cell Biol, 2010. 20(11): p. 680-7.
27. Epand, R.M., Proteins and cholesterol-rich domains. Biochim Biophys Acta, 2008.
1778(7-8): p. 1576-82.
28. Liscum, L. and N.J. Munn, Intracellular cholesterol transport. Biochim Biophys Acta,
1999. 1438(1): p. 19-37.
29. Soccio, R.E. and J.L. Breslow, Intracellular cholesterol transport. Arterioscler Thromb
Vasc Biol, 2004. 24(7): p. 1150-60.
30. Mukherjee, S., et al., Cholesterol distribution in living cells: fluorescence imaging using
dehydroergosterol as a fluorescent cholesterol analog. Biophys J, 1998. 75(4): p. 1915-
25.
121
31. Lange, Y., Disposition of intracellular cholesterol in human fibroblasts. J Lipid Res,
1991. 32(2): p. 329-39.
32. Rone, M.B., J. Fan, and V. Papadopoulos, Cholesterol transport in steroid biosynthesis:
role of protein-protein interactions and implications in disease states. Biochim Biophys
Acta, 2009. 1791(7): p. 646-58.
33. Epand, R.M., et al., An apolipoprotein AI mimetic peptide: membrane interactions and
the role of cholesterol. Biochemistry, 2004. 43(17): p. 5073-83.
34. Resh, M.D., Membrane targeting of lipid modified signal transduction proteins. Subcell
Biochem, 2004. 37: p. 217-32.
35. Chakrabandhu, K., et al., Palmitoylation is required for efficient Fas cell death signaling.
EMBO J, 2007. 26(1): p. 209-20.
36. Morandat, S., M. Bortolato, and B. Roux, Cholesterol-dependent insertion of
glycosylphosphatidylinositol-anchored enzyme. Biochim Biophys Acta, 2002. 1564(2): p.
473-78.
37. Wang, J., et al., Evidence for segregation of heterologous GPI-anchored proteins into
separate lipid rafts within the plasma membrane. J Membr Biol, 2002. 189(1): p. 35-43.
38. Li, H. and V. Papadopoulos, Peripheral-type benzodiazepine receptor function in
cholesterol transport. Identification of a putative cholesterol recognition/interaction
amino acid sequence and consensus pattern. Endocrinology, 1998. 139(12): p. 4991-7.
39. Papadopoulos, V., J. Fan, and B. Zirkin, Translocator protein (18 kDa): an update on its
function in steroidogenesis. J Neuroendocrinol, 2018. 30(2).
40. Papadopoulos, V. and W.L. Miller, Role of mitochondria in steroidogenesis. Best Pract
Res Clin Endocrinol Metab, 2012. 26(6): p. 771-90.
122
41. Lev, S., Non-vesicular lipid transport by lipid-transfer proteins and beyond. Nat Rev Mol
Cell Biol, 2010. 11(10): p. 739-50.
42. Baumann, N.A., et al., Transport of newly synthesized sterol to the sterol-enriched
plasma membrane occurs via nonvesicular equilibration. Biochemistry, 2005. 44(15): p.
5816-26.
43. Urbani, L. and R.D. Simoni, Cholesterol and vesicular stomatitis virus G protein take
separate routes from the endoplasmic reticulum to the plasma membrane. J Biol Chem,
1990. 265(4): p. 1919-23.
44. Kaplan, M.R. and R.D. Simoni, Transport of cholesterol from the endoplasmic reticulum
to the plasma membrane. J Cell Biol, 1985. 101(2): p. 446-53.
45. DeGrella, R.F. and R.D. Simoni, Intracellular transport of cholesterol to the plasma
membrane. J Biol Chem, 1982. 257(23): p. 14256-62.
46. Lev, S., Nonvesicular lipid transfer from the endoplasmic reticulum. Cold Spring Harb
Perspect Biol, 2012. 4(10).
47. Issop, L., M.B. Rone, and V. Papadopoulos, Organelle plasticity and interactions in
cholesterol transport and steroid biosynthesis. Mol Cell Endocrinol, 2013. 371(1-2): p.
34-46.
48. Miller, W.L., Steroidogenic acute regulatory protein (StAR), a novel mitochondrial
cholesterol transporter. Biochim Biophys Acta, 2007. 1771(6): p. 663-76.
49. Jefcoate, C., High-flux mitochondrial cholesterol trafficking, a specialized function of the
adrenal cortex. J Clin Invest, 2002. 110(7): p. 881-90.
123
50. Liu, J., M.B. Rone, and V. Papadopoulos, Protein-protein interactions mediate
mitochondrial cholesterol transport and steroid biosynthesis. J Biol Chem, 2006.
281(50): p. 38879-93.
51. Hauet, T., et al., Peripheral-type benzodiazepine receptor-mediated action of
steroidogenic acute regulatory protein on cholesterol entry into leydig cell mitochondria.
Mol Endocrinol, 2005. 19(2): p. 540-54.
52. Li, H., et al., Identification, localization, and function in steroidogenesis of PAP7: a
peripheral-type benzodiazepine receptor- and PKA (RIalpha)-associated protein. Mol
Endocrinol, 2001. 15(12): p. 2211-28.
53. Papadopoulos, V., et al., Targeted disruption of the peripheral-type benzodiazepine
receptor gene inhibits steroidogenesis in the R2C Leydig tumor cell line. J Biol Chem,
1997. 272(51): p. 32129-35.
54. Rone, M.B., et al., Identification of a dynamic mitochondrial protein complex driving
cholesterol import, trafficking, and metabolism to steroid hormones. Mol Endocrinol,
2012. 26(11): p. 1868-82.
55. Bose, M., et al., Steroidogenic activity of StAR requires contact with mitochondrial
VDAC1 and phosphate carrier protein. J Biol Chem, 2008. 283(14): p. 8837-45.
56. Stocco, D.M. and B.J. Clark, Regulation of the acute production of steroids in
steroidogenic cells. Endocr Rev, 1996. 17(3): p. 221-44.
57. Epstein, L.F. and N.R. Orme-Johnson, Regulation of steroid hormone biosynthesis.
Identification of precursors of a phosphoprotein targeted to the mitochondrion in
stimulated rat adrenal cortex cells. J Biol Chem, 1991. 266(29): p. 19739-45.
124
58. Stocco, D.M. and T.C. Sodeman, The 30-kDa mitochondrial proteins induced by
hormone stimulation in MA-10 mouse Leydig tumor cells are processed from larger
precursors. J Biol Chem, 1991. 266(29): p. 19731-8.
59. Clark, B.J., et al., The purification, cloning, and expression of a novel luteinizing
hormone-induced mitochondrial protein in MA-10 mouse Leydig tumor cells.
Characterization of the steroidogenic acute regulatory protein (StAR). J Biol Chem,
1994. 269(45): p. 28314-22.
60. Lin, D., et al., Role of steroidogenic acute regulatory protein in adrenal and gonadal
steroidogenesis. Science, 1995. 267(5205): p. 1828-31.
61. Sugawara, T., et al., Human steroidogenic acute regulatory protein: functional activity in
COS-1 cells, tissue-specific expression, and mapping of the structural gene to 8p11.2 and
a pseudogene to chromosome 13. Proc Natl Acad Sci U S A, 1995. 92(11): p. 4778-82.
62. Tee, M.K., et al., T-->A transversion 11 bp from a splice acceptor site in the human gene
for steroidogenic acute regulatory protein causes congenital lipoid adrenal hyperplasia.
Hum Mol Genet, 1995. 4(12): p. 2299-305.
63. Stocco, D.M., Clinical disorders associated with abnormal cholesterol transport:
mutations in the steroidogenic acute regulatory protein. Mol Cell Endocrinol, 2002.
191(1): p. 19-25.
64. Caron, K.M., et al., Targeted disruption of the mouse gene encoding steroidogenic acute
regulatory protein provides insights into congenital lipoid adrenal hyperplasia. Proc Natl
Acad Sci U S A, 1997. 94(21): p. 11540-5.
65. Strauss, J.F., et al., START domain proteins and the intracellular trafficking of
cholesterol in steroidogenic cells. Mol Cell Endocrinol, 2003. 202(1-2): p. 59-65.
125
66. Tsujishita, Y. and J.H. Hurley, Structure and lipid transport mechanism of a StAR-related
domain. Nat Struct Biol, 2000. 7(5): p. 408-14.
67. Aghazadeh, Y., et al., Hormone-induced 14-3-3γ adaptor protein regulates steroidogenic
acute regulatory protein activity and steroid biosynthesis in MA-10 Leydig cells. J Biol
Chem, 2012. 287(19): p. 15380-94.
68. Artemenko, I.P., et al., Mitochondrial processing of newly synthesized steroidogenic
acute regulatory protein (StAR), but not total StAR, mediates cholesterol transfer to
cytochrome P450 side chain cleavage enzyme in adrenal cells. J Biol Chem, 2001.
276(49): p. 46583-96.
69. Clark, B.J., et al., Hormonal and developmental regulation of the steroidogenic acute
regulatory protein. Mol Endocrinol, 1995. 9(10): p. 1346-55.
70. Clark, B.J. and E.A. Hudson, StAR Protein Stability in Y1 and Kin-8 Mouse
Adrenocortical Cells. Biology (Basel), 2015. 4(1): p. 200-15.
71. Granot, Z., et al., Proteolysis of normal and mutated steroidogenic acute regulatory
proteins in the mitochondria: the fate of unwanted proteins. Mol Endocrinol, 2003.
17(12): p. 2461-76.
72. Arakane, F., et al., Steroidogenic acute regulatory protein (StAR) retains activity in the
absence of its mitochondrial import sequence: implications for the mechanism of StAR
action. Proc Natl Acad Sci U S A, 1996. 93(24): p. 13731-6.
73. Bose, H.S., V.R. Lingappa, and W.L. Miller, Rapid regulation of steroidogenesis by
mitochondrial protein import. Nature, 2002. 417(6884): p. 87-91.
74. Arakane, F., et al., Steroidogenic acute regulatory protein (StAR) acts on the outside of
mitochondria to stimulate steroidogenesis. Endocr Res, 1998. 24(3-4): p. 463-8.
126
75. Bose, H.S., V.R. Lingappa, and W.L. Miller, The steroidogenic acute regulatory protein,
StAR, works only at the outer mitochondrial membrane. Endocr Res, 2002. 28(4): p. 295-
308.
76. Arakane, F., et al., Phosphorylation of steroidogenic acute regulatory protein (StAR)
modulates its steroidogenic activity. J Biol Chem, 1997. 272(51): p. 32656-62.
77. Jo, Y., et al., Involvement of protein kinase C and cyclic adenosine 3',5'-monophosphate-
dependent kinase in steroidogenic acute regulatory protein expression and steroid
biosynthesis in Leydig cells. Biol Reprod, 2005. 73(2): p. 244-55.
78. Dyson, M.T., et al., Mitochondrial A-kinase anchoring protein 121 binds type II protein
kinase A and enhances steroidogenic acute regulatory protein-mediated steroidogenesis
in MA-10 mouse leydig tumor cells. Biol Reprod, 2008. 78(2): p. 267-77.
79. Tuckey, R.C., et al., Transfer of cholesterol between phospholipid vesicles mediated by
the steroidogenic acute regulatory protein (StAR). J Biol Chem, 2002. 277(49): p. 47123-
8.
80. Baker, B.Y., et al., Cholesterol binding does not predict activity of the steroidogenic
acute regulatory protein, StAR. J Biol Chem, 2007. 282(14): p. 10223-32.
81. Romanowski, M.J., et al., Crystal structure of the Mus musculus cholesterol-regulated
START protein 4 (StarD4) containing a StAR-related lipid transfer domain. Proc Natl
Acad Sci U S A, 2002. 99(10): p. 6949-54.
82. Mathieu, A.P., et al., Insights into steroidogenic acute regulatory protein (StAR)-
dependent cholesterol transfer in mitochondria: evidence from molecular modeling and
structure-based thermodynamics supporting the existence of partially unfolded states of
StAR. J Mol Endocrinol, 2002. 29(3): p. 327-45.
127
83. Miller, W.L., StAR search--what we know about how the steroidogenic acute regulatory
protein mediates mitochondrial cholesterol import. Mol Endocrinol, 2007. 21(3): p. 589-
601.
84. Gallegos, A.M., et al., Sterol carrier protein-2 expression alters plasma membrane lipid
distribution and cholesterol dynamics. Biochemistry, 2001. 40(21): p. 6493-6506.
85. Puglielli, L., et al., Sterol carrier protein-2 is involved in cholesterol transfer from the
endoplasmic reticulum to the plasma membrane in human fibroblasts. J Biol Chem,
1995. 270(32): p. 18723-6.
86. Li, N.C., J.J. Fan, and V. Papadopoulos, Sterol Carrier Protein-2, a Nonspecific Lipid-
Transfer Protein, in Intracellular Cholesterol Trafficking in Testicular Leydig Cells. Plos
One, 2016. 11(2).
87. Ohba, T., et al., Human sterol carrier protein x/sterol carrier protein 2 gene has two
promoters. Biochemistry, 1995. 34(33): p. 10660-8.
88. Atshaves, B.P., et al., Expression and intracellular processing of the 58 kDa sterol
carrier protein-2/3-oxoacyl-CoA thiolase in transfected mouse L-cell fibroblasts. J Lipid
Res, 1999. 40(4): p. 610-22.
89. Moncecchi, D., et al., Sterol carrier protein-2 expression in mouse L-cell fibroblasts
alters cholesterol uptake. Biochim Biophys Acta, 1996. 1302(2): p. 110-6.
90. Vahouny, G.V., et al., Sterol carrier and lipid transfer proteins. Adv Lipid Res, 1987. 22:
p. 83-113.
91. Wanders, R.J., et al., Sterol carrier protein X (SCPx) is a peroxisomal branched-chain
beta-ketothiolase specifically reacting with 3-oxo-pristanoyl-CoA: a new, unique role for
128
SCPx in branched-chain fatty acid metabolism in peroxisomes. Biochem Biophys Res
Commun, 1997. 236(3): p. 565-9.
92. Mendis-Handagama, S.M., et al., Leydig cell peroxisomes and sterol carrier protein-2 in
luteinizing hormone-deprived rats. Endocrinology, 1992. 131(6): p. 2839-45.
93. Keller, G.A., et al., Subcellular localization of sterol carrier protein-2 in rat hepatocytes:
its primary localization to peroxisomes. J Cell Biol, 1989. 108(4): p. 1353-61.
94. Mori, T., et al., Molecular cloning and deduced amino acid sequence of nonspecific lipid
transfer protein (sterol carrier protein 2) of rat liver: a higher molecular mass (60 kDa)
protein contains the primary sequence of nonspecific lipid transfer protein as its C-
terminal part. Proc Natl Acad Sci U S A, 1991. 88(10): p. 4338-42.
95. Westerman, J. and K.W. Wirtz, The primary structure of the nonspecific lipid transfer
protein (sterol carrier protein 2) from bovine liver. Biochem Biophys Res Commun,
1985. 127(1): p. 333-8.
96. Stolowich, N., et al., Holo-sterol carrier protein-2. (13)C NMR investigation of
cholesterol and fatty acid binding sites. J Biol Chem, 1999. 274(50): p. 35425-33.
97. Colles, S.M., et al., Cholesterol interaction with recombinant human sterol carrier
protein-2. Lipids, 1995. 30(9): p. 795-803.
98. Frolov, A., et al., Spontaneous and protein-mediated sterol transfer between intracellular
membranes. J Biol Chem, 1996. 271(27): p. 16075-83.
99. Gallegos, A.M., et al., Gene structure, intracellular localization, and functional roles of
sterol carrier protein-2. Progress in Lipid Research, 2001. 40(6): p. 498-563.
100. Atshaves, B.P., et al., Sterol carrier protein-2 alters high density lipoprotein-mediated
cholesterol efflux. J Biol Chem, 2000. 275(47): p. 36852-61.
129
101. Murphy, E.J. and F. Schroeder, Sterol carrier protein-2 mediated cholesterol
esterification in transfected L-cell fibroblasts. Biochim Biophys Acta, 1997. 1345(3): p.
283-92.
102. Baum, C.L., et al., Sterol carrier protein-2 overexpression enhances sterol cycling and
inhibits cholesterol ester synthesis and high density lipoprotein cholesterol secretion. J
Biol Chem, 1997. 272(10): p. 6490-8.
103. Zanlungo, S., et al., Sterol carrier protein 2 gene transfer changes lipid metabolism and
enterohepatic sterol circulation in mice. Gastroenterology, 2000. 119(6): p. 1708-19.
104. Seedorf, U., et al., Defective peroxisomal catabolism of branched fatty acyl coenzyme A
in mice lacking the sterol carrier protein-2 sterol carrier protein-x gene function. Genes
& Development, 1998. 12(8): p. 1189-1201.
105. Puglielli, L., et al., Modulation of intrahepatic cholesterol trafficking: evidence by in vivo
antisense treatment for the involvement of sterol carrier protein-2 in newly synthesized
cholesterol transport into rat bile. Biochem J, 1996. 317 ( Pt 3): p. 681-7.
106. Chanderbhan, R., et al., Sterol carrier protein2. Delivery of cholesterol from adrenal
lipid droplets to mitochondria for pregnenolone synthesis. J Biol Chem, 1982. 257(15): p.
8928-34.
107. Yamamoto, R., et al., Cloning and expression of a cDNA encoding human sterol carrier
protein 2. Proc Natl Acad Sci U S A, 1991. 88(2): p. 463-7.
108. Rennert, H., et al., Regulated expression of sterol carrier protein 2 in the ovary: a key
role for cyclic AMP. Biochemistry, 1991. 30(47): p. 11280-5.
130
109. Trzeciak, W.H., et al., Studies on the synthesis of sterol carrier protein-2 in rat
adrenocortical cells in monolayer culture. Regulation by ACTH and dibutyryl cyclic
3',5'-AMP. J Biol Chem, 1987. 262(8): p. 3713-7.
110. McLean, M.P., et al., Estradiol regulation of sterol carrier protein-2 independent of
cytochrome P450 side-chain cleavage expression in the rat corpus luteum.
Endocrinology, 1989. 125(3): p. 1337-44.
111. van Noort, M., et al., Localization and hormonal regulation of the non-specific lipid
transfer protein (sterol carrier protein2) in the rat testis. J Endocrinol, 1986. 109(3): p.
R13-6.
112. Gallegos, A.M., et al., A potential role for sterol carrier protein-2 in cholesterol transfer
to mitochondria. Chem Phys Lipids, 2000. 105(1): p. 9-29.
113. Schroeder, F., et al., PROBING THE LIGAND-BINDING SITES OF FATTY-ACID AND
STEROL CARRIER PROTEINS - EFFECTS OF ETHANOL. Biochemistry, 1995. 34(37):
p. 11919-11927.
114. McArthur, M.J., et al., Cellular uptake and intracellular trafficking of long chain fatty
acids. J Lipid Res, 1999. 40(8): p. 1371-83.
115. Murphy, E.J., Sterol carrier protein-2 expression increases NBD-stearate uptake and
cytoplasmic diffusion in L cells. Am J Physiol, 1998. 275(2): p. G237-43.
116. Atshaves, B.P., S.M. Storey, and F. Schroeder, Sterol carrier protein-2/sterol carrier
protein-x expression differentially alters fatty acid metabolism in L cell fibroblasts. J
Lipid Res, 2003. 44(9): p. 1751-62.
131
117. Murphy, E.J., T. Stiles, and F. Schroeder, Sterol carrier protein-2 expression alters
phospholipid content and fatty acyl composition in L-cell fibroblasts. J Lipid Res, 2000.
41(5): p. 788-96.
118. Gossett, R.E., et al., Acyl-CoA binding proteins: multiplicity and function. Lipids, 1996.
31(9): p. 895-918.
119. Knudsen, J., et al., Role of acylCoA binding protein in acylCoA transport, metabolism
and cell signaling. Mol Cell Biochem, 1999. 192(1-2): p. 95-103.
120. Starodub, O., et al., Sterol carrier protein-2 localization in endoplasmic reticulum and
role in phospholipid formation. Am J Physiol Cell Physiol, 2000. 279(4): p. C1259-69.
121. Schroeder, F., et al., Sterol carrier protein-2 functions in phosphatidylinositol transfer
and signaling. Biochemistry, 2003. 42(11): p. 3189-202.
122. Atshaves, B.P., et al., Effect of sterol carrier protein-2 expression on sphingolipid
distribution in plasma membrane lipid rafts/caveolae. Lipids, 2007. 42(10): p. 871-84.
123. Bloj, B. and D.B. Zilversmit, Accelerated transfer of neutral glycosphingolipids and
ganglioside GM1 by a purified lipid transfer protein. J Biol Chem, 1981. 256(12): p.
5988-91.
124. Fong, T.H., S.M. Wang, and H.S. Lin, Immunocytochemical demonstration of a lipid
droplet-specific capsule in cultured Leydig cells of the golden hamsters. J Cell Biochem,
1996. 63(3): p. 366-73.
125. Seedorf, U., et al., Sterol carrier protein X is peroxisomal 3-oxoacyl coenzyme A thiolase
with intrinsic sterol carrier and lipid transfer activity. J Biol Chem, 1994. 269(33): p.
21277-83.
132
126. Ferdinandusse, S., et al., Bile acids: the role of peroxisomes. J Lipid Res, 2009. 50(11): p.
2139-47.
127. Atshaves, B.P., et al., Effect of SCP-x gene ablation on branched-chain fatty acid
metabolism. Am J Physiol Gastrointest Liver Physiol, 2007. 292(3): p. G939-51.
128. Ferdinandusse, S., et al., Mutations in the gene encoding peroxisomal sterol carrier
protein X (SCPx) cause leukencephalopathy with dystonia and motor neuropathy.
American Journal of Human Genetics, 2006. 78(6): p. 1046-1052.
129. Horvath, R., et al., SCP2 MUTATIONS AND NEURODEGENERATION WITH BRAIN
IRON ACCUMULATION. Neurology, 2015. 85(21): p. 1909-1911.
130. Epstein, L.F. and N.R. Orme-Johnson, Acute action of luteinizing hormone on mouse
Leydig cells: accumulation of mitochondrial phosphoproteins and stimulation of
testosterone synthesis. Mol Cell Endocrinol, 1991. 81(1-3): p. 113-26.
131. Luo, L., et al., Leydig cell protein synthesis and steroidogenesis in response to acute
stimulation by luteinizing hormone in rats. Biol Reprod, 1998. 59(2): p. 263-70.
132. Tugaeva, K.V. and N.N. Sluchanko, Steroidogenic Acute Regulatory Protein: Structure,
Functioning, and Regulation. Biochemistry (Mosc), 2019. 84(Suppl 1): p. S233-S253.
133. Bose, H.S., et al., The pathophysiology and genetics of congenital lipoid adrenal
hyperplasia. N Engl J Med, 1996. 335(25): p. 1870-8.
134. Akula, N., et al., Identification of small-molecule inhibitors of the steroidogenic acute
regulatory protein (STARD1) by structure-based design. Bioorg Med Chem Lett, 2012.
22(12): p. 4139-43.
135. Papadopoulos, V., et al., Translocator protein-mediated pharmacology of cholesterol
transport and steroidogenesis. Mol Cell Endocrinol, 2015. 408: p. 90-8.
133
136. Chung, J.Y., et al., Cholesterol accumulation, lipid droplet formation, and steroid
production in Leydig cells: Role of translocator protein (18-kDa). Andrology, 2020. 8(3):
p. 719-730.
137. Pagac, M., et al., SEIPIN Regulates Lipid Droplet Expansion and Adipocyte Development
by Modulating the Activity of Glycerol-3-phosphate Acyltransferase. Cell Rep, 2016.
17(6): p. 1546-1559.
138. Itabe, H., et al., Perilipins: a diversity of intracellular lipid droplet proteins. Lipids
Health Dis, 2017. 16(1): p. 83.
139. Liao, Y.C. and S.H. Lo, Deleted in liver cancer-1 (DLC-1): a tumor suppressor not just
for liver. Int J Biochem Cell Biol, 2008. 40(5): p. 843-7.
140. Oude Weernink, P.A., M. López de Jesús, and M. Schmidt, Phospholipase D signaling:
orchestration by PIP2 and small GTPases. Naunyn Schmiedebergs Arch Pharmacol,
2007. 374(5-6): p. 399-411.
141. Gutiérrez-Martínez, E., et al., Lipid phosphate phosphatase 3 participates in transport
carrier formation and protein trafficking in the early secretory pathway. J Cell Sci, 2013.
126(Pt 12): p. 2641-55.
142. Lee, J. and C. Jefcoate, Monitoring of Dual CRISPR/Cas9-Mediated Steroidogenic Acute
Regulatory Protein Gene Deletion and Cholesterol Accumulation Using High-Resolution
Fluorescence. Front Endocrinol (Lausanne), 2017. 8: p. 289.
143. Besterman, J.M., V. Duronio, and P. Cuatrecasas, Rapid formation of diacylglycerol from
phosphatidylcholine: a pathway for generation of a second messenger. Proc Natl Acad
Sci U S A, 1986. 83(18): p. 6785-9.
134
144. Qiu, Y., et al., Steroidogenic acute regulatory protein (StAR) overexpression attenuates
HFD-induced hepatic steatosis and insulin resistance. Biochim Biophys Acta Mol Basis
Dis, 2017. 1863(4): p. 978-990.
145. Hadjian, A.J., M. Culty, and E.M. Chambaz, Steroidogenic effect of exogenous
phospholipase C on bovine adrenal fasciculata cells. Biochem Biophys Res Commun,
1984. 123(1): p. 33-40.
146. Papadopoulos, V., S. Carreau, and M.A. Drosdowsky, Effect of phorbol ester and
phospholipase C on LH-stimulated steroidogenesis in purified rat Leydig cells. FEBS
Lett, 1985. 188(2): p. 312-316.
147. Chaudhary, L.R. and D.M. Stocco, Stimulation of progesterone production by phorbol-
12-myristate-13-acetate in MA-10 Leydig tumor cells. Biochimie, 1988. 70(10): p. 1353-
60.
148. Boni, L.T. and R.R. Rando, The nature of protein kinase C activation by physically
defined phospholipid vesicles and diacylglycerols. J Biol Chem, 1985. 260(19): p. 10819-
25.
149. Zechner, R., et al., FAT SIGNALS--lipases and lipolysis in lipid metabolism and
signaling. Cell Metab, 2012. 15(3): p. 279-91.
150. Fredrikson, G. and P. Belfrage, Positional specificity of hormone-sensitive lipase from rat
adipose tissue. J Biol Chem, 1983. 258(23): p. 14253-6.
151. Hatano, M., et al., SF-1 deficiency causes lipid accumulation in Leydig cells via
suppression of STAR and CYP11A1. Endocrine, 2016. 54(2): p. 484-496.
135
152. Miyoshi, H., et al., Perilipin promotes hormone-sensitive lipase-mediated adipocyte
lipolysis via phosphorylation-dependent and -independent mechanisms. J Biol Chem,
2006. 281(23): p. 15837-44.
153. Federico, L., et al., Lipid phosphate phosphatase 3 regulates adipocyte sphingolipid
synthesis, but not developmental adipogenesis or diet-induced obesity in mice. PLoS One,
2018. 13(6): p. e0198063.
154. Homma, Y. and Y. Emori, A dual functional signal mediator showing RhoGAP and
phospholipase C-delta stimulating activities. Embo j, 1995. 14(2): p. 286-91.
155. Ng, D.C., et al., Mitochondrial targeting of growth suppressor protein DLC2 through the
START domain. FEBS Lett, 2006. 580(1): p. 191-8.
156. Ascoli, M.A., Characterization of Several Clonal Lines of Cultured Leydig Tumor Cells:
Gonadotropin Receptors and Steroidogenic Responses. Endocrinology, 1981. 108: p. 88-
95.
157. Bassi, G., S.K. Sidhu, and S. Mishra, The Expanding Role of Mitochondria, Autophagy
and Lipophagy in Steroidogenesis. Cells, 2021. 10(8).
158. Christenson, L.K. and J.F. Strauss, Steroidogenic acute regulatory protein: an update on
its regulation and mechanism of action. Arch Med Res, 2001. 32(6): p. 576-86.
159. Stocco, D.M., et al., A brief history of the search for the protein(s) involved in the acute
regulation of steroidogenesis. Mol Cell Endocrinol, 2017. 441: p. 7-16.
160. Alberta, J.A., et al., Mitochondrial localization of a phosphoprotein that rapidly
accumulates in adrenal cortex cells exposed to adrenocorticotropic hormone or to cAMP.
J Biol Chem, 1989. 264(4): p. 2368-72.
136
161. Miller, W.L., Congenital lipoid adrenal hyperplasia: the human gene knockout for the
steroidogenic acute regulatory protein. J Mol Endocrinol, 1997. 19(3): p. 227-40.
162. Clark, B.J., et al., Inhibition of transcription affects synthesis of steroidogenic acute
regulatory protein and steroidogenesis in MA-10 mouse Leydig tumor cells.
Endocrinology, 1997. 138(11): p. 4893-901.
163. Arakane, F., et al., The mechanism of action of steroidogenic acute regulatory protein
(StAR). StAR acts on the outside of mitochondria to stimulate steroidogenesis. J Biol
Chem, 1998. 273(26): p. 16339-45.
164. Medar, M.L.J., et al., Dependence of Leydig Cell's Mitochondrial Physiology on
Luteinizing Hormone Signaling. Life (Basel), 2020. 11(1).
165. Ishii, T., et al., The roles of circulating high-density lipoproteins and trophic hormones in
the phenotype of knockout mice lacking the steroidogenic acute regulatory protein. Mol
Endocrinol, 2002. 16(10): p. 2297-309.
166. Castillo, A.F., et al., The role of mitochondrial fusion and StAR phosphorylation in the
regulation of StAR activity and steroidogenesis. Mol Cell Endocrinol, 2015. 408: p. 73-9.
167. Rademaker, G., et al., Myoferlin controls mitochondrial structure and activity in
pancreatic ductal adenocarcinoma, and affects tumor aggressiveness. Oncogene, 2018.
37(32): p. 4398-4412.
168. Anania, S., et al., Myoferlin Is a Yet Unknown Interactor of the Mitochondrial Dynamics'
Machinery in Pancreas Cancer Cells. Cancers (Basel), 2020. 12(6).
169. Tan, E.P., et al., Sustained. J Biol Chem, 2017. 292(36): p. 14940-14962.
170. Divakaruni, A.S., et al., Analysis and interpretation of microplate-based oxygen
consumption and pH data. Methods Enzymol, 2014. 547: p. 309-54.
137
171. Vincent, A.E., et al., The Spectrum of Mitochondrial Ultrastructural Defects in
Mitochondrial Myopathy. Sci Rep, 2016. 6: p. 30610.
172. Prince, F.P., Lamellar and tubular associations of the mitochondrial cristae: unique
forms of the cristae present in steroid-producing cells. Mitochondrion, 2002. 1(4): p.
381-9.
173. StatPearls. 2022.
174. Rosselin, M., P. Nunes-Hasler, and N. Demaurex, Ultrastructural Characterization of
Flashing Mitochondria. Contact (Thousand Oaks), 2018. 1(1): p. 1-14.
175. Cserép, C., et al., Mitochondrial Ultrastructure Is Coupled to Synaptic Performance at
Axonal Release Sites. eNeuro, 2018. 5(1).
176. Sasaki, S., Determination of altered mitochondria ultrastructure by electron microscopy.
Methods Mol Biol, 2010. 648: p. 279-90.
177. Midzak, A.S., et al., ATP synthesis, mitochondrial function, and steroid biosynthesis in
rodent primary and tumor Leydig cells. Biol Reprod, 2011. 84(5): p. 976-85.
178. Midzak, A.S., et al., Effect of myxothiazol on Leydig cell steroidogenesis: inhibition of
luteinizing hormone-mediated testosterone synthesis but stimulation of basal
steroidogenesis. Endocrinology, 2007. 148(6): p. 2583-90.
179. Shiratori, R., et al., Glycolytic suppression dramatically changes the intracellular
metabolic profile of multiple cancer cell lines in a mitochondrial metabolism-dependent
manner. Sci Rep, 2019. 9(1): p. 18699.
180. Schrade, A., et al., GATA4 is a key regulator of steroidogenesis and glycolysis in mouse
Leydig cells. Endocrinology, 2015. 156(5): p. 1860-72.
138
181. Weiss, J.N., et al., Role of the mitochondrial permeability transition in myocardial
disease. Circ Res, 2003. 93(4): p. 292-301.
182. Guo, C., et al., Oxidative stress, mitochondrial damage and neurodegenerative diseases.
Neural Regen Res, 2013. 8(21): p. 2003-14.
183. Saikumar, P., et al., Mechanisms of cell death in hypoxia/reoxygenation injury.
Oncogene, 1998. 17(25): p. 3341-9.
184. Kowaltowski, A.J. and A.E. Vercesi, Mitochondrial damage induced by conditions of
oxidative stress. Free Radic Biol Med, 1999. 26(3-4): p. 463-71.
185. Wasilewski, M., et al., Optic atrophy 1-dependent mitochondrial remodeling controls
steroidogenesis in trophoblasts. Curr Biol, 2012. 22(13): p. 1228-34.
186. Shabalina, I.G., et al., Leydig cell steroidogenesis unexpectedly escapes mitochondrial
dysfunction in prematurely aging mice. FASEB J, 2015. 29(8): p. 3274-86.
187. Reddy, J.K. and T. Hashimoto, Peroxisomal beta-oxidation and peroxisome proliferator-
activated receptor alpha: An adaptive metabolic system, in Annual Review of Nutrition,
D.B. McCormick, D.M. Bier, and R.J. Cousins, Editors. 2001. p. 193-230.
188. Waterham, H.R., S. Ferdinandusse, and R.J.A. Wanders, Human disorders of peroxisome
metabolism and biogenesis. Biochimica Et Biophysica Acta-Molecular Cell Research,
2016. 1863(5): p. 922-933.
189. Puglielli, L., et al., STEROL CARRIER PROTEIN-2 IS INVOLVED IN CHOLESTEROL
TRANSFER FROM THE ENDOPLASMIC-RETICULUM TO THE PLASMA-
MEMBRANE IN HUMAN FIBROBLASTS. Journal of Biological Chemistry, 1995.
270(32): p. 18723-18726.
139
190. Gadella, T.W.J. and K.W.A. Wirtz, THE LOW-AFFINITY LIPID-BINDING SITE OF
THE NONSPECIFIC LIPID TRANSFER PROTEIN - IMPLICATIONS FOR ITS MODE
OF ACTION. Biochimica Et Biophysica Acta, 1991. 1070(1): p. 237-245.
191. Stolowich, N.J., et al., The sterol carrier protein-2 fatty acid binding site: An NMR,
circular dichroic, and fluorescence spectroscopic determination. Biochemistry, 1997.
36(7): p. 1719-1729.
192. Frolov, A., et al., Lipid specificity and location of the sterol carrier protein-2 fatty acid-
binding site: A fluorescence displacement and energy transfer study. Lipids, 1997.
32(11): p. 1201-1209.
193. Frolov, A., et al., Sterol carrier protein-2, a new fatty acyl coenzyme A-binding protein.
Journal of Biological Chemistry, 1996. 271(50): p. 31878-31884.
194. Schroeder, F., et al., Sterol carrier protein-2: New roles in regulating lipid rafts and
signaling. Biochimica Et Biophysica Acta-Molecular and Cell Biology of Lipids, 2007.
1771(6): p. 700-718.
195. Keller, G.A., et al., SUBCELLULAR-LOCALIZATION OF STEROL CARRIER
PROTEIN-2 IN RAT HEPATOCYTES - ITS PRIMARY LOCALIZATION TO
PEROXISOMES. Journal of Cell Biology, 1989. 108(4): p. 1353-1361.
196. Shickh, S., et al., Exome and genome sequencing in adults with undiagnosed disease: a
prospective cohort study. J Med Genet, 2021. 58(4): p. 275-283.
197. Quehenberger, O., et al., Lipidomics reveals a remarkable diversity of lipids in human
plasma. J Lipid Res, 2010. 51(11): p. 3299-305.
198. Nordgren, M., et al., Peroxisome degradation in mammals: mechanisms of action, recent
advances, and perspectives. Front Physiol, 2013. 4: p. 145.
140
199. Nsiah-Sefaa, A. and M. McKenzie, Combined defects in oxidative phosphorylation and
fatty acid β-oxidation in mitochondrial disease. Biosci Rep, 2016. 36(2).
200. Bougarne, N., et al., Molecular Actions of PPARα in Lipid Metabolism and Inflammation.
Endocr Rev, 2018. 39(5): p. 760-802.
201. Ellinghaus, P., et al., Phytanic acid activates the peroxisome proliferator-activated
receptor alpha (PPARalpha) in sterol carrier protein 2-/ sterol carrier protein x-deficient
mice. J Biol Chem, 1999. 274(5): p. 2766-72.
202. Chiang, J.Y., Bile acids: regulation of synthesis. J Lipid Res, 2009. 50(10): p. 1955-66.
203. Yunfeng, C., et al., Increased bile lithogenicity by SCP2 via HMGCR and CYP7A1
regulation in human hepatocytes. Turk J Gastroenterol, 2013. 24(3): p. 241-50.
204. Alessenko, A.V. and E. Albi, Exploring Sphingolipid Implications in Neurodegeneration.
Front Neurol, 2020. 11: p. 437.
205. Taniguchi, M. and T. Okazaki, Ceramide/Sphingomyelin Rheostat Regulated by
Sphingomyelin Synthases and Chronic Diseases in Murine Models. J Lipid Atheroscler,
2020. 9(3): p. 380-405.
206. Rosenson, R.S., Fenofibrate: treatment of hyperlipidemia and beyond. Expert Rev
Cardiovasc Ther, 2008. 6(10): p. 1319-30.
207. Dansen, T.B., et al., Regulation of sterol carrier protein gene expression by the forkhead
transcription factor FOXO3a. J Lipid Res, 2004. 45(1): p. 81-8.
208. Storey, S.M., et al., Impact of. J Lipid Res, 2017. 58(6): p. 1153-1165.
209. Wolfrum, C., et al., Phytanic acid is ligand and transcriptional activator of murine liver
fatty acid binding protein. J Lipid Res, 1999. 40(4): p. 708-14.
141
210. Tahri-Joutey, M., et al., Mechanisms Mediating the Regulation of Peroxisomal Fatty Acid
Beta-Oxidation by PPARα. Int J Mol Sci, 2021. 22(16).
211. Reddy, J.K., Nonalcoholic steatosis and steatohepatitis. III. Peroxisomal beta-oxidation,
PPAR alpha, and steatohepatitis. Am J Physiol Gastrointest Liver Physiol, 2001. 281(6):
p. G1333-9.
212. Wanders, R.J., H.R. Waterham, and S. Ferdinandusse, Metabolic Interplay between
Peroxisomes and Other Subcellular Organelles Including Mitochondria and the
Endoplasmic Reticulum. Front Cell Dev Biol, 2015. 3: p. 83.
213. Pawlak, M., P. Lefebvre, and B. Staels, Molecular mechanism of PPARα action and its
impact on lipid metabolism, inflammation and fibrosis in non-alcoholic fatty liver
disease. J Hepatol, 2015. 62(3): p. 720-33.
214. Kohjima, M., et al., Re-evaluation of fatty acid metabolism-related gene expression in
nonalcoholic fatty liver disease. Int J Mol Med, 2007. 20(3): p. 351-8.
215. Ouali, F., et al., Dietary lipids regulate beta-oxidation enzyme gene expression in the
developing rat kidney. Am J Physiol, 1998. 275(5): p. F777-84.
216. Latruffe, N., et al., Studies on regulation of the peroxisomal beta-oxidation at the 3-
ketothiolase step. Dissection of the rat liver thiolase B gene promoter. Adv Exp Med
Biol, 1999. 466: p. 253-9.
217. Milligan, S., et al., Ablating both Fabp1 and Scp2/Scpx (TKO) induces hepatic
phospholipid and cholesterol accumulation in high fat-fed mice. Biochim Biophys Acta
Mol Cell Biol Lipids, 2018. 1863(3): p. 323-338.
218. Zurkinden, L., et al., Downregulation of. Front Endocrinol (Lausanne), 2020. 11: p.
586980.
142
219. Bouhlel, M.A., B. Staels, and G. Chinetti-Gbaguidi, Peroxisome proliferator-activated
receptors--from active regulators of macrophage biology to pharmacological targets in
the treatment of cardiovascular disease. J Intern Med, 2008. 263(1): p. 28-42.
220. Quinn, C.M., et al., Expression and regulation of sterol 27-hydroxylase (CYP27A1) in
human macrophages: a role for RXR and PPARgamma ligands. Biochem J, 2005. 385(Pt
3): p. 823-30.
221. Klipsic, D., et al., Impact of SCP-2/SCP-x gene ablation and dietary cholesterol on
hepatic lipid accumulation. Am J Physiol Gastrointest Liver Physiol, 2015. 309(5): p.
G387-99.
222. Sarkar, J., et al., Transcription coactivator PRIP, the peroxisome proliferator-activated
receptor (PPAR)-interacting protein, is redundant for the function of nuclear receptors
PParalpha and CAR, the constitutive androstane receptor, in mouse liver. Gene Expr,
2007. 13(4-5): p. 255-69.
223. Viswakarma, N., et al., Coactivators in PPAR-Regulated Gene Expression. PPAR Res,
2010. 2010.
224. Jia, Y., et al., The natural carotenoid astaxanthin, a PPAR-α agonist and PPAR-γ
antagonist, reduces hepatic lipid accumulation by rewiring the transcriptome in lipid-
loaded hepatocytes. Mol Nutr Food Res, 2012. 56(6): p. 878-88.
225. Joseph, D.R., Structure, function, and regulation of androgen-binding protein/sex
hormone-binding globulin. Vitam Horm, 1994. 49: p. 197-280.
226. Claeysen, S., J. Bockaert, and P. Giannoni, Serotonin: A New Hope in Alzheimer's
Disease? ACS Chem Neurosci, 2015. 6(7): p. 940-3.
143
Abstract (if available)
Abstract
The trafficking of cholesterol between cellular compartments is tightly regulated, as cholesterol must be properly distributed to carry out its various key functions. One mechanism through which cholesterol is trafficked in the cell is via cholesterol transport proteins, involving the movement of cholesterol between membranes via binding to specific proteins. These include the steroidogenic acute regulatory protein (STAR), which is classically known to facilitate cholesterol transport to the mitochondria for steroid biosynthesis, and sterol carrier protein 2/sterol carrier protein-x (SC2/SCPx), which are non-specific lipid transfer proteins suggested to function in the transport of many lipids including cholesterol. Although much work has been done to investigate the roles of these proteins in cholesterol transport, many aspects of their functions are still unclear.
In order to elucidate the function of constitutive STAR, or STAR that is present under basal conditions as opposed to hormone-induced STAR, we used CRISPR/Cas9 technology to generate a STAR knockout (KO) MA-10 mouse tumor Leydig cell line (STARKO1). STAR KO cells had significantly increased levels of cholesteryl ester, diacylglycerol, and phosphatidylcholine compared to wild-type (WT) MA-10 cells. Additionally, many lipid-related genes were differentially expressed between WT MA-10 and STARKO1 cells. Further, we showed that mitochondrial structure and function were altered by STAR knockout and that reintroduction of STAR into STARKO1 cells exacerbated, rather than recovered mitochondrial structure and function. Additionally, the processing of STAR into its mature form was inhibited in STARKO1 cells with STAR overexpression suggesting that mitochondrial dysfunction alters STAR processing. Taken together, these data indicate that constitutive STAR may have roles in lipid metabolism and mitochondrial function which are independent of the role of hormone-induced STAR in cholesterol transport for steroidogenesis.
We also investigated the role(s) of SCPx in a patient with a mutation in SCP2 presenting with brainstem neurodegeneration and testicular defects. The patient’s SCP2 mutation led to decreased levels of SCPx, but normal levels of SCP2. RNA sequencing identified many differentially expressed genes that were altered between patient and control fibroblasts. Lipidomic analyses identified many species of free fatty acids, acylcarnitines, sterols, phospholipids, and sphingolipids that had altered levels between patient and control fibroblasts. Pathway analyses using transcriptomic and lipidomic data identified several metabolic pathways that were affected by the patient’s SCP2 mutation such as PPAR signaling, cholesterol metabolism, and fatty acid biosynthesis. We also identified two compounds, fenofibrate and 4-hydroxytamoxifen, that recovered SCPx levels and improved fatty acid levels in patient fibroblasts. Collectively, these data suggest that SCPx plays a role in the transport and/metabolism of many lipids that function in several key pathways, which may explain many of the clinical presentations associated with SCPx deficiency.
Thus, the data presented here suggest that classic examples of intracellular cholesterol transport proteins play a more general role in lipid transport and metabolism in addition to their respective roles in cholesterol trafficking.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Mitochondrial dynamics regulate Leydig cell health and integrity
PDF
Reevaluation of the pregnenolone biosynthesis pathway in the human brain
PDF
Role of translocator protein (18 kDa) in cell function and metabolism
PDF
The role of translocator protein (TSPO) in mediating the effects of mitotane in human adrenocortical carcinoma cells
PDF
Effect of acetaminophen and ibuprofen on spermatogenesis and cell signaling mechanisms
PDF
Transcriptional regulation of lipid metabolism and exploring the specific roles of AKT isoforms
PDF
Exploring the impact of elevated bile acids in metabolic dysfunction- associated steatotic liver disease in vitro
PDF
Eicosanoid regulation of spermatogonial stem cell development
PDF
Effects of endocrine disrupting chemicals and pharmaceuticals on rodent immature Sertoli cell functions
PDF
Disruption of immature Sertoli cell functions by endocrine disruptors and analgesic drugs
PDF
Fetal exposure to genistein and DEHP mixtures alters the expression of genes involved in critical testicular functions in adult rats
PDF
Phosphatase and tensin homolog deleted on chromosome 10 (PTEN) signaling regulates fatty acid beta-oxidation
PDF
Role of the NF-κB-CXCL1 signaling pathway in the development of metabolic dysfunction-associated steatohepatitits
PDF
Identification of functional pathways altered in testis by in utero exposure to low doses of genistein and DEHP
PDF
SGLT2 inhibitors reverse adipocyte differentiation and reduce daunorubicin metabolism in acute lymphoblastic leukemia treatment
PDF
Allopregnanolone promotes cholesterol and amyloid-beta clearance mechanisms
PDF
Metabolic shift in lung alveolar cell mitochondria after exposure to environmental toxicants
PDF
The regulation of fatty acid oxidation by estrogen related receptor alpha
PDF
The role of ATAD3A in the progression of non-alcoholic fatty liver disease (NAFLD)
PDF
Insulin sensitivity in cognition, Alzheimer's disease and brain aging
Asset Metadata
Creator
Galano, Melanie Rose
(author)
Core Title
Role of steroidogenic acute regulatory protein (STAR) and sterol carrier protein-x (SCPx) in the transport of cholesterol and other lipids
School
School of Pharmacy
Degree
Doctor of Philosophy
Degree Program
Molecular Pharmacology and Toxicology
Degree Conferral Date
2022-08
Publication Date
11/27/2022
Defense Date
05/02/2022
Publisher
University of Southern California. Libraries
(digital)
Tag
cholesterol transport,lipid metabolism,OAI-PMH Harvest,SCPx,Star,steroidogenesis
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Papadopoulos, Vassilios (
committee chair
), Culty, Martine (
committee member
), Zirkin, Barry (
committee member
)
Creator Email
galano@usc.edu,galanomelanie@gmail.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-oUC111336171
Unique identifier
UC111336171
Identifier
etd-GalanoMela-10730.pdf (filename)
Legacy Identifier
etd-GalanoMela-10730
Document Type
Dissertation
Rights
Galano, Melanie Rose
Internet Media Type
application/pdf
Type
texts
Source
20220527-usctheses-batch-944
(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. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.
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
Tags
cholesterol transport
lipid metabolism
SCPx
steroidogenesis